text stringlengths 8 5.77M |
|---|
Tutorial
========
Installation
------------
Use npm to install Ring.js:
```sh
npm install ring
```
Then inside your program:
```javascript
var ring = require("ring");
```
Or if you don't use CommonJS:
```html
<script type="text/javascript" src="node_modules/lodash/lodash.js"></script>
<script type="text/javascript" src="node_modules/ring/ring.js"></script>
```
Declaring Classes
-----------------
The `ring.create()` function is used to declare new classes:
```javascript
var A = ring.create({
saySomething: function() {
return "hello world";
},
});
var a = new A();
console.log(a.saySomething()); // prints "hello world" in the JavaScript console
```
`ring.create()` takes a dictionary as argument. The content of that dictionary will be added to the prototype of the
class.
Constructor
-----------
```javascript
var A = ring.create({
constructor: function(name) {
this.name = name;
},
});
var a = new A("Nicolas");
console.log(a.name); // prints "Nicolas"
```
When the dictionary of properties given to `ring.create()` contains a method named `constructor`, that method will be
called during the object construction with the arguments given to the class when using `new`.
Inheritance
-----------
To inherit from one or multiple classes, add a list of the classes to inherit as an argument of `ring.create()` *before*
the properties dictionary.
```javascript
var A = ring.create({
x: function() {
return "x";
},
});
var B = ring.create({
y: function() {
return "y";
},
});
var C = ring.create([A, B], {});
var c = new C();
console.log(c.x()); // prints "x";
console.log(c.y()); // prints "y";
```
Diamond inheritance (ie: the classes B and C both inherit from A and the class D inherits from B *and* C) is allowed by
Ring.js.
Alternatively you can use the more classical `$extend` method:
```javascript
var A = ring.create({});
var B = A.$extend({});
```
Super Method
------------
To call the super method when overriding a method, use `this.$super()`. Pass arguments to `this.$super()` to forward
arguments if necessary.
```javascript
var A = ring.create({
sayHello: function(name) {
return "Hello " + name;
},
});
var B = ring.create([A], {
sayHello: function(name) {
return this.$super(name) + ", nice to meet you";
},
});
var b = new B();
console.log(b.sayHello("Nicolas")); // prints "Hello Nicolas, nice to meet you"
```
Testing the Class of an Object
------------------------------
The standard procedure in JavaScript to test if an object is an instance of a class is to use `instanceof`. But, due to
the implementation of inheritance in Ring.js (which is necessary to have multiple inheritance), `instanceof` can not
work properly.
So you should use `ring.instance()` instead.
```javascript
var A = ring.create({});
var B = ring.create([A], {});
console.log(ring.instance(new A(), A)); // prints "true"
console.log(ring.instance(new B(), A)); // prints "true"
```
`ring.instance(obj, type)` will return `true` if `obj` is an instance of `type` or an instance of a sub-class of `type`.
`ring.instance()` can also be used with classes and objects that do not use the Ring.js class system. It will use the
`instanceof` operator in that case. It means you can safely replace all the usages of `instanceof` in any programs by
`ring.instance()` without causing problems.
Additionally, `ring.instance()` can be used to test the type of basic data types in JavaScript. To do so, use a string
as second argument instead of a class. Example:
```javascript
console.log(ring.instance("", "string")); // prints "true"
console.log(ring.instance(function() {}, "function")); // prints "true"
```
Creation of New Exception Types
-------------------------------
To create new exception types you are supposed to inherit from the standard `Error` class. But inheriting from that
standard class can be hard. That's why Ring.js contains a feature to bring a solution to that:
```javascript
var MyError = ring.create([ring.Error], {
name: "MyError",
});
try {
throw new MyError("an error occured");
} catch(e) {
console.log(ring.instance(e, Error)); // prints "true"
console.log(ring.instance(e, ring.Error)); // prints "true"
console.log(ring.instance(e, MyError)); // prints "true"
}
```
The `ring.Error` class is a special class using the Ring.js class system *but* it also inherits from the standard
`Error` class. You can create new exceptions classes inheriting from it and thus create complex exception hierarchies.
Please note the `name` property is a property of the standard `Error` class that can be useful for debugging. So all
exception classes should define one.
Compatibility with other class systems
--------------------------------------
With Ring.js, since version 2, it is possible to inherit from a class created with any other class system. Example
with a Backbone model:
```javascript
var A = Backbone.Model.extend({
initialize: function() {
this.a = "a";
}
});
var B = ring.create([A], {
constructor: function() {
this.$super();
}
});
```
Inheriting from a non-Ring class uses the same syntax than with a Ring class. `this.$super` is usable as always. When
overriding the constructor method, simply define a `constructor` function like you would with a normal Ring class.
To call the original implementation of the constructor, use the usual `this.$super`. It doesn't matter if you are
using a class system that uses another name for the constructor, like `initialize`, `init` or whatever. Just use
`constructor` and `this.$super()` like usual and Ring.js will do the rest.
Also note that, in the above example, `new B() instanceof A` will return `false`. `instanceof` still doesn't work in
this case so you should use `ring.instance()` instead.
Inheriting from a non-Ring class has some limitations. We could mostly resume it by saying that Ring does not "see"
the inheritance in other classes. Example:
```javascript
var A = ... some non-Ring class ...
var B = ... some non-Ring class inheriting from A ...
var C = ring.create([B], {...});
```
Instances of C will work as expected in this example. But still, ring doesn't understand the relation between `A` and
`B`, and `ring.instance(new C(), A)` will return `false` while `ring.instance(new C(), B)` return `true`.
|
AppalCart
AppalCART (Appalachian Campus Area Rapid Transit) is a free public bus network located in Boone, North Carolina. It provides free fixed route and paratransit service throughout Appalachian State University and Boone, as well as low-fare van service to other towns within Watauga County. In 2013, AppalCART reported a ridership of 1,712,873 passenger trips.
AppalCART is governed by an 8-member board. The current chairman of the board is Greg Lovins. It receives its funding from a mix of federal, state, local, and Appalachian State University funds.
Routes
AppalCART provides Watagua with thirteen routes listed in the table below:
References
External links
Official Website
Chauffeur Service
Category:Bus transportation in North Carolina |
1. Field of the Invention
The present invention relates to an image processing apparatus and an image processing method for performing frame interpolation processing in which an interpolation frame is inserted between frames of an image, and an image display apparatus and an image display method for displaying an image based on image data that has been subjected to such frame interpolation processing.
2. Description of the Related Art
Movement of an object on a displayed image of a hold-type display such as a liquid crystal display is performed on a frame-by-frame basis and is discontinuous one, whereas following of the movement of the object by human eyes (i.e., movement of human eyes) is continuous one. For this reason, a judder phenomenon, in which an object moving on a displayed image exhibits blur or an object on a displayed image exhibits jerky movement (i.e., stiff and unnatural movement), is apt to appear in the hold-type display.
To take measures against this, a method is proposed to insert an interpolation frame between frames of an image, thereby increasing number of the frames and smoothing the movement of the object on the displayed image. It is generally known that a typical method of generating an interpolation frame is a zero-order hold method in which the interpolation frame is generated from the same image as that of a frame preceding the interpolation frame by one frame or an average-value interpolation method in which the interpolation frame is generated from an average of an image of a frame preceding the interpolation frame by one frame and another image of a frame subsequent to the interpolation frame by one frame. However, in the zero-order hold method, since the same frame is displayed repeatedly, reduction of the image blur and judder is insufficient. Furthermore, in the average-value interpolation method, since an edge of the object moving on the displayed image sometimes exhibits a double image, reduction of the image blur and judder is insufficient.
Further, a television signal converted from a film picture such as a movie is a signal in which number of frames is increased by generating two frames or three frames from the same frame, thereby making the number of frames larger than that of the film picture. In this method, since the same frame is displayed repeatedly, the image blur and judder are apt to appear on an image which is displayed on the basis of the television signal converted from the film picture. Similarly, a television signal converted from an image generated by computer processing also is an image signal in which number of frames is increased by generating two frames from the same frame. Since the same frame is displayed repeatedly also in this method, the image blur and judder are apt to appear on an image which is displayed on the basis of the television signal converted from the film picture.
Furthermore, there is a highly developed method of generating an interpolation frame, including the steps of: finding pixel pairs with a high correlation, each of which is a set of a pixel on a former frame and another pixel on a later frame, where the pixel on the former frame and the pixel on the later frame are point-symmetric with reference to an interpolation pixel on an interpolation frame; generating the interpolation pixels using a plurality of pixel pairs; and generating the interpolation frame composed of the interpolation pixels. See patent document 1, Japanese Patent Kokai Publication No. 2006-129181 (paragraph 0025, FIG. 3). However, in this method, a correlation is detected on a pixel-by-pixel basis by finding the pixel pairs with a high correlation, each of which is a set of a pixel on the former frame and another pixel on the later frame, where the pixel on the former frame and the pixel on the later frame are point-symmetric with reference to an interpolation pixel on an interpolation frame, generating the interpolation pixels using a plurality of pixel pairs, and generating the interpolation frame composed of such interpolation pixels. Therefore, even if a high correlation is detected between two frames while these have quite different images, e.g., when an object appears and/or disappears suddenly on the former frame or the later frame with reference to an inserted position of the interpolation frame, the pixel pair with a high correlation between pixels is detected. As a result, there is a possibility that some inappropriate interpolation frames are generated and thus a disturbance appears on the displayed image.
As has been described above, since the conventional image processing method cannot reduce the image blur and judder to a sufficient degree or the disturbance sometimes occurs in the interpolation frame, there is a problem that images with high quality cannot be displayed. |
**To the Editor**: Picornaviruses (family *Picornaviridae*) are small, nonenveloped viruses with single-stranded, positive-sense genomic RNA. Picornaviruses are currently divided into 8 genera: *Enterovirus, Aphthovirus, Cardiovirus, Hepatovirus, Parechovirus, Erbovirus, Teschovirus,* and *Kobuvirus* ([@R1]). To date, the genus *Kobuvirus* consists of 2 officially recognized species, *Aichi virus* and *Bovine kobuvirus,* and 1 porcine kobuvirus as a candidate species ([@R2]--[@R4]). Aichi virus (strain A846/88) was first isolated in 1991 from feces of a person with acute gastroenteritis ([@R2]). Bovine kobuvirus (strain U-1) was detected in 2003 in bovine serum and fecal samples from clinically healthy cattle ([@R3]); in 2008, it was isolated from cattle with diarrhea ([@R5]). Aichi virus and bovine kobuvirus were first isolated in Japan. Porcine kobuvirus (strain S-1-HUN) was recently identified from domestic pigs in Hungary ([@R4]). Aichi viruses have been also detected in other countries in Asia ([@R6]), Europe ([@R7],[@R8]), South America ([@R7]), and northern Africa ([@R9]). Bovine kobuvirus, however, has not been detected outside Asia (Japan and Thailand) ([@R3],[@R5]).
Kobuvirus genomes are ≈8.2--8.4 kb and have a typical picornavirus genome organization, including leader (L) protein following structural (VP0, VP3, and VP1) and nonstructural (2A--2C and 3A--3D) regions ([@R1],[@R3],[@R4]). The genetic identity on coding regions of Aichi virus, bovine kobuvirus strain U-1, and porcine kobuvirus strain S-1-HUN is between 35% (L protein) and 74% (3D region) ([@R3],[@R4]). We report the detection of bovine kobuvirus in Europe.
In February 2002, a total of 32 fecal samples were collected from cattle (*Bos taurus*) in a closed herd of 870 animals in central Hungary; age groups were 1--9 days (n = 6), 14--17 days (n = 4), 6--7 months (n = 5), and 1--7.6 years (n = 17). In February 2008, 26 more samples were collected from animals \<20 days of age on this farm. On the sampling days, no diarrhea was reported.
Reverse transcription--PCR was performed by using a new generic kobuvirus primer (UNIV-kobu-F, forward, 5′-TGGAYTACAAG(/R)TGTTTTGATGC-3′, corresponding to nucleotides 7491--7512 of strain U-1 and UNIV-kobu-R, reverse, 5′-ATGTTGTTRATGATGGTGTTGA-3′, corresponding to nucleotides 7686--7707 of strain U-1). The primer design was based on the viral sequences of the Aichi virus (AB040749), bovine kobuvirus strain U-1 (bovine, AB084788), and bovine kobuvirus strain S-1-HUN (porcine, EU787450), which amplify a 216-nt region of 3D (RNA-dependent RNA polymerase region) of all species. The continuous 862-nt 3D and 3′ untranslated region (UTR) of the genome was determined by using 5′/3′RACE Kit (2nd Generation; Roche Diagnostics GmbH, Mannheim, Germany) and primers UNIV-kobu-F and Z20-F-7729 (5′-CCAACATCCTGACTTCTCTCCT-3′, corresponding to nucleotides 7729--7750 of strain U-1). PCR products were sequenced directly in both directions by using the BigDye Reaction Kit (Applied Biosystems, Warrington, UK), the PCR primers, and an automated sequencer (ABI PRISM 310 Genetic Analyzer; Applied Biosystems, Stafford, TX, USA). Phylogenetic analysis was conducted by using MEGA version 4.1 ([@R10]). The sequence of this bovine kobuvirus strain (kobuvirus/bovine/Aba-Z20/2002/Hungary) was submitted to GenBank under accession no. FJ225406.
Of the 32 samples collected in 2002, two (6.25%), from 1-year-old animals, were positive for bovine kobuvirus; however, no kobuvirus was found in the samples from 2008. The 2 partial 3D regions (216 nt) were genetically identical. Strain kobuvirus/bovine/Aba-Z20/2002/Hungary (FJ225406) had 89%--94% nucleotide and 96%--100% amino acid identities to the 19 known Asian bovine kobuvirus strains in GenBank. Strain Z20 had 93% and 95% nucleotide identities to U-1 in 3D/3′-UTR (862 nt) and 3′-UTR (174 nt) regions, respectively. Phylogenetic analysis of the overlapping partial 3D nucleotide sequences of bovine kobuvirus strain Z20 from Hungary, together with all published bovine kobuvirus strains available in the GenBank database, are shown in the Figure. Aichi virus and porcine kobuvirus were included in the tree as outlier viruses. The phylogenetic tree confirmed that strain Z20 belonged to bovine kobuviruses ([Figure](#F1){ref-type="fig"}).
![Phylogenetic tree of bovine kobuvirus (kobuvirus/bovine/Aba-Z20/2002/Hungary, in **boldface**) based on the 455-nt fragment of the kobuvirus 3D regions. The phylogenetic tree was constructed by using the neighbor-joining clustering method with distance calculation and the maximum composite likelihood correction for evolutionary rate with help of MEGA version 4.1 software ([@R10]). Bootstrap values (based on 1,000 replicates) are given for each node if \>50%. Reference strains were obtained from GenBank. Scale bar indicates nucleotide substitutions per site.](08-1427-F){#F1}
Our detection of bovine kobuviruses in Europe confirms a wider geographic presence of this type of picornavirus in cattle and suggests that bovine kobuvirus is common and potentially distributed worldwide. Genetic diversity was seen, based on the 3D regions of bovine kobuviruses; however, this region shows the highest genetic identity among the kobuvirus genetic regions ([@R3],[@R4]). Strain Z20 also confirms the 174-nt 3′-UTR region of bovine kobuvirus. At this time it is not clear what diseases (including gastroenteritis) are associated with bovine kobuvirus ([@R3],[@R5]). In addition to the bovine kobuvirus, 2 other RNA viruses that are transmitted by the fecal--oral route (genotypes GIII/1 and GIII/2 of bovine noroviruses and rotavirus) were detected at the same time from these apparently healthy animals. More epidemiologic and molecular studies are required to determine the relevance, distribution, and diversity of bovine kobuvirus in cattle.
*Suggested citation for this article*: Reuter G, Egyed L. Bovine kobuvirus in Europe \[letter\]. Emerg Infect Dis \[serial on the Internet\]. 2009 May \[*date cited*\]. Available from <http://www.cdc.gov/EID/content/15/5/822.htm>
This work was supported by grants from the Hungarian Scientific Research Fund (OTKA, F048433) and the project "Enteric Virus Emergence, New Tools" (EVENT, SP22-CT-2004-502571), funded by the European Union.
|
River Lossie
The River Lossie () is a river in north east Scotland. Ptolemy (c.90 – c.168), the Greco-Roman geographer, named it as ost. Loxa Fl. The river originates in the hills above Dallas, in Moray, and has its source 400 metres above sea-level. It enters the sea at Lossiemouth on the Moray Firth. By the time it moves through Elgin its rate of flow, in normal conditions, is best described as very slow. The gradient between Elgin and Lossiemouth is almost imperceptible with a total fall of less than 5 metres.
Settlements
(from south to north)
Dallas
Kellas
Paddockhaugh
Pittendreich
Elgin
Calcots
Lossiemouth
Lossie
0Lossie |
Q:
Estoy intentando pasar de Ilist a list
Estoy intentando pasar de Ilist a list para luego mostrarlo en un DataGridView pero me sale error... me dice que no es posible convertir implicitamente
List<Empleado> lista = new List<Empleado>();
lista = servicio.Consultar();
A:
Utiliza el metodo extensor ToList() sobre el tipo IList<Empleado>:
List<Empleado> empleados = servicio.Consultar().ToList();
IList<T> implementa IEnumerable<T> por lo que tiene disponibles todos los metodos extensores de System.Linq.Enumerable, entre ellos esta el metodo List<T> ToList(this IEnumerable<T> source) que convierte un IEnumerable<T> a List<T>.
A:
Buenas,
El asunto es que IList<> es una Interfaz, en cambio List<> es una clase, IList expone unos métodos públicos que la clase List implementa.
Para solucionar tu problema, tienes tres opciones, hacer que el método devuelva una List, o que la variable donde lo recibes sea IList, o pasarle la Interfaz en el constructor
IList<Empleado> lista;
lista = servicio.Consultar();
List<Empleado> lista = new List<Empleado>(servicio.Consultar());
Atte
|
Looking beyond 5-HT(3) receptors: a review of the wider role of serotonin in the pharmacology of nausea and vomiting.
The concept that 5-hydroxytryptamine (5-HT; serotonin) is involved in the emetic reflex was revealed using drugs that interfere with its synthesis, storage, release and metabolism ahead of the discovery of selective tools to modulate specific subtypes of receptors. This review comprehensively examines the fundamental role of serotonin in emesis control and highlights data indicating association of 5-HT1-4 receptors in the emetic reflex, whilst leaving open the possibility that 5-HT5-7 receptors may also be involved. The fact that each receptor subtype may mediate both emetic and anti-emetic effects is discussed in detail for the first time. These discussions are made in light of known species differences in emesis control, which has sometimes affected the perception of the translational value of data in regard to the development of novel anti-emetic for use in man. |
Introduction {#s1}
============
Next generation sequencing (NGS) technologies are revolutionizing the study of variation among individuals in a population. The ability of sequencing platforms such as AB SOLiD and Illumina (Solexa) to sequence one billion basepairs (gigabase) or more in a few days has enabled the cheap re-sequencing of human genomes, with the genomes of a Chinese individual [@pcbi.1000386-Wang1], a Yoruban individual [@pcbi.1000386-Bentley1], and matching tumor and healthy samples from a female individual [@pcbi.1000386-Ley1] sequenced in the last few months. These resequencing efforts have been enabled by the development of extremely efficient mapping tools, capable of aligning millions of short (25--70 bp) reads to the human genome [@pcbi.1000386-Bowtie.1]--[@pcbi.1000386-Ma1]. In order to accelerate the computation, most of these methods allow for only a fixed number of mismatches (usually two or three) between the reference genome and the read, and usually do not allow for the matching of reads with insertion/deletion (indel) polymorphisms. These methods are extremely effective for mapping reads to the human genome, most of which has a low polymorphism rate, and so the likelihood that a single read spans multiple SNPs is small. While matching with up to a few differences (allowing for a SNP and 1--2 errors) is sufficient in these regions, these methods fail when the polymorphism level is high.
NGS technologies are also opening the door to the study of population genomics of non-model individuals in other species. Various organisms have a wide range of polymorphism rates - from 0.1% in humans to 4.5% in the marine ascidian *Ciona savignyi*. The polymorphisms present in a particular species can be used to discern its evolutionary history and understand the selective pressures in various genomic loci. For example, the large amount of variation in *C. savignyi* (two individuals\' genomes are as different as Human and Macaque) was found to be due to a large effective population size [@pcbi.1000386-Small1]. The re-sequencing of species like *C. savignyi* (and regions of the human genome with high variability) requires methods for short read mapping that allow for a combination of several SNPs, indels, and sequencing errors within a single (short) read. Furthermore, due to larger-scale "structural" variation, only a fraction of the read may match to the genome, necessitating the use of local, rather than global, alignment methods.
Previous short read mapping tools typically allow for a fixed number of mismatches by separating a read into several sections and requiring some number of these to match perfectly, while others are allowed to vary [@pcbi.1000386-Bowtie.1],[@pcbi.1000386-Maq.1],[@pcbi.1000386-Li2]. An alternative approach generates a set of subsequences from the read (often represented as spaced seeds [@pcbi.1000386-Li1],[@pcbi.1000386-Ma1],[@pcbi.1000386-Buhler1]), again in such a manner that if a read were to match at a particular location with some number of mismatches, at least one of the subsequences would match the genome [@pcbi.1000386-mapreads.1],[@pcbi.1000386-Lin1]. While these methods are extremely fast, they were developed for genomes with relatively low levels of polymorphism, and typically cannot handle a highly polymorphic, non-model genome.
This becomes especially apparent when working with data from Applied Biosystem\'s SOLiD sequencing platform (AB SOLiD). AB SOLiD uses a di-base sequencing chemistry that generates one of four possible calls (colors) for each pair of nucleotides. While a sequencing error is a change of one color-call to another, a single SNP will change two adjacent color positions. Hence a read with two (non-adjacent) SNPs and a sequencing error will differ from the reference genome in five different positions. Simultaneously, the nature of the di-base sequencing code allows for the identification (and correction) of sequencing errors, so by carefully analyzing the exact sequence of matches and mismatches within a read, it is possible to determine that the read and the genome differ by two SNPs. While efficient mappers for color-space sequences have been developed [@pcbi.1000386-mapreads.1],[@pcbi.1000386-Ondov1], they translate the genome to color-space, and directly compare to the color-space read. The complexity of the color-space representation makes the identification of complex variation such as adjacent SNPs and short indels challenging or impossible with these tools.
In this paper we develop algorithms for the mapping of short reads to highly polymorphic genomes and methods for the analysis of the mappings. We demonstrate an algorithm for mapping short reads in the presence of a large amount of polymorphism. By employing a fast k-mer hashing step and a simple, very efficient implementation of the Smith-Waterman algorithm, our method conducts a full alignment of each read to all areas of the genome that are potentially homologous. Secondly, we introduce a novel, specialized algorithm for mapping di-base (color-space) reads, which allows for an accurate, non-heuristic alignment of AB SOLiD reads to a reference genome. Finally, we introduce methodology for evaluating the accuracy of discovered alignments. Because a read may match the genome in several locations with variable amounts of polymorphism, we develop a statistical method for scoring the hits, allowing for the selection of the most probable variants, and filtering of false positives.
Our methods are implemented as part of SHRiMP: the SHort Read Mapping Package. To demonstrate the usefulness of SHRiMP we re-sequenced a Japanese *Ciona savignyi* genome on the SOLiD platform. Preliminary estimates obtained in the course of sequencing the reference genome indicate that the SNP heterozygosity is 4.5%, whereas indel heterozygosity is 16.6%. This species represents the most challenging known test case for the detection of polymorphisms with short read technologies. We aligned the SOLiD reads of the Japanese individual to the *C. savignyi* reference genome using both SHRiMP and AB\'s read mapper. SHRiMP is able to identify 5-fold more SNPs than AB\'s mapper, while also capturing 70,000 indel variants.
Results/Discussion {#s2}
==================
This section is organized as follows: we begin with three methodological sections, in which we first present an overview of the algorithms used in SHRiMP for mapping short reads, explain our specialized algorithm for alignment of di-base sequencing (AB SOLiD) data, and present our framework for computing p-values and other statistics for alignment quality. The data flow for these methods is illustrated in [Figure 1](#pcbi-1000386-g001){ref-type="fig"}. In the last two subsections we will first show the application of SHRiMP to the resequencing of *Ciona savignyi* using the AB SOLiD sequencing technology and present results on the accuracy of the SHRiMP tool on simulated data.
{#pcbi-1000386-g001}
Read Mapping Algorithm {#s2a}
----------------------
The SHRiMP algorithm draws upon three recent developments in the field of sequence alignment: q-gram filter approaches, introduced by Rasmussen et al [@pcbi.1000386-Rasmussen1]; spaced seeds, introduced by Califano and Rigoutsos [@pcbi.1000386-Califano1] and popularized by the PatterHunter family of tools [@pcbi.1000386-Li1],[@pcbi.1000386-Ma1]; and specialized vector computing hardware to speed up the Smith-Waterman Algorithm [@pcbi.1000386-Rognes1]--[@pcbi.1000386-Wozniak1] to rapidly find the likely locations for the reads on the genome. Once these locations are identified, we conduct a thorough, Smith-Waterman-based algorithm to rigorously evaluate the alignments. In this section we will provide a brief exposition of the methods used to align short reads in SHRiMP (a more thorough description of each of these steps is in [Methods](#s3){ref-type="sec"}).
### Spaced seeds {#s2a1}
Most heuristic methods for local alignment rely on the identification of seeds -- short exact matches between the two sequences. The advantage of using exact matches is that they are easy to find using hash tables, suffix arrays, or related techniques. While classically seeds have been contiguous matches, more recently "spaced" seeds, where *predetermined* positions in the read are allowed not to match, have been shown to be more sensitive. Spaced seeds are often represented as a string of 1 s and 0 s, where 1 s indicate positions that must match, while 0 s indicate positions that may mismatch. We refer to the *length* or *span* of the seed as the total length of the string, and the *weight* of the seed as the number of 1 s in the string. For example, the seed "11100111" requires matches at positions 1--3 and 6--8, and has length 8 and weight 6. Because seeds with such small weight match extremely often, we require multiple seeds to match within a region before it is further considered, using a technique called Q-gram filtering.
### Q-gram filters {#s2a2}
While most older local alignment tools, such as BLAST, use a single matching seed to start a thorough comparison of the strings around the seed, more recently Rassmussen et al [@pcbi.1000386-Rasmussen1] introduced the use of q-gram filters, where multiple seeds are used to determine if a good match exists. This idea is also used in SHRiMP where we require a pre-determined number of seeds from a read to match within a window of the genome before we conduct a thorough comparison.
### Vectorized Smith-Waterman {#s2a3}
If a particular read has the required number of seeds matching to a window of the genome we conduct a rapid alignment of the two regions to verify the similarity. This alignment is done using the classical Smith-Waterman algorithm [@pcbi.1000386-Smith1], implemented using specialized "vector" instructions that are part of all modern CPUs. In order to speed up this stage we compute just the score of the optimal alignment, and not the alignment itself. For every read we store the locations of top hits, sorted by their score. The number of top hits to store is a parameter.
### Final alignment {#s2a4}
After we finish aligning all of the reads to all of the potential locations, we conduct a final, full alignment of each read to all of the top hits. This final alignment stage differs depending on the specifics of the sequencing technology. Within SHRiMP we have implemented separate final alignment modules for Illumina/Solexa data (this is done with the regular Smith-Waterman algorithm) and for color-space (di-base) data produced by the AB SOLiD instrument (described in the next section). Additionally we have an experimental module for alignment of two-pass sequencing data, where two reads are generated from every genomic location, which is described elsewhere [@pcbi.1000386-Yanovsky1].
Algorithm for Color-space Alignment {#s2b}
-----------------------------------
The AB SOLiD sequencing technology introduced a novel dibase sequencing technique, which reads overlapping pairs of letters and generates one of four colors (typically labelled 0--3) at every stage. Each base is interrogated twice: first as the right nucleotide of a pair, and then as the left one. The exact combinations of letters and the colors they generate are shown in [Figure 2A](#pcbi-1000386-g002){ref-type="fig"}. The sequencing code can be thought of as a finite state automaton (FSA), in which each previous letter is a state and each color code is a transition to the next letter state. This automaton is demonstrated in [Figure 2B](#pcbi-1000386-g002){ref-type="fig"}. It is notable that the sequence of colors is insufficient to reconstruct the DNA sequence, as reconstruction requires knowledge of the first letter of the sequence (or the last letter of the primer, which is fixed for a single run of the technology).
{#pcbi-1000386-g002}
The AB SOLiD sequencing technology has the remarkable property of differentiating between sequencing errors and biological SNPs (under the assumption that the reference genome has no sequencing errors): a SNP changes two adjacent readouts of the color-space code, while a sequencing error is unlikely to happen at two adjacent positions by chance (the technology does not sequence adjacent letters at adjacent time points). At the same time, however, the color-space code introduces certain complexities.
Let us consider a comparison done by first translating the color-space read code into the letter-space sequence. Notice that a single sequencing error would cause every position after the place of error to be mistranslated ([Figure 3B](#pcbi-1000386-g003){ref-type="fig"}). Consequently, most approaches have translated the letter-space genome into the corresponding color code. However, this is problematic: since the color-coding of every dibase pairing is not unique, a string of colors can represent one of several DNA strings, depending on the preceding base pair. For example, a string of zeroes could be translated as a poly-A, poly-C, poly-G or poly-T string.
{#pcbi-1000386-g003}
There is an additional drawback to translating the genome into color-space code: a sequence of matches and mismatches in color-space does not map uniquely into letter-space similarity. For example, a single SNP results in two sequential color-space mismatches. However, given two consecutive colors, there are 9 possible ways to generate two mismatches. Of these, only 3 correspond to a SNP, while the rest lead to DNA strings that completely differ from the reference. This is illustrated in [Figure 3D](#pcbi-1000386-g003){ref-type="fig"}.
We propose an alternate approach. Our key observation is that while a color-space error causes the rest of the sequence to be mistranslated, the genome will match one of the other three possible translations. This is illustrated in [Figure 4C](#pcbi-1000386-g004){ref-type="fig"}. Consequently, we adapt the classical dynamic programming algorithm to simultaneously align the genome to all four possible translations of the read, allowing the algorithm to move from one translation to another by paying a "crossover", or sequencing error penalty. If one wishes for a probabilistic interpretation of the algorithm, one can consider the FSA in [Figure 2B](#pcbi-1000386-g002){ref-type="fig"} to be a Hidden Markov Model, where the letter is the hidden state, and the color-space sequence is the output of the model. By taking the cross product of this HMM with the standard pair-HMM associated with the Smith-Waterman algorithm, we can allow all of the typical alignment parameters, including the error penalty, to be probabilistically motivated as the log of the probability of the event, and trained using the Expectation-Maximization algorithm. It is notable that our approach handles not only matches, mismatches, and sequencing errors, but also indels. Because the sequences are aligned in letter-space (to be precise, they are aligned and translated simultaneously), indels can be penalized using the standard affine gap penalty with no further modification of the algorithm.
{#pcbi-1000386-g004}
In the SHRiMP algorithm, we only apply the special color-space Smith-Waterman algorithm in the final stage. For the initial stages, we convert the genome from letter-space to color-space, and search for k-mer matches as well as perform vectorized Smith-Waterman strictly in color-space. In order to better incorporate SNPs in color-space data, we use a spaced seed that allows for two adjacent mismatching colors between the read and the reference genome.
Computing Statistics for Reads and Mate-pairs {#s2c}
---------------------------------------------
Once all of the reads are mapped, for every read and mate-pair we compute mapping confidence statistics. Initially these are computed for each read; however, they are then combined to compute likelihoods of accidental matches for mate-pairs.
### Computing statistics for single reads {#s2c1}
While a very thorough statistical theory for local alignments has been established [@pcbi.1000386-Karlin1], this theory assumes the comparison of infinite length strings, and hence is inappropriate for evaluating alignments of very short reads to a reference genome. Instead, we have designed confidence statistics that explicitly model short reads, and allow for the computation of confidences in the presence of short insertions and deletions. We estimate the confidence in the possible mappings of each read by using the following statistics (calculated by the PROBCALC program): *pchance* -- the probability that the hit occurred by chance -- and *pgenome* -- the probability that the hit was generated by the genome, given the observed rates of the various evolutionary and error events. For example, a good alignment would have a low *pchance* (close to 0) and a very high *pgenome* (close to 1). In this section we briefly expand on these two concepts, give them mathematical definitions, and merge them to formulate an overall alignment quality measurement. A detailed description is in [Methods](#s3){ref-type="sec"} (Computing Statistics: *pchance* and *pgenome*).
The *pchance* of a hit is the probability that the read will align with as good a score to a genome that has the same length, but random nucleotide composition with equal base frequencies (that is, the read will align as well *by chance*). In order to compute this, we count all of the possible k-mers with an equal number of changes as observed in the hit, and we call this number . For example, if we only have substitutions in our alignment (that is, no indels) and an alignment length of , then gives the number of unique strings to which the read can align with the specified number of substitutions. A more detailed discussion on the construction of , especially for the more complex count for indels, appears in Computing Statistics: *pchance* and *pgenome*. The term compares the number of unique strings with the given score (when aligned to the read) compared to all possible unique reads of length , and gives us the probability that a read matches by chance at any location. To compute the *pchance* statistic over the entire length of the genome, we assume independence of positions, and evaluate the likelihood that there is a match at any of the positions:where is the alignment length, is the genome length (2 corresponds to the two strands), and is a correction factor for mappings that are shorter than the length of the read, detailed in Computing Statistics: *pchance* and *pgenome*.
Our second computation, *pgenome*, defines the probability that a hit was generated by the genome via common evolutionary events characteristic of the genome - i.e. substitutions, indels and errors. First, we estimate the rate for each type of event via bootstrapping. Then, we compute the likelihood that the read will differ by as many events from the genome via a binomial probability that uses this estimation and our observations for the events in the current hit. For example, when considering the number of errors, we first estimate the average error rate over all hits, and then we can define the probability that the current read was created via this many errors bywhere is the number of observed errors in the current hit, and is the alignment length. We can similarly define and for substituion and indel events, respectively. Finally, we can form asMore specifics about the mathematical formulations are available in Computing Statistics: *pchance* and *pgenome*.
Finally, we define the quality measurement of this hit as the *normalized odds*, i.e. a probability odds ratio normalized over all of the hits of this read:This value represents a relative credibility of this hit compared to the others for a given read: A single hit would have a normalized odds score of 1, two equally good hits will both have of 0.5 for both, while for an exact match and a more distant one, the former will have a close to 1, and the latter close to 0.
### Computing statistics for mate-pairs {#s2c2}
SHRiMP also assigns mate-pair confidence values (akin to the read confidence values predicted by probcalc) by combining the confidence values for individual reads with emprically observed distributions of insert sizes in the library. We compute the distribution of the mapped distances (distance between the mapped positions of the two reads) for all mate-pairs, and save the average distance (see *Computing Mate Pairs with Statistics* for more details). Then, for each mate-pair mapping, we assign a pchance, pgenome and normodds score, similar in meaning to those used in the previous section:
- ***pchance*** **for mate-pairs**: assume is the pchance of a read that takes g, the length of the genome, as a parameter. Now, the pchance of a mate-pair read_1, read_2 is defined aswhere is the length of the genome used in probcalc, is the average mate-pair distance, and is the distance of the current mate-pair. That is, we ask the question: what is the probability that a read as good as the first read would align anywhere in the genome by chance, *and* that a second read will align by chance within the observed mate-pair distance?
- ***pgenome*** **for mate-pairs**: assume is the pgenome of a read. We can compute the pgenome of each mate-pair bywhere is the tail probability of the mate-pair distance distribution we computed (both tails, starting at the cutoff). Therefore, for a mate-pair with the distance really close to , the pgenome will be close to , otherwise, it will be penalized. Thus, following the difinition of pgenome, we will get a lower probability that the mate-pair was generated from the genome if the mate-pair distance is too big or too small compared to the average.
A discussion of the implementation steps are included in the SHRiMP README, and a more detailed discussion of the statistical values is included in *Computing Mate Pairs with Statistics*.
Validation {#s2d}
----------
In our experiments, we used SHRiMP to compare 135 million 35 bp reads from a tunicate *Ciona savignyi* to the reference genome [@pcbi.1000386-Small2]. The fragments were sequenced from sheared genomic DNA with an AB SOLiD 1.0 instrument. In the following sections we first describe the running time of SHRiMP at different parameter settings, and then evaluate the quality of our alignments compared to the Applied Biosystem\'s read mapping program.
### Running time analysis {#s2d1}
One of the advantages of the SHRiMP algorithm is the seamless parallelism provided by the fact that we can simply subdivide the reads into separate computational jobs, without affecting the results. This allows us to take full advantage of compute clusters regardless of the amount of memory available at each machine. We took a random subset consisting of 500,000 35 bp *C. Savignyi* reads and mapped them to the genome. The full read dataset and reference genome are available at <http://compbio.cs.toronto.edu/shrimp/misc/paper_ciona_reads_35mer.csfasta.tar.bz2> and <http://mendel.stanford.edu/sidowlab/CionaData/CionaSavignyi_v.2.1.fa.zip>, respectively.
The running times at several parameter settings are summarized in [Table 1](#pcbi-1000386-t001){ref-type="table"}. Note that from smallest to largest seed weight, we see a nearly two orders of magnitude difference in total run time, most of which is concentrated in the vectorized Smith-Waterman filter, and, to a lesser degree, in the spaced k-mer scan. The final, full color-space Smith-Waterman alignment took approximately constant time across all runs, as the average number of top scoring hits that reached the stage was nearly constant (24.49±0.5); however, proportional time increased as the filter stages became more efficient. While SHRiMP is somewhat slower than other short read mapping programs, it allows both for micro-indels in the alignments and a proper color-space alignment algorithm. SHRiMP is also very configurable in terms of sensitivity and running time trade-offs.
10.1371/journal.pcbi.1000386.t001
###### Running time of SHRiMP for mapping 500,000 35 bp SOLiD *C. savignyi* reads to the 180 Mb reference genome on a single Core2 2.66 GHz processor.
{#pcbi-1000386-t001-1}
K-mer (7,8) (8,9) (9,10) (10,11) (11,12) (12,13)
------------------------- ------------- --------- --------- ---------- ---------- ----------
\% K-mer Scan 10.1% 16.5% 18.9% 13.4% 9.8% 7.4%
\% Vectorized SW Filter 88.8% 75.4% 49.8% 30.2% 20.1% 14.9%
\% Full SW Alignment 1.1% 8.0% 30.7% 55.5% 68.8% 76.2%
Time 1 d21 h34 m 6 h18 m 1 h36 m 50 m28 s 37 m52 s 32 m32 s
In all cases, two k-mer hits were required within a 41 bp window to invoke the vectorized Smith-Waterman filter.
### *Ciona savignyi* polymorphism analysis {#s2d2}
The primary strength of SHRiMP and other mapping methods based on Smith-Waterman alignments is the ability to map reads containing complex patterns of sequence variation, including insertions, deletions and clusters of closely-space SNPs. Mappers that exclusively produce ungapped alignments can only find SNPs. Furthermore they are more likely to miss dense clusters of SNPs, since the overlapping reads contain many mismatches, and SNPs adjacent to an indel, since only a small fraction of the overlapping reads contain just the SNP. Finally, since SHRiMP produces local alignments, it can map a read even if either end overlaps a large indel or structural variant.
To evaluate the effectiveness of SHRiMP for detecting sequence variation we used it to find polymorphisms in a resequenced *Ciona savignyi* individual. *C. savignyi* is a challenging test case because of its very high polymorphism rate: the SNP heterozygosity is 4.5% and the average per-base indel heterozygosity is 16.6% (indel rate of 0.0072 events per base) [@pcbi.1000386-Small1]. We therefore expect that even short reads will frequently span multiple variant sites.
We used the AB SOLiD sequencing platform to generate 135 million reads of length 35 bp from a single *C. savignyi* individual. We then aligned those reads to the reference genome [@pcbi.1000386-Small2] with SHRiMP using lenient scoring thresholds so that reads with multiple variant sites could be mapped, and we selected the single highest-scoring alignment for each read (see [Methods](#s3){ref-type="sec"}). We discarded alignments in repetitive sequence by removing reads with multiple similarly scoring alignments ("non-unique" matches). The mapping took 48 hours using 250 2.33 GHz cores. [Table 2](#pcbi-1000386-t002){ref-type="table"} summarizes the mapping results.
10.1371/journal.pcbi.1000386.t002
###### Mapping results for 135 million 35 bp SOLiD reads from *Ciona savignyi* using SHRiMP and the SOLiD mapper provided by Applied Biosystems.
{#pcbi-1000386-t002-2}
SHRiMP SOLiD Mapper
------------------------------------------ -------------------- ---------------------
Uniquely-Mapped Reads 51,856,904 (38.5%) 15,268,771 (11.3%)
Non-Uniquely-Mapped Reads 64,252,692 (47.7%) 12,602,387 (9.4%)
Unmapped Reads 18,657,736 (13.8%) 106,896,174 (79.3%)
Average Coverage (Uniquely-Mapped Reads) 10.3 3.0
Median Coverage (Uniquely-Mapped Reads) 8 1
SNPs 2,119,720 383,099
Deletions (1--5 bp) 51,592 0
Insertions (1--5 bp) 19,970 0
Non-uniquely-mapped reads have at least two alignments, none of which is significantly better than the others (see [Methods](#s3){ref-type="sec"}). SNPs and indels have at least four supporting reads.
The alignment data contains noise due to two types of errors: sequencing errors and chance alignments. Chance alignments are a significant problem for short reads, particularly with the low alignment score thresholds necessary for mapping reads containing significant variation. Reads containing both sequence variation and sequencing errors are even more likely to map to the wrong position in the reference sequence. To combat the high false-positive rate, for the remaining analysis we focused on a high-quality subset of the data consisting of sequence variants supported by at least four independent reads.
Across the genome SHRiMP detected 2,119,720 SNPs supported by at least four reads. For comparison, we used the SOLiD aligner provided by Applied Biosystems to map the reads to the genome with up to three mismatches, where each mismatch can be either a single color-space mismatch or a pair of adjacent mismatches consistent with the presence of a SNP. Compared to the SOLiD mapper, SHRiMP mapped 4.2 times as many reads and found 5.5 times as many SNPs. The AB mapper, however, was a lot faster, requiring 255 CPU hours to complete the alignments, or roughly 50× faster than SHRiMP. While it is possible to run the mapper with greater sensitivity, allowing for more errors and SNPs, and thus more mapped reads, doing so would surrender much of the runtime advantage and still not overcome its fundamental inability to detect insertion and deletion polymorphisms. SHRiMP, on the other hand, is capable of handling indels, and detected tens of thousands of them.
SHRiMP detected 51,592 deletions and 19,970 insertions of size 1--5 bp. The observed ratio of 2.5× between insertions and deletions for the *C. savignyi* data is biased by the construction of the reference genome -- whenever the two haplomes differed, the reference agreed with the longer one. While there is a smaller inherent bias against detecting insertions (reads containing nucleotides not present in the reference) compared to deletions because a read spanning a deletion only incurs a gap penalty whereas an insertion both incurs a gap penalty and has fewer bases that match the reference. For simulated data (see next section) this bias was only ∼5% for single basepair indels (data not shown). The size distribution of the detected indels ([Figure 5A](#pcbi-1000386-g005){ref-type="fig"}) drops more rapidly with length than expected [@pcbi.1000386-Small1], but this detection bias against longer indels is not surprising since longer indels have lower alignments scores.
{#pcbi-1000386-g005}
Mapping *C. savignyi* sequence is challenging primarily because the population contains so much variation. [Figure 5B](#pcbi-1000386-g005){ref-type="fig"} shows the high frequency of closely spaced SNPs detected by SHRiMP. Mappers that can only detect nearly exact matches fail to map the reads overlapping these dense SNP clusters. Note that even though the reads are generated from the whole genome, a significant fraction of the non-repetitive *C. savignyi* genome is coding, making it is possible to see the typical three-periodicity of SNPs in coding regions. Furthermore SHRiMP recovers microindels, which are completely invisible to ungapped aligners and yet account for a significant fraction of sequence variation in *C. savignyi*.
### Analysis of simulated data {#s2d3}
In order to further validate the accuracy of the SHRiMP alignments we have designed simulated experiments, where we sampled random locations from the *C. savignyi* genome, introduced polymorphisms (SNPs and indels) at the rates previously observed in the C. savignyi genome [@pcbi.1000386-Small2], added sequencing errors at rates observed in our *C. savignyi* dataset (2--7%, depending on the position in the read), and mapped the reads back to the original genome. Each sampled read could have multiple SNPs and indels, though due to the low indel rate only a small fraction of the reads had multiple indels. We mapped the reads with SHRiMP and postprocessed with PROBCALC (*pchance*\<0.001). Considering only those reads that had a unique top hit, we computed the **precision** -- the fraction of reads for which this unique hit was correct, and **recall** -- the fraction of all reads that had a unique, correct hit. [Table 3](#pcbi-1000386-t003){ref-type="table"} shows the results of this analysis. For each read, we classified it based on the number of SNPs and the maximum indel length, and computed precision and recall for each class. With such polymorphism, we can expect the average read to have approximately 1.5 SNPs and 1.9 errors. SHRiMP was able to accurately map 76% of reads with 2 SNPs and 0 indels, at 84% precision, and nearly half of all reads with 2 SNPs and 3 bp indels at 74% precision.
10.1371/journal.pcbi.1000386.t003
###### Color-space mapping accuracy of SHRiMP.
{#pcbi-1000386-t003-3}
Number of SNPs
-------- --- ---------------- ------ ------ ------ ------ ------ ------ ------ ------ ------
0 85.7 83.2 84.8 81.3 83.5 76.6 80.6 65.2 75.6 46.8
Max 1 83.8 79.4 82.2 74.0 79.4 62.6 72.8 43.2 63.1 24.7
Indel 2 83.2 77.1 80.8 69.6 77.9 56.6 68.2 36.4 56.4 18.9
Length 3 80.7 71.0 79.6 64.2 73.6 48.3 66.5 31.5 57.1 16.6
4 78.0 65.4 76.5 56.1 71.4 41.9 60.6 23.9 50.3 12.4
5 75.9 58.9 73.0 48.1 69.7 36.6 57.0 21.3 46.0 12.7
Each cell shows the precision and recall for mapping simulated reads with varying amounts of polymorphism. SHRiMP was able to accurately map \>46% of all reads with either 4 SNPs or 5 bp indels, despite the large number of sequencing errors in our dataset (up to 7% towards the end of the read).
Methods {#s3}
=======
Details of the SHRiMP Algorithm {#s3a}
-------------------------------
The algorithm starts with a rapid k-mer hashing step to localize potential areas of similarity between the reads and the genome. All of the spaced k-mers present in the reads are indexed. Then for each k-mer in the genome, all of the matches of that particular k-mer among the reads are found. If a particular read has as many or more than a specified number of k-mer matches within a given window of the genome, we execute a vectorized Smith-Waterman step, described in the next section, to score and validate the similarity. The top highest-scoring regions are retained, filtered through a full backtracking Smith-Waterman algorithm, and output at the end of the program if their final scores meet a specified threshold. The SHRiMP algorithm is summarized in [Figure 6](#pcbi-1000386-g006){ref-type="fig"}.
{#pcbi-1000386-g006}
### Spaced seed filter {#s3a1}
We build an index of all spaced k-mers in the reads, and query this index with the genome. Our approach was taken primarily for simplicity: our algorithm can rapidly isolate which reads have several k-mer matches within a small window by maintaining a simple circular buffer of recent positions in the genome that matched the read. Since our targeted compute platform is a cluster of batch processing machines, indexing the reads means that we can easily control memory usage and parallelism by varying the read input size and splitting the read set accordingly. Data is only loaded at program invocation; we do not stream in new reads from disk as the algorithm runs.
### Vectorized Smith-Waterman implementation {#s3a2}
The SHRiMP approach relies on a rather liberal initial filtering step, followed by a rigorous, but very fast Smith-Waterman alignment process. By maximizing the speed of the Smith-Waterman comparison, we are permitted to let the algorithm test a larger number of potential regions.
Most contemporary mobile, desktop and server-class processors have special *vector* execution units, which perform multiple simultaneous data operations in a single instruction. For example, it is possible to add the eight individual, 16-bit elements of two 128-bit vectors in one machine instruction. Over the past decade, several methods have been devised to significantly enhance the execution speed of Smith-Waterman-type algorithms by parallelizing the computation of several cells of the dynamic programming matrix. The simplest such implementation computes the dynamic programming matrix using diagonals. Since each cell of the matrix can be computed once the cell immediately above, immediately to the left, and at the upper-left corner have been computed, one can compute each successive diagonal once the two prior diagonals have been completed. In this way, the problem can be parallelized across the length of supported diagonals (see [Figure 6B](#pcbi-1000386-g006){ref-type="fig"}). In most cases, this is a factor of 4 to 16. The only portion of such a 'Wozniak' approach that cannot be parallelized is the identification of match/mismatch scores for every cell of the matrix, which has to be done sequentially. These operations are expensive, necessitating 24 independent data loads for 8-cell vectors, and become increasingly problematic as vector sizes increase. Because memory loads cannot be 'vectorized', when the parallelism grows, so does the number of lookups. For example, with 16-cell vectors, the number of data loads doubles to 48.
We propose an alternate method, where the running time of the fully vectorized algorithm is independent of the number of matches and mismatches in the matrix, though it only supports fixed match/mismatch scores (rather than full scoring matrices). Our key observation is that it is possible to completely parallelize the score computation for every diagonal. [Figure 6B](#pcbi-1000386-g006){ref-type="fig"} demonstrates the essence of our algorithm: by storing one of the sequences backwards, we can align them in such a way that a small number of logical instructions obtain the positions of matches and mismatches for a given diagonal. We then construct a vector of match and mismatch scores for every cell of the diagonal without having to use expensive and un-vectorizable load instructions or a pre-compute a 'query profile'. In our tests, using a diagonal approach with our scoring scheme surpasses the performance of Wozniak\'s original algorithm and performs on par with Farrar\'s method [@pcbi.1000386-Farrar1]. [Table 4](#pcbi-1000386-t004){ref-type="table"} summarizes these results. The advantage of our method over Farrar\'s is that it is independent of the scores used for matches/mismatches/gaps, and it will scale better with larger vector sizes. A disadvantage is that we cannot support full scoring matrices and are restricted to match/mismatch scores, though this is less important for DNA alignment. Additionally, Farrar\'s method is much faster for large databases where most of the sequence is dissimilar to the query. However, this is never the case for SHRiMP as the seed scan phase targets only small, similar regions for dynamic programming. In these cases our algorithms perform similarly.
10.1371/journal.pcbi.1000386.t004
###### Performance (in millions of cells per second) of the various Smith-Waterman implementations, including a regular implementation (not vectorized), Wozniak\'s diagonal implementation with memory lookups, Farrar\'s method and our diagonal approach without score lookups.
{#pcbi-1000386-t004-4}
Processor type Unvectorized Wozniak Farrar SHRiMP
---------------- -------------- --------- -------- --------
**Xeon** 97 261 335 338
**Core 2** 105 285 533 537
We inserted each into SHRiMP, and used SHRiMP to align 50 thousand reads to a reference genome with default parameters. The improvements of the Core 2 architecture for vectored instructions lead to a significant speedup for our approach and Farrar\'s, while Wozniak\'s algorithm slight improvement is due to the slow match/mismatch lookups.
### Final pass {#s3a3}
The vectorized Smith-Waterman approach described above is used to rapidly determine if the read has a strong match to the local genomic sequence. The locations of the top hits for each read are stored in a heap data structure, which is updated after every invocation of the vectorized Smith-Waterman algorithm if the heap is not full, or if the attained score is greater than or equal to the lowest scoring top hit. Once the whole genome is processed, highest scoring matches are re-aligned using the appropriate full color- or letter-space Smith-Waterman algorithm. This is necessary, as the vectorized Smith-Waterman algorithm described above only computes the maximum score of an alignment, not the traceback, as this would require a much more complicated and costly implementation. Instead, at most only the top alignments for each read are re-aligned in the final step.
Computing Statistics: *pchance* and *pgenome* {#s3b}
---------------------------------------------
In *Computing Statistics for Single Reads*, we briefly introduced the concepts of the *pchance*, *pgenome* and *normalized odds* of a hit. In this section we expand on the details regarding the construction of *pchance* and *pgenome*. In these formulas we make use the following definitions:
- is the genome length
- is the alignment length (note this may be different from the read length, which is constant)
- is the number of substitutions (mismatches) in our alignment
- is the number of nucleotide insertions in our alignment, where the genome is the "original" sequence. For example, if the genome is AC-G and a read is ACTG, there is an insertion of a T.
- is the number of nucleotide deletions in our alignment. For example, if the genome is ACTG and a read is A-TG, there is a deletion of a C.
- is the number of insertion events (for example, for a single insertion of length 3 we have and .) is similar.
- : following the previous definition, will describe the number of permutations of insertion events. To determine the number of *distinguishable* permutations, we need to first look at the frequency of insertion events of a certain size, . For example, is we have 3 insertions of size 2, we need to divide the permutations by . Therefore, the *distinguishable* permutations of insertion events can be written as:
Below, we refer to this denominator term as . We similarly define .
- describes the number of ways to assign indistinguishable objects in indistinguishable bins, which is recursively defined by with and .
### pchance {#s3b1}
We begin with the mathematical formulation of *pchance* (defined above):where, as described before, is the number of possible unique sequences with the given edit distance as a fraction of all possible unique reads of length . Thus, gives us the probability that the current read has aligned by chance to a random genome of the size of a read. To this term, we add a correction factor of which accounts for all the possible places the alignment of size might match. For example, if the is 25 and we have a match of size 22, we should count for every position where this match could be found, that is 25−22+1 = 4. Finally, to get the probability that the current read has aligned by chance to a random genome of size (instead of size ), we get formula (7).
The factor that lies at the core of this calculation is , the number of possible unique sequences that would align to the read with the given edit distance. We have shown the definition of , which computes when there are no indels in the alignment:
However, the calculation of the number of references to which a read will map with a particular *indel* count, , depends on the sequence of that read and is significantly more complicated. We define a lower and upper bound on in this case: a lower bound (least number of unique sequences) occurs when the current read is one repeated nucleotide, for example \[AAAAAA\], and the higher bound occurs with the most change in nearby nucleotides, say \[ACGTAC\]. In the former case, we need to look at the deletion events from the genome to this read, consider all the combinations of that number of deletion events and deleted nucleotides, as well as all the places where these combinations may occur. This gives the formulaLooking for the upper bound, we note that the places and combinations of insertions also matters in generating unique sequences, therefore giving us two extra terms involving
In order to estimate the correct value for , we estimated the average complexity of the reads in our dataset (i.e., between the simplest \[AAAAA...\] and the most complex \[ACGTACGT...\]). And have found that the mean observed could be accurately estimated byFinally, we can approximate the total as
### pgenome {#s3b2}
In *Computing Statistics for Single Reads*, we defined our pgenome factor as , wherewith the rate of event (estimated via bootstrapping) and the number of observed events of type in the current alignment. We wrote as an approximation because there are small corrections to this formula for each probability that is part of pgenome. First, for the *error* term , the number of sites that can support errors is in fact one minus the read size, giving usWhen considering substitutions, we can have changes at any of the inner nucleotides, excluding erroneous sites:As before, when we look at alignments that involve indels, the formula becomes more complex. In the case of *pgenome*, we do not have to consider the various placements of insertion or deletion events, but we do have to consider, for fixed placements of events, the various combinations of the total number of insertions and deletions into a set number of events.
### Computing mate pairs with statistics {#s3b3}
In this section we provide several details for the implementation, usage and statistics of the matepair post-processing step introduced in *Computing Statistics for Mate-pairs*. We define a **good** matepair mapping as a mapping whose distance (between the two reads) are smaller than some chosen limit , and for which the read mappings are in a consistent orientation and strand(i.e. R~+~F~+~ or F~−~R~−~). First, probcalc_mp will compute a matepair distance and standard deviation by looking at all the connected forward and reverse reads - all matepairs - and adding the distance of any matepair with exactly one *good* mapping to a histogram. Optionally, one can choose to use only unique good mappings, or only use a certain number of mappings (say, the first 100,000) to speed up the program.
Next, we call a matepair **concordant** if it has at least one *good* mapping, and otherwise we call it **discordant**. Depending on the task, probcalc_mp can output all concordant matepairs, or all discordant matepairs. For each matepair mapping, probcalc_mp will compute the pgenome and pchance, as introduced in *Computing Mate Pairs with Statistics*.
Parameters {#s3c}
----------
For the *C. savignyi* polymorphism analysis we ran SHRiMP with the following parameters. We used the spaced seed "11110111" and required two hits per 40-base window to invoke the Smith-Waterman algorithm. The Smith-Waterman scoring parameters were set to +100 for a matching base, −90 for a mismatch, −250 and −100 to open and extend a gap respectively, and −300 for a crossover (sequencing error). The minimum Smith-Waterman score thresholds were 1000 for the vectorized first pass and 1275 for the final alignment pass. We discarded alignments with *pchance* less than 0.05, and to remove reads from known repetitive sequence we required *normodds* to be at least 0.8.
We thank Ziming Weng for preparing the *C. savignyi* library and Cheryl Smith for sequencing the library. We also thank Aston Wallis for proofing the manuscript.
The authors have declared that no competing interests exist.
This work was sponsored by Natural Sciences and Engineering Research Council (NSERC) of Canada Undergraduate Student Research Awards, Canadian Institute for Health Research (CIHR), Applied Biosystems, NSERC Discovery Grant, MITACS, and a Canada Foundation for Innovation equipment grant. Computational resources were provided by the Stanford BioX2 compute cluster, supported by NSF award CNS-0619926. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
[^1]: Conceived and designed the experiments: SMR PL AS MB. Performed the experiments: SMR PL AVD MF. Analyzed the data: SMR PL AVD MF AS MB. Contributed reagents/materials/analysis tools: AS. Wrote the paper: SMR PL AVD MF AS MB.
|
Infrared (IR) thermal cameras can be used in a number of different situations, for example, when inspecting or surveying complex electrical systems such as transformers, switchgears, etc., or water carrying systems such as heat exchangers, radiators, etc. IR cameras are used for capturing, displaying and storing thermal images. The thermal images may then be viewed and analyzed in order to, for example, find faulty electrical wirings or couplings, leaking water pipes, etc.
When viewing a thermal image captured by an IR camera in an IR camera display, there are a number of different view parameters that determines how the thermal image is presented to the user of the IR camera. For example, there may be a thermal image colour span, which the user may set in order to determine in between which temperature values the colour map of the presented thermal image should range. A further example is a thermal image colour level, which the user may set in order to determine around which temperature the colours should be centred.
Different settings of these view parameters in an IR camera may be used in situations, such as, for example, when trying to perform scene comparisons that have the same colour-to-temperature scaling, or in order to identify and show temperature gradients in thermal images with large temperature differences where the interesting temperature gradients is within a small portion of the entire thermal image.
These view parameters require default settings or that manual settings are inputted into the IR camera in order to present a thermal image that can be readily interpreted by the user. However, this requires that the user of the IR camera be experienced and understands exactly how to adjust the view parameters in order to get the information he needs. For an inexperienced user, the manual inputs may be difficult to understand and inefficient to use.
Furthermore, it may also be difficult for an inexperienced user to know how to adjust the settings of the view parameters in order to see other objects, particularly, if the IR camera is in a fusion mode where the temperatures that are bottomed (flattened) are replaced with a visual image, that is, the pixels having a temperature that are outside the thermal image colours span being replaced with visual image pixels. It may also be difficult to adjust the view parameter settings and interpret what you actually see when, for example, sweeping with the IR camera over an area with different temperature content looking for small temperature gradients at different temperature levels.
Automatic settings based on the entire thermal image content or a majority of the entire thermal image content is known. These automatic settings may work well in some situations, but work less well in others, such as, for example, when viewing an image view with large temperature differences or when the user is only interested in a specific portion of the thermal image. |
<?php
/**
* Zend Framework
*
* LICENSE
*
* This source file is subject to the new BSD license that is bundled
* with this package in the file LICENSE.txt.
* It is also available through the world-wide-web at this URL:
* http://framework.zend.com/license/new-bsd
* If you did not receive a copy of the license and are unable to
* obtain it through the world-wide-web, please send an email
* to license@zend.com so we can send you a copy immediately.
*
* @category Zend
* @package Zend_Service
* @subpackage Amazon
* @copyright Copyright (c) 2005-2010 Zend Technologies USA Inc. (http://www.zend.com)
* @license http://framework.zend.com/license/new-bsd New BSD License
* @version $Id$
*/
/**
* Amazon Ec2 Interface to allow easy creation of the Ec2 Components
*
* @category Zend
* @package Zend_Service
* @subpackage Amazon
* @copyright Copyright (c) 2005-2010 Zend Technologies USA Inc. (http://www.zend.com)
* @license http://framework.zend.com/license/new-bsd New BSD License
*/
class Zend_Service_Amazon_Ec2
{
/**
* Factory method to fetch what you want to work with.
*
* @param string $section Create the method that you want to work with
* @param string $key Override the default aws key
* @param string $secret_key Override the default aws secretkey
* @throws Zend_Service_Amazon_Ec2_Exception
* @return object
*/
public static function factory($section, $key = null, $secret_key = null)
{
switch(strtolower($section)) {
case 'keypair':
$class = 'Zend_Service_Amazon_Ec2_Keypair';
break;
case 'eip':
// break left out
case 'elasticip':
$class = 'Zend_Service_Amazon_Ec2_Elasticip';
break;
case 'ebs':
$class = 'Zend_Service_Amazon_Ec2_Ebs';
break;
case 'availabilityzones':
// break left out
case 'zones':
$class = 'Zend_Service_Amazon_Ec2_Availabilityzones';
break;
case 'ami':
// break left out
case 'image':
$class = 'Zend_Service_Amazon_Ec2_Image';
break;
case 'instance':
$class = 'Zend_Service_Amazon_Ec2_Instance';
break;
case 'security':
// break left out
case 'securitygroups':
$class = 'Zend_Service_Amazon_Ec2_Securitygroups';
break;
default:
throw new Zend_Service_Amazon_Ec2_Exception('Invalid Section: ' . $section);
break;
}
if (!class_exists($class)) {
//$1 'Zend/Loader.php';
Zend_Loader::loadClass($class);
}
return new $class($key, $secret_key);
}
}
|
Q:
Why Can't I Drag IBAction from Storyboard to Swift File?
I'm attempting to connect some buttons from a preferences view in my Swift OS X application.
Instead of putting all the IBActions and IBOutlets in my AppDelegate.swift file (seems messy, though I may be wrong), I'd like to simply have the actions and outlets interacting with another file. In this scenario I want to interface much of the userDefaults storage with the preferences window I create.
But whenever I attempt to drag a preference window button or anything else to the new swift file, it doesn't give me the same ability to drop the link and define the action/outlet.
If someone has any insight to this, or could help walk me through the process that would be fantastic. I have a bit of understanding of Xcode/cocoa and from what I've searched it may something to do with View Controllers, but I don't know where to go from here.
Thanks!
Tyler
A:
With swift you could make a new file with an extension of the app delegate class and add the interface builder stuff there.
extension AppDelegate {
//IB stuff here
}
|
Amazon lets gift-givers send Kindle books by email
Nov 19, 2010
A Kindle DX (R), a large-screen version of its popular Kindle electronic reader is seen during a press conference in New York 2009. Amazon.com on Friday began letting gift-givers send Kindle electronic books by email as the leading online shop positioned itself to cash in on the year-end holiday season.
Amazon.com on Friday began letting gift-givers send Kindle electronic books by email as the leading online shop positioned itself to cash in on the year-end holiday season.
Kindle digital books can be read on computers, smartphones, iPads, and iPod touch devices as well as on the eponymous tablet sold by Amazon.
"We are thrilled to make it easier than ever for our customers to give their favorite Kindle book to a friend or family member as a gift," said Kindle vice president Russ Grandinetti.
"We're making this functionality available in time for the holidays to offer an easy, stress-free holiday shopping option for anyone -- not just Kindle owners."
The massive library of gift books available at amazon.com/givekindlebooks includes "The Girl Who Kicked the Hornet's Nest" and "The Immortal Life of Henrietta Lacks."
US spending on e-books was expected to total 966 million dollars this year, up from 301 million dollars last year, and to reach 2.81 billion dollars in 2015, according to a report this month by Forrester Research Inc.
Forrester said the number of e-book readers in the United States was expected to grow from 3.7 million at the end of last year to 10.3 million at the end of this year, hitting 29.4 million in 2015.
Seven percent of online US adults who read books read e-books, a number that is expected to double a year from now, Forrester said.
Forrester also said Amazon's Kindle store "stands to benefit tremendously" from the rise in e-book reading because of its existing relationship with book buyers through Amazon.com.
At the end of 2014, Facebook reported 1.39 billion monthly active users. In the meantime, 500 million tweets were sent each day on Twitter. Indeed, social networks have come to dominate aspects of our lives. ...
Two former federal agents are accused of using their positions and savvy computer skills to siphon more than $1 million in digital currency from the online black market known as Silk Road while they and their agencies operated ... |
1.10.2010
Evangelio / José Antonio Ramos Sucre
Gospel
The mystical commotion had startled me. I was in the presence of an aerial vision. The symbols of faith gained a spiritual form and emitted voice. I fell on my knees under the radiant sky. A message of health, music from chaste silence, the earth surprised everyone, the inveterate aridity consoled. The escape of the devoted dream caused a unanimous lament in the far ends of the dark valley. The humble ones told themselves they had been hallucinated by a meteor of vain light and they complained about their shame and abandonment. |
587 N.E.2d 501 (1992)
225 Ill. App.3d 54
167 Ill.Dec. 232
The PEOPLE of the State of Illinois, Plaintiff-Appellee,
v.
John PEREZ, Defendant-Appellant.
No. 1-89-1268.
Appellate Court of Illinois, First District, First Division.
January 21, 1992.
*502 Michael J. Pelletier, Deputy Defender, Office of the State Appellate Defender, Chicago (James Chadd, Asst. Appellate Defender, of counsel), for defendant-appellant.
Jack O'Malley, Cook County State's Atty., Chicago (Renee Goldfarb, Steven Klaczynski, William P. Pistorius, Asst. State's Attys., of counsel), for plaintiff-appellee.
Presiding Justice BUCKLEY delivered the opinion of the court:
After a jury trial, defendant John Perez was found guilty of murder (Ill.Rev.Stat. 1987, ch. 38, par. 9-1-(a)(1)) and sentenced to a 25-year term of imprisonment. Defendant contends on appeal that the circuit *503 erred in denying his motion to suppress statements on the grounds they were the fruit of an illegal arrest and involuntary. Defendant further contends that he was denied a fair trial when the State improperly introduced testimony that a key witness made a prior consistent statement and then argued later in closing that this witness was more credible because he consistently told the same story. We affirm.
Prior to trial, a hearing was held regarding defendant's motion to suppress statements; the following evidence was produced:
Chicago police detective Robert O'Leary testified that on September 25, 1987, while investigating the homicide of the victim, Gerald Gains, he spoke with Garfield Birmingham. Birmingham told O'Leary that he sold drugs for a living from his apartment at 6241 North Kenmore, apartment # 214 in Chicago and that the victim, who had worked for him, was fired for using drugs. Birmingham further stated that he hired defendant and a man named Peter as bodyguards for his drug transactions. Defendant and Peter told Birmingham that they had a 9-millimeter chrome automatic handgun which they would point at customers while hiding during drug transactions. Birmingham gave O'Leary defendant's address.
Detective Lawrence Thezan testified that he worked the evening shift of September 25, 1987, and that he also spoke with Birmingham. Birmingham told Thezan that defendant was in the building where the victim was shot and that defendant knew about the incident. Birmingham informed Thezan where defendant lived and that a 9-millimeter chrome handgun was in the apartment where they sold drugs. At the time of this conversation, Thezan was aware that the bullet which killed the victim came from a large caliber, semi-automatic pistol. Based on this information, Thezan and his partner, Detective Wojtowicz, went to defendant's house. The detectives spoke with numerous individuals at the side of the building; a few of these individuals went into the building and retrieved defendant. The detectives identified themselves, told defendant why they were there and asked him if he would go to Area Six headquarters with them. Defendant agreed to go; he was not handcuffed.
At the station, now about 8 or 9 p.m., defendant was taken to an interview room on the second floor. There, the detectives explained to defendant what they knew, and asked him if he had any information about the murder. Defendant told the detectives that he was at apartment # 214 at 6241 North Kenmore with Peter on September 24, 1987. Defendant stated that Peter owned a 9-millimeter hand gun. Around 7 p.m. that evening, a Jamaican man known only as Earl-16 came to the apartment and asked if they had any beer. They did not, so Earl-16 and Peter gave defendant some money to purchase beer. The victim arrived and spoke with Earl-16. Defendant then left to go to the store.
Detective Thezan asked defendant if he knew where they could find Peter and Earl-16. Defendant said he knew that Peter was Jamaican but did not know where he lived or his phone number; if defendant ever needed to meet with Peter, he would see him in Jamaican bars. Defendant stated that he had no information on Earl-16.
After the detectives returned from their unsuccessful attempt to locate Peter and Earl-16, they again talked with defendant, who related that Peter kept the 9-millimeter gun in the toilet tank. Again, after defendant provided them with this information, the detectives left the station and went to the apartment to verify this lead. They were unable to locate the gun at the apartment but did find some papers with the victim's name on them. During the time the officers were away, defendant remained in the police station.
Detective O'Leary testified that he arrived at work around 8:30 a.m. on September 26 and spoke with the midnight shift. At 9:30 a.m., O'Leary and his partner, Detective Kajari, went and spoke with defendant. Defendant told them that a man named "Cratches" came over to the apartment and asked for a couple of beers. Cratches then told defendant and Peter that Earl-16 had shot someone, perhaps *504 Gaines. Defendant told the detectives that he could point out Earl-16's apartment. The detectives and defendant then proceeded to where Earl-16 lived. The detectives exited the vehicle to see the apartment manager who told them that a Jamaican named Herliz Vanzie lived in the building. The detectives went to his apartment but nobody was there. During this time, defendant remained unhandcuffed and alone in the back seat of the squad car.
On the way back to Area Six headquarters, the three stopped and got something to eat. At headquarters, the detectives checked their records on Herliz Vanzie and learned that he had the nickname of Earl. They again left the station and unsuccessfully tried to locate Earl-16. During this time, defendant remained in an interview room.
Detectives Thezan and Wojtowicz continued the investigation when their shift began around 4:30 p.m. on September 26. At this time, defendant was still at the police station, but according to Thezan was not in custody, was free to go at any time and did not ask to leave.
The detectives continued their search for Earl-16 during the night. Later that evening, Earl-16 was located and interviewed in a separate interview room. Thezan explained that defendant did not want to be seen by any individual being brought into the station. Earl-16 told the detectives about his presence in apartment # 214 and that Cratches was a friend of his. He gave the detectives information as to where Cratches could be located, but the information proved unuseful. Thezan then went and got some hamburgers for defendant. Later that night, Detectives Thezan and Wojtowicz again talked to Birmingham who came to Area Six. He said that he had been stabbed while looking for Peter. Birmingham was assisting the detectives in their search for Peter.
Detective O'Leary testified that when he started work on September 27 at 8:30 a.m., Cratches (Duane Sammons) was already there. Cratches was interviewed and stated that at Earl-16's request, he had gone over to apartment # 214 to get a couple of beers. Defendant and Peter were in the apartment. He got two beers and then returned to the apartment where Earl-16 was located. Following this interview, Detectives O'Leary and Thezan again tried to find Peter but were unsuccessful. Defendant remained in an interview room during this time.
When the detectives returned to the station, defendant told them that they might be able to find the gun under the refrigerator or in a vent on the stove at the apartment. No gun was found. The detectives brought food back for defendant, Earl-16 and Cratches, who were all in separate rooms. O'Leary stated that they did not let Earl-16 and Cratches know defendant was there in an attempt to protect defendant because he was the one that gave the detectives their names and other information regarding their possible involvement in the murder.
After eating, the detectives interviewed Earl-16 and Cratches again. Cratches stated that Earl-16 left Cratches' girl friend's apartment and went over to the apartment to get two beers. Cratches next went to the apartment and defendant opened the door pointing a 9-millimeter chrome handgun at him. Cratches told defendant to put the gun down. Defendant replied, "You guys are always ripping us off." Cratches responded that he is Jamaican and defendant put the gun down to his side. Cratches denied telling defendant anything about the shooting because he did not learn of it until the next day.
During Earl-16's interview, Earl-16 said that he went over to apartment # 214 to get a couple of beers; defendant, Peter and a female were in the apartment. They did not have any beer so he and Peter gave defendant some money. During this time, the victim came over and began arguing with defendant and Peter about some furniture of his that he wanted. Defendant told him that he would have to talk to Birmingham because it was his apartment. Defendant eventually left the apartment followed first by the woman and then a short while later by the victim. A few minutes after the victim had left, defendant came back *505 and went into the kitchen. He then left with his coat over his hand. Earl-16 left and went to Cratches' girl friend's apartment and told Cratches to go back and get the beer.
After these interviews, Detectives O'Leary and Thezan discussed discrepancies between the stories given by defendant, Earl-16 and Cratches. They then went back into the interview room with defendant and explained to him that they had been talking to witnesses who have been giving consistent stories which were inconsistent with his story. They told defendant what the discrepancies were and read defendant his rights. Defendant acknowledged that he understood those rights and wished to waive them. It was now 7:30 p.m., about 48 hours after the police first contacted defendant. Defendant then proceeded in the presence of the two detectives to make certain admissions with regard to the murder. After making the oral admission, defendant was placed under arrest and handcuffed for the first time. The detectives contacted Assistant State's Attorney Judith Mondello of Felony Review who came and took a statement from defendant. Defendant, O'Leary and Mondello all signed the bottom of each page of the written statement.
Detectives Thezan and O'Leary testified that no one at anytime threatened, pushed, shoved, struck, kicked, strangled or abused defendant in any way. Defendant was never promised anything in return for his statements, was never refused food, drink or bathroom facilities, was never locked in an interview room and remained free to leave the entire time. According to the detectives, defendant was cooperating with the police. Defendant would give the detectives information which they would attempt to confirm. Both testified that Earl-16 and Cratches were brought to the station based on information defendant provided.
Defendant testified on his own behalf. He stated that he was sitting at home on September 25, 1987, when three police officers arrived without a warrant. The officers grabbed him, patted him down and put him in the back of a police car. The officers drove a short distance, stopped the car, told defendant to get out, handcuffed him, placed him back in the car and drove to the station where he was locked in an interview room. According to defendant, while in the interview room, the officers gave him no food and would not let him use the bathroom. When he asked to use one, he could hear the officers laughing outside the door. Defendant was forced to use a garbage can as a toilet.
Defendant testified that on September 26, around 1 p.m., he left the interview room with Detectives Kajari and O'Leary to go looking for an individual he had told them about. Defendant testified he could point out a building where he had seen that individual exit. When they arrived at the building, defendant went inside with the detectives but was not handcuffed. They went in and talked to a lady manager. The three then went upstairs. One detective told the defendant to stand with him by the stairwell out of view of the door while the other detective went to the door and knocked. No one answered.
On the way back to the station, defendant told the detectives he was hungry and so they stopped at a gas station. Defendant testified that he went in alone and bought cigarettes, a Snickers bar and a 7-Up with his own money. Defendant stated, "I didn't try to run from them and nothing, I try to assist them." After they left the gas station, they went to another police station where the detectives did some research. Defendant stated: "I try to relax, try to assist them with whatever I could help them." They left this police station around 3 or 4 p.m. and returned to Area Six headquarters. Defendant testified that he was again placed in a locked interview room. Sometime later that evening a uniformed police officer brought him a Big Mac and a 7-Up. Defendant also stated he ate again early the next morning.
On the morning of the 27th, defendant was transferred to another room where he was handcuffed to the wall. The detective kept telling him they knew he committed the murder. Defendant denied this and *506 alleged that Detective Kajari started beating him. When the assistant State's Attorney came, O'Leary told her what to write down. She wrote it on a yellow piece of paper. Defendant claimed he never told the assistant State's Attorney what to write. O'Leary told defendant he had to sign the statement, but he would not. Defendant stated that he never received his Miranda rights. After he refused to sign, the assistant State's Attorney ran out of the room. A detective then told defendant that if he signed the statement they would let him go. When the assistant State's Attorney came back in, she also told defendant that he could go home if he signed the statement. Defendant testified that he signed the statement only because he had been beaten and denied food and access to the bathroom during the preceding 48 hours.
During cross-examination, defendant stated that he did not sign the statement. Defendant claimed that the police must have photocopied his signature and tried to "prefabricate" it so it would look like he signed the statement. Defendant later stated that he did sign the bottom of all three pages. Defendant testified that, even though the statement said he was given food, drink, cigarettes and allowed to use the bathroom, this was not true and he only signed it because he was forced.
Defendant testified that in the early morning on September 27, Detective Kajari came in and beat him for approximately 15 to 30 minutes. Kajari then left but came back about 40 minutes later and beat him again for approximately 20 to 30 minutes. Defendant said these beatings occurred throughout the day and up until the assistant State's Attorney came. Defendant said that during one session he was hit in the left eye.
Defendant said he told the assistant State's Attorney that he did not commit the murder, but she wrote what Detective O'Leary told her to write. Defendant said that he told the police that he heard Cratches tell Peter that Earl-16 did the shooting. This is why he was helping the police find Earl-16. Defendant said he was at the apartment because Birmingham took him there instead of dropping him off at his sister's house.
The State called three witnesses in rebuttal. Carlos Martinez, a paramedic for the Cook County Department of Corrections, testified that he conducted the intake procedures for the defendant on September 28, 1987. Martinez took defendant's medical history and did a visual inspection of defendant from the waist up. Defendant told Martinez that he received a "blunt trauma" on September 26 or 27. Defendant stated that he had no complaints or physical problems and that he was in good health. Defendant never told Martinez that he was beaten by police or that he had any pain in his lower back. Martinez noted that defendant had some swelling on his right eye but did not note any bruises to defendant's upper torso.
Dr. Aaron Hamb, a physician at Cermak Health Services of the Cook County Department of Corrections, testified that defendant complained of pain to the coccyx (tail bone) area during a March 1988 examination. Defendant related to Hamb that Chicago police officers had pulled a chair out from beneath him and that this incident caused his pain. At the examination, Hamb looked through defendant's records and found no record of the type of incident defendant claimed. An X-ray was taken, and the radiologist reported that defendant's tail bone had a possible injury.
Assistant State's Attorney Judith Mondello testified that she worked for Felony Review on September 27, 1988. On that date, she went to Area Six at 9:45 p.m. and first spoke with Detectives Thezan and O'Leary. Mondello then reviewed all the available police reports and interviewed the witnesses. She then went into the interview room where defendant was sitting. She introduced herself as a lawyer working for the police and then immediately advised defendant of his rights. Defendant indicated that he understood each of those rights and that he was willing to give a statement. Accordingly, Mondello sat with him and drafted one. She denied writing *507 the statement based on what O'Leary told her.
Mondello testified that the statement was taken on a form used by the State's Attorney's office. At the top of this form are defendant's pre-printed rights and then a space for the statement. Mondello stated that she did not use a yellow pad to take down defendant's statement. Mondello stated that defendant read aloud the rights written on the form and then she read his statement to him. Defendant's signature appears at the top of the handwritten statement, under the pre-printed rights section and also at the bottom of each page of the statement. After reading the statement to defendant, she told him that if there was anything he wanted changed he should tell her.
Mondello testified that defendant never indicated that he was beaten or was denied the use of bathroom facilities. She never left the room between the time she took the statement and when defendant signed it. She did not notice any bruises or discoloring of defendant's face. Defendant told her that he received food while at the station and that he indeed shot the victim. At the close of the evidence, the court denied defendant's motion.
The trial began on March 14, 1989. Duane Sammons, nicknamed Cratches, testified that on September 24, 1987, cocaine was being sold from apartment #214 at 6241 North Kenmore. Sammons stated that the victim had lived in that apartment in the past. He identified defendant and stated that defendant was living there on September 24, 1987. Around 7 p.m. that evening, Sammons was at his girl friend's house on Kenmore when Earl-16 came over. Earl-16 told Sammons to go down to Birmingham's apartment to get some beer. When Sammons got to the apartment, defendant answered the door pointing a chrome gun at him. Sammons told defendant to put the gun down and defendant put it to his side. Sammons told defendant that Earl-16 sent him for some beers. Peter, who was also in the apartment, gave Sammons some beers. Defendant was clutching the gun the whole time. Sammons denied ever telling defendant that Earl-16 shot the victim and testified that he did not learn of the shooting until the next day.
Following the testimony of several occurrence witnesses, Mondello read defendant's statement to the jury. In his statement, defendant said that he was at apartment # 214 on September 24, 1987, when Earl-16 came over that evening. Defendant agreed to buy beer for him and he accepted money from Earl-16 and Peter. Just before defendant left, the victim arrived at the apartment and followed defendant into the hallway. The victim started to insult defendant. Defendant went back into the apartment and got a gun from Peter. Defendant went into the hallway and pointed the gun at the victim; the victim ran. Defendant chased him into the stairwell and up the stairs. Defendant grabbed the back of the victim's shirt and he fell. Defendant pointed the gun at his head and shot him once. Defendant took the gun back to the apartment and went to buy beer. Defendant returned with the beer. Earl-16 was gone; Cratches came by later and picked up some beer. Defendant and Peter then went to another apartment in the building where they waited until the police left. Defendant and Peter then went to stay at a friend's house. Defendant said the victim never had a gun. Defendant also stated he was well treated by the police and was not threatened or made any promises. He stated he was given food, cigarettes and drinks and was allowed to use the bathroom. Mondello testified that defendant was not handcuffed and never complained of any injuries.
Defendant testified at trial that on September 24, 1987, he was on the south side of Chicago. Garfield Birmingham offered him a ride home but took defendant to the north side of Chicago instead. They arrived at apartment # 214 at around 2 p.m. Defendant sat around the apartment until Birmingham returned. Later that evening, defendant went to the store to purchase beer. When he returned, only Peter was present. Defendant left the apartment around 8 p.m. and went home. Defendant denied shooting anyone.
*508 Defendant also testified about his treatment by the police. While his testimony is similar to his previous testimony, several inconsistencies can be noted. At trial, defendant stated that Thezan told Mondello what to write down but at the hearing he said that O'Leary told her what to write. Second, defendant stated at the hearing he was struck in the right eye while at trial he said he was struck in the left. Third, at trial defendant stated that O'Leary was the one who beat him yet at the hearing stated that it was Kajari who beat him.
In rebuttal, Carlos Martinez and Dr. Aaron Hamb testified as they did before. Officers Kajari, O'Leary and Thezan testified about defendant's treatment and that there were no garbage cans in the interview room.
Following closing arguments, defendant was found guilty of murder and sentenced to 25 years imprisonment. Defendant brings this appeal.
Defendant first contends that his admissions to the police should have been suppressed because they were the fruit of an illegal arrest. Defendant asserts that the existence of an arrest is demonstrated by the length of his stay and by the facts that the police placed him in an interrogation room, left him there for long durations, questioned defendant on numerous occasions, and never told him that he was not under arrest or free to leave at any time. Defendant notes that no probable cause existed to support his detention. The State responds that the length of defendant's stay and the other circumstances to which defendant cites reflect that he was an informant cooperating with the police in a potentially dangerous murder investigation.
When determining whether an arrest has been made, courts focus on the intent of the police officers and the understanding of the individual being questioned at the time he is detained. (People v. Wipfler (1977), 68 Ill.2d 158, 11 Ill.Dec. 262, 368 N.E.2d 870; People v. Johnson (1989), 187 Ill.App.3d 756, 135 Ill.Dec. 896, 544 N.E.2d 392.) In assessing the officer's intent, the court will consider whether there was any formal declaration of arrest and whether other routine procedures associated with an arrest were present, such as searching, booking, handcuffing, fingerprinting and photographing. (Johnson, 187 Ill.App.3d at 769, 135 Ill.Dec. at 904, 544 N.E.2d at 400.) In assessing the understanding of the individual being questioned, the court is not concerned with what the particular individual thought; rather, the court is concerned with whether a reasonable person, innocent of any crime, would have concluded he was not free to leave in similar circumstances. (Johnson, 187 Ill.App.3d at 769, 135 Ill.Dec. at 904, 544 N.E.2d at 400.) A reviewing court will not overturn a trial court's denial of a motion to quash arrest unless it is manifestly erroneous. Johnson, 187 Ill.App.3d at 770, 135 Ill.Dec. at 904, 544 N.E.2d at 400.
In this case, we believe the circuit court's denial of defendant's motion to suppress was not manifestly erroneous. Defendant's entire case rests on his testimony that he was kept against his will. This testimony was contradicted by all the State's witnesses. More importantly, a review of defendant's testimony reveals that he was not the most credible witness. During cross-examination, defendant's testimony became convoluted and conflicting. At the end of cross-examination at the suppression hearing and at trial, no re-direct was attempted. Defendant's lack of credibility must have weighed heavily against him with the judge and jury.
Aside from defendant's lack of credibility, evidence exists to support the State's assertion on appeal that the police treated defendant as an informant who was cooperating with the police in a potentially dangerous investigation and who remained at police headquarters for those 48 hours for his protection. The evidence further supports that defendant believed himself to be acting in such a capacity.
Garfield Birmingham, who was cooperating with the police, gave information which led the police to defendant's residence. There, the officers asked defendant to accompany them to Area Six headquarters. Defendant agreed. During the next 48 *509 hours, the evidence showed that defendant voluntarily assisted the police. He was never handcuffed, searched, booked, finger-printed and never received his Miranda rights prior to interviews or subjected to other procedures indicative of arrest.
Defendant, on numerous occasions, gave police information which they attempted to verify. This information included key investigatory facts such as the location of the murder weapon and the whereabouts of individuals potentially involved. On one occasion, defendant accompanied the officers to the residence of Earl-16. Earl-16 did in fact live in that building but was not home at the time. Defendant's own testimony reflects that he went upstairs with the two detectives and waited to the side with one of them while the other detective knocked on the door. On the return trip to headquarters, defendant testified he was allowed to purchase food at a gas station outside the presence of the police. Defendant testified at the hearing that he did not run away from the police because he was trying to "assist" them. We find that the above evidence is more reflective of a defendant cooperating with the police than one who is the subject of a murder investigation.
Defendant in his brief puts great emphasis on his 48-hour stay at police headquarters. Defendant cites numerous cases where lengthy stays and continual questioning have been found improper in the absence of probable cause. (See Dunaway v. New York (1979), 442 U.S. 200, 99 S.Ct. 2248, 60 L.Ed.2d 824; People v. Townes (1982), 91 Ill.2d 32, 61 Ill.Dec. 614, 435 N.E.2d 103, cert. denied (1982), 459 U.S. 878, 103 S.Ct. 174, 74 L.Ed.2d 143; People v. Travis (1984), 122 Ill.App.3d 671, 78 Ill. Dec. 535, 462 N.E.2d 654; People v. Fitzpatrick (1982), 107 Ill.App.3d 876, 63 Ill. Dec. 484, 438 N.E.2d 222; People v. Sturdivant (1981), 99 Ill.App.3d 370, 54 Ill.Dec. 829, 425 N.E.2d 1046.) These cases are distinguishable; unlike the present case, no testimony existed in those cases to reflect a defendant who was cooperating with the police in an ongoing and dangerous murder investigation.
Based on the evidence presented, the circuit court was entitled to find that over the course of 48 hours defendant provided information to the police which led to the discovery of other individuals. The detectives testified that defendant did not want these other individuals to know of his cooperation. The police, therefore, kept him secluded from these persons while the investigation continued. Only when these individuals gave versions of the night's events which conflicted with defendant's did defendant come under police scrutiny as a suspect. Defendant was then read his rights, elected to waive those rights, and thereafter made the confession in question.
For these reasons, we believe it was not manifestly erroneous for the circuit court to hold that defendant was not subjected to an illegal arrest.
Defendant next contends that his statements should have been suppressed on the grounds that they were involuntary. The State responds that the evidence demonstrates that defendant's statements were voluntary.
The State bears the burden of establishing by a preponderance of the evidence that a confessions is voluntary. (People v. Caballero (1984), 102 Ill.2d 23, 33, 79 Ill.Dec. 625, 629, 464 N.E.2d 223, 227, cert. denied (1984), 469 U.S. 963, 105 S.Ct. 362, 83 L.Ed.2d 298; People v. Mackey (1990), 207 Ill.App.3d 839, 859, 152 Ill. Dec. 762, 775, 566 N.E.2d 449, 462.) In making this determination, courts utilize the test of whether, considering the totality of the circumstances, the confession is found to have been made freely and voluntarily and without compulsion (Mackey, 207 Ill.App.3d at 859-60, 152 Ill.Dec. at 775, 566 N.E.2d at 462, citing People v. Prude (1977), 66 Ill.2d 470, 6 Ill.Dec. 689, 363 N.E.2d 371, cert denied (1977), 434 U.S. 930, 98 S.Ct. 418, 54 L.Ed.2d 291), or whether defendant's will was overborne at the time he confessed so as not to be the product of rational intellect and free will. (Mackey, 207 Ill.App.3d at 860, 152 Ill.Dec. at 775, 566 N.E.2d at 462, citing Townsend v. Sain (1963), 372 U.S. 293, 83 S.Ct. 745, 9 L.Ed.2d 770; People v. O'Leary (1970), 45 *510 Ill.2d 122, 257 N.E.2d 112.) A reviewing court will not disturb the trial court's determination in this regard unless it is against the manifest weight of the evidence (Caballero, 102 Ill.2d at 33, 79 Ill.Dec. at 629-30, 464 N.E.2d at 227-28; People v. Payton (1984), 122 Ill.App.3d 1030, 1033, 78 Ill.Dec. 424, 426, 462 N.E.2d 543, 545), and the court may consider the evidence heard by the trial court at the suppression hearing and at trial when evaluating the voluntariness of the confession. Caballero, 102 Ill.2d at 35-36, 79 Ill.Dec. at 631, 464 N.E.2d at 229.
In this case, the evidence demonstrates that defendant's statements were voluntary. At a minimum, under the applicable standard of review, the circuit court's determination is not against the manifest weight of the evidence.
According to the police, defendant was not given food the night he was brought in because it was already 8 p.m. and defendant made no request for food. The next day, after visiting Earl-16's residence, defendant testified that, at around noon or 1 p.m., he was allowed to purchase a candy bar and a 7-Up. Later that evening, he was given a Big-Mac and a 7-Up. The next day, September 27, defendant was again given food. Consistent with this evidence is defendant's own statement, read at trial, that he was provided food.
As for defendant's allegation of physical abuse and denial of washroom facilities, defendant's testimony is contradicted by that given by the various police officers, Assistant State's Attorney Mondello and Carlos Martinez. Defendant's own statement alleged that he was never beaten and was indeed provided washroom facilities. No evidence exits that defendant was denied the opportunity to sleep. Moreover, the physical injuries to which Carlos Martinez and Dr. Aaron Hamb testified do not necessarily support defendant's claimed version of his treatment. Martinez specifically asked defendant if he had any physical complaints or problems; defendant responded negatively. Martinez also failed to observe any physical signs of mistreatment which would support defendant's claim of frequent and harsh beatings.
The question of the voluntariness of defendant's statements became one of balancing the credibility of the various witnesses. Defendant's testimony was contradicted by numerous State witnesses and lacked clear, objective support. Moreover, defendant's credibility was called into issue in light of the numerous inconsistencies between his trial testimony and that given at the hearing. Given this court's inability to substitute its judgment for that of the circuit court, and the manifest weight of the evidence supporting the court's finding of voluntariness, we reject defendant's claim that his admissions were involuntary.
Turning now to defendant's final contention, defendant claims that he was denied a fair trial when the State argued in closing argument that Duane Sammons, or Cratches, consistently told the same story to the police and, therefore, was a more credible witness.
We dispose of this contention on the grounds of waiver. Defendant neither objected to the State's comments during closing argument nor specifically mentioned this error in his post-trial motion. The alleged error is, therefore, waived. (People v. Enoch (1988), 122 Ill.2d 176, 119 Ill.Dec. 265, 522 N.E.2d 1124, cert. denied (1988), 488 U.S. 917, 109 S.Ct. 274, 102 L.Ed.2d 263.) Moreover, the plain error exception does not require this court to consider the issue. The evidence in this case is not closely balanced, and the alleged error is not of such magnitude as to deprive defendant a fair trial. For these reasons, we deem defendant's argument waived.
For the foregoing reasons, the judgment of the circuit court of Cook County is affirmed.
AFFIRMED.
CAMPBELL and O'CONNOR, JJ., concur.
|
Q:
CSS grayscale filter and background-blend-mode at same time?
I'm trying to treat an image the same way it is in a photoshop file - desaturating the image to grayscale, and then applying a color overlay with a multiply blend mode. To this end, I am styling a CSS background image with...
.someclass
{
/* grayscale */
-webkit-filter: grayscale(1);
filter: gray;
filter: grayscale(1);
filter: url(desaturate.svg#greyscale);
/* multiply */
background-color: rgba(190, 50, 50, 0.65);
background-blend-mode: multiply;
}
The problem with this is that the grayscale filter ends up desaturating red color for the blend mode. In other words, the multiply is occurring first and then the grayscale. How do I switch it so that the grayscale is applied first and then the blend mode is applied second?
I have created a fiddle (http://jsfiddle.net/g54LcoL1/1/) for the code and a screenshot (made in Photoshop) of what I would expect the the fiddle result to look like. The bottom most image, div.grayscale.multiply, should be colored red.
A:
You can not do it with filter, but you can do it staying with blend mode for everything
the grayscale equivalent in blend is luminosity, with the image as source and a solid white as backdrop
So the background images, from bottom to top, are:
white (as background-color)
your image
solid red (that must be specified as a gradient for now)
and the blend modes are luminosity and multiply
.test {
width: 400px;
height: 400px;
background-image: linear-gradient(0deg, red, red), url("http://cuteoverload.files.wordpress.com/kitteh_puddle-1312.jpg");
background-color: white;
background-blend-mode: multiply, luminosity;
background-size: cover;
}
<div class="test"></div>
|
For the menu above to work, you need to enable JavaScript on your browser. See the "Help" section via the menus below...
Browse
Search
Buy
Customers
Get Help
Catalogue
Publishing
The Princess and the Pea
by Peter Bond
Read the complete script on line. All the scripts on this site are copyrighted and may not be printed, quoted or performed without the permission of Lazy Bee Scripts.
A full length pantomime adaptation of the Hans Christian Andersen fairytale. Princess Zabaglioni went missing as a baby, but Fairy Fettucine is sure she still lives as a peasant and will one day become a Princess again - just as Prince Tiramisu of Trattoria is looking for a bride as it happens. Meanwhile, the evil wizard Vermicelli persuades the Queen to outlaw chocolate.Cast: 5F, 5M, 9 Either. No chorus - though opportunities for extras in ballroom scene. Estimated run time: 1 hour 29 minutes. |
Colorectal cancer is the third most frequent type of cancer in the world having an occurrence of about 1 million new cases every year. The incidents of cancer are considerably more frequent in the industrial part of the world.
Current techniques for mechanically performing anastomosis of hollow organs use circular mechanical staplers, which execute the connection of the tissue edges of the dissected hollow organ by metallic or plastic staples. A wide variety of surgical staplers have been developed for gastric, oesophageal and intestinal surgery. In performing surgical anastomotic stapling, generally two pieces of the hollow organ are joined by a ring of staples with a closed loopstapler. End to end anastomoses are generally performed by intraluminal surgical staplers that deliver a pair of staggered rings of staples. During this process, a circular knife blade is used to separate the tissue that is held within the circular ring. The separated tissue is then removed with the stapler to form a circular opening within the lumen along the stapling line.
A major issue regarding anastomosis healing is the blood circulation of the anastomosis during the healing process. Despite substantial development of surgical techniques during the last decades, morbidity and mortality after resections in the gastrointestinal tract, e.g. due to anastomotic leakage, remain as serious problems. Ischemia and inflammation, which are natural parts of the healing process, may cause leakage and secondary infection that may be fatal for the patient in the stapling area. Therefore, it has become common practice to relieve the pressure from the anastomosis by performing a deviating stoma, especially when the anastomosis is carried out in the lower part of colon and in rectum. By relieving pressure and faecal stream from the anastomosis during the healing process, the leakage incident may be reduced and fatal consequences of anastomotic dehiscence can be avoided. The inconvenience for the patient is obvious, since the patient must have a temporary stoma for a time period of about 3-6 months, and then has to undergo a second surgery in order to close the stoma. Unfortunately in many cases, the closure of the stoma cannot be reversed and the patient is forced to live with a permanent stoma leading to lower quality of life associated with increased costs.
Another problem arising from stapling of anastomoses is anastomotic stenosis. The critical area for healing is the contact area between the two ends of the hollow structure to be connected. The connection has to be liquid proof, and the cross section of the lumen should be as wide and flexible as the original lumen. The size of the stapler determines the size of the lumen and thus the contact area between the ends. Surgical staplers create a smaller and more rigid opening compared to the cross section of the original lumen due to the staples inside the hollow structure connecting the two ends thereof, i.e. a collar may be formed that may lead to stenosis. For solving this problem repeated need for dilatation is required.
Another disadvantage associated with mechanical staplers, is that there is no fast, simple and reliable method to control anastomotic insufficiency, which at late discovery can result in abdominal sepsis.
Furthermore, staplers require an incision in the intestine in order to insert the instrument into the bowel lumen. This additional incision increases the duration of the operation and the risks associated with surgery, e.g. secondary infections and anastomotic leakage.
The stapler itself is a critical link, since there are several severe problems connected with the use of mechanical staplers in surgical anastomotic stapling, such as anastomotic leakage and anastomotic stenosis. Other disadvantages are high consumption of time and expensive instruments for the performance.
U.S. Pat. No. 5,697,943 discloses a surgical instrument for carrying and attaching separate components of an anastomosis device to the end of tissue of a tubular hollow organ. The instrument includes an elongated housing having a proximal end and a distal end, first and second compression anastomosis device components, a supporting member operatively associated with the distal end of the housing, and an assembling structure associated with the supporting member and operable from the proximal end of the housing for assembling the first and second compression anastomosis device components within tubular body tissue. |
Since the advent of the wheel, man has applied that structure in an incalculable variety of manner and use. One such use was the development of the inflatable tire, which tire could be used with great success in association with the automobile, trucks, tractors and great earth-moving equipment. Such tires range in size from the smallest to those weighing over four hundred pounds. Since such tires currently exist as interchangeable commodities, and since the tires may not be expected to last indefinitely, a problem arises concerning the movement of the tires from one place to another, in their own right. At times, this problem is preceded by the concern of even getting the tire off the ground. This problem becomes particularly acute with respect to farmers and the like who may own hoisting equipment but cannot afford a large capital outlay in order to possess a device whose sole use is to lift large tractor tires. It has also been observed that when a tire from a tractor or earth moving equipment needs replacement or repair in the field, many tire repair companies lack a device which readily engages a tire and in combination with a crane readily moves it and positions it. |
"I am a fashion designer refusing this association," Nicolas Ghesquière said on Instagram after the president attended a ribbon-cutting for a new Louis Vuitton workshop
President Donald Trump recently lent his support to Louis Vuitton, but an artistic director for the luxury brand isn’t reciprocating the gesture.
Days after Trump and his daughter Ivanka Trump attended a ribbon-cutting ceremony for the opening of a new Louis Vuitton workshop in Texas, Nicolas Ghesquière slammed the president on social media, specifically pointing to Trump’s lack of support for LGBTQ rights.
Get push notifications with news, features and more.
“Standing against any political action. I am a fashion designer refusing this association,” Ghesquière, the artistic director of women’s collections at the famed fashion house, wrote on Instagram.
“Trump is a joke,” he added in a hashtag alongside another that simply read “homophobia.”
The LVMH group, which owns the brand, has not publicly commented on Ghesquière’s stance. But he has received support from colleague Camille Miceli, the creative director for accessories, who responded with a trio of applause and heart emojis.
Out Magazine Editor-in-Chief Phillip Picardi added, “You better say it!!! Proud of you,” while model Teddy Quinlivan, who is transgender, wrote: “BRAVO👏🏻 Thank you for standing on the right side of history 🙌🏼🙌🏼❤️❤️.”
Last Thursday, Bernard Arnault, the founder and chairman of LVMH, joined the Trumps at the ribbon-cutting ceremony.
“We are very honored to have the president of the United States,” Arnault said during the event, according to the The New York Times. “I am not here to judge his types of policies.”
“I have no political role. I am a business person. I try to tell him what I think for the success of the economy of the country, and the success of what we are doing” Arnault added.
In an earlier interview, Michael Burke, chief executive and chairman of Louis Vuitton, commented on the fact that the president also hosted a campaign rally and a fundraiser on the day of his visit, saying, “We have nothing to do with the timing,” the Times reported.
Image zoom From left: Alexandre Arnault, Michael Burke, Bernard Arnault, Donald Trump and Ivanka Trump NICHOLAS KAMM/AFP via Getty
At the event, Trump, who also mispronounced the brand’s name, spoke about how he and Arnault first met after his election “to discuss my vision for creating an American manufacturing renaissance.”
“Bernard had faith in that vision — the vision of doing something right here in this incredible state — and promised to bring a brand-new factory to the United States. And today, wow, has he delivered. He’s really delivered. This is some place,” said the president, noting that the workshop will employ 500 people.
During his speech, Trump also took a moment to lavish praise on daughter Ivanka and his son-in-law Jared Kushner, who was also present. |
Q:
How to fix inconsistent labelling of edges/nodes in Python's networkx?
I want to define an objective function for the max-cut problem on graphs. I use the following expression, 0.5*sum([w[i,j]*(1-spin[i]*spin[j]) for i,j in G.edges]) where G is a networkx graph, w is the numpy matrix generated from that graph (connectivity matrix), and spin is an array with entries -1 or 1 to denote which side of the partition the node is in.
All good I thought, but it turns out that the labelling of the edges is not consistent with the labelling of the nodes, see code below. For example in the graph G1, edge weight w[1,5] is 0 even though the edge (1,5) is in the graph. Any recommendations on how to fix this?
Cheers
import networkx as nx
G1 = nx.random_regular_graph(3,6, seed = 1)
G2 = nx.random_regular_graph(3,6, seed = 1)
# labelling seems not to be conserved when transforming to matrix and back
G2 = nx.to_numpy_matrix(G2)
G2 = nx.from_numpy_matrix(G2)
print(nx.to_numpy_matrix(G1))
print(G1.edges)
print(nx.to_numpy_matrix(G2))
print(G2.edges)
Output
[[0. 1. 0. 1. 0. 1.]
[1. 0. 1. 0. 1. 0.]
[0. 1. 0. 1. 0. 1.]
[1. 0. 1. 0. 1. 0.]
[0. 1. 0. 1. 0. 1.]
[1. 0. 1. 0. 1. 0.]] # matrix of G1
[(0, 1), (0, 4), (0, 3), (1, 2), (1, 5), (2, 3), (2, 4), (4, 5), (5, 3)] # edges of G1
[[0. 1. 0. 1. 0. 1.]
[1. 0. 1. 0. 1. 0.]
[0. 1. 0. 1. 0. 1.]
[1. 0. 1. 0. 1. 0.]
[0. 1. 0. 1. 0. 1.]
[1. 0. 1. 0. 1. 0.]] # matrix of G2
[(0, 1), (0, 3), (0, 5), (1, 2), (1, 4), (2, 3), (2, 5), (3, 4), (4, 5)] # edges of G2
A:
As described in to_numpy_matrix documentation, the method implicitly uses the ordering of G.nodes(), if the parameter nodelist is not used.
Using the following code should fix the ordering
nx.to_numpy_matrix(G, nodelist=sorted(G))
|
The Coast Guard (film)
The Coast Guard () is a 2002 South Korean film directed by Kim Ki-duk. The film deals with military atrocities and the absurdities of borders and conflicts.
Plot
Private Kang is a member of the South Korean marine corps who is eager to shoot a North Korean spy during his time guarding the South Korean coastline near the Korean Demilitarized Zone. One evening he shoots and kills a South Korean civilian who has strayed into a forbidden zone to have sex with his girlfriend. Kang and the girlfriend of the dead civilian both have mental breakdowns. The woman believes the members of the coast guard are her dead lover, and engages in sexual affairs with them. Though commended after the shooting, Kang is dismissed from service. He then returns to kill other members of his unit. He then goes to the South Korean capital city of Seoul, where kills people at random with his bayonet before being confronted by armed policemen. Gunfire then erupts.
Cast
Jang Dong-gun as Private Kang
Kim Jung-hak as Private Kim
Park Ji-ah as Mi-young
Yoo Hae-jin as Cheol-gu
Jeong Jin as So Cho-jang
Kim Ku-taek as Sergeant Jang
Kim Kang-woo as Private Jo
Park Yoon-jae as Private Yoon
Kim Tae-woo as Private Seo
Kim Young-jae as medic
Kim Mi-sung as tourist woman #1
In-seong as agent 2
No Jun-ho as man stabbed with bayonet
Won Deok-hyun as kid
Shim Hoon-ki as coast guard 2
Jung Hee-tae as agent 7
Jeon Sung-ae as Young-kil's mother
Park Sung-il as Private Kang's friend
Choi Min as provost
Awards
Wins
38th Karlovy Vary International Film Festival 2003:
FIPRESCI International Critics Award: "For the strong and innovative depiction of the illusion of power which destroys humanity on both sides of the fence."
NETPAC Award
Town of Karlovy Vary Award
Nominations
38th Karlovy Vary International Film Festival 2003:
Crystal Globe
Notes
External links
Category:2002 films
Category:2000s action drama films
Category:Films directed by Kim Ki-duk
Category:South Korean independent films
Category:Korean-language films
Category:South Korean films
Category:South Korean war drama films
Category:South Korean action drama films |
By resting halfback Cooper Cronk, Australia coach Mal Meninga has opened the door for Matt Moylan to be handed the bench utility role in this Saturday's Four Nations clash with New Zealand.
The World Cup holders on Thursday named only an 19-man squad for the Ricoh Arena, Coventry, clash with the five players not required also including winger Josh Mansour who has suffered a nine-month knee injury. The side was listed in alphabetical order with positions not specified.
Sidelined: Cooper. Credit:Mark Kolbe
But the absence of Cronk, who scored two tries against Scotland last week, would allow Michael Morgan to start alongside his North Queensland club-mate, Johnathan Thurston in the halves.
That, in turn, could leave last week's man of the match – Moylan – on the bench where he would effectively trial for Morgan's regular spot ahead of the clash with England on Sunday week. |
The present invention relates to a device for coupling multiple connectors, of the type that comprises two connector supporting plates, particularly for pressure die-casting dies.
Devices for coupling the supporting plates of multiple connectors are already commercially available.
The plates, one fixed and one movable, are connected to threaded members, and the threaded member that is connected to the movable plate is formed by a rod which is provided with a threaded section and is supported for rotation by a movable ring integrated in the corresponding plate.
The movable ring is supported by a fixed frame by means of sliding guides that ensure the assembly thereof with a framework which forms the support for a motor which is designed to turn the rod.
The rotation rod is connected to the output shaft of the motor by means is of an intermediate shaft provided with a threaded section which cooperates with an internal thread of the frame in order to provide the longitudinal movement of the rod before the threaded section thereof cooperates with the other threaded member.
The rotation rod is connected in an angular manner to the intermediate shaft by means of a sliding joint which allows the rod to interrupt its advancement while it continues to be rotationally actuated.
Elastic elements cooperate with the guides, allowing to block the advancement of the movable ring, but not the rotation of the rod if the two threaded members do not couple as soon as they make contact.
In any case, this kind of device is not free from drawbacks.
The coupling between the threaded members in fact occurs in a rough manner, except in unusual cases, since the shoulders of the threads make contact first and collide with each other; at this point the geometry of the device allows to stop the translatory motion of the plates until the two threads mate.
Accordingly, the movements of the plates are not gradual and the geometry is particularly complicated.
The aim of the present invention is to provide a device for coupling multiple connectors which solves the problems of conventional devices.
Within the scope of this aim, a consequent object of the present invention is to provide a device which allows the coupling of connectors gradually and not discontinuously.
Another object of the present invention is to provide a device which does not have a complicated geometry.
Another object of the present invention is to provide a device which allows to perform the approach and centering movement with limited force.
This aim and these and other objects which will become better apparent hereinafter are achieved by a device for coupling multiple connectors, comprising two connector supporting plates, one fixed and one movable, two threaded members which are supported by the two plates, one of said members having a combined rotary and translatory motion for the coupling and uncoupling of said plates, characterized in that the member provided with combined rotary and translatory motion is associated with a fluidactuated actuator for the approach and spacing of said threaded members. |
Aberystwyth League
The Aberystwyth & District Football League (currently the Cynghrair Cambrian Tyres Aberystwyth League) is a football league in Mid Wales, sitting at the fifth and sixth levels of the Welsh football league system.
The league hosts several cup competitions: J. Emrys Morgan Cup, Dai 'Dynamo' Davies Cup, League Cup, Len & Julie Newman Memorial Trophy, Second Division Trophy and Consolation Cup.
Teams promoted from Division One may enter the Mid Wales League if standards and facilities fall into line with the regulations of the Mid Wales League.
Member clubs for 2019–20 season
Division One
Aberdyfi
Bont
Bow Street reserves
Llanilar
Padarn United
Penrhyncoch reserves
Talybont
Tregaron Turfs
U.W.A. reserves
Division Two
Borth United reserves (withdrew from league January 2020)
Corris United
Dolgellau Athletic reserves
Llanilar reserves
Padarn United reserves
Penparcau reserves
U.W.A. thirds
League champions - Top division
Information sourced from the Welsh Soccer Archive unless referenced.
1930s
1934–35: Aberayron
1935–36: Trefechain
1940s
1946–47: Season competition declared void
1947–48: Pontrhydfendigaid & District (Bont )
1948–49: Aberystwyth Rovers
1949-50: Bont
1950s
1950–51: Aberayron
1951–52: Trefechain
1952–53: Trefechain
1953–54: Aberystwyth Rovers
1954–55: Aberayron
1955–56: Tregaron Turfs
1956–57: University College of Wales reserves
1957–58: YMCA
1958–59: University College of Wales reserves
1959–60: Dewi Stars
1960s
1960–61: Dewi Stars
1961–62: University College of Wales reserves
1962–63: University College of Wales reserves
1963–64: University College of Wales reserves
1964–65: Bont
1965–66: Penparcau
1966–67: Bont
1967–68: Bont
1968–69: Penparcau
1969–70: Bont
1970s
1970–71: Bont
1971–72: CPD Penrhyncoch
1972–73: Phoenix
1973–74: Llanilar
1974–75: CPD Penrhyncoch
1975–76: CPD Penrhyncoch
1976–77: CPD Penrhyncoch
1977–78: CPD Penrhyncoch
1978–79: Aber Athletics Club
1979–80: Bryncrug
1980s
1980–81: Dolgellau Athletic
1981–82: Barmouth & Dyffryn United
1982–83: Bow Street
1983–84: Aberystwyth Town reserves
1984–85: Bryncrug
1985–86: Penparcau
1986–87: Bryncrug
1987–88: Padarn United
1988–89: Penparcau
1989–90: Penparcau
1990s
1990–91: Penparcau
1991–92: Machynlleth
1992–93: Bow Street
1993–94: U. W. A.
1994–95: Bow Street
1995–96: Bow Street
1996–97: Padarn United
1997–98: CPD Penrhyncoch reserves
1998–99: CPD Penrhyncoch reserves
1999–2000: Penparcau
2000s
2000–01: League not completed - Foot & Mouth outbreak
2001–02: Bow Street
2002–03: Bow Street
2003–04: Bow Street
2004–05: Penparcau
2005–06: CPD Penrhyncoch reserves
2006–07: Bow Street
2007–08: Penparcau
2008–09: Dolgellau Athletic 'A'
2009–10: Bont
2010s
2010–11: Aberdyfi
2011–12: CPD Penrhyncoch reserves
2012–13: Tregaron Turfs
2013–14: Borth United
2014–15: Talybont
2015–16: Dolgellau Athletic
2016–17: Bont
2017–18: Bow Street reserves
2018–19: CPD Penrhyncoch thirds
2019–20:
Number of titles by winning clubs
CPD Penrhyncoch – 10 titles
Bow Street – 9 titles
Penparcau – 9 titles
Bont – 7 titles
U. W. A./ University College of Wales – 6 titles
Aberayron – 3 titles
Bryncrug – 3 titles
Dolgellau Athletic – 3 titles
Trefechain – 3 titles
Dewi Stars– 2 titles
Padarn United – 2 titles
Tregaron Turfs – 2 titles
Aber Athletics Club – 1 title
Aberdyfi – 1 title
Aberystwyth Rovers – 1 title
Aberystwyth Town reserves – 1 title
Barmouth & Dyffryn United – 1 title
Borth United – 1 title
Llanilar – 1 title
Machynlleth – 1 title
Phoenix – 1 title
Talybont – 1 title
YMCA – 1 title
See also
Football in Wales
List of football clubs in Wales
External links
League website
References
5
Wales |
#!/usr/bin/env ruby
$LOAD_PATH.unshift(File.expand_path(File.dirname(__FILE__)) + '/../lib/')
require 'gimli'
if ARGV.flags.version?
puts "Version: #{Gimli::VERSION}"
exit
end
config = Gimli.configure do |config|
config.file = ARGV.flags.file
config.recursive = ARGV.flags.recursive?
config.merge = ARGV.flags.merge?
config.debug = ARGV.flags.debug?
config.wkhtmltopdf_parameters = ARGV.flags.wkhtmltopdfparameters
config.remove_front_matter = ARGV.flags.removefrontmatter
config.output_filename = ARGV.flags.outputfilename
config.output_dir = ARGV.flags.outputdir
config.stylesheet = ARGV.flags.stylesheet
config.cover = ARGV.flags.cover
end
Gimli.process! config
|
Brucella infection in total hip replacement: case report and review of the literature.
A 47-y-old female underwent revision of a left total hip replacement because of loose prosthesis. Routine intraoperative culture of the hip site grew Brucella sp. The patient was treated with a combination of vibramycin and rifampicin for 5 months. At 4-y follow-up, her condition is good. Though prosthetic infection with Brucella spp. is an extremly rare condition (only 1 case each of femur and hip, and 3 cases of knee had been previously reported in the English literature), brucella infection of prosthetic joints should be considered in brucella endemic areas. |
Q:
Extracting Words of specific length in R using regular expressions
I have a code like (I got it here):
m<- c("Hello! #London is gr8. I really likewhatishappening here! The alcomb of Mount Everest is excellent! the aforementioned place is amazing! #Wow")
x<- gsub("\\<[a-z]\\{4,10\\}\\>","",m)
x
I tried other ways of doing it, like
m<- c("Hello! #London is gr8. I really likewhatishappening here! The alcomb of Mount Everest is excellent! the aforementioned place is amazing! #Wow")
x<- gsub("[^(\\b.{4,10}\\b)]","",m)
x
I need to remove words which are lesser than 4 or greater than 10 in length. Where am I going wrong?
A:
gsub("\\b[a-zA-Z0-9]{4,10}\\b", "", m)
"! # is gr8. I likewhatishappening ! The of is ! the aforementioned is ! #Wow"
Let's explain the regular expression terms :
\b matches at a position that is called a "word boundary". This match is zero-length.
[a-zA-Z0-9] :alphanumeric
{4,10} :{min,max}
if you want to get the negation of this so , you put it between() and you take //1
gsub("([\\b[a-zA-Z0-9]{4,10}\\b])", "//1", m)
"Hello! #London is gr8. I really likewhatishappening here! The alcomb of Mount Everest is excellent! the aforementioned place is amazing! #Wow"
It is funny to see that words with 4 letters exist in the 2 regexpr.
|
/*
* Licensed to Elasticsearch B.V. under one or more contributor
* license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright
* ownership. Elasticsearch B.V. licenses this file to you under
* the Apache License, Version 2.0 (the "License"); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
import React, {
ReactNode,
FunctionComponent,
useState,
useEffect,
} from 'react';
import { EuiBreakpointSize, getBreakpoint } from '../../services/breakpoint';
import { throttle } from '../color_picker/utils';
export type EuiShowForBreakpoints = EuiBreakpointSize;
export interface EuiShowForProps {
/**
* Required otherwise nothing ever gets returned
*/
children: ReactNode;
/**
* List of all the responsive sizes to show the children for.
* Array of #EuiBreakpointSize
*/
sizes: EuiShowForBreakpoints[] | 'all' | 'none';
}
export const EuiShowFor: FunctionComponent<EuiShowForProps> = ({
children,
sizes,
}) => {
const [currentBreakpoint, setCurrentBreakpoint] = useState(
getBreakpoint(typeof window === 'undefined' ? -Infinity : window.innerWidth)
);
const functionToCallOnWindowResize = throttle(() => {
const newBreakpoint = getBreakpoint(window.innerWidth);
if (newBreakpoint !== currentBreakpoint) {
setCurrentBreakpoint(newBreakpoint);
}
// reacts every 50ms to resize changes and always gets the final update
}, 50);
// Add window resize handlers
useEffect(() => {
window.addEventListener('resize', functionToCallOnWindowResize);
return () => {
window.removeEventListener('resize', functionToCallOnWindowResize);
};
}, [sizes, functionToCallOnWindowResize]);
if (
sizes === 'all' ||
sizes.includes(currentBreakpoint as EuiBreakpointSize)
) {
return <>{children}</>;
}
return null;
};
|
“Here, where the air is loaded with iodine and where the ultra-violet ray is ever-present in our smiling sunshine, your health and happiness is our business.”
Sun Fun in New Jersey
(1946 publication of the New Jersey Resort Association)
Tuesday, March 25, 2014
Velma, Chapter Two
for Liz French
I poked at Velma's smoldering cigarette butt in the ashtray
with the point of a pencil. Why do women always do that?
Velma crossed her shapely left leg over her right,
then changed to her right over her left. I didn't pretend not to look.
"Well?" asked Velma.
I said, "Velma, what part of 'I don't do divorce work'
don't you understand?"
Enjoyed this, Bob. Very much. I hardly know Liz French, though I chat with her for a few minutes, and get to spend a little time in her company, at each year's WFMU marathon. A little reunion that makes me content.
I'm impressed, I have to say. Actually hardly ever do I encounter a weblog that's equally educative and entertaining, and let me inform you, you may have strike the nail on the head. Your concept is outstanding the issue is one thing that not sufficient individuals are speaking intelligently about. I am very joyful that I stumbled throughout this in my find for anything referring to this.seaside heights motel |
Story highlights North Korea seems "not as stable as we thought," one analyst says
Some observers warn a provocative move like a missile or nuclear test could follow
Jang Song Thaek was married to Kim's aunt, was vice chairman of the top military body
North Korea state media say Jang was convicted and executed
As the shock sinks in of North Korea's extraordinary announcement of the execution of leader Kim Jong Un's uncle and former protector, government officials and analysts are trying to decipher what the brutal move means.
The ruthless disposal of Jang Song Thaek -- Kim's uncle by marriage who had, until recently, been regarded as the second-most powerful figure in the secretive, nuclear-armed nation -- has serious implications for North Korea, its neighbors and the United States, observers said.
But exactly what is going on inside the notoriously opaque North Korea regime remains as murky as ever.
"We don't have a clear sense of this at all," said Victor Cha, a senior adviser at the Center for Strategic and International Studies who represented the United States in nuclear talks with North Korea.
Some saw the execution, which North Korean state media reported early Friday, as a chilling demonstration of total control by Kim, the young leader who came to power two years ago.
"I think what he's telling people -- the United States, South Korea, China, others -- is that he is his own man, that you are going to have to deal with him," said Philip Yun, executive director of the Ploughshares Fund, a nuclear nonproliferation group.
JUST WATCHED Uncle of North Korea leader executed Replay More Videos ... MUST WATCH Uncle of North Korea leader executed 01:29
JUST WATCHED N. Korean media: Kim's uncle 'worse than a dog' Replay More Videos ... MUST WATCH N. Korean media: Kim's uncle 'worse than a dog' 02:53
JUST WATCHED Chang: Regime like Richard III with nukes Replay More Videos ... MUST WATCH Chang: Regime like Richard III with nukes 02:35
Who is Jang?
Jang, who was married to Kim's aunt, was vice chairman of North Korea's top military body and had often been pictured beside the young leader, who is believed to be around 30. He was considered to be the regent who secured Kim's assumption of power after the 2011 death of his father, Kim Jong Il.
But in a lengthy article foaming with outraged rhetoric, North Korea's official news agency on Friday accused Jang of trying to overthrow the state, describing him as "despicable human scum."
One big question is whether Kim acted out of strength, consolidating the power he has amassed over the past two years, or out of fear his uncle was building a rival force inside the regime.
Kim already removed the country's top general last year, Cha noted. By taking down Jang, he's axed a powerful figure from the country's dominant Workers' Party.
"It makes you wonder: If he's consolidating his power, what is he building it around?" Cha said. "He's basically attacking the two most important institutions in North Korea, which is the party and the military."
A U.S. official said, "Executing someone with Jang's pedigree would be a dramatic statement that Kim Jong Un intends to be ruthless in consolidating his control.
"The public airing of the power play under way -- which is highly unusual -- is probably sending shockwaves through North Korea's leadership cadre."
Provocative moves
Few analysts interpreted the execution, which took place days after the North had said Jang had been dramatically removed from his government posts, as a healthy sign.
"If two weeks ago, we thought that North Korea was somewhat stable, I think today people feel that it's not as stable as we thought it was," said Cha, author of "The Impossible State: North Korea, Past and Future."
Suh Sang-ki, a lawmaker in South Korea's governing Saenuri Party who sits on a parliamentary intelligence committee, said the decision to kill Jang suggests Kim's power is weaker than that of his father.
Photos: Photos: A glimpse inside North Korea Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – North Korean citizens bow before the portraits of the founding father Kim Il-Sung, left, and his son Kim Jong-Il, in Pyongyang, North Korea on Monday, April 9, 2012. April 15 marked the 100-year anniversary of the founder's birth and journalists were allowed inside the country. Hide Caption 1 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – North Korean technicians check the Unha-3 rocket at Tangachai-ri space center on Sunday, April 8. Hide Caption 2 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – A controller is seen from the window of a train along the railway on the west coast Sunday. A controversial missile launch is expected to take place in the coming days. Pyongyang insists it has no bad intentions and invited foreign journalists to view its launch site. Hide Caption 3 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – Citizens dance on Monday during a rehearsal for the commemoration of Kim Il-Sung's 100th birthday anniversary. Japan, the United States and South Korea see the launch -- which would violate U.N. Security Council resolutions -- as a cover for a long-range ballistic missile test. And a South Korean intelligence report says it's likely to precede a nuclear test, as it did in 2006 and 2009. Hide Caption 4 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – North Korean soldiers are seen from the window of a train along the railway heading from Pyongyang to the North Pyongan Province on the west coast. Hide Caption 5 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – A woman plays the piano and entertains in a downtown Pyongyang restaurant. U.S. President Obama said the real consequence for North Korea, should it go through with the launch, is that the country's leaders will miss an opportunity. "I hope that at some point the North Koreans make the decision that it is in their interests to figure out how to feed their people and improve their economy rather than have big parades where they show off weapons," he said in March. Hide Caption 6 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – Two women on the the train prepare food for the journalists traveling across the country. Hide Caption 7 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – The dress rehearsal for the celebration continues in the capital. Hide Caption 8 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – A pin with the face of Kim Il-Sung is affixed to the uniform of a North Korean soldier standing guard at the space center in Pyongyang on Wednesday, April 11. Hide Caption 9 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – A woman ties the branches of apple trees on a farm near Pyongyang on Tuesday, April 10. Hide Caption 10 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – Workers and farms are seen through the window of a train as it passes through the country. Hide Caption 11 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – Bicycles line the road as citizens work the land between Pyongyang and the North Phyongan province. Hide Caption 12 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – A North Korean soldier is seen from the window of a train as he walks near a small town along the railway heading from Pyongyang to the North Pyongan Province on the west coast. Hide Caption 13 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – Employees work in a textile factory in Pyongyang. Hide Caption 14 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – People line the street as they wait for a bus. Hide Caption 15 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – A dance troupe performs during the opening ceremony of the Spring Arts Festival in Pyongyang. Hide Caption 16 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – A band performs during the opening of the Spring Arts Festival. Hide Caption 17 of 18 Photos: Photos: A glimpse inside North Korea A glimpse inside North Korea – A building adorned with a huge portrait of the late president Kim Il-Sung is cleaned by workers in Pyongyang. Hide Caption 18 of 18
Photos: The photos N. Korea banned Photos: The photos N. Korea banned Photos North Korea didn't want you to see – A stern looking North Korean guard by the Chinese border customs office. This image was deleted by North Korean officials. Hide Caption 1 of 17 Photos: The photos N. Korea banned Deleting the offensive photos – Writer Johan Nylander and his guide, Ko Chang Ho, watch as a North Korean guard deletes 90 photos deemed unacceptable. Nylander was able to recover the photos with the help of an IT specialist -- the images that follow are an edited selection. Hide Caption 2 of 17 Photos: The photos N. Korea banned Hello, Dear Leader – This propaganda monument of "Dear Leader" Kim Jong-Il by a countryside road, not far from the border to China, was deleted by authorities. North Korea required images of leaders be full body shots. Hide Caption 3 of 17 Photos: The photos N. Korea banned Waiting for a train – People standing by the train track, while a guard is monitoring the bike race. Hide Caption 4 of 17 Photos: The photos N. Korea banned Watching the race – In the city of Rason, people are leaning out of windows to get a glimpse of the Western cyclists. Hide Caption 5 of 17 Photos: The photos N. Korea banned Pedestrian peasants – A woman and a man walking by the side of the road lined with cornfields. Hide Caption 6 of 17 Photos: The photos N. Korea banned Village life – Villagers waving by the race path. Hide Caption 7 of 17 Photos: The photos N. Korea banned Heavy security – Guards and custom officials by the border to China. Hide Caption 8 of 17 Photos: The photos N. Korea banned Secret volleyball court? – By the border checkpoint next to the Tumen River, North Korean customs officials can play volleyball. Officials prohibited any photos of North Korean military bases. Hide Caption 9 of 17 Photos: The photos N. Korea banned Photos North Korea didn't want you to see – Peasants and villagers standing by the road to look at the Western cyclists Hide Caption 10 of 17 Photos: The photos N. Korea banned Keeping watch – Guard keeping an eye on the bikers next to a small village. Hide Caption 11 of 17 Photos: The photos N. Korea banned Photos North Korea didn't want you to see – Kids playing outside village houses. Hide Caption 12 of 17 Photos: The photos N. Korea banned Waiting for the cyclists – Spectators waiting for the bikers to reach the finish line. In the background the "Great" and "Dear Leaders" Kim Il Sung and his son, Kim Jong-Il. Hide Caption 13 of 17 Photos: The photos N. Korea banned Standing on bikes to see cyclists – Huge crowds -- some of whom standing on their own bikes -- as they await cyclists by the race finish line in Rason. Hide Caption 14 of 17 Photos: The photos N. Korea banned Document check – Custom official and tourist bureau guide checking foreigners' passports. Hide Caption 15 of 17 Photos: The photos N. Korea banned Water checkpoint – Guides from the local tourist bureau handing out water bottles to bikers, monitored by a guard in the background. Hide Caption 16 of 17 Photos: The photos N. Korea banned Writer and his minder – Journalist Johan Nylander and his North Korean guide, Ko Chang Ho. EDITOR'S NOTE: This image was not among those deleted by North Korean officials. Hide Caption 17 of 17
Photos: Kim Jong Un and North Korea's military Photos: Kim Jong Un and North Korea's military North Korean leader Kim Jong Un meets with North Korea's first female fighter jet pilots in this undated photo released by the country's state media on Monday, June 22. He called the women "heroes of Korea" and "flowers of the sky." Hide Caption 1 of 55 Photos: Kim Jong Un and North Korea's military Kim stands on the snow-covered top of Mount Paektu in North Korea in a photo taken by North Korean newspaper Rodong Sinmun on April 18 and released the next day by South Korean news agency Yonhap. Kim scaled the country's highest mountain, North Korean state-run media reported, arriving at the summit to tell soldiers that the hike provides mental energy more powerful than nuclear weapons. Hide Caption 2 of 55 Photos: Kim Jong Un and North Korea's military Kim Jong Un, center, poses with soldiers on the snow-covered top of Mount Paektu in an April 18 photo released by South Korean news agency Yonhap. Hide Caption 3 of 55 Photos: Kim Jong Un and North Korea's military Kim visits the Kumsusan Palace of the Sun in Pyongyang, North Korea, on April 15 to celebrate the 103rd birth anniversary of his grandfather, North Korean founder Kim Il Sung. Hide Caption 4 of 55 Photos: Kim Jong Un and North Korea's military Kim inspects a drill for seizing an island at an undisclosed location in North Korea in an undated picture released by North Korea's official Korean Central News Agency on February 21. Hide Caption 5 of 55 Photos: Kim Jong Un and North Korea's military Kim speaks during a meeting of the Political Bureau of the Central Committee of the Workers' Party of Korea in Pyongyang, North Korea, in this photo released February 19 by the state-run Korean Central News Agency. Hide Caption 6 of 55 Photos: Kim Jong Un and North Korea's military A picture released by the North Korean Central News Agency shows North Korean leader Kim Jong Un appearing without his cane at an event with military commanders in Pyongyang on Tuesday, November 4. Kim, who recently disappeared from public view for about six weeks, had a cyst removed from his right ankle, a lawmaker told CNN. Hide Caption 7 of 55 Photos: Kim Jong Un and North Korea's military Kim is seen walking with a cane in this image released Thursday, October 30, by the state-run Korean Central News Agency. Hide Caption 8 of 55 Photos: Kim Jong Un and North Korea's military Kim sits in the pilot's seat of a fighter jet during the inspection. Hide Caption 9 of 55 Photos: Kim Jong Un and North Korea's military This undated photo, released Tuesday, October 14, by the KCNA, shows Kim inspecting a housing complex in Pyongyang, North Korea. International speculation about Kim went into overdrive after he failed to attend events on Friday, October 10, the 65th anniversary of the Workers' Party. He hadn't been seen in public since he reportedly attended a concert with his wife on September 3. Hide Caption 10 of 55 Photos: Kim Jong Un and North Korea's military A picture released by the KCNA shows Kim and his wife watching a performance by the Moranbong Band on Wednesday, September 3, in Pyongyang. Hide Caption 11 of 55 Photos: Kim Jong Un and North Korea's military Kim tours a front-line military unit in this image released Wednesday, July 16, by the KCNA. Hide Caption 12 of 55 Photos: Kim Jong Un and North Korea's military Kim poses for a photo as he oversees a tactical rocket-firing drill in June. Hide Caption 13 of 55 Photos: Kim Jong Un and North Korea's military Kim watches a tactical rocket-firing drill in June. Hide Caption 14 of 55 Photos: Kim Jong Un and North Korea's military A North Korean soldier patrols the bank of the Yalu River, which separates the North Korean town of Sinuiju from the Chinese border town of Dandong, on Saturday, April 26. Hide Caption 15 of 55 Photos: Kim Jong Un and North Korea's military In this photo released Thursday, April 24, by the Korean Central News Agency, Kim smiles with female soldiers after inspecting a rocket-launching drill at an undisclosed location. Hide Caption 16 of 55 Photos: Kim Jong Un and North Korea's military A picture released Tuesday, March 18, by the KCNA shows Kim attending a shooting practice at a military academy in Pyongyang. Hide Caption 17 of 55 Photos: Kim Jong Un and North Korea's military A North Korean soldier uses binoculars on Thursday, February 6, to look at South Korea from the border village of Panmunjom, which has separated the two Koreas since the Korean War. Hide Caption 18 of 55 Photos: Kim Jong Un and North Korea's military A North Korean soldier kicks a pole along the banks of the Yalu River on Tuesday, February 4. Hide Caption 19 of 55 Photos: Kim Jong Un and North Korea's military A photo released by the KCNA on Thursday, January 23, shows the North Korean leader inspecting an army unit during a winter drill. Hide Caption 20 of 55 Photos: Kim Jong Un and North Korea's military Kim inspects the command of an army unit in this undated photo released Sunday, January 12, by the KCNA. Hide Caption 21 of 55 Photos: Kim Jong Un and North Korea's military Kim visits an army unit in this undated photo. Hide Caption 22 of 55 Photos: Kim Jong Un and North Korea's military Kim inspects a military factory in this undated picture released by the KCNA in May 2013. Hide Caption 23 of 55 Photos: Kim Jong Un and North Korea's military Kim visits the Ministry of People's Security in 2013 as part of the country's May Day celebrations. Hide Caption 24 of 55 Photos: Kim Jong Un and North Korea's military A North Korean soldier, near Sinuiju, gestures to stop photographers from taking photos in April 2013. Hide Caption 25 of 55 Photos: Kim Jong Un and North Korea's military North Korean soldiers patrol near the Yalu River in April 2013. Hide Caption 26 of 55 Photos: Kim Jong Un and North Korea's military Kim is briefed by his generals in this undated photo. On the wall is a map titled "Plan for the strategic forces to target mainland U.S." Hide Caption 27 of 55 Photos: Kim Jong Un and North Korea's military Kim works during a briefing in this undated photo. Hide Caption 28 of 55 Photos: Kim Jong Un and North Korea's military In this KCNA photo, Kim inspects naval drills at an undisclosed location on North Korea's east coast in March 2013. Hide Caption 29 of 55 Photos: Kim Jong Un and North Korea's military Kim, with North Korean soldiers, makes his way to an observation post in March 2013. Hide Caption 30 of 55 Photos: Kim Jong Un and North Korea's military Kim uses a pair of binoculars to look south from the Jangjae Islet Defense Detachment, near South Korea's Taeyonphyong Island, in March 2013. Hide Caption 31 of 55 Photos: Kim Jong Un and North Korea's military Kim is greeted by a soldier's family as he inspects the Jangjae Islet Defense Detachment in March 2013. Hide Caption 32 of 55 Photos: Kim Jong Un and North Korea's military Kim is surrounded by soldiers during a visit to the Mu Islet Hero Defense Detachment, also near Taeyonphyong Island, in March 2013. Hide Caption 33 of 55 Photos: Kim Jong Un and North Korea's military Kim arrives at Jangjae Islet by boat to meet with soldiers of the Jangjae Islet Defense Detachment in March 2013. Hide Caption 34 of 55 Photos: Kim Jong Un and North Korea's military Soldiers in the North Korean army train at an undisclosed location in March 2013. Hide Caption 35 of 55 Photos: Kim Jong Un and North Korea's military In a photo released by the official North Korean news agency in December 2012, Kim celebrates a rocket's launch with staff from the satellite control center in Pyongyang. Hide Caption 36 of 55 Photos: Kim Jong Un and North Korea's military Kim, center, poses in this undated picture released by North Korea's official news agency in November 2012. Hide Caption 37 of 55 Photos: Kim Jong Un and North Korea's military Kim visits the Rungna People's Pleasure Ground, under construction in Pyongyang, in a photo released in July 2012 by the KCNA. Hide Caption 38 of 55 Photos: Kim Jong Un and North Korea's military A crowd watches as statues of North Korean founder Kim Il Sung and his son Kim Jong Il are unveiled during a ceremony in Pyongyang in April 2012. Hide Caption 39 of 55 Photos: Kim Jong Un and North Korea's military A North Korean soldier stands guard in front of an UNHA III rocket at the Tangachai-ri Space Center in April 2012. Hide Caption 40 of 55 Photos: Kim Jong Un and North Korea's military In April 2012, Pyongyang launched a long-range rocket that broke apart and fell into the sea. Here, the UNHA III rocket is pictured on its launch pad in Tang Chung Ri, North Korea. Hide Caption 41 of 55 Photos: Kim Jong Un and North Korea's military – A closer look at the UNHA III rocket on its launch pad in Tang Chung Ri, North Korea. Hide Caption 42 of 55 Photos: Kim Jong Un and North Korea's military A military vehicle participates in a parade in Pyongyang in April 2012. Hide Caption 43 of 55 Photos: Kim Jong Un and North Korea's military North Korean soldiers relax at the end of an official ceremony attended by leader Kim Jong Un at a stadium in Pyongyang in April 2012. Hide Caption 44 of 55 Photos: Kim Jong Un and North Korea's military Kim Jong Un applauds as he watches a military parade in Pyongyang in April 2012. Hide Caption 45 of 55 Photos: Kim Jong Un and North Korea's military A North Korean soldier stands on a balcony in Pyongyang in April 2012. Hide Caption 46 of 55 Photos: Kim Jong Un and North Korea's military North Korean soldiers march during a military parade in Pyongyang in April 2012. Hide Caption 47 of 55 Photos: Kim Jong Un and North Korea's military Soldiers board a bus outside a theater in Pyongyang in April 2012. Hide Caption 48 of 55 Photos: Kim Jong Un and North Korea's military North Korean performers sit below a screen showing images of leader Kim Jong Un in Pyongyang in April 2012. Hide Caption 49 of 55 Photos: Kim Jong Un and North Korea's military North Korean soldiers salute during a military parade in Pyongyang in April 2012. Hide Caption 50 of 55 Photos: Kim Jong Un and North Korea's military North Korean soldiers listen to a speech during an official ceremony attended by leader Kim Jong Un at a stadium in Pyongyang in April 2012. Hide Caption 51 of 55 Photos: Kim Jong Un and North Korea's military Members of a North Korean military band gather following an official ceremony at the Kim Il Sung stadium in Pyongyang in April 2012. Hide Caption 52 of 55 Photos: Kim Jong Un and North Korea's military North Korean military personnel watch a performance in Pyongyang in April 2012. Hide Caption 53 of 55 Photos: Kim Jong Un and North Korea's military A North Korean controller is seen along the railway line between the Pyongyang and North Pyongan provinces in April 2012. Hide Caption 54 of 55 Photos: Kim Jong Un and North Korea's military A North Korean military honor guard stands at attention at Pyongyang's airport in May 2001. Hide Caption 55 of 55
In a statement issued after a phone briefing from South Korea's National Intelligence Service, Suh said the execution appeared to be a pre-emptive effort to prevent any internal unrest over Jang's ouster.
Analysts said North Korea was likely to continue with the provocative moves under Kim that have strained its relations with South Korea, the United States and others.
"I think there's going to be a clear amount of brinksmanship," said Yun of the Ploughshares Fund. "I think if we continue to wait for him to do things, he's going to continue to shoot missiles, and he'll probably at some point decide to test a nuclear weapon."
Missile and nukes
North Korea carried out a long-range rocket launch a year ago and an underground nuclear test, its third so far, in February. The U.N. sanctions that followed were met by a barrage of threatening rhetoric from Pyongyang, directed at South Korea and the United States, which ratcheted up tensions in the region.
The situation has calmed since, and the North and South have resumed dialogue. The two sides have agreed to meet next week in their joint industrial zone on the North's side of the border.
But with the anniversary of Kim Jong Il's death, last year's rocket launch and now Jang's execution, Seoul is keeping a close eye on Pyongyang's actions, officials said.
The South Korean defense ministry said Friday that no unusual activities by the North Korean military had been detected.
"December has always been a month in which something happens with North Korea," Cha said. "And we're only halfway through it."
In Washington, a State Department official acknowledged having seen the report of Jang's execution. "While we cannot independently verify this development, we have no reason to doubt the official KCNA report," deputy spokeswoman Marie Harf said in a statement, referring to North Korea's state news agency.
"If confirmed, this is another example of the extreme brutality of the North Korean regime. We are following developments in North Korea closely and consulting with our allies and partners in the region," Harf added.
China, whose senior officials were considered to have close ties to Jang, described the recent developments as North Korea's "internal affairs."
Beijing hopes and believes that relations between the two countries "will continue (to) advance healthily and steadily," Hong Lei, a foreign ministry spokesman, said at a regular briefing Friday.
'Worse than a dog'
The official North Korean report on the execution said a special military tribunal had been held Thursday against Jang, who was accused of trying to overthrow the state "by all sorts of intrigues and despicable methods."
It added, "All the crimes committed by the accused were proved in the course of hearing and were admitted by him."
Once his guilt was established, Jang was immediately executed, it said.
The KCNA report described Jang as a "traitor for all ages" and "worse than a dog," saying he had betrayed his party and leader.
Jang and his allies were accused of double-dealing behind the scenes, "dreaming different dreams" and selling the country's resources at cheap prices, thereby threatening North Korea's economic development, according to a KCNA statement this week.
"Jang desperately worked to form a faction within the party by creating illusion about him and winning those weak in faith and flatterers to his side," the statement said.
It also accused Jang of womanizing, drug use, gambling, eating at expensive restaurants and undergoing medical treatment in a foreign country.
Friday's KCNA report said Jang distributed pornographic pictures among his confidants and took at least 4.6 million euros ($6.3 million) "from his secret coffers and squandered it in 2009 alone." |
is anyone working on this? it has been a long time since the last post. now that i have new elements installed (HOORAY< I DID IT!!!) i need to fire those 20 bowls. some have the green underglaze on them and i have covered them in a different glaze this time. if it works, i will share that recipe.
i am sending out the recipe i found that works to cover amoco velvet green #353 and allows it to remain green. it is from George Wettlaufer's 1976 book, "getting into pots" and he describes it as having a waxy surface. the original recipe called for a feldspar called C-6, a soda spar. when i was unable to find any information on this product (even in the materials list posted on Digitalfire) i called mr. wettlaufer and asked what he would suggest to replace it. his reply was Kona F-4.
this covers colors nicely EXCEPT my mixed underglaze yellow which had floating chunks of yellow suspended above the clay and below the glaze which looked like someone threw yellow sand at a clear paint. unlike other glazes i have tested, this one does not "flatten" texture which is the most important quality for me. if you want to try it, test, test test.
have you been trying any of these glazes? i really wish i could send you a picture.
i was so pleased that 3 of the green underglazed pieces came out well enough to take to the empty bowl supper on friday. i approached one young man who had one of them and asked why he picked it out of the 400 bowls that were on offer. he said he liked the color. hooray! unfortunately, he dropped it on the driveway as he was leaving. we had some extra bowls and we gave him a green one from another potter so he at least had a green one.
I am so glad you found a solution. I am very busy getting glaze firings done for holiday shows. Wll pick up playtime in a couple of months. I wonder if the soda vs potash made the diff. There is also very little magnesium in the "good" glaze. |
#!/usr/bin/make -f
# -*- coding: utf-8 -*-
#
# LinOTP - the open source solution for two factor authentication
# Copyright (C) 2016 - 2019 KeyIdentity GmbH
#
# This file is part of LinOTP server.
#
# This program is free software: you can redistribute it and/or
# modify it under the terms of the GNU Affero General Public
# License, version 3, as published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the
# GNU Affero General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
#
#
# E-mail: linotp@keyidentity.com
# Contact: www.linotp.org
# Support: www.keyidentity.com
#
#
# LinOTP toplevel makefile
#
#
# If you are running in a local development environment, you can
# set these environment variables to configure make behaviour:
# export http_proxy=http://proxy.hostname:port
# export no_proxy=localhost,127.0.0.1,.my.local.domain
# export DOCKER_REGISTRY_URL=registry.local.domain
PYTHON:=python3
# This directory is used as destination for the various parts of
# the build phase. The various install targets default to this directory
# but can be overriden by DESTDIR
BUILDDIR:=$(PWD)/build
# Targets to operate on LinOTPd and its dependent projects shipped
# in this repository
LINOTPD_PROJS := linotpd
# These variables let you set the amount of stuff LinOTP is logging.
#
# LINOTP_LOGLEVEL controls the amount of logging in general while
# LINOTP_CONSOLE_LOGLEVEL controls logging to the console (as opposed
# to logstash -- logstash always gets whatever LINOTP_LOGLEVEL lets
# through, so LINOTP_CONSOLE_LOGLEVEL can be used to have less stuff
# show up on the console than in logstash).
# SQLALCHEMY_LOGLEVEL controls the amount of logging done by SQLAlchemy
# (who would have guessed); DEBUG will log SQL queries and results,
# INFO will log just queries (no results) and WARN will log neither.
# APACHE_LOGLEVEL limits the amount of stuff Apache writes to its error
# output; normally anything that is written to the LinOTP console goes
# through here, too, so there isn't a lot of sense in setting this
# differently to LINOTP_CONSOLE_LOGLEVEL unless you're doing nonstandard
# trickery and/or use a different (and unsupported by us) web server
# than Apache to run LinOTP.
export LINOTP_LOGLEVEL=INFO
export LINOTP_CONSOLE_LOGLEVEL=DEBUG
export SQLALCHEMY_LOGLEVEL=ERROR
export APACHE_LOGLEVEL=DEBUG
#############################################################################################
# Recursive targets
#
# These invoke make in the project subdirectories
#
# install
# clean
#
# Each target will be expanded into the subdirectory targets
#
# e.g. build -> build.subdirmake -> build.smsprovider + build.useridresolver + build.linotpd
#############################################################################################
# Targets that should recurse into linotp project directories
LINOTPD_TARGETS := install clean
.PHONY: $(LINOTPD_TARGETS)
INSTALL_TARGETS := $(addsuffix .install,$(LINOTPD_PROJS))
CLEAN_TARGETS := $(addsuffix .clean,$(LINOTPD_PROJS))
MAKEFILE_TARGETS := $(INSTALL_TARGETS) $(CLEAN_TARGETS)
.PHONY: $(MAKEFILE_TARGETS)
$(MAKEFILE_TARGETS):
# Invoke makefile target in subdirectory/src
$(MAKE) -C $(basename $@)/src $(subst $(basename $@).,,$@)
# Subdirectory make that should invoke target in all subproject directories
.PHONY: %.subdirmake
%.subdirmake : linotpd.% ;
# Add dependencies for main targets
# build -> build.subdirmake
# clean -> clean.subdirmake
# etc.
.SECONDEXPANSION:
$(LINOTPD_TARGETS): $$(basename $$@).subdirmake
clean:
if [ -d $(BUILDDIR) ]; then rm -rf $(BUILDDIR) ;fi
if [ -d RELEASE ]; then rm -rf RELEASE; fi
# Run a command in a list of directories
# $(call run-in-directories,DIRS,COMMAND)
run-in-directories = \
echo run-in-directories:$(1) ;\
for P in $(1) ;\
do \
cmd="cd $$P/src && $(2)" ;\
echo \\n$$cmd ;\
( eval $$cmd ) || exit $? ;\
done
# Run a command in all linotpd directories
run-in-linotpd-projs = $(call run-in-directories,$(LINOTPD_PROJS),$(1))
#################
# Targets invoking setup.py
#
# Installation of packages in 'develop mode'.
.PHONY: develop
develop:
$(call run-in-linotpd-projs,$(PYTHON) setup.py $@)
###############################################################################
# Unit test targets
#
#
# These targets can be run directly from a development
# environment, within a container or an installed system
#
# unittests - just the unit tests
# integrationtests - selenium integration tests
# test - all tests
###############################################################################
test: unittests integrationtests functionaltests
unittests:
$(MAKE) -C linotpd/src/linotp/tests/unit $@
# Functional tests. Additional arguments can be
# supplied with FUNCTIONALTESTS_ARGS
functionaltests:
$(MAKE) -C linotpd/src/linotp/tests/functional $@
# integrationtests - selenium integration tests
# Use the SELENIUMTESTS_ARGS to supply test arguments
integrationtests:
$(MAKE) -C linotpd/src/linotp/tests/integration $@
.PHONY: test unittests functionaltests integrationtests
###############################################################################
# Packaging targets
#
#
# These targets run the various commands needed
# to create packages of linotp
#
# builddeb: Generate .debs
# deb-install: Build .debs and install to DESTDIR
###############################################################################
DEBPKG_PROJS := linotpd
BUILDARCH = $(shell dpkg-architecture -q DEB_BUILD_ARCH)
CHANGELOG = "$(shell cd linotpd/src ; dpkg-parsechangelog)"
# Output is placed in DESTDIR, but this
# can be overriden
ifndef DESTDIR
DESTDIR = $(BUILDDIR)
endif
.PHONY: builddeb
builddeb:
# builddeb: Run debuild in each directory to generate .deb
$(call run-in-directories,$(DEBPKG_PROJS),$(MAKE) builddeb)
.PHONY: deb-install
deb-install: builddeb
# deb-install: move the built .deb, .changes and related files into an archive directory and
# generate Packages file
mkdir -pv $(DESTDIR)
cp $(foreach dir,$(DEBPKG_PROJS),$(dir)/build/*.deb) $(DESTDIR)
find $(foreach dir,$(DEBPKG_PROJS),$(dir)) -type f -regex '.+\.changes' -o -regex '.+\.dsc' -o -regex '.+\.tar\..+' -o -regex '.+\.buildinfo' | xargs -iXXX -n1 cp XXX $(DESTDIR)
find $(DESTDIR)
cd $(DESTDIR) && dpkg-scanpackages -m . > Packages
######################################################################################################
# Docker container targets
#
# These targets are for building and running docker containers
# for integration and builds
#
# Container name | Dockerfile location | Purpose
# ---------------------------------------------------------------------------------------------------
# linotp-builder | Dockerfile.builder | Container ready to build linotp packages
# linotp | linotpd/src | Runs linotp in apache
# selenium-test | linotpd/src/tests/integration | Run LinOTP Selenium tests against selenium remote
# linotp-unit | linotpd/src/linotp/tests/unit | Run LinOTP Unit tests
######################################################################################################
# Extra arguments can be passed to docker build
DOCKER_BUILD_ARGS=
# List of tags to add to built linotp images, using the '-t' flag to docker-build
DOCKER_TAGS=latest
# Override to change the mirror used for image building
DEBIAN_MIRROR=deb.debian.org
# Override to change the Debian release used to build with
DEBIAN_RELEASE_NAME=buster
BASE_IMAGE=debian:$(DEBIAN_RELEASE_NAME)
# Pass proxy environment variables through to docker build by default
DOCKER_PROXY_BUILD_ARGS= --build-arg=http_proxy --build-arg=https_proxy --build-arg=no_proxy
# Arguments passed to Docker build commands
DOCKER_BUILD_ARGS+= --build-arg BASE_IMAGE=$(BASE_IMAGE) \
--build-arg DEBIAN_MIRROR=$(DEBIAN_MIRROR) \
--build-arg DEBIAN_RELEASE_NAME=$(DEBIAN_RELEASE_NAME) \
--build-arg DEPENDENCY_SOURCE=$(DEPENDENCY_SOURCE) \
--build-arg DEPENDENCY_COMPONENT=$(DEPENDENCY_COMPONENT) \
--build-arg DEPENDENCY_GPG_KEYID=$(DEPENDENCY_GPG_KEYID) \
--build-arg DEPENDENCY_GPG_KEY_URL=$(DEPENDENCY_GPG_KEY_URL)
# Default Docker run arguments.
# Extra run arguments can be given here. It can also be used to
# override runtime parameters. For example, to specify a port mapping:
# make docker-run-linotp-sqlite DOCKER_RUN_ARGS='-p 1234:80'
DOCKER_RUN_ARGS=
DOCKER_BUILD = docker build $(DOCKER_BUILD_ARGS) $(DOCKER_PROXY_BUILD_ARGS)
DOCKER_RUN = docker run $(DOCKER_RUN_ARGS)
TESTS_DIR=linotpd/src/linotp/tests
SELENIUM_TESTS_DIR=$(TESTS_DIR)/integration
UNIT_TESTS_DIR=$(TESTS_DIR)/unit
FUNCTIONAL_TESTS_DIR=$(TESTS_DIR)/functional
## Toplevel targets
# Toplevel target to build all containers
docker-build-all: docker-build-debs docker-build-linotp docker-build-linotp-test-image docker-build-linotp-softhsm
# Toplevel target to build linotp container
docker-linotp: docker-build-debs docker-build-linotp
# Build and run Selenium /integration tests
docker-selenium: docker-build-all docker-run-selenium
# Build and run Unit tests
docker-unit: docker-build-linotp docker-build-linotp-test-image docker-run-linotp-unit
docker-functional: docker-build-linotp docker-build-linotp-test-image docker-run-linotp-functional-test
docker-pylint: docker-run-linotp-pylint
.PHONY: docker-build-all docker-linotp docker-run-selenium docker-unit docker-pylint docker-functional
# This is expanded during build to add image tags
DOCKER_TAG_ARGS=$(foreach tag,$(DOCKER_TAGS),-t $(DOCKER_IMAGE):$(tag))
# The linotp builder container contains all build dependencies
# needed to build linotp, plus a copy of the linotp
# sources under /pkg/linotp
#
# To use this container as a playground to test build linotp:
# docker run -it linotp-builder
.PHONY: docker-build-linotp-builder
docker-build-linotp-builder: DOCKER_IMAGE=linotp-builder
docker-build-linotp-builder:
$(DOCKER_BUILD) \
-f Dockerfile.builder \
$(DOCKER_TAG_ARGS) \
-t $(DOCKER_IMAGE) \
.
# A unique name to reference containers for this build
DOCKER_CONTAINER_TIMESTAMP := $(shell date +%H%M%S-%N)
NAME_PREFIX := linotpbuilder-$(DOCKER_CONTAINER_TIMESTAMP)
DOCKER_CONTAINER_NAME = $(NAME_PREFIX)
.PHONY: docker-build-debs
docker-build-debs: docker-build-linotp-builder
# Force rebuild of debs
rm -f $(BUILDDIR)/apt/Packages
$(MAKE) $(BUILDDIR)/apt/Packages
# Build the debian packages in a container, then extract them from the image
$(BUILDDIR)/apt/Packages:
$(DOCKER_RUN) \
--detach \
--rm \
--name $(DOCKER_CONTAINER_NAME)-apt \
linotp-builder \
sleep 3600
docker cp . $(DOCKER_CONTAINER_NAME)-apt:/build
docker exec \
$(DOCKER_CONTAINER_NAME)-apt \
make deb-install DESTDIR=/build/apt DEBUILD_OPTS=\"$(DEBUILD_OPTS)\"
docker cp \
$(DOCKER_CONTAINER_NAME)-apt:/build/apt $(DESTDIR)
docker rm -f $(DOCKER_CONTAINER_NAME)-apt
# Build just the linotp image. The builder-linotp is required but will not be
# built by this target - use 'make docker-linotp' to build the dependencies first
.PHONY: docker-build-linotp
docker-build-linotp: DOCKER_IMAGE=linotp
docker-build-linotp: $(BUILDDIR)/dockerfy $(BUILDDIR)/apt/Packages
cp linotpd/src/Dockerfile \
linotpd/src/config/*.tmpl \
linotpd/src/tools/* \
linotpd/src/linotp/tests/integration/testdata/se_mypasswd \
$(BUILDDIR)
cp -r linotpd/src/config/docker-initscripts.d $(BUILDDIR)
# We show the files sent to Docker context here to aid in debugging
find $(BUILDDIR)
$(DOCKER_BUILD) \
$(DOCKER_TAG_ARGS) \
-t $(DOCKER_IMAGE) \
$(BUILDDIR)
# Build testing Docker Container
# This container is based on the linotp image and includes additional
# dependencies for testing targets.
# It needs an existing linotp image available
# which can be built by make docker-build-linotp
.PHONY: docker-build-testenv
docker-build-linotp-test-image: DOCKER_IMAGE=linotp-testenv
docker-build-linotp-test-image:
cd $(TESTS_DIR) \
&& $(DOCKER_BUILD) \
$(DOCKER_TAG_ARGS) \
-t $(DOCKER_IMAGE) .
# Build Softhsm test container
.PHONY: docker-build-linotp-softhsm
docker-build-linotp-softhsm: DOCKER_IMAGE=linotp-softhsm
docker-build-linotp-softhsm: BASE_IMAGE=linotp:latest
docker-build-linotp-softhsm:
cd $(SELENIUM_TESTS_DIR) \
&& $(DOCKER_BUILD) \
$(DOCKER_TAG_ARGS) \
-f Dockerfile.softhsm \
-t $(DOCKER_IMAGE) .
# Run Selenium based smoketest against LinOTP configured with
# softhsm security module
.PHONY: docker-run-softhsm-smoketest
docker-run-softhsm-smoketest:
cd $(SELENIUM_TESTS_DIR) \
&& docker-compose \
-f docker-compose.yml -f docker-compose-softhsm.yml \
run \
--rm \
-e PYTESTARGS="-m smoketest ${PYTESTARGS}" \
selenium_tester
cd $(SELENIUM_TESTS_DIR) \
&& docker-compose down
# Pass PYTESTARGS=test_manage.py for picking a specific test file
# PYTESTARGS="-k testname" for picking a specific test
#
# e.g.
# make docker-run-selenium PYTESTARGS=test_manage.py
.PHONY: docker-run-selenium
docker-run-selenium:
cd $(SELENIUM_TESTS_DIR) \
&& docker-compose run \
--rm \
-e PYTESTARGS="${PYTESTARGS}" \
selenium_tester
cd $(SELENIUM_TESTS_DIR) \
&& docker-compose down
# Remove all selenium test relevant containers/images
# We do not remove the LinOTP image:
# - Maybe built an up-2-date image some pipeline steps before test execution.
.PHONY: docker-selenium-clean
docker-selenium-clean:
# This container triggers the python test scripts
docker stop $$(docker ps -a -q --filter "name=integration_selenium_tester_run") 2>/dev/null || echo "Stop integration_selenium_tester_run_*"
# This container receives the selenium webdriver instructions
docker stop $$(docker ps -a -q --filter "name=integration_selenium") 2>/dev/null || echo "Stop integration_selenium_*"
docker stop $$(docker ps -a -q --filter "name=integration_linotp") 2>/dev/null || echo "Stop integration_linotp_*"
docker stop $$(docker ps -a -q --filter "name=integration_db") 2>/dev/null || echo "Stop integration_db_*"
docker rm -f $$(docker ps -a -q --filter "name=integration_selenium_tester_run") 2>/dev/null || echo "Remove container integration_selenium_tester_run_*"
docker rm -f $$(docker ps -a -q --filter "name=integration_selenium") 2>/dev/null || echo "Remove container integration_selenium_*"
docker rm -f $$(docker ps -a -q --filter "name=integration_linotp") 2>/dev/null || echo "Remove container integration_linotp_*"
docker rm -f $$(docker ps -a -q --filter "name=integration_db") 2>/dev/null || echo "Remove container integration_db_*"
docker rmi -f integration_selenium_tester 2>/dev/null || echo "Remove image integration_selenium_tester"
docker rmi -f selenium_tester 2>/dev/null || echo "Remove image selenium_tester"
docker rmi -f selenium/standalone-chrome-debug 2>/dev/null || echo "Remove image selenium/standalone-chrome-debug"
docker rmi -f mysql 2>/dev/null || echo "Removed image mysql"
docker images
docker ps -a
.PHONY: docker-run-linotp-sqlite
docker-run-linotp-sqlite: docker-build-linotp
# Run linotp in a standalone container
cd linotpd/src \
&& $(DOCKER_RUN) -it \
-e HEALTHCHECK_PORT=80 \
-e LINOTP_LOGLEVEL=$(LINOTP_LOGLEVEL) \
-e LINOTP_CONSOLE_LOGLEVEL=$(LINOTP_CONSOLE_LOGLEVEL) \
-e SQLALCHEMY_LOGLEVEL=$(SQLALCHEMY_LOGLEVEL) \
-e APACHE_LOGLEVEL=$(APACHE_LOGLEVEL) \
linotp
# Dockerfy tool
.PHONY: get-dockerfy
get-dockerfy: $(BUILDDIR)/dockerfy
DOCKERFY_URL=https://github.com/SocialCodeInc/dockerfy/releases/download/1.1.0/dockerfy-linux-amd64-1.1.0.tar.gz
DOCKERFY_SHA256=813d47ebf2e63c966655dd5349a29600ba94deac7a57c132bf624c56ba210445
# Obtain dockerfy binary
# TODO: Build from source
$(BUILDDIR)/dockerfy:
mkdir -pv $(BUILDDIR)/dockerfy-tmp
wget --directory-prefix=$(BUILDDIR)/dockerfy-tmp $(DOCKERFY_URL)
echo "${DOCKERFY_SHA256} " $(BUILDDIR)/dockerfy-tmp/dockerfy-linux-amd64*gz \
| sha256sum -c -
tar -C $(BUILDDIR) -xvf $(BUILDDIR)/dockerfy-tmp/dockerfy-linux-amd64*.gz
rm -r $(BUILDDIR)/dockerfy-tmp
#
# # Unit tests
#
# Run Unit tests. Use $PYTESTARGS for additional pytest settings
.PHONY: docker-run-linotp-unit
docker-run-linotp-unit:
cd $(UNIT_TESTS_DIR) \
&& $(DOCKER_RUN) \
--name=$(DOCKER_CONTAINER_NAME)-unit \
--volume=$(PWD):/linotpsrc:ro \
-t linotp_test_env \
/usr/bin/make test PYTESTARGS="$(PYTESTARGS)"
#jenkins pipeline uses this make rule
.PHONY: docker-run-linotp-unit-pipeline
PYTESTARGS=-v -p no:cacheprovider --junitxml=/tmp/pytests.xml
docker-run-linotp-unit-pipeline: docker-run-linotp-unit
docker cp $(DOCKER_CONTAINER_NAME)-unit:/tmp/pytests.xml $(PWD)
docker rm $(DOCKER_CONTAINER_NAME)-unit
#
# # Pylint
#
# Run Pylint Code Analysis
.PHONY: docker-run-linotp-pylint
docker-run-linotp-pylint: docker-build-linotp-test-image
$(DOCKER_RUN) \
--name=$(DOCKER_CONTAINER_NAME)-pylint \
--volume=$(PWD):/linotpsrc \
-w="/linotpsrc" \
--entrypoint="" \
--env "LANG=C.UTF-8" \
-t linotp_test_env \
pylint --output-format=parseable --reports=y --rcfile=.pylintrc \
--disable=E1101,maybe-no-member --ignore tests,functional,integration linotp > pylint.log; exit 0
#
# # Functional Tests
#
# NIGHTLY variable controls, if certain long-runnig tests are skipped
#
# NIGHTLY="no" or unset: long-running tests are skipped
# NIGHTLY="yes" all tests are executed
#
# Example:
# $ export NIGHTLY="yes"
# $ make docker-run-linotp-functional-test
FUNCTIONAL_DOCKER_CONTAINER_NAME=linotp-$(DOCKER_CONTAINER_TIMESTAMP)-functional
FUNCTIONAL_MYSQL_CONTAINER_NAME=mysql-$(DOCKER_CONTAINER_TIMESTAMP)-functional
.PHONY: docker-run-linotp-functional-test
docker-run-linotp-functional-test:
cd $(FUNCTIONAL_TESTS_DIR) && \
NIGHTLY=${NIGHTLY} \
FUNCTIONAL_DOCKER_CONTAINER_NAME=$(FUNCTIONAL_DOCKER_CONTAINER_NAME) \
FUNCTIONAL_MYSQL_CONTAINER_NAME=$(FUNCTIONAL_MYSQL_CONTAINER_NAME) \
PYTESTARGS="$(PYTESTARGS)" \
docker-compose --project-directory $(PWD) up \
--abort-on-container-exit \
--force-recreate
docker rm $(FUNCTIONAL_DOCKER_CONTAINER_NAME) $(FUNCTIONAL_MYSQL_CONTAINER_NAME)
|
Even the hard-core skeptics believe in magic, says Matthew Hutson in his new book The 7 Laws of Magical Thinking: How Irrational Beliefs Keep us Happy, Healthy and Sane which has just been released in paperback.
Most of us have some sentimental objects that would seem to lose their importance if replaced by an exact copy. We imbue our pets with human personality traits. We are disgusted at the thought of eating a cake that looks like fecal matter. We expect that what goes around comes around. All of these are examples of magical thinking, Hutson argues. A skeptic and an atheist, Hutson claims that ‘our ongoing flirtation with supernaturalism is a relationship that we depend on for survival.’ I’m not convinced. In a lively discussion, we delve into magical thinking, its pitfalls and potential benefits.
Matthew Hutson is a former editor at Psychology Today, and has a B.Sc. in cognitive neuroscience from Brown University and an M.S. in science writing from MIT. His work has appeared in Wired, Discover, Scientific American Mind, Popular Mechanics, The Boston Globe, The New York Times and the New York Times Magazine.
Sign up now for Point of Inquiry updates. New POI episodes and updates sent right to you. It's as easy as typing in your email. Your email isn't shared with anyone else. Just news and updates. Point of Inquiry
Links Mentioned in this Episode
magicalthinkingbook.com
|
#!/usr/bin/env bash
. path.sh
data_dir=data/ptb
mkdir -p $data_dir
echo "*** Downloading PTB text data ***"
(
data_path=https://raw.githubusercontent.com/townie/PTB-dataset-from-Tomas-Mikolov-s-webpage/master/data/
cd $data_dir
[ ! -f $data_dir/ptb.txt ] && wget -nv $data_path/ptb.train.txt || exit 1;
mv ptb.train.txt ptb.txt
[ ! -f $data_dir/dev.txt ] && wget -nv $data_path/ptb.valid.txt || exit 1;
mv ptb.valid.txt dev.txt
[ ! -f $data_dir/test.txt ] && wget -nv $data_path/ptb.test.txt || exit 1;
mv ptb.test.txt test.txt
)
echo "*** Finished downloading PTB text data ***"
|
Faster big-data analysis - breck
http://news.mit.edu/2017/faster-big-data-analysis-tensor-algebra-1031
======
breck
Here's the paper: [http://people.csail.mit.edu/fred/taco-
tools.pdf](http://people.csail.mit.edu/fred/taco-tools.pdf)
|
Not Applicable.
Not Applicable
1. Field of the Invention
The present invention relates to an apparatus, a bar which is a beam to lash cargo to, herein called the cargo lash to bar, for fastening and restraining cargo to a transport vessel or vehicle, especially as used in lashing down mobile equipment to the deck of an intermodal cargo carrying roll-on/roll-off ocean going ship known as a RoRo, or for fastening to a tanker ship""s weather deck.
2. Description of Related Art
The transportation of cargo is increasingly geared to intermodal and unitized handling. Most ships are equipped for intermodal shipping that consists primarily of freight containers that are eight feet high, eight feet wide, and twenty feet long. This standard is referred to as a TEU or twenty foot equivalent unit and is particularly evident in the international ISO 668 specification. Many variety of containers, and flatracks have been developed for transport of lumber and liquid tanks, and are just two examples of intermodal freight handling equipment that maintain the TEU although height of specialty containers frequently varies. The intermodal nature of freight containers makes them readily usable without adaptation or modification for transport over sea, over land by trucking, or by rail. Mobile equipment and other break bulk cargo are frequently transported on vessels designed to carry intermodal cargo.
Currently when an armored tank or other irregular shaped mobile equipment is loaded on the deck of a ship which is designed for a variety of cargo including intermodal freight containers, the mobile equipment will be chained down and connected to a socket in the deck. Current cargo restraint lashing rigging consists of D-rings welded to a ship""s deck and D-rings welded to twist lock bases, breech bases, clover leaf bases, dovetail bases or other commercially available or proprietary locking mounting bases for mounting a lashing line to corresponding deck sockets. U.S. Pat. Nos. 4,457,650 and 3,860,209 illustrate the styles of fastening rings, cavities, and sockets and the various bases types which are used to fasten to a ship""s deck. All of these embodiments suffer from the same problem, that they mount and seat into the sockets built into the deck that are spaced too far apart for convenient, safe, and secure
Spacing of these sockets is usually substantially twenty feet (or forty feet) apart fore and aft and two adjacent eight feet across athwartship in a configuration reciprocal to the mounting of an intermodal freight container. As a result the cargomen who are stevedores, longshoremen, and ship""s crew are frequently faced with the problem on the ship of inadequate tie down locations adjacent to the designated fastening points of most mobile equipment. The result is multiple chains being stretched at odd angles to reach a single mounting location on the deck. This aggravates the problems of lashing and can lead to lashing gear, D-ring, or deck sockets and bases breaking or tearing with the equipment dangerously coming loose at sea. Often the twist lock sockets or D-rings are covered by cargo or wheels or tracks of a piece of mobile equipment, which must be accommodated by lashing to a farther point and putting extra shear load on a neighboring D-ring or omitting a lashing entirely.
In some locations on ships, D-rings are welded to a ship""s deck generating different problems. Welding to the deck is a permanent attachment which can be an encumbrance later on and be an obstruction for man and machines that causes a safety hazard. Also, dragging a welder and all of it""s equipment around a ship""s deck to weld down and retrofit individual D-rings or repair broken D-rings is slow and labor intensive. Welding is a severe optical hazard to ship""s crew about the deck. Welding also heats the metal of the deck in excess of 3,000 degrees F., which causes localized changes in the crystal structure of the base metal making it brittle and reducing the strength. A ship""s deck cannot be readily annealed. Weld on D-rings are not able to be welded to the lower deck of most ships because the bottom side of the lower deck is the fuel hold or fuel tank. To weld to the lower deck requires the costly procedure of evacuating the space with an inert gas or risking major fire or explosion.
The erectable secondary deck shown in U.S. Pat. No. 4,329,935 of Jonasson using twist locks and taking advantage of the regular spacing of twist lock mounting apertures in the deck of a ship displays the practicality of the present invention to take advantage of the same regularly ordered twist lock socket apertures.
Folding flatracks of the type shown in U.S. Pat. No. 5,398,832 of Clive-Smith reveal the great need for versatile lashing of cargo so that cargo will securely endure a voyage aboard a waterborne vessel. Clive-Smith discloses an improved flatrack with d-rings for lashing cargo to keep a load fastened to the flatrack wherein the lashing chain begins, ends, and is brought taught by a binder as an internal force to the flatrack. That is to say that cargo native to the flatrack is held fast to the flatrack. The present invention on the other hand provides a new cargo lashing apparatus for lashing cargo to which has a completely different mode of functionality from Clive-Smith, wherein an external force is applied to the apparatus of the present invention and an object external to the cargo lash to bar is restrained to the deck of a ship. Clive-Smith is beneficial to the understanding of the great difficulty in facilitating the safe and secure arrival of cargo aboard the modern intermodal ships. These are new solutions to the broad array of challenges that shippers face in these days of intermodal cargo movement.
The versatility of the twist lock sockets apertures at their respective regular spacing as an erection attachment point is further evidenced by the novel portable hand rail of Bel, U.S. Pat. No. 4,655,153. As with the previous citation, the preponderance of examples show that the deck sockets can be utilized far beyond simple freight containers. All these however fail to grasp the enormity of the advantage to be gained in cargo fastening, restraining, and securing of the present invention.
People in the mobile equipment shipping industry are clear, there is a lot of money being spent shoreside for load preplanning and logistical processes to make efficient stowage but when mobile equipment cargo arrives onboard the ship it is not loaded according to the plan because of the impediments of the use of old style lashing gear exacerbating the problems of inadequate deck tie down lashing points. That serious difficulty of consistent, safe, and secure lashing of mobile equipment cargo is clearly enunciated in U.S. MTMCTEA Ref 97-55-22 Marine Lifting and Lashing Handbook on page 1-1, which says, xe2x80x9cAs we saw during Desert Shield/Storm, this often leads to inconsistent and excessive lashing procedures that wastes time, money, and manpower hoursxe2x80x9d.
The cargo lash to bar is a beam which is a continuous structural member that can be fastened and unfastened to a ship""s deck and quickly handled aboard a ship to facilitate multiple convenient and accessible lashing point apertures along the length of the span of the member to ensure secure cargo lashing and restraining especially of irregular shaped loads. The key to the application of the present invention is the use of standard twist lock bolt connections. Such twist locks are commercially available in a variety of mounting base configurations. The cargo lash to bar is constructed to use a standard twist lock rotating locking bolt connections at each end of the bar for releasable locking to its mounting locking surface. These twist locks are an integral part of a RoRo ship""s current inventory of lashing gear. The cargo lash to bar comes with the mounting apertures in its mounting locking surface on the bottom, that receive the twist lock body and rotating locking bolt.
The result is a versatile new piece of cargo lashing gear. Accordingly, several objects and advantages of the present invention are: a) the primary object of the cargo lash to bar invention is to provide to ships and other cargo transport vehicles or handlers a device that speeds up the loading and deployment of irregular shaped cargo and mobile equipment by aiding their task of securing a load by providing a structural member to attach to with lashing aperture locations heretofore unavailable along the entire length between typical deck mounting sockets, b) a further advantage to the use of the cargo lash to bar is that more D-rings and aperture holes for lashing makes for less overloaded rings which provides security from losing multiple pieces of mobile equipment at sea adding a safety benefit to the crew tasked with resecuring a loose load, c) a further advantage of the lash to bar is that if the lashings break loose on a piece of mobile equipment cargo by fraying or other means, the beam will act as dunnage to restrict movement reducing potential damage to a ship and adjacent cargo, d) a further advantage of the cargo lash to bar is the extensive use of D-rings which have a smooth contoured surface that can be used with chains, cable, hemp rope, or especially light weight and high strength nylon and kevlar synthetic fibrous straps or webbing, e) a further advantage to the use of the cargo lash to bar is that it allows the stevedores to make a more systematic approach to the deck location of mobile equipment and actually position equipment according to the shoreside prestowage plan, f) a further advantage is the simplification of load positioning for consistent repeatable lashing of mobile equipment, g) a further advantage is reclamation of deck space previously occupied by flailed lashings allowing a ship to carry more cargo, h) a further advantage to the use of the cargo lash to bar when it lays fore to aft across the twenty foot span of a ships deck spanning deck fastening twist lock sockets, mobile equipment can readily be loaded in between a pair of cargo lash to bars like driving a car into a stall in a striped asphalt parking lot since most mobile equipment is loaded on a RoRo fore and aft in the same orientation as a container, i) a further object of the cargo lash to bar is to provide a new means to assist the ordinary lashing of standard containers attached one on top of another by placing a cargo lash to bar under the end of a bottom container to lash the upper containers freeing up the valuable deck space which previously required using heavy and cumbersome lashing bars and associated lashing gear, j) a further advantage is to remove the hazard of welding on a ship""s deck and facilitate welding repairs at a safe location, even by off-ship contractors, k) a further advantage is the ability to use the cargo lash to bar as a safe article on which to weld fasteners for restraining irregular cargo to the lower deck of a ship immediately above a fuel tank or a fuel soaked wooden deck, l) a further advantage of the cargo lash to bar is use as a shoring beam by itself or with multiple lash to bars stacked on top of the other, m) a further advantage of the cargo lash to bar is to use a pair of the apparatus for supporting or securing cargo above a clear span over a hatch or damaged portion of a ships deck, n) a further advantage is the ability to deploy a pair of cargo lash to bars to utilize space on top of a container, or stack of containers, or stowage flat rack even while the container is being drawn by a semi truck, or rail car, o) a further object of the cargo lash to bar is a lashing apparatus that can be readily stacked one on top of another for unitized storage when this is a desirable feature, p) a further object of the cargo lash to bar is a member that can be manipulated and moved securely and quickly by a fork lift truck, q) a further advantage of the lash to bar is the ability to keep mobile equipment cargo separated far enough apart for a person to walk between the mobile cargo especially during the stevedoring loading process to allow access from front to back of a ship to tend to cargo while a ship is at sea to access the distant areas of a loaded cargo hold in case of medical emergency or fire.
Further objects and advantages of the present invention will become apparent from a consideration of the drawings and ensuing description. |
ARRL DX Bulletin ARLD042 (2004)
SB DX @ ARL $ARLD042
ARLD042 DX news
ZCZC AE42
QST de W1AW
DX Bulletin 42 ARLD042
From ARRL Headquarters
Newington CT October 21, 2004
To all radio amateurs
SB DX ARL ARLD042
ARLD042 DX news
This week's bulletin was made possible with information provided by
F5MUX, K7ZZ, NF4A, VE7SV, QRZ DX, the OPDX Bulletin, The Daily DX,
425DXnews, DXNL, WA7BNM and Contest Corral from QST. Thanks to all.
MAURITIUS, 3B. Mart, DL6UAA is QRV as 3B8MM until November 16. QSL
to home call.
CROATIA, 9A. Paul, N4PN and Charlie, NF4A are QRV as 9A4PN and
9A5PC, respectively. Activity is on all bands. They will also
participate in the upcoming CQ WW contest as part of the 9A1A
Multi/Multi effort. QSL to home calls. QSL contest call via
operators' instructions.
SIERRA LEONE, 9L. Ivo, 9A3A is QRV as 9L1ADA from Freetown until
November 1. Activity is on 160 to 10 meters. QSL via operator's
instructions.
BHUTAN, A5. Clipperton DX Club members will be QRV as A52CDX from
Thimphu and Jakar from October 24 to November 11. QSL via F9DK.
MARTINIQUE, FM. Lee, F5MUX will be QRV as TO7X from October 26 to
November 11. He will also participate in the upcoming CQ WW contest
as a Single Op/All Band/High Power entry. QSL to home call.
ST. PIERRE AND MIQUELON, FP. A nine-member team will be QRV as
FP/VE7SV from October 23 to November 2. Activity will be on 160 to
6 meters, including the newer bands, using CW and SSB, and possibly
other modes as well. They will also participate in the upcoming CQ
WW contest. QSL via operators' instructions.
WALLIS AND FUTUNA ISLANDS, FW. FW7AQR has been QRV using RTTY on 15
meters around 2330 and 0100z. QSL via JA7AQR.
DOMINICA, J7. Bill, W4WX, Clarence, W9AAZ, Larry, W1LR, Cory, N1WON
and Vance, N5VL will be QRV as J75WX, J79AA, J79LR, J79CM and J79VL,
respectively, from October 26 to November 2. The will participate
in the upcoming CQ WW contest as J75L. QSL contest call via KR4DA
and all others to home calls.
MINAMI TORISHIMA, JD1. Yuji is QRV as JR6TYH/JD1 and is here until
mid December. QSL via bureau.
AMERICAN SAMOA, KH8. Lee, KH6BZF will be QRV as WH8/KH6BZF during
his spare time from October 22 to 28. He plans to be active on 80
to 15 meters, and possibly 12 and 10 meters, depending on
conditions. QSL direct to home call.
SOUTH SHETLAND ISLANDS. Alex, R1ANF has been QRV using PSK on 17
meters around 2145z and RTTY on 20 meters around 0015z. QSL via
RK1PWA.
POLAND, SP. Juergen, DL5CE and Dieter, DL7VAF will be QRV as SO1CE
and SO1VAF, respectively, from Wolin Island, IOTA EU-132, from
October 23 to 24. Activity will be on 160 to 10 meters using SSB,
RTTY and PSK31. QSL to home calls.
WEST KIRIBATI, T30. Tom, K7ZZ will be QRV as T30T from October 25
to November 9. Activity will be on 160 to 10 meters, including the
newer bands. He will also participate in the upcoming CQ WW contest
as a Single Op/All Band entry. QSL direct to home call.
TUVALU, T2. Ulli, DL2AH is QRV as T20AH from Funafuti Island, IOTA
OC-015, until October 24. Activity is on 20 to 10 meters using SSB,
RTTY and PSK31. QSL to home call.
CAMEROON, TJ. Lionel, F5PSA is QRV as TJ3SL during his spare time
and plans to be here until February 2006. QSL to home call.
CANADA, VE. Jeffrey, VE3JFF will be QRV as VC3W from October 23 to
December 23 to celebrate the 175th anniversary of the opening of the
first Welland Canal. He also plans to operate from a couple of
islands located in the old and current canals. QSL to home call.
CAMBODIA, XU. Jaak, ES1FB is QRV as XU7ACE until November 5. QSL
to home call.
IRAQ, YI. SP8HKT and SP3GTS are QRV as YI9KT and YI9GT,
respectively, from the Polish base near Al-Chilla. They are active
in their spare time and are stationed here until February 2005. QSL
via operators' instructions.
THIS WEEKEND ON THE RADIO. The QRP ARCI Fall QSO Party, USI W/VE
Islands QSO Party, 50 MHz Fall Sprint and the 4th Annual FISTS Coast
to Coast Contest are all scheduled for this weekend. Please see
October QST, page 99, and the ARRL and WA7BNM contest websites for
details.
NNNN
/EX |
Q:
Simple MPI_Gather test with memcpy error
I am learning MPI, and trying to create examples of some of the functions. I've gotten several to work, but I am having issues with MPI_Gather. I had a much more complex fitting test, but I trimmed it down to the most simple code. I am still, however, getting the following error:
root@master:/home/sgeadmin# mpirun ./expfitTest5
Assertion failed in file src/mpid/ch3/src/ch3u_request.c at line 584: FALSE
memcpy argument memory ranges overlap, dst_=0x1187e30 src_=0x1187e40 len_=400
internal ABORT - process 0
I am running one master instance and two node instances through AWS EC2. I have all the appropriate libraries installed, as I've gotten other MPI examples to work. My program is:
int main()
{
int world_size, world_rank;
int nFits = 100;
double arrCount[100];
double *rBuf = NULL;
MPI_Init(NULL,NULL);
MPI_Comm_size(MPI_COMM_WORLD, &world_size);
MPI_Comm_rank(MPI_COMM_WORLD, &world_rank);
assert(world_size!=1);
int nElements = nFits/(world_size-1);
if(world_rank>0){
for(int k = 0; k < nElements; k++)
{
arrCount[k] = k;
}}
MPI_Barrier(MPI_COMM_WORLD);
if(world_rank==0)
{
rBuf = (double*) malloc( nFits*sizeof(double));
}
MPI_Gather(arrCount, nElements, MPI_DOUBLE, rBuf, nElements, MPI_DOUBLE, 0, MPI_COMM_WORLD);
if(world_rank==0){
for(int i = 0; i < nFits; i++)
{
cout<<rBuf[i]<<"\n";
}}
MPI_Finalize();
exit(0);
}
Is there something I am not understanding in malloc or MPI_Gather? I've compared my code to other samples, and can't find any differences.
A:
The root process in a gather operation does participate in the operation. I.e. it sends data to it's own receive buffer. That also means you must allocate memory for it's part in the receive buffer.
Now you could use MPI_Gatherv and specify a recvcounts[0]/sendcount at root of 0 to follow your example closely. But usually you would prefer to write an MPI application in a way that the root participates equally in the operation, i.e. int nElements = nFits/world_size.
|
your ray of sunshine
Vacation time = family time
During this past long weekend, I had the blessing of spending some quality time with a small portion of my already tiny family.
My eldest brother, Kuya Miko, is 13 years older than me. The memories I have growing up rarely have him involved because of our large age gap. When I was just starting school, he was already wrapping up his last year in high school. The age gap and his tendency to separate himself from our parents led to a brother-sister relationship that was just barely there. For the longest time, I felt like I knew nothing about him and vice versa. I loved him, but mostly out of the fact that I knew I had to love my own brother. There was a lack of feeling behind the statement “I love you” when it came to him. At points I felt like he was that random relative I would see during major holidays and would only hold polite conversation with for half an hour once or twice a year. I knew I had him in my life, it just felt like sometimes there wasn’t much of an impact from having him involved at all.
I have noticed that the older I get the easier it becomes to talk to him. I now have reasons behind my mumbled “love you’s” to him and I know a few more things about him than when I was in elementary school. We’ve finally reached this level of connection that I’ve always wanted while growing up, and I couldn’t be happier about it. He has provided so much wisdom for me to listen to and follow, and I’ve genuinely witnessed how much growth there’s been in our relationship with each other.
Over the weekend, we headed down to Southern California with his fiancé, Ate Mariane, and their son, Zeus. From the moment our vacation started to the very end, I noticed just how much the three of them make sense as a little family unit. Due to my Kuya’s tendency to keep to himself more often than less, I never really understood the dynamic he had with his fiancé and son. I was able to witness the amount of care and love they have for each other, and for their son, and it was beautiful to see. They’re a loving family in the midst of our larger family, and I’m content in thought to know that my Kuya Miko will be developing a future with the both of them involved in every aspect.
The time spent at Knott’s Berry Farm, Disneyland, and all of the little Southern California hotspots was a really joyous occasion since I was able to understand these 3 members of my family a lot more than before. I’m grateful for them, and I know that no matter what the future holds, at least I have these goofy, loving family members involved throughout everything. |
Kihihi Garuga Golf Course is located in South Western Uganda. It has 18 holes totaling to par 71 with length of 7189m yards. With an airstrip in Kihihi, visitors can fly-in, enjoy a game of Golf, and sample the countryside tourism cites and activites. The Garuga course is a big boost to area tourism as it is situated situated right at the heart of Kigezi Tourism Circuit. It boarders Kihihi airstrip.
Tourism
Kihihi Garuga Golf Course is sorrounded by a lot of Tourism cites and activities which include:-
gorilla trekking and Batwa Forest experiences in Bwindi Impenetrable Forest National Park,Its mist-covered hillsides are blanketed by one of Uganda's oldest and most biologically diverse rainforests, which dates back over 25,000 years and contains almost 400 species of plants. More famously, this “impenetrable forest” also protects an estimated 320 mountain gorillas– roughly half of the world’s population, including several habituated groups, which can be tracked
tree climbing lions at Ishasha in Queen Elizabeth National Park,Queen Elizabeth National Park is understandably Uganda’s most popular tourist destination. The park’s diverse ecosystems, which include sprawling savanna, shady, humid forests, sparkling lakes and fertile wetlands, make it the ideal habitat for classic big game, ten primate species including chimpanzees and over 600 species of birds.
Tea plantations
Lake Bunyonyi, Lake Bunyonyi ("Place of many little birds") lies in south western Uganda between Kisoro and Kabale close to the border with Rwanda. Located at 1,962 m (6,437 ft) above sea level, it is about 25 km (15.5 mi) long and 7 km (1.35 mi) wide. The depth of the lake is rumored to vary between 44 m (144 ft) and 900 m (2,952 ft), which if true would make the lake the second deepest in Africa. It is one of the few lakes in the region that is rumored to be free of bilharzia and safe for swimming. |
Home Appraisals
When you set out to look at houses and the help of a professional, knowledgeable real estate agent, you have to know if the homes you are looking at are worth the prices the sellers are asking. This is a tedious but important part to buying a new home and it could be easy to overpay for a piece of property. A home appraisal is a review of a house that determines a house’s actual market value. Your agent understands market values and how to sell houses, so make good use of their expertise in this regard.
A formal home appraisal is a further step where a trained professional comes in and looks at a house to determine a formal valuation of the property. This cannot happen until you have applied for financing with your lender. This all follows the accepting of the offer by the seller and a purchase agreement is signed. The formal appraisal will be ordered by your lender to act as a neutral third party to assess the true value of the home. This protects their interests in case the buyer defaults on the loan in the end.
If a buyer defaults on a mortgage, the lender must take possession of the house and sell it immediately. The formal home appraisal will give them adequate information to help them do this. Their goal is to create a loan that is less than that value, so that in cases of repossession, they can sell the house to make up for the loan and the money it took for expenses of reselling. This is their biggest order to the buyer and once they are reassured of the value of the home, the commitment letter is that much closer to finalization. As always, apply to multiple lenders for a loan to maximize the best possible interest rate, which will save you thousands in the years to come over the period of the mortgage.
Acquiring the services of a home appraiser (or surveyor), whether it is by yourself or your lender (usually both) is key to providing everyone with accurate market information about the value of your potential home. Without this information, the bank has no idea what to loan you and you have no idea if you’re receiving a good deal or being ripped off. This is just one of the many legal checks and balances set in place to make the business of dealing in real estate fun and fair for every party involved.
It is a fundamental goal of IGRE to enhance and impact our local economy and culture. We mentor and train our agents consistently, volunteer and give back to those in need, and engage with our local economy every chance we get. |
Shooting blanks: Ca2+-free signaling.
Insect flight muscle is capable of very high oscillatory frequencies. In this issue of Structure, De Nicola and colleagues (De Nicola et al., 2007) describe the structure of the Ca2+ binding protein that regulates asynchronous contraction, casting light on the mechanism of stretch activation. |
Q:
LaTex vs. Word vs. etc
I'm a student and I write many labs, reports, letters, and fun little things. I use Latex for school because I don't have to worry about formatting, i.e, its like "programming" my paper and can scale to any type of print media (ex. books, newsletters, mailings, thesis papers).
If I were to take my writing to the next level and continue to improve my writing, would investing in Latex path be worth the time, or should I just stick with Word or some other platform?
A:
Any text-based "markup" format -- LaTeX, HTML, various XML schemas like DocBook, etc -- will serve you better than binary formats like Word, Pages, FrameMaker, etc. (I am aware that some of these tools export XML or SGML.) The reasons include:
Decoupling from editors. You can use your favorite tool to edit any of these, which gives you more flexibility.
Conversion to other formats is probably easier and at least no worse. If you want to take your Word document and convert it to HTML for your blog, probably you're going to end up cutting and pasting and re-adding the format directives. Converting from LaTeX to HTML, on the other hand, can probably be largely scripted (hedge because I'm very rusty with LaTeX). If your document is 5 pages long maybe you don't care much; if it's 500 pages, or you're going to be doing this a lot, you do.
It's a better foundation for producing output for multiple platforms. Depending on what you're doing you might only need to change a style sheet to go from book-style output to newsletter-style output, for instance.
|
"Hey, guys." "Where are Marshall and Lily?" "Something terrible happened." "Are they okay?" "Are they in the hospital?" "No." "Somewhere much... much worse." "Long Island!" "Tonight, we're moving out there for good." "Yeah." "Here are your spare keys to our new house." "Long Island?" "I don't understand." "You can get spray tans here." "ROBIN:" "I never let myself believe this day would really come." "It's like when they canceled" "Party of Five for the second time." "I mean when they..." "canceled sports." "Guys, we'll still see each other all the time, and we'll talk on the phone." "And I know people say that having kids means you won't see your friends anymore, but..." "Oh, God, this is good-bye, isn't it?" "No." "Stop it." "Guys, this doesn't change anything." "We are only a 46-minute train ride away." "Just pop out whenever you want." "So..." "I guess this seat is open." "(whimpering)" "I'll go get a chair." "So that's it." "No more Marshall." "No more Lily." "They're gone." "Which means... no lame married couple shooting down all my amazing ideas." "Guys, great news." "I'm the new leader of the gang!" "♪ How I Met Your Mother 7x14 ♪ 46 Minutes Original Air Date on January 16, 2012" "== sync, corrected by elderman ==" "NARRATOR:" "Marshall and Lily were almost done decorating their new home." "I like it there much better." "Yeah." "There was only one problem." "MICKEY:" "Wrong!" "This can't go here." "Kids, when he found out she was pregnant," "Lily's estranged father unexpectedly showed up on her doorstep." "It was a gesture she appreciated." "But after two weeks, they'd had just about enough." "That has to go there to cover up the splintery floorboard." "And you can't hang this picture here." "This isn't plaster;" "it's dried toothpaste." "Guys, you're so lucky I'm here." "I grew up in this house." "I know it like the back of my hand." "And this lamp absolutely cannot be here." "Why not?" "Blocks my view of the Widow Rodriguez doing her Jane Fonda workouts." "Isn't she a little old, Dad?" "Oh, you didn't see her when I was a kid." "In my mind, she'll always be 54." "I know he means well, but I can't take it anymore." "He won't leave, and-and he's acting like this is his house." "No, I know." "And these annoying notes he keeps leaving around?" ""I need absolute silence when practicing the drums."" ""Do not touch the fudge in my nightstand."" ""We need ant traps for my nightstand."" "I find myself hiding out here just because it's the only room in the house where he can't boss us around." "MICKEY (over intercom):" "Great news." "I put new batteries in the old intercom system." "Now we can communicate at all times." "Mickey out." "Look at what's become of our booth." "Looks like my old shop teacher's hand-- just sort of missing something." "Enough!" "I am sick of you two wallowing in sadness, instead of doing the healthy thing and pretending anyone who leaves you never existed in the first place." "I hate to agree with Barney's near-paralyzing abandonment issues, but he has a point." "You can't just stop living because two of your friends moved away." "Precisely." "Now, as new group leader, I say we go out and do something that we never would have done with Marshall and Lily here." "Ooh." "Remember that amazing idea I had that one time?" "Let's go to a strip club." "Nope." "So what do you say?" "Let's declare our independence with an on-da-peen dance." "On-the-peen dance." "No, we got it." "We got it." "Let's go to a strip club!" "TED:" "Well..." "I'm going to miss them anywhere." "Might as well see some cans while I'm at it." "I'll take it!" "Robin, Kevin, you in?" "NARRATOR:" "Kids, early in any relationship, there's a phase where you don't say no to anything." "Because you want to seem interesting... adventurous... and open-minded." "I call it Early Relationship Chicken." "Well, I'm open to anything." "Oh, mos' def." "So..." "So..." "I don't want to go to a strip club." "I don't want to go to a strip club." "But I don't want him to think I'm some prude." "Man, we've been saying "so" for a while." "Yeah!" "Yeah!" "(both grunt)" "Yes!" "Tonight is going to be legen" "(imitating Lily):" "Wait, are we sure it's a good idea to go to a strip club?" "Shut up, Lily!" "I'm in charge now-- dary!" "I've got no cell reception." "I wanted to call Ted." "This feels weird." "The last time I lived this far from him was my semester abroad." "You never did a semester abroad." "That's what I called it when we lived on opposite sides of campus." "Wrong!" "Please tell me you're not plugging that in there." "Yes, Mickey, that's exactly what I'm doing." "All right, fine." "I won't say anything." "It's... just that this was my room when I growing up, so I know it a little better than you." "Lot of great memories here." "Mostly just discovering my body." "That's it." "Mickey, since you got here, you've been nothing but judgmental and pushy and strangely obsessed with your adolescent sex life." "Tomorrow morning, you're out of here." "Well, I-I've just been trying to help you..." "We don't need your help!" "I..." "(door closes)" "Oh, it-it's okay, baby." "It had to be done." "(sighs)" "(electrical buzzing)" "(beep)" "MICKEY (over intercom):" "Too bad the one person who could stop you from making mistakes like that was told he has to move out next month." "Tomorrow!" "Tomorrow!" "Would either of you like a dance?" "(stammering)" "I'm always ready to rock." "(chuckles) And I'm always ready to roll." "So..." "So..." "I'm not ready to rock." "Rolling makes me dizzy." "Hell, yeah!" "Yeah!" "I'm buying my man a lap dance, 'cause that's just how I do!" "Party like a rock star!" "(trailing off):" "Party like a rock star..." "TED:" "Uh, hey, guys." "It's Ted." "Um... you're probably busy with the move still." "Um..." "Give me a call if you want to gab." "Sorry about using the word "gab."" "Ted, stop pining over Marshall and Lily." "Have some self-respect." "Now, put this fiver in your mouth so... that stripper with the lazy eye can vacuum it up with..." "Barney!" "Wh-what?" "I'm allowed to miss them, okay?" "They're my two best friends." "I'm your two best friends!" "And we're here tonight to show you there's no need to miss something when you can replace it with something better." "ANNOUNCER:" "Gentlemen, say hello to J-J-J-Jasmine!" "(Barney guffawing)" "Remember her?" "NARRATOR:" "So, a while back, we discovered the most amazing thing:" "Lily had a doppelganger who was a Russian stripper." "Look at her, Ted." "She's just like old, less-good Lily, but instead of bossing us around, she shows us her boobs." "No touching." "Okay, a little bossing us around." "I call her Better Lily." "Hey, Better Lily, please dance for my friend." "Barney, I don't need a lap dance." "I need to talk to my friends." "Who says you can't do both?" "Talk to her." "She's got ears." "Yes, she does have ears." "You want to touch my girlfriend's boobs?" "50 bucks." "Wait" " Better Lily is dating this behemoth?" "Sound like anyone you know?" "That's New Marshall." "This is our gang now." "Hey, guys, Ted again." "Starting to think you forgot about me." "Or replaced me with some cool neighbor." "What's his name?" "!" "(chuckling):" "I'm totally kidding." "Anyway, uh... just trying to remember what your faces look like." "Ted, when they serve alcohol at a strip club, you're not allowed to show your vagina." "100 bucks, and you will see the world, my friend." "They're already ignoring my calls?" "Look, I'm realistic." "I knew I'd have to get used to going to the dentist alone, but I didn't think they'd cut me out completely." "If they don't need us, we don't need them." "Yes!" "Finally!" "By the way, did you see how she opened that bottle?" "Whoa." "Lights out in the suburbs is... is really dark." "MARSHALL:" "Where's the fuse box?" "LILY:" "I don't know." "(beep)" "(singsongy):" "I know where it is." "Maybe I could fix it for you, if I wasn't so busy packing." "MARSHALL:" "You know what?" "I bet the fuse box is in the basement." "I'll just..." "I'll go find it myself." "How hard can it be?" "(Marshall groans)" "As hard as that desk you just walked into?" "(guffawing)" "Good burn, Mickey." "You deserve a little nightstand fudge for that one." "KEVIN:" "Where do I look?" "I mean, I can't look at her chest." "Damn it!" "I just did." "Maybe Robin didn't notice." "Man, he's really up in those jugs." "Shake it, baby!" "A playa's gotta play!" "This is weird." "I'll just look her in the eyes." "Is he falling in love with this stripper?" "Quick, do something cool." "Make it rain, son!" "Money ain't no thang!" "Damn it, there was a 20 in there." "(chimes tinkling)" "MICKEY (over intercom):" "Hello, Mr. Eriksen." "From the sound of the wind chimes," "I hear you've made it to the first floor." "Okay... you know what, Mickey?" "You can save the creepy game master routine." "It's only kind of terrifying me." "Oh, but the game has just begun." "I call it "Try Not to Bang into All the Furniture and Stuff on Your Way to the Fuse Box in the Basement, Marshall."" "That's just a working title." "Hey!" "Got a little boob glitter on your eyebrow, Mac Daddy." "Straight pimpin'." "Barney, you got a little bit, too." "Where?" "Did I get it?" "Yeah, you got it." "Who needs Marshall and Lily?" "We got a new gang, and we're all going to hang out together forever." "My shift is done." "Good-bye." "Wait, stop!" "Technically, this song's not over, so we still own you." "Plus we're best friends!" "Where to next?" "We go to underground poker game in mostly-abandoned insane asylum." "Yes!" "This is gonna be awesome, and when it is, I want you all to remember who led you there:" "Barney Stinson, new group leader." "(giggles) Yay!" "I'm winning all your chippies!" "Let's not drink when we're..." "I'm thirsty; it's good." "It's good." "(intercom beeps)" "LILY (over intercom):" "Marshall, I just remembered" "I saw a box of matches in the drawer by the trash can." "Okay, thanks, baby." "Lily, I can't find them." "Looking for these?" "(evil chuckle over intercom)" "Oh, right, you can't see me." "Um, well, I'm burning the matches you so desperately need." "(evil laugh)" "Aah, ooh, ow!" "(over intercom):" "Ow." "Hah!" "Oh!" "(laughs)" "Mickey... guess who found the basement stairs?" "I'll be down at that fuse box in no time, and when I get there," "I'm sure that I'll figure out how fuse boxes work, so it'll be fine." "MICKEY (muffled):" "Hey, Marshall," "I know I've been messing with you, but seriously, that second step, don't..." "I don't want to hear it, Mickey." "(yells)" "(yelling and crashing)" "Oh, my knee!" "(cymbals crash)" "Ow!" "(sighs) And watch out for my drum set." "I win again!" "I was bluffing." "I didn't even have enough cards." "Ted, please stop winning." "Hey, look, it's the New York City skyline." "♪ We built Chip City" "Uh, uh!" "♪ We built Chip City on all your dough ♪" "♪ Built Chip City Hit it!" "(high-pitched whine)" "You know Butterfly-knife." "I'd expect this from Face Tattoo... or Jagged Cheek Scar, or Larry, but not you." "Why don't I get nickname?" "Because you have too many things, Larry." "You get one thing!" "(squeals)" "(speaking Russian)" "See, that's why you bring New Marshall." "(high-pitched laugh)" "We go to party in slaughterhouse." "You come?" "Please, God, no." "(sobbing):" "Don't let me be in charge of the gang anymore." "I can't believe Dr. Kevin doesn't remember me from our sessions three years ago." "Is that the lunatic that stabbed all those prison guards?" "We gotta hit that party!" "(quietly):" "I'll never find love." "We got to hit that party!" "BOTH (weakly):" "Sweet." "Lily?" "I think we made a mistake moving out to the suburbs." "I miss our home." "I miss our booth with our friends." "(cymbal crashes)" "And I'm pretty sure I have a drumstick... somewhere bad." "Dad, why are you acting like this?" "Well, I was just trying to be helpful, but you guys want me out by the Fourth of July." "Tomorrow." "(sighs)" "You know what a helpful father, not to mention grandfather, might do?" "Get these lights back on." "(sighs quietly)" "(intercom beeps)" "Okay, Marshall, listen up." "I'm gonna get you to that fuse box." "Can't you just come down here and do it for me, please?" "I don't know anything about this house." "No, Marshall... you don't know anything about your house." "All right, we got a lot of work to do." "Wait here, we get you on VPI list for party." "You give us, uh, $200 per person." "Cover charge." "That's pretty steep for a cover charge." "Is very good party." "Is party for, uh, New York "Yunkees" and Coca-Cola." "Oh, my God." "A Coca-Cola Yankees party?" "That sounds like a real thing." "Okay, now, Marshall, take three steps forward, while ducking underneath the broken ceiling fan." "Now look to your left and spit." "(spits, sizzles)" "That's the water heater." "Want to stay away from that." "That baby is hotter than the Widow Rodriguez in a unitard." "Okay, now past the washer and dryer is the hardest part." "You're gonna have to belly-crawl underneath the ping-pong table and don't so much as nudge it." "Why not?" "Because there are 900 dominos set up in the shape of Barbara Eden." "Big, big I Dream Of Jeannie fan." "I'm clear." "(thud, dominos falling)" "(dominos continue falling)" "I don't know what that sound is, Mickey." "That's the sound of Barbara Eden never banging me." "That's what that is." "Dad, get it back." "(quietly):" "Come on." "(beep) Okay, Marshall, take three steps forward, and you should be at the fuse box." "Do you think he made it?" "I did it!" "No, Marshall... you did it." "NARRATOR:" "Fun fact: that night inspired" "Mickey's one and only successful board game" "♪ With splintery floorboards and rusty nails ♪" "♪ Make sure you don't get impaled ♪" "♪ Lites Out!" "MICKEY:" "Brought to you by Aldrin Games!" "Well, judging from how many clients that hooker has serviced," "I'd say we've been here almost an hour." "Ted, we've been robbed." "Better Lily and New Marshall would never rob us." "They're our best friends." "(squealing)" "(speaking Russian)" "Damn it, Ted." "New Marshall is an escaped zoo bear and the only thing that Better Lily is better at than our Lily is over-the-pants hand stuff." "I'm assuming." "I'll admit it." "I can't lead this group." "We need Marshall and Lily." "But they're gone, aren't they?" "I mean, it'll go from seeing them a couple times a week, to a couple times a month, and then it'll just be holiday parties." "Then the years pass, and you find out Marshall's dead, and you're there for Lily emotionally at first, but then it becomes sexual, and you feel guilty, but maybe that guilt just makes it dirtier and better..." "No, I won't let that happen." "Especially that last part." "Look, I know I've been a little drunk and naive tonight, but here's what's gonna happen." "First, we're gonna leave a note for Better Lily and Arvydas, they're worriers." "Oh, buddy." "Then, we're gonna go see Only Marshall and One-Of-A-Kind Lily because they're our best friends and they're only 46 minutes away." "Are you with me?" "Kind of resent the power grab, but okay, let's go!" "Hey, do you want to come to Long Island with us?" "Uh, well, still a few more hours before the sun comes up." "BOTH:" "So..." "No." "No?" "Is that bad?" "(exhales) No, it's great." "Ooh, that felt so good." "I hate doing things and going anywhere!" "Oh, new experiences suck!" "(laughing)" "While we're at it, I hate high-fiving you!" "I know." "We're adults who are sleeping together, not teens pledging a frat." "Oh, man, I really liked the high-five." "NARRATOR:" "The next morning, Marshall and Lily woke up to a surprise." "(wind chimes clatter)" "Hey, what are... what are you guys doing here?" "Just wanted to make sure this key worked." "(sighs) We missed you." "Give mama a hug." "ALL:" "Oh..." "Don't check your voicemail." "Y-You might have a couple weird ones." "You hear this?" "!" "This is the sound of me moving on." "(whooping)" "(giggles)" "Yeah, I'm a stripper." "Okay, who wants pancakes?" "(gasps)" "Which two of you want pancakes?" "I'll make more." "Oh, good-good-good, 'cause I want pancakes." "There you go." "Thank you, okay." "Enjoy them." "Hey, honey, look," "I'm sorry I was a jerk before." "Okay?" "I'll leave right after breakfast." "I'll crash with some friends." "Dad, you know what?" "You can stay." "Oh, thank God." "I got no friends." "I'll be here two weeks, tops." "NARRATOR:" "It wasn't two weeks." "Would the five of us always live within a few minutes of that booth?" "No, that's life, kids, but here's what I discovered." "Our booth was wherever the five of us were together." "== sync, corrected by elderman ==" "That's right, Widow Rodriguez." "Stretch it." "Work those gams." "Ooh, yeah." "Grab that left arm." "Clutch that chest." "Fall down really fast." "Oh, my God!" "Somebody call 911!" |
women’s role
Muhtar is a boy living in a farming village in Sambelia district. Nowadays his father is no longer farming but working in the big city and rarely coming home. His mother still works on the farm but it does not… read more
Pak Rahman is a farmer who lives in a remote village. He has a daughter named Fatima. She is a very clever girl. She helps her parents everyday in the garden. She is 7 years old, time to go to… read more
Wangari Maathai, a human rights advocate and Nobel Peace Prize Laureate, started a movement to plant more than 30 million trees and generate nearly 1 million jobs.
By Rachel Nuwer, September 26, 2011
Wangari Maathai, who won the Nobel Peace… read more
BANGKOK, 13 October 2009 (IRIN) –
Women are being excluded from the debate over climate change, despite being most at risk, and governments should do more to ensure their situations and views are represented, campaigners and experts say.
So far,… read more |
Jon Kyl Has Just the Rape Analogy for This Cairo Embassy Mess
Share
Rather than simply paying his respects to the dead from last night’s attack in Libya, Republican Senator Jon Kyl somehow managed to make things worse and mention rape. In reference to the controversial statement released — before the killings in Libya — by the U.S. embassy in Cairo, the one already fumbled by Mitt Romney, Kyl explained, “It’s like the judge telling the woman who got raped, ‘You asked for it because of the way you dressed.’ OK? That’s the same thing. ‘Well America, you should be the ones to apologize, you should have known this would happen, you should have done — what I don’t know — but it’s your fault that it happened.’” Nope, that’s not the same thing at all. Not even a littlebit. |
Join This Site
To join this site, ensure you have logged in with your Wikidot Account and click the button below. Enter the password provided in the news files on game into the box on the next page and click the button there, too.
"I'm pretty sure her name is Magma, Garth." Raven points out, glancing between the young lady and her friend. Roy's arrival to bind up Fiddler is met by Raven moving her foot from the downed villains chest. "Given her skill set.." Pointing to some of the cooled rock formations that were previously super heated rock, "And Am-Magma not making any sense.. unless you're a seven foot tall green wall of muscle."
Her indigo eyes fall on Amara, "Hello, ''Magma''…" Stressing it with a side glance at Tempest, there's no smile on her face but she very well ''may'' be teasing him. "…I'm Raven."
Now that the scene has calmed down some, she's a moment to feel the heightened fears of those gathered not far away watching the heroes… one in particular whose fear is nearly palpable. She searches for the source, eyes flickering around steadily from one face to another.
*
"It's just Magma," Amara starts to say, just as Raven makes the correction, shooting a brief, grateful smile toward the other woman. "A pleasure to meet you, Raven," she adds with a polite nod. "Garth, Flash, Arsenal." She leans forward to get a glimpse at the newest arrival, frowning slightly. "I would imagine he is not exactly local here."
*
"Magma. Got it." Roy whistles, as he eyes the young lady. Picking up the Fiddler, and securing him to something safe (like a pillar), Roy ticks his tongue. "I'll talk to the police." Perk of being an Agent of Checkmate. "But go ahead and take him to HQ!" he calls after Flash. Who, likely, has already zipped off in a hurry with the visitor. "We'll find out later, I suppose," he says in response to Amara, although he does stumble over the last two words as she leans forward. "Uh… hey, Raven, what's up?" he asks, looking away hurriedly.
*
"Really? Well, yeah, that does make more sense." Garth replies to Raven about his mixup of Magma and Am-Magma. Little does he know that Amara nearly gave her real name to a total stranger. He just figured that it was a codename. A lot of them make no sense to him. Like what does Psylocke, Starfire, or Harley Quinn mean? These humans and their codenames, he shakes his head at his own thought.
The look in his direction and the lack of a smile from Raven as she introduces herself to Magma does not go unnoticed, but it does go unremarked. "So, Arsenal, have you decided to completely give up your identity, or just tempting fate?" He knows in this company Roy probably has nothing to fear, but he did just perform some heroic feats in front of the public. Mind controlled public, but still, one of them might have had a smart phone recording it, or the Opera House itself.
With Magma referring to the Gravity Kid, Tempest shakes his head, “He's new to me, and I suspect, new to all of us. I'm sure Flash will figure that out after he's been patched up by the medics back at the…," and he was about to say United Nations, but corrects himself, “headquarters."
*
Amy would definitely look like a complete wreck. Then again more than a few regular people who were here at the wrong time probably do. But she's somehow on her feet, and begins to force herself up the side of the hall, putting a hand on the wall now and then to keep herself from leaning.
*
Raven looks to Roy when he steps up beside her and nods towards a group of people who are, perhaps understandably, handling this event poorly. "One of them has strong…" She's not sure exactly what the word for it is, still so far away and distant, like trying to pluck an idea out of someone’s hat from a mile away. She could reach out and bring it all in, but that could be dangerous… for herself and for everyone present.
Instead she floats a few feet off the ground and hovers towards the group of civilians who have collected near the back of the amphitheatre, "I can understand that everything you have just seen must be really harrowing." In her dull, perhaps even uncaring, sounding voice. As luck would have it, she's gotten far closer to Amy now. Casting indigo coloured eyes across the woman and moving on to inspect a pair of elderly woman and her husband who were, not terribly long ago, slap fighting each other like the Batman and Joker.
*
Amara arches a brow slightly when Roy's gaze dips, though she sticks to a shake of her head and an amused smile rather than pointing it out. When Raven floats over toward the group, she tilts her head, watching the confrontation. "She reminds me very much of a good friend," she murmurs to Garth, looking between the group. "You all have been fighting together for some time, haven't you?"
*
Roy's being good, really. Just moving over there with Raven. Because there's no place to look where Raven is involved, unless you wanted to spend time gazing deeply into the dark depths of her hood. And doing that the one time in years past got Roy a deep, deep stare followed by a 'Whatever you're looking for isn't in here, Harper'.
So no, Roy was being good and helping to check different people, following Raven's lead, and then going ahead. "Excuse me," he says to Amy. "You ok?"
*
With the survivors beginning to empty out of the Opera House, Tempest notes a seemingly shell-shocked woman propping herself up against the wall, trying to keep herself from leaning. He goes to her, looking her in the eye, "are you all right, miss?” Of course she's not all right after that ordeal, who would be. And Garth is no medic. Fortunately, there seem to be some coming in, and he gestures to one of them with a wave of his hand, “Guys, can you give me a hand over here?" The paramedics go to Amy, trying to see how she is, one of them will point a flashlight in her eye, check her pulse, ask her all the usual questions to see how she is and what she needs.
He looks up at Raven as she floats above, calming the people and giving them some sound advice. She can be awfully sympathetic when she wants to be. But just in case Amy is worried at the floating woman, he says, "That's Raven. This," and he’ll point to Arsenal, who checking on her, "is Arsenal. We're the good guys, girls too. And my name is Tempest." He’ll point out the other ones he knows in turn, giving their codenames so that the woman feels safe and secure before heading back towards the group, leaving her for a moment with the paramedic.
Just in time to hear Magma describe Raven as reminiscent to someone she knew. "Oh, no, not really. We used to, but we've only recently formed a team. We're called the New Warriors." He hopes she doesn't wince at the name, as he suggested it to the group after that bank robbery with the Eeensy Eight.
*
Amy looks up at the people talking to her. "I'm…" she says, looking at everyone. Her movements are slow, but she is cooperating with getting herself treated. "I'm.. I don't think I'm hurt," she says softly, slowly. Taking slow, deep breaths, she rubs her eyes again. "Thank you," she says to the people who says they were helping, and to the paramedics who decide she doesn't need to be taken anywhere. She then leans on the wall, and just watches the group of powerful people.
*
Raven settles back down near to Amy, looking for all the world like the very last person anyone would come to for emotional support. She reaches up slowly and pulls back the hood of her costume so that raven black hair can spill out across her nearly grey pale shoulders. "May I touch you?" She asks the woman, her voice something that would be digitally spit out of a computer. Her friends, standing around her helpfully assuring the masses that there is no harm to come from this group may give her some form of reassurance by proximity, but if anyone is hoping for a warm smile from Raven…
They'll need to find a comfortable seat and possibly order some dinner. "You have a lot of pain." Said to Amy, indigo eyes not necessarily glowing, but when do purple eyes have to glow? The gem in her forehead, however, does shimmer faintly. "I can take it away if you wish."
*
"The Warriors," Amara echoes, quirking a brow at Garth with a wry smile. "And who is Kronos?" Someone who casually knows their mythology? That's at least a little unusual. As the medics start to arrive, she finds a relatively whole seat to settle for a moment, eyes drooping half-closed as she quietly makes some repairs to the ground where she was drawing up magma earlier.
*
"It's all right," Roy replies, patting Amy on the shoulder as he offers the whole warmth of human kindness that was wholly missing from Raven's demeanor. "She's Raven. Just don't ask her to say 'Nevermore'. I tried that years ago, and all she did was give me crap about how she never thought I could read."
*
While Tempest gets Magma's reference to the son of Zeus and leader of the Titans, he shrugs his shoulders, "we haven't really picked one yet. Though I firmly believe that any one of us could fill that role. We're all leaders in our own way." He's been spending too much time under the sea as a High Councillor, he's starting to sound like a politician.
He smiles to Roy, who covers for Raven's… well, Raven-ness. Satisfied that Amy is all right, the paramedic who was with her moves on to the next person. A small group of public servants litter the opera house, checking on people, tending to wounds, doing their job well. But Amy, she gets the Warriors for company as each in turn seems to have focused their attention on her, and each other.
But at Raven's suggestion, he hesitates, not really wanting to influence things, but thinking it might be a bad idea. He opens his mouth as if to speak, but instead, chooses to close it, saying nothing on the matter. Pain is an integral part of life, as important, if not more, than pleasure. Without pain and pleasure, is life really worth living? A balance must be maintained. Oh, why does he think such things, plagued by thoughts on philosophy, and so rarely ever giving voice to it?
*
Amy tenses up a little. Watching Raven, she leans back a bit, her back and the back of her head against the wall, which forces her to stand up straight. She takes a deep breath, but decides Raven must be one of the good people, since nobody's doing anything about this. Taking a deep breath, she gives a little nod. "Um… okay," she says softly, not really sure what that means, but she assumes it means some sort of physical pain. Soreness from the blasts that may come later. She has recent experience with that. She relaxes slightly when others join in, which confirms her thought that no, this isn't some bad person sneaking in and causing more trouble.
*
I'm still not entirely convinced you can." Raven says with unintentional venom in her tone, more out of defensive intent than really meaning anything harsh to Roy. Her eyes, however, remain entirely on Amy. She seems so unapologetic about the wholeness of her staring or the cool manner in which she raises her hand to press her fingers lightly against the woman's forehead and jaw. Despite the cool demeanour of her expression, the touch is surprisingly gentle and when it is made the raven haired woman closes her eyes and opens her carefully constructed defensive shields to accept the rush of emotional pain that she expects to flood into her from Amy.
The sorrows of the experience is stolen away, the fear and terror of seeing the explosion no longer Amy's to feel, but taken from her and made Raven's. The pale woman's face bundles into a network of wrinkles across her brow, her black lips press tightly together as a faint glow beats softly to the rhythm of Raven's heart beneath the fingers touching Amy's face.
Whatever anguish plagues her, Raven will take it, until she no longer wishes her to. Until she is either whole or wholly afraid of what the ''hero'' is doing and jerks away from her.
*
Amara's smile quirks at Tempest's answer. "That sounds very…Roman." It's an odd sort of compliment, but her tone certainly suggests that's what it is. For all the oddity of what's going on between Raven and Amy, Amara doesn't seem concerned, once Amy consents, at least. She's spent a considerable portion of her life at Xavier's, surrounded by extremely powerful telepaths. She's no stranger to these things. Not that she's volunteering, either.
*
"Hey, I read to my daughter," Roy retorts, keeping a hand lightly on Amy's shoulder, doing his best to make it reassuring. "Do you like green eggs and ham? I do, I do, I like them, Sam-I-Am!"
Eyes shifts back to Raven as she starts absorbing the anguish, Roy quirks an eyebrow, before trying once more to peer into the depths of Raven's hood. "Hey, Raven? You okay in there?"
*
A twinge of sadness fills Tempest's thoughts as Amy agrees to have Raven take away her pain. He would never do it, but to each their own. She's an adult, fully capable of making her own decisions, and for her, that was the correct one. But it is not what Tempest would choose. Maybe it's because he's had a life of pain, from the moment of his birth, he's been an outsider, exhaled from Atlantis, left to die alone… but he didn’t. He survived and grew to be the man he is today.
With a lean of his neck to the right, he reaches up to slowly brush his hands against his forehead, running the tips of his fingers over the scars, a reminder of the pain he has endured in his still young life. And as he contemplates life, Raven absorbs Amy's pain, taking it unto herself. Garth watches with interest, but there is so much more at play than what he can see on the surface.
Though his introspection is broken by Magma's words. He doesn't understand her meaning, not being familiar with her people, but he takes it as a compliment, “thank you. Is that where you're from, Roma?" Her accent makes it difficult for him to place her. It's a human tongue, but it doesn't quite sound like any language he's ever heard. Bits and pieces yes, but hers seems unique, almost as if it were progenitor language.
So much for keeping a straight face. Garth bursts into laughter, hearing Roy rhyme, the sort of things he would teach to his little girl, Lian. "And I'm sure that she loves those books, R… Arsenal. I’m sure she does." That Lian, so adorable, he could just eat her up and tickle her until she squeals like the little girl she is.
*
Amy is kind of surprised. She expected some sort of physical pain relief, but instead… she's feeling emotionally drained. Instead of broken, she's instead getting tired, so her reaction to the surprise is a bit muted. All the stress, all of how she's been on edge, constantly afraid of someone popping up out of the blue and fighting or killing. All the jumpiness every time there's a loud noise, wondering if some stranger around her is abruptly being killed, it's just… turning into nothing. Her posture relaxes, her face falls from strain to passivity. Staring at Raven… she closes her eyes a moment, takes a breath, and then asks quietly, in a flat voice "What… what is happening?"
*
The gem set into Raven's forehead is the first indicator that whatever it is she's taken away from Amy was of far greater potency than she had originally imagined. The Chakra is at a constant shimmer as it helps the woman deal with her daily emotional state, small things like frustrations and doubts, but suddenly it is like someone turned on a lighthouse's beam and shined it straight up into the sky.
The woman's eyes snap open as it floods into her. If the rhythmic pulse of energy coming from her hand is an indication of Raven's heartbeat, it's suddenly sky rocketed and her jaw is quivering, but she doesn't let Amy go. She doesn't scold her hand away like a child touching a hot plat. "You have… a lot of pain…" Said quietly, tears rolling down her cheeks as she stares at Amy.
When she finally moves her hand away, she immediately reaches up to pull her hood back up over her hair and turns away from the woman. Not out of dismissal, but to hide behind the protective shadows.
It has been a very long time since Raven cried.
"Whoever did this to you, they are a monster." She says with every bit the lack of empathy as she spoke earlier, but there's no short of genuine regard in her flat near robotic tone. "I am very sorry you had to…" Swallowing, she shakes her head and leans against the back of one of the seats with one hand supporting her weight.
*
The words themselves convey what Raven must be feeling, even with the lack of inflection in her voice. Now it's Raven's turn to feel Roy's hand on her shoulder. "Hey, hey… was… this isn't normal, is it?" Or at least, much more than expected for being terrorized by the Fiddler. "What's going on?" Lowering his voice, Roy leans forward. "Rachel…?" he asks quietly, for her ears only. Unspoken, but implict in his tone: 'We're here.'
*
"Nova Roma, to be precise," Amara nods to Garth's question, still watching Raven and Amy. "It is in Brazil. The Amazon. Your friend needs…" Raven's reaction isn't entirely unknown to Amara either, and the young woman unfolds from her own seat to move toward her, brows furrowed in concern. When Roy moves, though, she pauses, not wanting to intrude. Which doesn't mean she isn't on stand-by.
*
Whatever Raven is up to, it seems to be working. Amy looks to be relaxed. But then Garth notices how it affects Raven. He can't help but think that her reaction was not planned. But then, with one single drop of water, any thought that she got more than she bargained for is confirmed. Even before it pours out of the duct in Raven's violet eye, he could sense it coming. It was too close to the surface. “Raven…” he says, concern in his voice.
But he trusts her, and there’s not much he could do to help. It's a little beyond his area of expertise, but he is there, with the rest of their friends. The moral support would have to be enough. And then he penny drops, and he looks to Amy. Would could it be? It sounds serious. But for now, it will remain between Raven and their new friend… did she give her name? In all the confusion, he couldn't remember if she had given it. Probably best not to ask. That’d have to be viewed as rude, right?
He's thankful to have Roy as a friend. Even before he could think of it, Roy was already there, comforting Raven, with a hand on her shoulder. He thinks about adding a 'we all are’, but thinks better of it, letting Roy's words sink in, and their looks, their presence, speak for them.
For a moment, he forgets about Magma. But he is thankful for her presence, to raise the tone to a nicer, and brighter area. "Nova Roma, Brazil?" The two concepts confuse him. He had thought they spoke Portugese in Brazil. With them all looking at Raven, it is clear that even though they don't know these two new women as well as each other, but they are definitely keepers.
*
Amy shakes her head. Staring at Raven now, she begins to realize what was going on. Walking over to Raven, she bends over a little, and looks. "If you… if you're feeling what I am… it's because of this… and because of Joker and his um, friend." Amy didn't quite catch Harley's name, when she had a gun to Amy's head for use as a hostage. Well, before leaving her on a roof with blood all over her and exploding said roof not that far from her. "It… well thank you but.. I hope you're okay."
*
Raven closes her world off within the comforting silence of the shadow of her hood. The world outside a distant memory as she surrounds her mind with a blissful calming silence, outwardly squeezing the cushion of the chair supporting her weight and only hearing Roy's assurances as if over some great distant and through a thick fog. Her posture is tense, near rigid and her entire demeanour has shrunk in on itself like a turtle collapsing into its shell for protection.
When her eyes open only moments have passed, but to her it feels as some great eternity. Indigo eyes are bloodshot and rimmed red from the river of tears, but she looks every bit as stoic when she turns her head to regard Amy. "The Joker." and his little friend. Raven doesn't keep tabs on Gotham City, but she's heard that name before…
"I will be fine." Her voice is deadened, even more so than usual, as she wrestled a river worth of anguish down into a small mental compartment. She releases a quiet, steady sigh and looks between her two friends and the two women beside her.
"What?" She snaps, since everyone is staring at her…
*
"It is complicated," Amara says with a brief, rueful smile for Garth. She looks back to Raven, but when the other woman snaps, simply raises a hand in an accepting gesture, looking back to Garth as if it were an entirely natural change of topic. "We've been apart from the rest of the world for some time. Shortly after the death of Julius Caesar, our people sailed away from Rome and founded our city in the Amazon. We've been there ever since, preserving the ideals of the Republic. It's only in the last few years that we've begun to re-establish contact with the outside world. It has been…difficult, to say the least."
Leaning over, Roy tilts his head down to try and get a better look at Raven's face. Never mind invading her personal space, in this case, she'd probably stand a better chance of grounding herself if she saw a friendly face.
Of course, once she turns her head to face Amy, Roy backs off quickly. Not until she asks, does Roy finally smile, before the smile transforms into a crooked half-grin. "Yo, short, dark, and cute. Feeling better?"
*
At the mention of the Joker, Garth winces a little. He had a run in with them shortly before he caught up with Nightwing, hoping to bring him into the fold. "Harley Quinn," he states, guessing that she is the friend Amy was talking about. "A woman in a black and red jester outfit? Then that's Harley Quinn. Unless of course the Joker's started dressing his henchpeople like that." Stranger things have happened.
It's good to see that Raven hasn’t been affected by the ordeal. When she snaps at everyone staring at her with a deadpanned ‘what’, that's exactly what he'd have expected her to say. But then, that’s why he’s a High Councillor and not a counsellor. Though similar in sound, their differences are profound in meaning.
To Magma, he smiles, "I’d love to hear it, when we're not in the middle of a crisis. Or cleaning up a crisis." Not that they’re doing much of the cleanup. The emergency personnel are doing the hard work, sorting out lives, helping people in duress. But it seems that she gives him the cliff notes. "Sounds fascinating." And though a short response, his tone seems serious, not sarcastic.
Watching over at the way Roy is attending to Raven, he wonders to himself if romance could be in the cards? Probably a one-sided crush, but there's something there. Not that he blames Roy. Raven’s very appealing… in a somewhat unconventional way. Still, she's remarkable.
*
Amy nods at Garth. "Yeah…" She expects to wince even thinking about that night again, but surprisingly.. it's just nothing. She smiles very weakly when Raven seems to be fine. "You're.. stronger than I am," she says quietly. "If you're totally fine, feeling what I was feeling, you're amazing." She then looks around to see how others are reacting to Raven, to see if there's more that she should know about, but doesn't.
*
Raven's eyes snap to Roy and if looks could kill… It softens fairly quickly though. Not much beyond a bland stare, mind, but that's got to be better than staring daggers right? "Do not call me that. Ever again." She warns, but the venom is all out of her now. Even if she'd intended for their to be any in it, she's just sort of saying it without inflection.
In the end she just looks tired. Too tired. She breaths out another sigh and looks to the civil servants arrived to take the ''heroes'' places as shepherd’s for the wounded. "I need to go." She informs everyone and takes a step back away from them, whipping around to hold one hand straight out where a pulsating purple/black hole appears which swallows her without giving anyone an opportunity to put up any reasons why she shouldn't.
It snaps closed behind her without any circumstance or sound.
*
"She is very much like my roommate," Amara observes as Raven makes her getaway, apparently unfazed by the teleportation or the quick departure. As it seems the worst of the danger is past, though, she takes a step back from the others, looking toward where she stashed her shoes. "Whom I should probably call, actually. The bus was late enough getting here before it turned into a war zone. I can only imagine how late they'll be running to go home."
*
"Whatever you say, Spookie," Roy replies, deciding to keep the tone light and keep Raven nettled at him rather than brood over the Joker. "So how about…"
And then Raven pops out, and Roy squints at where she was, before he looks back at the others. "Was it my breath? I'm pretty sure I took a Mentos before I came here."
*
If he knew what to do, he would help Amy. But he's not sure what there is that he can do? Raven’s probably already done wonders, though Garth will never understand how or why. Such things are beyond mortal Atlanteans. "Quinn's some piece of work. As crazy as the Joker, and twice as deadly." While he seemed unable to aim, she hit him with a few bullets when he last met them. Thankfully for him, his skin is bullet resistant. But it still stings like hell.
"That’s our Raven," he says after she politely informs them that she must go, and then she disappears into a portal. "One of these days, I’m going to learn how she does that.” He probably could. He's a sorceror too, though Idyllist magic is somewhat different from what she practices. For one, Idyllist magic is entirely physical. No chants, words, or speech of any kind.
"If you’d like, Magma, I could drive you home? My car was parked outside, and it looks like I won’t be giving Raven a lift. That goes for the rest of you, by the way. Plenty of room.” Not really. He drives a sports car, but there is still enough room for the four of them. Five would have been pushing it.
“I wouldn’t worry about it Arsenal. Some women like to play hard to get. It’s the aggressive ones you have to be worried about.” Not that he’s had much experience in that department.
*
Amy physically looks almost like a different person. Her face calm, her motions smoother, she watches the spot where Raven was, and just stares off into space there. "Who was that?" she asks aloud, for anyone who would listen. "She didn't tell me her name."
*
"Thank you, but it's all the way out in Westchester," Amara shakes her head to Garth, smile faint. "I wouldn't want to impose." Or admit that she's still rooming at what is ostensibly a private school, for that matter. Roy gets a shake of her head, just enough to not quite hide her amused smile.
*
"Huh?" Now Roy blinks at Garth. "What're you talking about, Tempest?" Eyes narrow, as he considers the statement, before eye rolling. "This is the -same- stuff we did when we were younger, it's not like it means anything. It's just Spookie being Spookie." And besides, one would like to keep Raven nice and even before she sprouted four more eyes. Not that he was volunteering -that- tidbit in front of the others.
"Anyway, I've got my own ride, but I'm going to talk to the police and turn the fiddler over first. Enjoy yourself." A wink at Amara, followed by a look at Amy. "Uh… you, I think, probably need to talk to a counselor. Even if Raven took care of most of it."
*
Although Garth had identified Raven to Amy, she can be forgiven for not picking up on it. She's still in shock, he introduced her to a number of people, and there was the whole ordeal of having her pain taken away. It’s a lot to handle if you’re not used to this kind of thing. “That was Raven. She's very nice. A bit… distant, but nice.”
Garth chooses not to address Roy's protestations. He thinks they’d make a good couple, if there really was something to it. He could be wrong. Maybe it’s just concern. Until recently, he hadn’t seen either of them in a while. And Roy has Lian. For all he knows, he’s back with the mother. Now what was her name? Something English, Carlisle, Chester, Coventry, something like that, Westchester? Nah, but he’ll remember it, eventually.
So, Raven made her exit. Roy’s got his own ride, and Amy’s probably going to get herself checked out by a paramedic, again, just to be safe? Looking to Amara, he says in a kind tone, “Westchester? That’s doable. I have a friend out in Salem Centre I’ve been meaning to visit. A bit earlier than I had planned, but yeah, that’s doable. Up to you?”
*
Amy nods at those telling her to get checked out again. She doesn’t think she's hurt, and she feels, well, tired, drained from all the eventfulness, but better than she has since the whole thing at the fair in Gotham. "Raven. She helped me a lot, whatever she did."
*
"Are you certain?" Amara hesitates, checking the time. "I would hate to call her so late," she admits. "I'm sure she's busy." And besides, trips through Limbo always go so…strangely. "If it isn't any trouble," she finally concedes, "I would appreciate it. I just need to pick up my things." She holds up a finger, picking her way barefoot through the damage toward where she stashed her purse and shoes. Women. Priorities.
*
Waving a farewell to the others, Roy settles in next to the Fiddler, waiting for the police.
"So…" the redhead comments. "Don't suppose you know how to play 'Jailhouse Rock', do you?"
*
"She does that kind of thing,” Garth says to Amy, explaining how she helps out perfect strangers, and then leaves before they can even really thank her. She’s a weird one, but they love her. She’s great in her own reclusive way. They just wish they could see more of her. Maybe being a Warrior will be good for her, bring her out of that shell.
"Yeah,” he answers to Amara, not finding a problem in going for a long drive with a hero, “I mean, as long as you promise not to melt my seat covers or anything like that.” She did turn into some kind of fiery lava being, and he’s pretty sure that it’s not covered by his extended warranty. Somehow, the sight of Magma looking for her purse and shoes, moving rubble, debris, and web cushions, just tickles Garth. It’s funny. Yep, women. Priorities. Nuff said.
*
Amy nods slowly to what Garth says, then she turns to leave. "Well, tell her thank you for me," she says slowly, before starting to resume her climb up toward the exit, much smoother now, relaxed.
*
Looking to Amy, Garth nods his head. He glances towards the main part of the theatre, noting that Amara's still not back with her shoes and purse. Women. They must be really nice shoes and purse. And how come her dress didn't burn up when she went all fiery? He'd have to talk to her about that. After all, he did have a couple of hours to figure it out. Maybe he'd even get her first name. Something with an Am he imagines.
But they still have to wait for her to find the purse and shoes. It could take a while, and he'd offer to help, but it's kind of hard for more than one person to search an aisle, not like he can fly or anything. He wishes he could fly. That'd be cool. How does Namor do it?
"Sure, will do Amy," he replies, though he has a sinking suspicion that she'll be able to thank Raven in person. She has the look about her. There's more to her than meets the eye.
With Roy dealing with the Fiddler, Flash already back at the headquarters with Gravity Kid, Raven probably at home googling craigslist and google, and he able to take Magma home, he thought to himself, this was a pretty good night. Sure, there was some super villainy, but in the end, everything worked out. There didn't seem to be any fatalities.
Little does he know that the owner of that multimillion dollar oboe was going to be coming after him for wanton destruction of a loaner. Oh well, that'll be for tomorrow. Tonight, the Warriors did well. There would be many more days like this. |
---
abstract: |
In a recent result, Khot and Saket \[FOCS 2014\] proved the quasi-NP-hardness of coloring a $2$-colorable $12$-uniform hypergraph with $2^{{(\log n)}^{\Omega(1)}}$ colors. This result was proved using a novel outer PCP verifier which had a strong soundness guarantee. We reduce the arity in their result by modifying their 12-query inner verifier to an 8-query inner verifier based on the hypergraph coloring hardness reductions of Guruswami [[*et. al.*]{}]{} \[STOC 2014\]. More precisely, we prove quasi-NP-hardness of the following problems on $n$-vertex hypergraphs.
- Coloring a $2$-colorable $8$-uniform hypergraph with $2^{(\log n)^{\Omega(1)}}$ colors.
- Coloring a $4$-colorable $4$-uniform hypergraph with $2^{(\log n)^{\Omega(1)}}$ colors.
author:
- 'Girish Varma[^1]'
bibliography:
- 'jrnl-names-abb.bib'
- 'prahladhbib.bib'
- 'crossref.bib'
- 'doc.bib'
title: 'Reducing uniformity in Khot-Saket hypergraph coloring hardness reductions'
---
Introduction
============
The discovery of the low-degree long code aka short code by Barak [[*et. al.*]{}]{} [@BarakGHMRS2012] has over the last one year led to a sequence of results improving our understanding of the hardness of constant colorable hypergraphs with as few colors as possible. A $k$-uniform hypergraph is a collection of vertices and hyperedges such that every hyperedge is a subset of $k$ vertices. A hypergraph is said to be $q$-colorable if the vertices of the hypergraph can be colored with at most $q$ colors such that no hyperedge is monochromatic. An independent set in a hypergraph is a collections of vertices such that no hyperedge is wholly contained within the collection. Note that a hypergraph is $q$-colorable iff it the set of vertices can be partitioned into at most $q$ independent sets.
Prior to the low-degree long code, all hardness reductions for hypergraph coloring where proved using the long code [@BellareGS1998] which resulted in a huge disparity between the best known positive and negative results for hypergraph coloring: the best known approximation algorithms require at least $n^{\Omega(1)}$ colors to color a constant colorable (hyper)graph while the inapproximability results could only rule out at best $(\log n)^{O(1)}$ colors. The situation was redeemed by introduction of the low-degree long code, a derandomization of the long code, which was then adapted by Dinur and Guruswami [@DinurG2013] toward proving inapproximability results. Building on the Dinur-Guruswami framework, Guruswami [[*et. al.*]{}]{} [@GuruswamiHHSV2014] showed that it is quasi-NP-hard to color a 2-colorable 8-uniform hypergraph with $2^{2^{\Omega(\sqrt{\log \log n})}}$ colors. Both the Dinur-Guruswami and Guruswami [[*et. al.*]{}]{} results were obtained by modifying the innermost PCP verifier to work with the low-degree long code. Shortly thereafter, in a remarkable improvement, Khot and Saket [@KhotS2014b] showed that it is quasi-NP-hard to color a 2-colorable 12-uniform hypergraph with $2^{(\log n)^{\Omega(1)}}$ colors. They obtained this result by using an 12-query inner PCP verifier based on the quadratic code, ie., a low-degree long code with degree two. However, to use a quadratic code based inner verifier, they needed an outer PCP verifier with a significantly stronger soundness guarantee than the standard outer PCP verifier obtained from parallel repetition of the PCP Theorem. In particular, they needed an outer PCP verifier, which in the soundness case, would not be satisfied by a short list of proofs even in [ *superposition*]{}[^2]. The construction of this outer PCP verifier with this stronger soundness guarantee is the main technical ingredient in the result of Khot and Saket [@KhotS2014b]. We show that this outer PCP verifier of Khot and Saket can in fact be combined with a 8-query inner PCP verifier based on the Guruswami [[*et. al.*]{}]{} inner PCP verifier to obtain a hardness result for 2-colorable 8-uniform hypergraphs. More precisely, we show the following.
\[thm:2c8u\] For every constant $\epsilon>0$ there is a quasi-polynomial time reduction from ${3\text{-}\mathsf{SAT}}$ to a $8$-uniform hypergraph $\mathcal G$ on $n$ vertices such that,
1. YES Case: If the ${3\text{-}\mathsf{SAT}}$ instance is satisfiable then $\mathcal G$ is $2$-colorable.
2. NO Case: If the ${3\text{-}\mathsf{SAT}}$ instance is unsatisfiable then $\mathcal G$ does not have an independent set of relative size $2^{-(\log n)^{\frac{1}{20} -\epsilon}}$.
Guruswami [[*et. al.*]{}]{} [@GuruswamiHHSV2014] also proved how to reduce the uniformity in certain reductions from 8 to 4 at the cost of increasing the number of colors from 2 to 4. We note that a similar trick can be performed in our setting to obtain the following result.
\[thm:4c4u\] For every constant $\epsilon>0$ there is a quasi-polynomial time reduction from ${3\text{-}\mathsf{SAT}}$ to a $4$-uniform hypergraph $\mathcal G$ on $n$ vertices such that,
1. YES Case: If the ${3\text{-}\mathsf{SAT}}$ instance is satisfiable then $\mathcal G$ is $4$-colorable.
2. NO Case: If the ${3\text{-}\mathsf{SAT}}$ instance is unsatisfiable then $\mathcal G$ does not have an independent set of relative size $2^{-(\log n)^{\frac{1}{20} -\epsilon}}$.
We remark that the analyses of the inner verifier in both the above theorems is simpler than the analyses of the corresponding inner verifiers in Guruswami [[*et. al.*]{}]{} and Khot-Saket results. Furthermore, in the language of covering complexity [^3] introduced by Guruswami, H[å]{}stad and Sudan [@GuruswamiHS2000], (the proof of) demonstrates a Boolean 4CSP for which it is quasi-NP-hard to distinguish between covering number of 2 vs. $(\log n)^{\Omega(1)}$.
Preliminaries
=============
Let ${\mathbb{F}}$ denote the field $GF(2)$. Let ${\mathbb{F}}^{m\times m}$ be the vector space of $m\times m$ matrices over the field ${\mathbb{F}}$. Our inner verifier is based on the quadratic code, which is a specialization of the low degree long code to degree $2$.
The quadratic code of $x \in {\mathbb{F}}^m$ is a function $A_x:{\mathbb{F}}^{m\times m} \rightarrow {\mathbb{F}}$ defined as $A_x(X):= \langle X, x\otimes x \rangle$.
Our reductions makes use of the following outer PCP verifier of Khot and Saket [@KhotS2014b]. As stated in the introduction, these instances have stronger soundness conditions which make them amenable for composition with a quadratic code based inner verifier.
\[thm:quad-label-cover\] There is a quasi-polynomial time reduction from an instance of ${3\text{-}\mathsf{SAT}}$ to a bi-regular instance $(U,V,E,\Pi)$ of Label Cover such that
- Vertex sets $U$ and $V$ are bounded in size by $N$.
- The label sets are ${\mathbb{F}}_2^{r\times r},{\mathbb{F}}_2^{m\times m}$ for $U$ and $V$ respectively.
- For $e \in E$, the map $\pi^e:{\mathbb{F}}_2^{m\times m} \rightarrow {\mathbb{F}}_2^{r\times r}$ is a linear transformation that maps symmetric matrices to symmetric matrices[^4]. For an $r\times r$ matrix $X$, $X\circ \pi^e$ is the unique $m \times m$ matrix such that $\langle X \circ \pi^e, Y \rangle = \langle X , \pi^e Y \rangle$.
- For each vertex $v \in V$, there is a constraint $C_v$ that is a a conjunction of homogeneous linear equations on the entries of the $m \times m$ matrix label.
- $\delta \leq 2^{- \log^{1/3} N}$ and $k \geq (\log N)^{1/9}$.
The reduction satisfies:
1. Completeness : If the ${3\text{-}\mathsf{SAT}}$ instance is satisfiable then there is a labeling $x_u \otimes x_u$ for $u \in U$ and $y_v \otimes y_v$ for $v\in V$ such that
- for each $v \in V$, $y_v \in {\mathbb{F}}_2^m$ has the $m$[^th^ ]{}coordinate $1$ and $y_v\otimes y_v$ satisfies the constraint $C_v$,
- for each $(u,v) \in E$, $\pi_{u,v}(y_v \otimes y_v) = x_u \otimes x_u$.
2. Soundness : If the ${3\text{-}\mathsf{SAT}}$ instance is not satisfiable then the following cannot hold: There are symmetric matrices $M_u \in {\mathbb{F}}_2^{r\times r}, M_v \in {\mathbb{F}}_2^{m\times m}$ for $u\in U, v\in V$ of rank $\leq k$ such that
- for each $v \in V$, $M_v \in {\mathbb{F}}_2^{m \times m}$ has the $(m,m)$[^th^ ]{}coordinate $1$ and $M_v$ satisfies the constraint $C_v$,
- for $\delta$ fraction of edges $e$, $\pi_e(M_v) = M_u$.
3. Smoothness : For any $v \in V$ and any symmetric non-zero matrix $M_v$ with rank $\leq k$, over a random choice of an edge $e$ incident on $v$, $$\Pr[\pi_e(M_v) = 0] \leq \delta/2.$$
For our reduction in the case of $3$-uniform hypergraphs, we need to construct a multi-layered label cover from the bipartite instance described above. We also need all the matrix labels to be over the field ${\mathbb{F}}_3$. Multilayered label cover has been used widely used for proving hypergraph coloring hardness results in \cite{}.
The parameter $\ell$ will denote the number of layers in the multi-layered label cover that is constructed. Let $(U,V,E,\Pi)$ be an instance of the bipartite label cover with the matrices in the label set over ${\mathbb{F}}_3$.
- For $0 \leq i < \ell$, the vertices in the $i$th layer are $V_i = U^{i}\times V^{\ell - i}$.
- The label set $\mathcal{L}_i$ for $V_i$ is $({\mathbb{F}}_3^{r\times r})^i \times ({\mathbb{F}}_3^{m\times m})^{\ell - i}$. We will think of $\mathcal{L}_i$ as the set of block diagonal matrices having dimension $m_i :=i\times r+ (\ell - i) m$, with the previously mentioned matrices in the diagonals.
- There is an edge between $u=(u_1,\cdots, u_\ell) \in V_i$ and $v =(v_1,\cdots, v_\ell)\in V_j$, if $(u_k,v_k)\in E$ for $k \in \{i+1, \cdots, j\} $ and $u_k = v_k$ otherwise.
- The projection constraint is a product of constraints for each coordinate. For $k \in \{i+1, \cdots, j\}$, it is the constraint $\pi_{u_k,v_k} \in \Pi$ corresponding to the edge $(u_k,v_k)\in E$ and identity otherwise.
The construction of the $\ell$-layered Label cover satisfies:
1. Completeness : Given a labeling for the bipartite instance $x_u \otimes x_u$ for $u \in U$ and $y_v \otimes y_v$ for $v\in V$ that satisfies all the constraints, the labeling $(x_{u_1} \otimes x_{u_1},\cdots,x_{u_i} \otimes x_{u_i}, y_{v_{i+1}} \otimes y_{v_{i+1}}, \cdots, y_{v_\ell} \otimes y_{v_\ell})$ for the vertex $(u_1,\cdots, u_i,v_{i+1},\cdots, v_\ell) \in V_i, \forall i \in \{0,\cdots \ell -1\}$ satisfies all the constraint in the multi-layered label cover instance.
2. Soundness : If the bipartite label cover was obtained from a unsatisfiable instance of ${3\text{-}\mathsf{SAT}}$ then in the multi-layered label cover for every $0\leq i < j <\ell$, the following cannot hold: There are symmetric matrices $M_u \in \mathcal{L}_i, M_v \in \mathcal{L}_j$ for $u\in V_i, v\in V_j$ of rank $\leq k$ such that
- for any vertex $u=(u_1,\cdots,u_i,v_{i+1},\cdots,v_\ell)$, the diagonal blocks in $M_u$ corresponding to $v_{i+1},\cdots,v_\ell$ have the $(m,m)$th coordinate $1$ and satisfies the constraints $C_{v_i}$.
- for $\delta$ fraction of edges $e=(u,v)$, $\pi_e(M_v) = M_u$.
3. Smoothness : For any $u \in V_i$ and any symmetric non-zero matrix $M_v \in \mathcal{L}_i $ with rank $\leq k$, over a random choice of an edge $e$ incident on $u$, $$\Pr[\pi_e(M_v) = 0] \leq \delta/2.$$
4. Weakly Dense : For any $\delta >0$, given $\ell' \geq 2/\delta$ layers $l_1 < l_2 < \cdots < l_{\ell'}$ and given any sets $S_i \subseteq V_{l_i}$ with $|S_{l_i}| \geq \delta |V_{l_i}|$, there always exists two layers $i$ and $j$ such that the edges between $S_i$ and $S_j$ is at least $\delta^2/4$ fraction of the edges between the layers.
$2$-colorable $8$-uniform hypergraphs
=====================================
In this section we prove . Our reduction starts from the label cover instances given by . Let $(U,V,E,\Pi)$ be an instance of the label cover. We will construct a hypergraph $\mathcal G=(\mathcal V, \mathcal E)$. For $v\in V$, let $\mathcal H_v \subseteq {\mathbb{F}}_2^{m\times m}$ be the dual of the subspace of the set matrices that are symmetric and which satisfies the constraint $C_v$. The set of vertices $\mathcal V$ will be the same as $V\times\left( {\mathbb{F}}_2^{m\times m}/\mathcal H_v\right)$. Any $2$-coloring of $\mathcal G$ is a collection of functions $A'_v:{\mathbb{F}}_2^{m\times m}/\mathcal H_v \rightarrow \{0,1\}$ for $v\in V$. For any such function, we can uniquely extend it to get $A_v:{\mathbb{F}}_2^{m\times m} \rightarrow \{0,1\}$ which is constant over cosets of $\mathcal H_v$. This method is called folding and it ensures that $A_v$ satisfies the following: if $\alpha \in {\mathbb{F}}_2^{m\times m}$ is such that $\widehat A_v(\alpha)$ is non-zero, then $\alpha$ is symmetric and satisfies $C_v$.
####
The set of edges $\mathcal E$ will be defined by the test mentioned below, which checks whether a supposed $2$-coloring $A'_v:{\mathbb{F}}_2^{m\times m}/\mathcal H_v \rightarrow \{0,1\}$ is valid. There is an edge in $\mathcal E$ between any set of vertices in $\mathcal V$ that are queried together by the test. The test will be querying the extended functions $A_v$ at matrices in ${\mathbb{F}}_2^{m\times m}$ instead of $A'_v$. So a query to $A_v$ at $X \in {\mathbb{F}}_2^{m\times m}$ corresponds to a query to $A'_v$ at the coset of $\mathcal H_v$ that contains $X$.
#### $2$-Colorable $8$-Uniform Test $\mathcal T_{2,8}$
1. Choose $u \in U$ uniformly at random and $v,w \in V$ uniformly and independently at random from the neighbors of $u$. Let $\pi,\sigma:{\mathbb{F}}_2^{m \times m} \rightarrow {\mathbb{F}}_2^{r \times r}$ be the projections corresponding to the edges $(u,v),(u,w)$ respectively. Uniformly and independently at random choose $X_1,X_2, Y_1,Y_2 \in {\mathbb{F}}_2^{m \times m}$ and $\overline x, \overline y, \overline z , \overline x', \overline y', \overline z' \in {\mathbb{F}}_2^m$ and $F \in {\mathbb{F}}_2^{r\times r}$. Let $\overline e_m \in {\mathbb{F}}_2^m$ be the vector with only the $m$[^th^ ]{}entry $1$ and the rest is $0$.
2. Accept if and only if the following $8$ values are not all equal : $$\begin{aligned}
&A_v(X_1)&&A_v(X_3) &\text{ where } X_3 &:= X_1+\overline x \otimes \overline y +F\circ \pi\\
&A_v(X_2)& &A_v(X_4) &\text{ where } X_4 &:= X_2 + (\overline x + \overline e_m) \otimes \overline z + F \circ \pi\\
&A_w(Y_1)&&A_w(Y_3) &\text{ where } Y_3 &:= Y_1+\overline x'\otimes \overline y'+F\circ \sigma + \overline e_m \otimes \overline e_m\\
&A_w(Y_2)&&A_w(Y_4) &\text{ where } Y_4 &:= Y_2 + (\overline x'+\overline e_m)\otimes \overline z' +F \circ \sigma + \overline e_m \otimes \overline e_m\end{aligned}$$
YES Case {#sec:yes-28}
--------
Let $\overline y_v \otimes \overline y_v$ for $v\in V$ and $\overline x_u \otimes \overline x_u$ for $u\in U$ be a perfectly satisfying labeling of the label cover instance. That is, for every $(u,v)\in E, \pi_{u,v}(\overline y_v\otimes\overline y_v) = \overline x_u \otimes \overline x_u$. Such a labeling is guaranteed by the YES instance of label cover, with the additional property that the $m$[^th^ ]{}coordinate of $\overline y_v$ is $1$. Consider the following $2$-coloring of $\mathcal G$: for each $v \in V$, $A_v(X):= \langle X,\overline y_v \otimes \overline y_v \rangle$. Note that such a function is constant over cosets of $\mathcal H_v$. Let $$\begin{aligned}
&x_1 := \langle X_1 , \overline y_v \otimes \overline y_v \rangle& ~~~~~&x_2 := \langle X_2 , \overline y_v \otimes \overline y_v \rangle\\
&y_1 := \langle Y_1 , \overline y_w \otimes \overline y_w \rangle& ~~~~~&y_2 := \langle Y_2 , \overline y_w \otimes \overline y_w \rangle\end{aligned}$$ and $f := \langle F, \overline x_u \otimes \overline x_u \rangle$. Note that $\langle F\circ \pi_{u,v}, \overline y_v\otimes \overline y_v \rangle = \langle F, \pi_{u,v}(y_v \otimes y_v)\rangle = \langle F, \overline x_u \otimes \overline x_u\rangle$ , and $\langle \overline e_m \otimes \overline e_m ,\overline y_v \otimes \overline y_v \rangle = \langle \overline e_m , \overline y_v \rangle = 1$. Using these, the assignments to the $8$ query locations are: $$\begin{aligned}
&x_1& ~~~~~&x_1+\langle \overline y_v, \overline x\rangle \langle \overline y_v, \overline y \rangle + f\\
&x_2& ~~~~~&x_2 + \left(\langle\overline y_v, \overline x\rangle +1\right) \langle \overline y_v, \overline z\rangle + f \\
&y_1& ~~~~~&y_1+\langle \overline y_w, \overline x'\rangle \langle\overline y_w, \overline y' \rangle + f + 1\\
&y_2& ~~~~~&y_2 + \left(\langle\overline y_w, \overline x'\rangle +1\right) \langle\overline y_w, \overline z'\rangle + f + 1\end{aligned}$$ It is easy to see at least one of the $4$ rows are always not equal. Hence $A$ is a valid $2$-coloring of $\mathcal G$.
NO Case {#sec:no-28}
-------
Suppose the reduction was applied to a NO instance of label cover. Let $k$ and $\delta$ be the parameters specified by .
\[lem:soundness-28\] If there is an independent set in $\mathcal G$ of relative size $s$ then $$s^8 \leq \delta +\frac{1}{2^{k/2+1}}.$$
The proof of the lemma is similar to Section 8.2 in Khot and Saket [@KhotS2014b]. Consider any set $A \subseteq \mathcal V$ of fractional size $s$. For every $v\in V$, let $A_v:{\mathbb{F}}_2^{m\times m} \rightarrow \{0,1\}$ be the indicator function that is extended such that it is constant over cosets of $\mathcal H_v$. $A$ is an independent set if and only if $$\label{eqn:indep}
\Theta := \operatorname*{\mathbb{E}}_{u,v,w} ~\operatorname*{\mathbb{E}}_{X_i,Y_i \in \mathcal T_{2,8}}~\prod_{i=1}^4A_v(X_i)A_w(Y_i) = 0.$$ Now we do the Fourier expansion and take expectations over $X_1,X_2,Y_1,Y_2$ to obtain the following: $$\begin{aligned}
\Theta = \operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{\alpha_1, \alpha_2\\ \beta_1,\beta_2} \in {\mathbb{F}}_2^{m\times m}} &\operatorname*{\mathbb{E}}_{F,\overline x,\overline x'} \Bigg[ \widehat A_v(\alpha_1)^2 \operatorname*{\mathbb{E}}_{\overline y}\left[\chi_{\alpha_1}(\overline x \otimes \overline y)\right]\chi_{\alpha_1}(F\circ \pi)\\
& \widehat A_v(\alpha_2)^2 \operatorname*{\mathbb{E}}_{\overline z}\left[\chi_{\alpha_2}((\overline x +\overline e_m) \otimes \overline z)\right]\chi_{\alpha_2}(F\circ \pi)\\
&\widehat A_w(\beta_1)^2 \operatorname*{\mathbb{E}}_{\overline y'}\left[\chi_{\beta_1}(\overline x' \otimes \overline y' )\right]\chi_{\beta_1}(F\circ \sigma) \chi_{\beta_1}(\overline e_m \otimes \overline e_m)\\
& \underbrace{\widehat A_w(\beta_2)^2 \operatorname*{\mathbb{E}}_{\overline z'}\left[\chi_{\beta_2}((\overline x'+\overline e_m) \otimes \overline z')\right]\chi_{\beta_2}(F\circ \sigma) \chi_{\beta_2}(\overline e_m \otimes \overline e_m) \Bigg]}_{=:\operatorname{Term}_{u,v,w}(\alpha_1,\alpha_2,\beta_1,\beta_2)}\end{aligned}$$
Note that since $F\in {\mathbb{F}}_2^{r\times r}$ is chosen uniformly at random, $$\operatorname*{\mathbb{E}}_F \chi_{\alpha_1}(F\circ \pi) \chi_{\alpha_2}(F\circ \pi) \chi_{\beta_1}(F \circ \sigma) \chi_{\beta_2}(F\circ \sigma) =\operatorname*{\mathbb{E}}_F (-1)^{\langle \pi(\alpha_1 +\alpha_2), F \rangle + \langle \sigma(\beta_1 +\beta_2), F \rangle}$$ is zero unless $\pi(\alpha_1 +\alpha_2) = \sigma(\beta_1 +\beta_2)$. Let $\nu(\alpha):= (-1)^{\langle \alpha, \overline e_m \otimes \overline e_m \rangle}$. Now taking expectations over $\overline x,\overline y, \overline z, \overline x',\overline y',\overline z'$, and noting that $\langle\alpha, x \otimes y \rangle = \langle \alpha x, y \rangle$, we obtain $$\label{eqn:sim-term}
\begin{aligned}
\operatorname{Term}_{u,v,w}(\alpha_1, \alpha_2, \beta_1,\beta_2) = (-1)^{\nu(\beta_1+\beta_2)}& \widehat A_v(\alpha_1)^2 \widehat A_v(\alpha_2)^2 \widehat A_w(\beta_1)^2 \widehat A_w(\beta_2)^2\\
& \Pr_{\overline x} \left[ \alpha_1 \overline x = 0 \wedge \alpha_2 \overline x = \alpha_2 e_m \right] \cdot\\
& \Pr_{\overline x'} \left[ \beta_1 \overline x' = 0 \wedge \beta_2 \overline x' = \beta_2 e_m \right]
\end{aligned}$$
when $\pi(\alpha_1 +\alpha_2) = \sigma(\beta_1 +\beta_2)$ and $0$ otherwise. Define: $$\begin{aligned}
\Theta_0 &= \operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{\operatorname{rank}(\alpha_1+\alpha_2), \operatorname{rank}(\beta_1+\beta_2) \leq k\\ \pi(\alpha_1+\alpha_2)= \sigma(\beta_1+\beta_2) \\ \nu(\beta_1+\beta_2) = 0 } } \operatorname{Term}_{u,v,w}(\alpha_1, \alpha_2, \beta_1,\beta_2)\\
\Theta_1 &= \operatorname*{\mathbb{E}}_{u,v,w}\sum_{\substack{\operatorname{rank}(\alpha_1+\alpha_2), \operatorname{rank}(\beta_1+\beta_2) \leq k\\ \pi(\alpha_1+\alpha_2)= \sigma(\beta_1+\beta_2) \\ \nu(\beta_1+\beta_2) = 1 } } \operatorname{Term}_{u,v,w}(\alpha_1, \alpha_2, \beta_1,\beta_2)\\
\Theta_2 &= \operatorname*{\mathbb{E}}_{u,v,w}\sum_{\substack {\max\left\{ \operatorname{rank}(\alpha_1+\alpha_2) ,\operatorname{rank}(\beta_1+\beta_2)\right\} > k\\ \pi(\alpha_1+\alpha_2)= \sigma(\beta_1+\beta_2)} } \operatorname{Term}_{u,v,w}(\alpha_1, \alpha_2, \beta_1,\beta_2)\end{aligned}$$ We lower bound $\Theta_0$ by $s^8$, upper bound $|\Theta_1|$ by $\delta$ and $|\Theta_2|$ by $1/2^{k/2+1}$ below. Along with , this will prove .
### Lower bound on Theta0
Note that all terms in $\Theta_0$ are positive. Now consider the term corresponding to $\alpha_1=\alpha_2=\beta_1=\beta_2=0$. $$\operatorname*{\mathbb{E}}_{u,v,w}\widehat A^4_v(0) \widehat A^4_w(0) = \operatorname*{\mathbb{E}}_u \left(\operatorname*{\mathbb{E}}_v \widehat A^4_v(0) \right)^2 \geq \left(\operatorname*{\mathbb{E}}_{uv} \widehat A_v(0)\right )^8 \geq s^8.$$
### Upper bound on Theta1
We can upper bound $|\Theta_1|$ by $$\label{eqn:fourier-decod}
\operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{ \operatorname{rank}(\alpha_1+\alpha_2),\operatorname{rank}(\beta_1+\beta_2) \leq k,\\ \pi(\alpha_1+\alpha_2) = \sigma(\beta_1+\beta_2),\\ \nu(\beta_1+\beta_2)=1 }} \widehat A^2_v(\alpha_1) \widehat A^2_v(\alpha_2) \widehat A_w^2(\beta_1) \widehat A_w^2(\beta_2).$$ Consider the following strategy for labeling vertices $u\in U$ and $v\in V$. For $u\in U$, pick a random neighbor $v$, choose $(\alpha_1,\alpha_2)$ with probability $\widehat A^2_v(\alpha_1) \widehat A^2_v(\alpha_2)$ and set its label to $\pi(\alpha_1+\alpha_2)$. For $w\in V$, choose $(\beta_1,\beta_2)$ with probability $\widehat A_w^2(\beta_1) \widehat A_w^2(\beta_2)$ and set its label to $\beta_1+\beta_2$. Since $A_w$ is folded, both $\beta_1$ and $\beta_2$ are symmetric and satisfies $C_v$. Since these constraints are homogeneous, $\beta_1+\beta_2$ is also symmetric and satisfies $C_v$. Also $\pi$ maps symmetric matrices to symmetric matrices. Note that gives the probability that a random edge $(u,w)$ of the label cover is satisfied by this labeling. Hence and $|\Theta_1|$ are upper bounded by $\delta$.
### Upper bound on Theta2
Note that if the $\operatorname{rank}(\alpha) >k$, for any fixed $b$, $\Pr_{\overline x}[\alpha x = b] \leq 1/2^{k+1}$. All terms in $\Theta_2$ has $\max \{ \operatorname{rank}(\alpha_1), \operatorname{rank}(\alpha_2),\operatorname{rank}(\beta_1), \operatorname{rank}(\beta_2) \} > k/2.$ From we have that, for any fixed choice of $u,v,w$ each term in $\Theta_2$ has absolute value at most $1/2^{k/2+1}$. Since $A,B$ are $\{0,1\}$ valued functions, sum of their squared coefficients is upper bounded by $1$ (i.e. Parseval’s inequality). Thus $|\Theta_2| \leq 1/2^{k/2+1}$.
We already saw in that an YES instance of label cover is mapped to a $2$-colorable hypergraph. Since $k= (\log N)^{1/8 - 2\epsilon}$ and $\delta = 2^{-(\log N)^{1/4 -2\epsilon}}$, $s \leq2^{-(\log N)^{1/8 -3\epsilon}}.$ Also the number of vertices in $\mathcal G$, $$n \leq N 2^{m^2} \leq N \cdot 2^{(\log N)^{10/4 +2\epsilon}}.$$ From and above, a NO instance of label cover is mapped to a hypergraph $\mathcal G$ that has no independent set of relative size $2^{-(\log n)^{1/20-4\epsilon}}$.
$4$-colorable $4$-uniform hypergraphs
=====================================
In this section, we modify the reduction in the previous section, so that the uniformity of the hypergraph produced is decreased to $4$ at the cost of increasing the number of colors required in the YES case to $4$. This method was proposed by Guruswami [[*et. al.*]{}]{}[@GuruswamiHHSV2014]. The hypergraph $\mathcal G=(\mathcal V, \mathcal E)$ constructed will have vertices $$\mathcal V= V\times\left( {\mathbb{F}}_2^{m\times m}\times {\mathbb{F}}_2^{m\times m}/\mathcal H_v \times \mathcal H_v\right).$$ Any $4$-coloring of $\mathcal G$ can be expressed as a collection of functions $$A'_v:\left( {\mathbb{F}}_2^{m\times m}\times {\mathbb{F}}_2^{m\times m}/\mathcal H_v \times \mathcal H_v\right) \rightarrow \{0,1\}^2, \text{ for } v \in V.$$ We can uniquely extend such functions to get $A_v:{\mathbb{F}}_2^{m\times m} \times {\mathbb{F}}_2^{m\times m} \rightarrow \{0,1\}^2$ which is constant over cosets of $\mathcal H_v \times \mathcal H_v$. This ensures that $A$ satisfies the following: if $\alpha = (\alpha_1,\alpha_2) \in {\mathbb{F}}_2^{m\times m} \times {\mathbb{F}}_2^{m\times m}$ is such that $\widehat A(\alpha)$ is non-zero, then $\alpha_1,\alpha_2$ are both symmetric and satisfies $C_v$. The set of edges $\mathcal E$ will be defined by the test mentioned below.
#### $4$-Colorable $4$-Uniform Test
1. Sample $v,w$ and $\{ X_i ,Y_i \}_{i=1}^4$ from the distribution $\mathcal T_{2,8}$ as described by the test in the previous section.
2. Accept if and only if the following $4$ values are not all equal : $$\begin{aligned}
A_v(X_1,X_2)~~~A_v(X_3, X_4)~~~A_w(Y_1,Y_2)~~~A_w(Y_3, Y_4)\end{aligned}$$
YES Case {#sec:yes-44}
--------
Given a perfectly satisfying labeling $\overline y_v \otimes \overline y_v$ for $v\in V$ and $\overline x_u \otimes \overline x_u$ for $u\in U$, we define the following $4$-coloring for $\mathcal G$: for each $v \in V$, $$A_v(X_1,X_2):= \left(\langle X_1,\overline y_v \otimes \overline y_v \rangle,\langle X_2,\overline y_v \otimes \overline y_v \rangle \right).$$ Note that such a function is constant over cosets of $\mathcal H_v$. Using the arguments from , it is easy to see that $A$ is a valid $4$-coloring of $\mathcal G$.
NO Case {#sec:no-44}
-------
The analysis of the NO case is similar to .
$2$-colorable $4$-uniform Hypergraphs
=====================================
In this section, we propose a reduction to $4$-uniform hypergraphs, where the YES case is $2$-colorable. As described in the previous sections, the reduction is defined by a test on functions $A_v:{\mathbb{F}}_2^{m\times m} \rightarrow \{0,1\}$ that is folded over cosets of $\mathcal H_v$ for each $v$ in the “big" side of the label cover instance.
#### $2$-Colorable $4$-Uniform Test $\mathcal T_{2,4}$
1. Choose $u \in U$ uniformly at random and $v,w \in V$ uniformly and independently at random from the neighbors of $u$. Let $\pi_1,\pi_2:{\mathbb{F}}_2^{m \times m} \rightarrow {\mathbb{F}}_2^{r \times r}$ be the projections corresponding to the edges $(u,v),(u,w)$ respectively and $\sigma_1,\sigma_2:{\mathbb{F}}_2^m \rightarrow {\mathbb{F}}_2^r$ be the linear operators obtained from Theorem 7.1 in [@KhotS2014b] such that $\pi_i(\alpha):= \rho_i \alpha \rho_i^T$. Uniformly and independently at random choose $X_1, Y_1 \in {\mathbb{F}}_2^{m \times m}$ and $\overline x \in {\mathbb{F}}_2^r, \overline y, \overline z \in {\mathbb{F}}_2^m$. Let $\overline e_m \in {\mathbb{F}}_2^m$ be the vector with only the $m$[^th^ ]{}entry $1$ and the rest is $0$.
2. Accept if and only if the following are not all equal: $$\begin{aligned}
&A_v(X_1)&&A_v(X_2)&\text{ where } X_2 &:= X_1+ (\sigma_1^T \overline x) \otimes \overline y + \overline e_m \otimes \overline e_m\\
&A_w(Y_1)&&A_w(Y_2)&\text{ where } Y_2 &:= Y_1+( \sigma_2^T \overline x+ \overline e_m) \otimes \overline z + \overline e_m \otimes \overline e_m\end{aligned}$$
YES case {#yes-case}
--------
Let $\overline y_v \otimes \overline y_v$ for $v\in V$ and $\overline x_u \otimes \overline x_u$ for $u\in U$ be a perfectly satisfying labeling of the label cover instance. Also the $m$[^th^ ]{}coordinate of $\overline y_v$ is $1$. Consider the following $2$-coloring of $\mathcal G$: for each $v \in V$, $A_v(X):= \langle X,\overline y_v \otimes \overline y_v \rangle$. Note that such a function is constant over cosets of $\mathcal H_v$. Let $$\begin{aligned}
&x_1 := \langle X_1 , \overline y_v \otimes \overline y_v \rangle& ~~~~~&y_1 := \langle Y_1 , \overline y_w \otimes \overline y_w \rangle.\end{aligned}$$ The assignments to the $4$ query locations are: $$\begin{aligned}
&x_1& ~~~~~&x_1+\langle \overline x_u, \overline x\rangle \langle \overline y_v, \overline y \rangle + 1\\
&y_1& ~~~~~&y_1+ (\langle \overline x_u, \overline x\rangle +1) \langle\overline y_w, \overline z \rangle + 1\end{aligned}$$ It is easy to see at least one of the $2$ rows are always not equal. Hence $A$ is a valid $2$-coloring of $\mathcal G$.
NO Case {#no-case}
-------
Suppose the reduction was applied to a NO instance of label cover. Let $k$ and $\delta$ be the parameters specified by Theorem 7.2 in [@KhotS2014b].
\[lem:soundness-24\] If there is an independent set in $\mathcal G$ of relative size $s$ then $$\operatorname{poly}(s) \leq O(\delta) +2^{-\Omega(k)}.$$
Consider any set $A \subseteq \mathcal V$ of fractional size $s$. For every $v\in V$, let $A_v:{\mathbb{F}}_2^{m\times m} \rightarrow \{0,1\}$ be the indicator function that is extended such that it is constant over cosets of $\mathcal H_v$. $A$ is an independent set if and only if $$\label{eqn:indep-24}
\Theta := \operatorname*{\mathbb{E}}_{u,v,w} ~\operatorname*{\mathbb{E}}_{X_i,Y_i \in \mathcal T_{2,4}} A_v(X_1)A_v(X_2)A_w(Y_1)A_w(Y_2) = 0.$$ Now we do the Fourier expansion and take expectations over $X_1,Y_1$ to obtain the following: $$\begin{aligned}
\Theta = \operatorname*{\mathbb{E}}_{u,v,w} \sum_{\alpha, \beta \in {\mathbb{F}}_2^{m\times m}} &\operatorname*{\mathbb{E}}_{\overline x} \Bigg[ \widehat A_v(\alpha)^2 \operatorname*{\mathbb{E}}_{\overline y}\left[\chi_{\alpha}(\rho^T_1\overline x \otimes \overline y)\right]\chi_{\alpha}(\overline e_m\otimes \overline e_m)\\
&\underbrace{\widehat A_w(\beta)^2 \operatorname*{\mathbb{E}}_{\overline z}\left[\chi_{\beta}((\rho^T_2\overline x +\overline e_m) \otimes \overline z )\right] \chi_{\beta}(\overline e_m \otimes \overline e_m) \Bigg]}_{=:\operatorname{Term}_{u,v,w}(\alpha,\beta)}\end{aligned}$$
Taking expectations over $\overline x, \overline y, \overline z$, we get $$\label{eqn:sim-term}
\begin{aligned}
\operatorname{Term}_{u,v,w}(\alpha, \beta) = (-1)^{\nu(\alpha+\beta)} \widehat A_v(\alpha)^2 \widehat A_w(\beta)^2 \Pr_{\overline x} \left[ (\rho_1 \alpha)^T \overline x = 0 \wedge (\rho_2\beta)^T \overline x = \beta^T \overline e_m \right]
\end{aligned}$$ Notice that terms with $\nu(\alpha +\beta)=0$ are all positive and $$\label{eqn:large-rank-24}
\max \{ \operatorname{rank}(\rho_1 \alpha) , \operatorname{rank}(\rho_2 \beta) \} > k \Rightarrow|\operatorname{Term}_{u,v,w}(\alpha,\beta)| \leq 1/2^{k+1}.$$
Now consider the following expectation $$\begin{aligned}
\Theta_v &:= \operatorname*{\mathbb{E}}_{X,\overline x, \overline y} A_v(X) A_v\left(X + (\sigma^T \overline x) \otimes \overline y + \overline e_m \otimes \overline e_m \right)\\
&= \sum_\alpha (-1)^{\nu(\alpha)}\widehat A_v(\alpha)^2 \operatorname*{\mathbb{E}}_{\overline x, \overline y}\chi_\alpha \left(\sigma^T \overline x \otimes \overline y\right)\\
&=\sum_\alpha \underbrace{(-1)^{\nu(\alpha)}\widehat A_v(\alpha)^2 \Pr_{\overline x}\left[(\sigma \alpha)^T \overline x = 0 \right]}_{=: \operatorname{Term}_v(\alpha)}\end{aligned}$$
If $\operatorname*{\mathbb{E}}A_v = s_v$ then $$\Theta_v \geq \operatorname{poly}(s_v)$$
If the conjecture is true, we have that $$\label{eqn:uncorrelated}
\Theta':= \operatorname*{\mathbb{E}}_{u,v,w} \Theta_v \Theta_w \geq \operatorname{poly}(s).$$ Now consider, $$\begin{aligned}
\Theta' - \Theta = \operatorname*{\mathbb{E}}_{u,v,w} \sum_{\alpha,\beta} \underbrace{\left( \operatorname{Term}_v(\alpha)\operatorname{Term}_w(\beta) - \operatorname{Term}_{u,v,w}(\alpha,\beta)\right)}_{:=\operatorname{Term}'_{u,v,w}(\alpha,\beta)} \geq \operatorname{poly}(s)\end{aligned}$$ since $\Theta = 0$ from . Now define: $$\begin{aligned}
\Theta_0 &= \operatorname*{\mathbb{E}}_{u,v,w}\sum_{\substack{(\operatorname{rank}(\alpha) \leq k \wedge \operatorname{rank}(\rho_1\alpha) \leq \sqrt k) \wedge \\ (\operatorname{rank}(\beta) \leq k \wedge \operatorname{rank}(\rho_2\beta) \leq \sqrt k)} } \operatorname{Term}'_{u,v,w}(\alpha, \beta)\\
\Theta_1 &= \operatorname*{\mathbb{E}}_{u,v,w}\sum_{\substack{(\operatorname{rank}(\alpha) \geq k \wedge \operatorname{rank}(\rho_1\alpha) \leq\sqrt k) \vee \\ (\operatorname{rank}(\beta) \geq k \wedge \operatorname{rank}(\rho_2\beta) \leq \sqrt k)} } \operatorname{Term}'_{u,v,w}(\alpha, \beta)\\
\Theta_2 &= \operatorname*{\mathbb{E}}_{u,v,w}\sum_{\substack {\max\left\{ \operatorname{rank}(\rho\alpha) ,\operatorname{rank}(\rho\beta)\right\} > \sqrt k} } \operatorname{Term}'_{u,v,w}(\alpha, \beta)\end{aligned}$$ From , $|\Theta_2| \leq 2^{-\sqrt k }$.
$|\Theta_1|$ is small similar to property (b) in Theorem 2.4 in [@Saket2013].
If $|\Theta_0| \geq \operatorname{poly}(s)$ then there is a labeling to the label cover instance that satisfies $\operatorname{poly}(s)$ fraction of its constraints.
$3$-colorable $3$-uniform Hypergraphs
=====================================
In this section, we give a reduction to $3$-uniform hypergraphs, where the YES case is $3$-colorable. We will be using the multi-layered version of the label cover over ${\mathbb{F}}_3$ for this reduction. As before, we will construct a hypergraph $\mathcal G=(\mathcal V, \mathcal E)$. For $v = (u_1,
\cdots,u_i,v_{i+1},\cdots ,v_\ell) \in V_i$, let $\mathcal H_v \subseteq {\mathbb{F}}_3^{m_i\times m_i}$ be the dual of the subspace of the set matrices in $\mathcal L_i$ that are symmetric and diagonal blocks corresponding to $v_{i+1},\cdots, v_\ell$ satisfies the constraints $C_{v_i}$’s. The reduction is given by a test on functions $A_v: {\mathbb{F}}_3^{m_i \times m_i} \rightarrow \{0,1\}$ folded over $\mathcal H_v$ for $v\in V_i, i \in \{0,\cdots, \ell -1\}$.
#### $3$-Colorable $3$-Uniform Test $\mathcal T_{3,3}$
1. Choose two layers $0\leq i < j < \ell$ uniformly at random and then choose $(u,v) \in E_{ij}$. Let $m:= m_i, r:= m_j$ and $\pi:{\mathbb{F}}_3^{m \times m} \rightarrow {\mathbb{F}}_3^{r \times r}$ be the projection corresponding to the edges $(u,v)$. Uniformly and independently at random choose $X \in {\mathbb{F}}_3^{r\times r}, Y \in {\mathbb{F}}_3^{m \times m}$ and $\overline y \in {\mathbb{F}}_3^m$. Let $\overline e_m \in {\mathbb{F}}_3^m$ be the vector with only the $m$[^th^ ]{}entry $1$ and the rest is $0$.
2. Accept if and only if the following are not all equal: $$\begin{aligned}
&A_u(X)&&B_v(Y)&&B_v(Y') &\text{ where } Y' &:= \overline y \otimes \overline y + \overline e_m \otimes \overline e_m - Y - X\circ \pi\end{aligned}$$
YES Case {#sec:yes-33}
--------
Let $\overline x_v \otimes \overline x_v$ for $v\in V_i, i \in \{0,\cdots,\ell-1\}$ perfectly satisfying labeling of the label cover instance. Consider the following $3$-coloring of $\mathcal G$: for each $v \in V$, $A_v(X):= \langle X,\overline y_v \otimes \overline y_v \rangle$. Note that such a function is constant over cosets of $\mathcal H_v$. Let $$\begin{aligned}
&x_1 := \langle X_1 , \overline y_v \otimes \overline y_v \rangle& ~~~~~&x_2 := \langle X_2 , \overline y_v \otimes \overline y_v \rangle\\
&y_1 := \langle Y_1 , \overline y_w \otimes \overline y_w \rangle& ~~~~~&y_2 := \langle Y_2 , \overline y_w \otimes \overline y_w \rangle\end{aligned}$$ and $f := \langle F, \overline x_u \otimes \overline x_u \rangle$. Note that $\langle F\circ \pi_{u,v}, \overline y_v\otimes \overline y_v \rangle = \langle F, \pi_{u,v}(y_v \otimes y_v)\rangle = \langle F, x_u \otimes x_u\rangle$ , and $\langle \overline e_m \otimes \overline e_m ,\overline y_v \otimes \overline y_v \rangle = \langle \overline e_m , \overline y_v \rangle = 1$. Using these, the assignments to the $8$ query locations are: $$\begin{aligned}
&x_1& ~~~~~&x_1+\langle \overline y_v, \overline x\rangle \langle \overline y_v, \overline y \rangle + f\\
&x_2& ~~~~~&x_2 + \left(\langle\overline y_v, \overline x\rangle +1\right) \langle \overline y_v, \overline z\rangle + f \\
&y_1& ~~~~~&y_1+\langle \overline y_w, \overline x'\rangle \langle\overline y_w, \overline y' \rangle + f + 1\\
&y_2& ~~~~~&y_2 + \left(\langle\overline y_w, \overline x'\rangle +1\right) \langle\overline y_w, \overline z'\rangle + f + 1\end{aligned}$$ It is easy to see at least one of the $4$ rows are always not equal. Hence $A$ is a valid $2$-coloring of $\mathcal G$.
NO Case {#sec:no-28}
-------
Suppose the reduction was applied to a NO instance of label cover. Let $k$ and $\delta$ be the parameters specified by Theorem 7.2 in [@KhotS2014b].
\[lem:soundness-28\] If there is an independent set in $\mathcal G$ of relative size $s$ then $$s^8 \leq \delta +\frac{1}{2^{k/2+1}}.$$
The proof of the lemma is similar to Section 8.2 in [@KhotS2014b]. Consider any set $A \subseteq \mathcal V$ of fractional size $s$. For every $v\in V$, let $A_v:{\mathbb{F}}_2^{m\times m} \rightarrow \{0,1\}$ be the indicator function that is extended such that it is constant over cosets of $\mathcal H_v$. $A$ is an independent set if and only if $$\label{eqn:indep}
\Theta := \operatorname*{\mathbb{E}}_{u,v,w} ~\operatorname*{\mathbb{E}}_{X_i,Y_i \in \mathcal T_{2,8}}~\prod_{i=1}^4A_v(X_i)A_w(Y_i) = 0.$$ Now we do the Fourier expansion and take expectations over $X_1,X_2,Y_1,Y_2$ to obtain the following: $$\begin{aligned}
\Theta = \operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{\alpha_1, \alpha_2\\ \beta_1,\beta_2} \in {\mathbb{F}}_2^{m\times m}} &\operatorname*{\mathbb{E}}_{F,\overline x,\overline x'} \Bigg[ \widehat A_v(\alpha_1)^2 \operatorname*{\mathbb{E}}_{\overline y}\left[\chi_{\alpha_1}(\overline x \otimes \overline y)\right]\chi_{\alpha_1}(F\circ \pi)\\
& \widehat A_v(\alpha_2)^2 \operatorname*{\mathbb{E}}_{\overline z}\left[\chi_{\alpha_2}((\overline x +\overline e_m) \otimes \overline z)\right]\chi_{\alpha_2}(F\circ \pi)\\
&\widehat A_w(\beta_1)^2 \operatorname*{\mathbb{E}}_{\overline y'}\left[\chi_{\beta_1}(\overline x' \otimes \overline y' )\right]\chi_{\beta_1}(F\circ \sigma) \chi_{\beta_1}(\overline e_m \otimes \overline e_m)\\
& \underbrace{\widehat A_w(\beta_2)^2 \operatorname*{\mathbb{E}}_{\overline z'}\left[\chi_{\beta_2}((\overline x'+\overline e_m) \otimes \overline z')\right]\chi_{\beta_2}(F\circ \sigma) \chi_{\beta_2}(\overline e_m \otimes \overline e_m) \Bigg]}_{=:\operatorname{Term}_{u,v,w}(\alpha_1,\alpha_2,\beta_1,\beta_2)}\end{aligned}$$
Note that since $F\in {\mathbb{F}}_2^{r\times r}$ is chosen uniformly at random, $$\operatorname*{\mathbb{E}}_F \chi_{\alpha_1}(F\circ \pi) \chi_{\alpha_2}(F\circ \pi) \chi_{\beta_1}(F \circ \sigma) \chi_{\beta_2}(F\circ \sigma) =\operatorname*{\mathbb{E}}_F (-1)^{\langle \pi(\alpha_1 +\alpha_2), F \rangle + \langle \sigma(\beta_1 +\beta_2), F \rangle}$$ is zero unless $\pi(\alpha_1 +\alpha_2) = \sigma(\beta_1 +\beta_2)$. Let $\nu(\alpha):= (-1)^{\langle \alpha, \overline e_m \otimes \overline e_m \rangle}$. Now taking expectations over $\overline x,\overline y, \overline z, \overline x',\overline y',\overline z'$, and noting that $\langle\alpha, x \otimes y \rangle = \langle \alpha x, y \rangle$, we obtain $$\label{eqn:sim-term}
\begin{aligned}
\operatorname{Term}_{u,v,w}(\alpha_1, \alpha_2, \beta_1,\beta_2) = (-1)^{\nu(\beta_1+\beta_2)}& \widehat A_v(\alpha_1)^2 \widehat A_v(\alpha_2)^2 \widehat A_w(\beta_1)^2 \widehat A_w(\beta_2)^2\\
& \Pr_{\overline x} \left[ \alpha_1 \overline x = 0 \wedge \alpha_2 \overline x = \alpha_2 e_m \right] \cdot\\
& \Pr_{\overline x'} \left[ \beta_1 \overline x' = 0 \wedge \beta_2 \overline x' = \beta_2 e_m \right]
\end{aligned}$$
when $\pi(\alpha_1 +\alpha_2) = \sigma(\beta_1 +\beta_2)$ and $0$ otherwise. Define: $$\begin{aligned}
\Theta_0 &= \operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{\operatorname{rank}(\alpha_1+\alpha_2), \operatorname{rank}(\beta_1+\beta_2) \leq k\\ \pi(\alpha_1+\alpha_2)= \sigma(\beta_1+\beta_2) \\ \nu(\beta_1+\beta_2) = 0 } } \operatorname{Term}_{u,v,w}(\alpha_1, \alpha_2, \beta_1,\beta_2)\\
\Theta_1 &= \operatorname*{\mathbb{E}}_{u,v,w}\sum_{\substack{\operatorname{rank}(\alpha_1+\alpha_2), \operatorname{rank}(\beta_1+\beta_2) \leq k\\ \pi(\alpha_1+\alpha_2)= \sigma(\beta_1+\beta_2) \\ \nu(\beta_1+\beta_2) = 1 } } \operatorname{Term}_{u,v,w}(\alpha_1, \alpha_2, \beta_1,\beta_2)\\
\Theta_2 &= \operatorname*{\mathbb{E}}_{u,v,w}\sum_{\substack {\max\left\{ \operatorname{rank}(\alpha_1+\alpha_2) ,\operatorname{rank}(\beta_1+\beta_2)\right\} > k\\ \pi(\alpha_1+\alpha_2)= \sigma(\beta_1+\beta_2)} } \operatorname{Term}_{u,v,w}(\alpha_1, \alpha_2, \beta_1,\beta_2)\end{aligned}$$ We lower bound $\Theta_0$ by $s^8$, upper bound $|\Theta_1|$ by $\delta$ and $|\Theta_2|$ by $1/2^{k/2+1}$ below. Along with , this will prove .
### Lower bound on $\Theta_0$
Note that all terms in $\Theta_0$ are positive. Now consider the term corresponding to $\alpha_1=\alpha_2=\beta_1=\beta_2=0$. $$\operatorname*{\mathbb{E}}_{u,v,w}\widehat A^4_v(0) \widehat A^4_w(0) = \operatorname*{\mathbb{E}}_u \left(\operatorname*{\mathbb{E}}_v \widehat A^4_v(0) \right)^2 \geq \left(\operatorname*{\mathbb{E}}_{uv} \widehat A_v(0)\right )^8 \geq s^8.$$
### Upper bound on $|\Theta_1|$
We can upper bound $|\Theta_1|$ by $$\label{eqn:fourier-decod}
\operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{ \operatorname{rank}(\alpha_1+\alpha_2),\operatorname{rank}(\beta_1+\beta_2) \leq k,\\ \pi(\alpha_1+\alpha_2) = \sigma(\beta_1+\beta_2),\\ \nu(\beta_1+\beta_2)=1 }} \widehat A^2_v(\alpha_1) \widehat A^2_v(\alpha_2) \widehat A_w^2(\beta_1) \widehat A_w^2(\beta_2).$$ Consider the following strategy for labeling vertices $u\in U$ and $v\in V$. For $u\in U$, pick a random neighbor $v$, choose $(\alpha_1,\alpha_2)$ with probability $\widehat A^2_v(\alpha_1) \widehat A^2_v(\alpha_2)$ and set its label to $\pi(\alpha_1+\alpha_2)$. For $w\in V$, choose $(\beta_1,\beta_2)$ with probability $\widehat A_w^2(\beta_1) \widehat A_w^2(\beta_2)$ and set its label to $\beta_1+\beta_2$. Since $A_w$ is folded, both $\beta_1$ and $\beta_2$ are symmetric and satisfies $C_v$. Since these constraints are homogeneous, $\beta_1+\beta_2$ is also symmetric and satisfies $C_v$. Also $\pi$ maps symmetric matrices to symmetric matrices. Note that gives the probability that a random edge $(u,w)$ of the label cover is satisfied by this labeling. Hence and $|\Theta_1|$ are upper bounded by $\delta$.
### Upper bound on $|\Theta_2|$
Note that if the $\operatorname{rank}(\alpha) >k$, for any fixed $b$, $\Pr_{\overline x}[\alpha x = b] \leq 1/2^{k+1}$. All terms in $\Theta_2$ has $\max \{ \operatorname{rank}(\alpha_1), \operatorname{rank}(\alpha_2),\operatorname{rank}(\beta_1), \operatorname{rank}(\beta_2) \} > k/2.$ From we have that, for any fixed choice of $u,v,w$ each term in $\Theta_2$ has absolute value at most $1/2^{k/2+1}$. Since $A,B$ are $\{0,1\}$ valued functions, sum of their squared coefficients is upper bounded by $1$ (i.e. Parseval’s inequality). Thus $|\Theta_2| \leq 1/2^{k/2+1}$.
We already saw in that an YES instance of label cover is mapped to a $2$-colorable hypergraph. Since $k= (\log N)^{1/8 - 2\epsilon}$ and $\delta = 2^{-(\log N)^{1/4 -2\epsilon}}$, $s \leq2^{-(\log N)^{1/8 -3\epsilon}}.$ Also the number of vertices in $\mathcal G$, $$n \leq N 2^{m^2} \leq N \cdot 2^{(\log N)^{10/4 +2\epsilon}}.$$ From and above, a NO instance of label cover is mapped to a hypergraph $\mathcal G$ that has no independent set of relative size $2^{-(\log n)^{1/20-4\epsilon}}$.
Old
===
\[lem:strong-smoothness\] For any $v \in V$ and symmetric matrix $\alpha \in {\mathbb{F}}_2^{m \times m}$ such that $k:=\operatorname{rank}(\alpha)$ and $k'_{u,v} := \operatorname{rank}(\pi^{(u,v)}\alpha)$ $$\Pr_{u \in N(v)}[ k'_{u,v} < k ] \leq 2^k\delta/2.$$
For a symmetric matrix $\alpha \in {\mathbb{F}}_2^{m\times m}$ and a number $k \leq m$, the following conditions are equivalent:
1. $ \operatorname{rank}(\alpha) = k$,
2. there are linearly independent $z_1, \cdots, z_k \in {\mathbb{F}}_2^m$ and numbers $s,t \geq 0$ with $k=s+2t$ such that $$\alpha = \sum_{i=1}^s z_i \otimes z_i + \sum_{j=1}^t ( z_{s+2j}\otimes z_{s+2j-1} +z_{s+2j-1}\otimes z_{s+2j}).$$
$1 \Rightarrow 2$ is already proved in \[KhotSaketb\], Lemma 2.1. To see the reverse direction, assume that $\alpha$ has the said decomposition in terms of outer products of linearly independent $z_1 \cdots z_k$. It is easy to come up with vectors $x_1,\cdots,x_k$ such that $\langle x_i , z_j\rangle = 1[i=j]$. Notice that $\forall i \in [s],~\alpha x_i = z_i$, for $i = s+2j,~\alpha x_i = z_{s+2j-1}$ and for $i = s+2j-1, \alpha x_i = z_{s+2j}$. Hence $\operatorname{rank}(\alpha) = k$.
Fix an edge $(u,v)$. Suppose $\operatorname{rank}(\alpha)=k$ and $\operatorname{rank}(\pi \alpha) < k$. Then from the claim above, there are some linearly independent $z_1,\cdots z_k \in {\mathbb{F}}_2^m$ such that
$$\alpha = \sum_{i=1}^s z_i \otimes z_i + \sum_{j=1}^t ( z_{s+2j}\otimes z_{s+2j-1} +z_{s+2j-1}\otimes z_{s+2j}).$$
Let $\rho:{\mathbb{F}}_2^m \rightarrow {\mathbb{F}}_2^r$ be the matrix of the linear operator such that $\pi \alpha = \rho \alpha \rho^T$ as given in the proof of Theorem 7.2 from Theorem 7.1 in \[KhotSaketb\] and $x_i = \rho z_i$. Then nce $z_i$ are linearly independent and $a_i$ is a non zero linear combination $y_u:=\sum_i a_i z_i \neq 0$. Since $y_u$ can take only $2^k-1$ values when $u$ is varied over neighbors of $v$. For each $y\neq 0$, from the smoothness condition in Theorem 7.1, $\Pr_u[\rho_u y = 0 ] \leq \delta/2$. Hence by union bound we have that $\Pr_u[ k_{u,v} < k] \leq \Pr_u[ \rho_u y_u = 0 ] \leq (2^k-1)\delta/2$ $$\Pr_u [y_u = y^*] \Pr_u [ \rho_u y_u = 0] \leq \Pr_u[\rho y* = 0] \leq \delta/2.$$
From Equation \[eqn:fourier-exp\] and Lemma \[lem:strong-smoothness\], we have that $$\begin{aligned}
\operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{\alpha,\beta \text{ symmetric },\\ \operatorname{rank}(\alpha),\operatorname{rank}(\beta) \leq r } } 1[\substack{\operatorname{rank}(\rho\alpha) = \operatorname{rank}(\alpha) \leq r \wedge \\ \operatorname{rank}(\rho'\beta)= \operatorname{rank}(\beta) \leq r }]\operatorname{Term}_{u,v,w}(\alpha,\beta) -\delta/2 + \sum_{\substack{\alpha,\beta \text{ symmetric },\\ \operatorname{rank}(\alpha),\operatorname{rank}(\beta) > r } } \operatorname{Term}_{u,v,w}(\alpha,\beta) \leq 0\end{aligned}$$
Now lets define
$$\begin{aligned}
\theta_0 &:= \operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{\alpha,\beta \text{ symmetric },\\ \operatorname{rank}(\alpha),\operatorname{rank}(\beta) \leq k\\ \nu(\alpha+\beta)=0}} 1[\substack{\operatorname{rank}(\rho\alpha) = \operatorname{rank}(\alpha) \leq r \wedge \\ \operatorname{rank}(\rho'\beta)= \operatorname{rank}(\beta) \leq r }]\operatorname{Term}_{u,v,w}(\alpha,\beta)\\
\theta_1 &:= \operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{\alpha,\beta \text{ symmetric },\\ \operatorname{rank}(\alpha),\operatorname{rank}(\beta) \leq k\\ \nu(\alpha+\beta)=1}} 1[\substack{\operatorname{rank}(\rho\alpha) = \operatorname{rank}(\alpha) \leq r \wedge \\ \operatorname{rank}(\rho'\beta)= \operatorname{rank}(\beta) \leq r }] \operatorname{Term}_{u,v,w}(\alpha,\beta)\\
\theta_2 &:= \operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{\alpha,\beta \text{ symmetric },\\ (k< \operatorname{rank}(\beta) \leq r) ~ \vee ~ (k< \operatorname{rank}(\beta)\leq r) }} 1[\substack{\operatorname{rank}(\rho\alpha) = \operatorname{rank}(\alpha) \leq r \wedge \\ \operatorname{rank}(\rho'\beta)= \operatorname{rank}(\beta) \leq r }] \operatorname{Term}_{u,v,w}(\alpha,\beta)\\
\theta_3 &:= \operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{\alpha,\beta \text{ symmetric },\\ (r< \operatorname{rank}(\beta) ) ~ \vee ~ (r< \operatorname{rank}(\beta)) }} 1[\substack{\operatorname{rank}(\rho\alpha) \leq k \vee \\ \operatorname{rank}(\rho'\beta)\leq k }] \operatorname{Term}_{u,v,w}(\alpha,\beta)\\
\theta_4 &:= \operatorname*{\mathbb{E}}_{u,v,w} \sum_{\substack{\alpha,\beta \text{ symmetric },\\ (r< \operatorname{rank}(\beta) ) ~ \vee ~ (r< \operatorname{rank}(\beta)) }} 1[\substack{\operatorname{rank}(\rho\alpha) > k \vee \\ \operatorname{rank}(\rho'\beta)> k }] \operatorname{Term}_{u,v,w}(\alpha,\beta)\end{aligned}$$
$\theta_0 \geq s^4$
Since $\nu(\alpha+\beta)=0$, all terms in $\theta_0$ are positive and in particular it contains the term $\alpha=\beta=0$ which is $s^4$.
$|\theta_1| \leq 2\delta$
$$|\theta_1| \leq \operatorname*{\mathbb{E}}_{u,v,w} 2\sum_{\substack{\alpha,\beta \text{ symmetric },\\ \operatorname{rank}(\alpha),\operatorname{rank}(\beta) \leq k\\ \nu(\alpha)=1}} A_v^2(\alpha) A_w^2(\beta)$$ We can do the Fourier decoding argument and bound the above by the soundness $\delta$ of the Label cover.
$|\theta_2|,|\theta_4| \leq 1/2^k$
Since $$\Pr\left[ \overline x \in \ker((\alpha \sigma)^*) \cap \left( \beta e_m + \ker((\beta \sigma')^*)\right) \right] \leq \frac{1}{2^{\max \{ \operatorname{rank}(\alpha \sigma^*), \operatorname{rank}(\beta \sigma^*) \}}}$$ the claim follows.
The outer verifier satisfies the additional property that $\Pr_{u \in N(v)}[\operatorname{rank}(\rho\alpha) < k] \leq O(\delta)$ for matrix $\alpha$ with rank at least $r$. Then $|\theta_4|\leq O(\delta)$
Hence we have that $$s^4- 2\delta - 1/2^k \leq \delta/2.$$ Parameters are set exactly as in \[KhotSaketb\] to get the hardness result.
Acknowledgements
================
The author would like to thank Prahladh Harsha, Subhash Khot, Rishi Saket and Srikanth Srinivasan for helpful discussions.
[^1]: Tata Institute of Fundamental Research, India. Supported by Google India under the Google India PhD Fellowship Award. Email : girishrv@tifr.res.in
[^2]: We will not require the exact definition of [*satisfying in superposition*]{} for this note. See for the details of the Khot-Saket outer PCP verifier.
[^3]: The covering number of a CSP is the minimal number of assignments to the vertices so that each hyperedge is covered by at least one assignment.
[^4]: The property that $\pi$ maps symmetric matrices to symmetric matrices is easy to see from the proof of [[@KhotS2014b Theorem 7.2]]{}.
|
The results are in for the All-Girls-Nationals held in Dallas, Texas from April 25 to 27. With 5.5/6, Medina Parrilla won the
Girls 18 & Under along with the scholarship to the University of Texas at Dallas, valued at over
65,000$. Angelica Berrios, pictured above in blue, earned the only perfect score of the event in
the Girls 16 & Under. In the Girls 14 & Under three New Yorkers, Rochelle Ballantyne, Catherine Hochman and Jasmine Fermin (pictured above with Angelica) tied for first with 5/6. Ballantyne took the first place trophy on tiebreak. In the Girls 12 & Under, Caroline Zhu won clear first place with 5.5/6. 2007 American World Youth representative Simone Liao won clear first with 5.5/6 in the largest section of the tournament, Girls 10 & Under. Brianna A. Guillen of host state Texas earned clear first in the Girls 8 & Under, with 5.5/6. See full results on MSA link. |
2010 East Texas church burnings
In January and February 2010, 10 churches were burned in East Texas.
Two local men, Jason Bourque and Daniel McAllister, were arrested, pleaded guilty and were jailed indefinitely.
Timeline
January 1 – Little Hope Baptist Church, Canton – ruled accident until investigation, later ruled arson
January 1 – Faith Church, Athens – ruled arson
January 12 – Lake Athens Baptist Church – burned
January 12 – Grace Community Church, Athens – burned
January 16 – Tyland Baptist Church in Tyler – torched
January 17 – First Church of Christ, Scientist in Tyler – burned to the ground
January 20 – Prairie Creek Fellowship of Lindale on Highway 69 – arson
February 4 – Russell Memorial United Methodist Church in Wills Point (Van Zandt County) – destroyed the sanctuary (ATF soon confirmed the fire was the result of arson)
8:30 PM on February 8 – Dover Baptist Church on Highway 110 outside Lindale – mostly destroyed
9:30 on February 8 – Clear Spring Missionary Baptist Church, on CR 426 near the Smith-Van Zandt county line – found burning
Suspects
A sketch was released of three persons of interest.
On February 21, 2010, Jason Robert Bourque, 19, of Lindale, and Daniel George McAllister, 21, of Ben Wheeler were charged in connection with the Dover Baptist Church burning on February 8. Their bond was set at $10 million. Because they targeted places of worship, the crime is a first-degree felony carrying a maximum penalty of 99 years to life.
Bourque was raised by his devout Christian maternal grandparents, while McAllister was homeschooled for religious reasons. Both men had started to question their faith, Bourque following his dropping-out from the University of Texas, and McAllister after the death of his mother and trouble finding work.
Faced with overwhelming evidence, both men pleaded guilty. On January 14, 2011, Judge Christi Kennedy sentenced Bourque to life and 20 years in prison, and McAllister to a life sentence.
On February 11, 2011, Bourque was interviewed by KLTV 7 from Smith County Jail. He blamed the drug Chantix, which he used to aid his quitting smoking, for psychotic episodes. He also claimed that McAllister had led the wave, targeting churches as he found them corrupt. Bourque stated that God had forgiven him.
Cultural legacy
Theo Love's documentary, Little Hope Was Arson, interviews community members in East Texas reacting to the burning of the 10 churches.
See also
Church arson
References
Category:2010 crimes in the United States
Category:2010 fires
Category:2010 in Texas
Category:Attacks on churches
Category:Anti-Christian sentiment in the United States
Category:Arson in Texas
Category:Attacks in the United States in 2010
Category:Crimes involving Satanism or the occult
Category:January 2010 events in the United States
Category:February 2010 events in the United States
Category:Van Zandt County, Texas
Category:Henderson County, Texas
Category:Tyler, Texas
Category:Smith County, Texas
Category:Attacks on religious buildings and structures in the United States |
We want to review how to divide these numbers
written in scientific notation on the TI-84 graphing calculator.
And then we'll write the result in both
scientific notation
as well as decimal notation. So it's a pretty straightforward process,
as long as we know how to type scientific notation into the graphing calculator.
So for our numerator we have
four point two times ten to the fourth, so we'll type in four point two.
And then we want this key here that has two capitol E's in blue,
so we'll press second, comma. And now we just type in the exponent on ten,
so we'll type in four. So this is how we
represent four point two
times ten to the fourth on this graphing
calculator, and then we'll divide this by
eight point four times ten to the negative three.
So we type in eight point four, second,
comma, negative three. Be careful not to enter minus three,
where you will get an error. So our quotient is given here decimal notation,
so we have one, two, three, four,
five, six zeros. So we have
five followed by six zeros, or five million.
But now we also want to write this in scientific notation.
Remember in scientific notation
this number here has to be greater than, or equal to one,
and less than ten, so if we write five million without the commas it'll be helpful.
The decimal is right here,
but for scientific notation it would have to be here.
So we're going to have five times
ten raised to some power, and the power is determined by how many places we move
the decimal.
So we moved it one, two,
three, four, five, six, and that's going to be a positive six.
The way remember this is that when you
multiply
something by ten to a positive power, this
number's going to get larger,
and we did start with a large number, so the exponent does have to be positive.
Let's go ahead and just try one more for
practice. So our numerator is six
times ten to the negative three, so we type in six,
second, comma, negative three. Then we'll divide this by
three times ten to the second, entered
as three,
second, comma, two. We don't want to use this
E in our notations, it's going to be two times ten to the negative five power.
And now we also want to convert this to decimal notation this time.
So if we use the number two where the
decimal point is here,
multiplying by ten to a negative power is going to make this number smaller,
so I know I have to move the decimal
point left
five places.
So one,
two, three, four, five. So we have to add four zeros here.
So in decimal notation we would have point
zero, zero, zero, zero, two.
And often you'll see a zero in the one's place value.
So this would be tenths, hundredths,
thousandths, ten thousandths,
hundred thousandths. This would be two hundred thousandths.
|
30 Ill. App.3d 229 (1975)
332 N.E.2d 538
THE PEOPLE OF THE STATE OF ILLINOIS, Plaintiff-Appellee,
v.
STANLEY THOMS, Defendant-Appellant.
No. 59995.
Illinois Appellate Court First District (5th Division).
June 13, 1975.
Brunswick, Jemilo & Richardson, of Blue Island, and Tully & Roddy, of Chicago (Richard S. Jemilo and Joseph V. Roddy, of counsel), for appellant.
Bernard Carey, State's Attorney, of Chicago (Patrick T. Driscoll, Jr., and Thomas D. Rafter, Assistant State's Attorneys, of counsel), for the People.
Judgment affirmed.
Mr. PRESIDING JUSTICE BARRETT delivered the opinion of the court:
Stanley Thoms, defendant, was found guilty after a bench trial of the crimes of bribery and official misconduct (Ill. Rev. Stat. 1969, ch. 38, pars. 33-1(d), (e), and 33-3(d)). He was placed on probation for a period of 2 years with the condition that he serve periodic imprisonment during the first year of his probation. Defendant appeals, arguing that the evidence was insufficient to establish his guilt beyond a reasonable doubt.
At trial, Louis B. Gemeinhardt testified that he and his partner, Edward Feeley, owned the Hollywood Motel located in Alsip, Illinois. Gemeinhardt testified that approximately 2 weeks after purchasing the motel, in February 1963, he received a message to contact the defendant. Subsequently, he met the defendant at defendant's home. At that time defendant informed him that running a motel was a very hazardous business since they had to deal with drunks, undesirables and women. Defendant offered to use his influence to take care of the property and see they did not run into any kind of problems. Defendant informed him *230 that the previous owner had been paying him $200 a month and it would be to his advantage if he continued the payments. Gemeinhardt testified that he told defendant that $200 was too much money, and after some negotiation defendant agreed to accept $100 a month. Gemeinhardt told him that he would have to discuss the matter with his partner and would get back to the defendant. In early March 1963, Gemeinhardt again returned to defendant's home and paid him the first installment of $100. During the next 5 to 7 months, Gemeinhardt paid the defendant $100 per month. The payments were all made in defendant's home except on one occasion they were made in his office at the police department. Gemeinhardt testified that Tom DiPaolo was the manager of the motel from 1963 until 1965. In 1966 Clarence Bastan was hired as the new manager.
Thomas DiPaolo testified that in February 1963, he became the manager of the Hollywood Motel in Alsip, Illinois. DiPaolo testified in May 1963, he went to defendant's home and gave defendant $50 upon orders of his employer, Mr. Gemeinhardt. DiPaolo testified that for the next 15 months while he was manager of the motel he would go to defendant's home on the first of every month and pay him $50 cash.
Clarence Bastan testified that on April 7, 1966, he became the manager of the Hollywood Motel in Alsip, Illinois. Shortly after assuming his duties on May 3, 1966, he met with the defendant in his office at the Alsip Police Station. At that time he gave the defendant $50. Each month thereafter he would meet with the defendant at defendant's home and pay him $50 cash. The payments were made to the defendant during the remainder of 1966, all of 1967 and until May of 1968. Bastan testified that in April or May 1968 business was down and he paid the defendant only $40 per month. Payments of $40 per month were made during the remainder of 1968. Bastan testified that in August 1968 he became the lessee of the Hollywood Motel. In December of 1968 or January of 1969, Bastan stopped making payments to the defendant. Bastan testified that all of the payments to defendant were entered on the daily sheets of the motel which were kept in the ordinary course of business. Bastan testified that 3 months after he stopped making the payments to the defendant, an Alsip squad car would be sitting on the north side of the driveway of the motel for an hour to and hour and a half at a time. The car would go away and would then return a short time later. Bastan called the mayor of Alsip to complain about the squad cars. The mayor told him that he would contact the chief of police. Bastan testified that when he called the mayor back the mayor stated that the police chief had informed him that Bastan had girls working at the motel.
Lawrence J. Kowalczyk testified that he is the owner of the Key Motel *231 in Alsip, Illinois. He first purchased the motel in 1960. The manager of the motel is Robert Dahms. Kowalczyk testified that in spring of 1963 he received a summons from the village of Alsip. The charges were dismissed by a justice of the peace who suggested that Kowalczyk stop by the police chief's office. Kowalczyk testified that he went into the defendant's office. At that time defendant told him, "We better take care of the boys." Kowalczyk testified that he asked if $50 would be sufficient and defendant stated that it would. Kowalczyk then gave the defendant $50.
Robert Dahms, the manager of the Key Motel in Alsip, Illinois, testified that he first became the manager of the Key Motel in Alsip, Illinois, in April 1963. Approximately a month after he began his employment he met the defendant in the parking lot of the Key Motel. At that time defendant stated that the motel would be O.K. if he were taken care of. Dahms gave defendant $50. The transaction was recorded on the daily sheets of the Key Motel. In December 1963, Dahms met with defendant at defendant's home and again gave 10 envelopes with $10 apiece in each envelope. Each envelope had the name of one of the police officers of the village written on it. Dahms testified that the police officers never received the envelopes with their names on it. Dahms testified that from 1964 until 1970 the Key Motel paid defendant $300 a year. Each payment occurred at defendant's home and was registered on the daily sheets of the motel. Dahms testified that prior to making the payments to the defendant a squad car would be parked across from the motel with radar equipment. After making the agreement with the defendant no other squad cars were parked across the street from the motel.
The daily sheets of the Key Motel and Hollywood Motel were admitted into evidence. They substantiated the testimony of the owners and managers of the motels.
It was stipulated that between 1960 and 1970 the defendant, Stanley Thoms, was the chief of police of the Alsip Police Department, Alsip, Illinois, and that the indictment was returned within 1 year after the responsible parties were notified of the alleged violations.
Stanley Thoms, defendant, testified that from 1960 to 1971 he was the chief of police for the village of Alsip. Defendant testified that prior to the trial he had never met Mr. Kowalczyk. Defendant denied that he ever received any money from Kowalczyk or Gemeinhardt or DiPaolo. Defendant testified that in 1965 Dahms came into the police station and asked him to be made a member of the auxiliary police force. Defendant stated that Dahms' age was against him and he was not capable of going through a strenuous training program. During the next 4 years Dahms approached him on 10 more occasions about being placed on the auxiliary *232 police force. Defendant testified that Dahms would become angry every time he was turned down. On July 15, 1971, as defendant was walking out of a Police and Fire Board meeting he passed by Dahms who stated that he had never given defendant a dime and wanted him to know that he would never testify against him. Dahms stated that he had to lie to cover up his own shortages. Defendant testified that Dahms again made similar statements on July 17, 1971, and March 14, 1972. Defendant denied that Dahms ever paid him any money.
Defendant testified that the Thursday prior to trial he met Bastan at the Gas City Gas Station in Calumet Park, Illinois. At that time Bastan told him that he was sorry, that he didn't think that the matter would go his far. Bastan stated that he only wanted to get defendant fired.
Raymond L. Termunde testified that from 1961 until 1972 he was mayor of the village of Alsip. Termunde testified that on March 14, 1972, he and the defendant were having lunch at the Pink Cloud Motel and Restaurant in Alsip, Illinois. As they began to exit the restaurant they met Dahms. At that time defendant asked Dahms why he had testified against him. Dahms replied that he never gave "Stan" a dime and that he had done it to cover up. Termunde testified that on several occasions Dahms had approached him about getting on as an auxiliary police officer. Termunde testified that defendant's reputation in the community for honesty and integrity was good.
It was stipulated that if Mrs. Joyce Arnold, a schoolteacher in Lemont, Illinois, and John Alsterda were called to testify they would testify that defendant's reputation in the community for honesty and integrity was good.
OPINION
Defendant's only contention on appeal is that the evidence did not establish his guilt beyond a reasonable doubt. Defendant argues that the evidence failed to establish that he had any criminal purpose or intent, that he was acting in his official capacity, that the remuneration was accepted knowingly, or that he accepted money for the purpose of causing the village of Alsip police vehicles to either be present or not present outside the motels involved.
Defendant was convicted of bribery and official misconduct. Section 33-1(d), (e), of the Criminal Code (Ill. Rev. Stat. 1969, ch. 38, pars. 33-1(d), (e)), under which defendant was indicted, defines bribery as:
"A person commits bribery when:
* * *
(d) He receives, retains or agrees to accept any property or personal advantage which he is not authorized by law to accept *233 knowing that such property or personal advantage was promised or tendered with intent to cause him to influence the performance of any act related to the employment or function of any public officer, public employee or juror; or
(e) He solicits any property or personal advantage which he is not authorized by law to accept pursuant to an understanding that he shall influence the performance of any act related to the employment or function of any public officer, public employee or juror."
Section 33-3(d) of the Criminal Code (Ill. Rev. Stat. 1969, ch. 38, par. 33-3(d)), under which defendant was also indicted, defines official misconduct as:
"A public officer or employee commits misconduct when, in his official capacity, he commits any of the following acts:
* * *
(d) Solicits or knowingly accepts for the performance of any act a fee or reward which he knows is not authorized by law."
Defendant in seeking a reversal relies upon People v. Jordan, 15 Ill. App.3d 672, 304 N.E.2d 713. In Jordan, this court reversed where a Chicago police officer had been convicted of bribery and official misconduct. The evidence adduced at trial demonstrated that an ambulance responded to a call for help. Upon arriving at the scene, the defendant, a Chicago police officer, was present with his partner. The defendant accepted a $10 bill from the ambulance driver, and the ambulance proceeded on its way. This court held that to sustain a conviction under the bribery statute or the official misconduct statute, the State must show more than the acceptance of a fee or reward by an employee. Since the evidence failed to establish that defendant was acting in his official capacity or that he had accepted the remuneration knowingly, or that he had accepted anything for the performance of a specified act, the conviction was reversed.
1 In the case at bar, the defendant was the chief of police of the village of Alsip from 1960 to 1970. The testimony of the State's witnesses established that in the case of both motels the defendant actively solicited the money using his position as chief of police. Louis Gemeinhardt, the owner of the Hollywood Motel, testified that several weeks after purchasing the motel in February of 1963, he met the defendant at defendant's home. At that time defendant told him what a hazardous business running a motel was and offered to use his influence as police chief to take care of the property and to see that he did not run into any kind of problems. Thereafter, for a period of 6 years, the defendant accepted monthly payments from Gemeinhardt and the managers of the motel.
*234 Lawrence Kowalczyk, the owner of the Key Motel, testified that shortly after purchasing that motel he received a summons from the village of Alsip. After his case was dismissed the justice of the peace suggested that he stop at the police chief's office. Kowalczyk testified that he met the defendant, who was the chief of police, in his office. At that time defendant told him, "We better take care of the boys." Thereafter, and for a period of 6 years, defendant accepted payments from Kowalczyk and the motel managers.
This testimony clearly established that defendant was using his office as chief of police of the village of Alsip in order to extract money from both motels. Defendant in both occasions actively solicited the money and accepted payments over a period of 6 years. Defendant argues that the testimony regarding the original solicitations in 1963 should not be considered on appeal since the statute of limitations has run as to those offenses. However, this evidence was admissible as showing a continuing course of conduct by the defendant. Defendant solicited and accepted the payments over a long period of time and there can be no question that he received the funds knowingly. Both motels had problems with squad cars parking near the motels when funds were not being paid to the defendant. However, as long as payments were being made to the defendant the squad cars disappeared. It is obvious from the evidence that the motel owners and managers did not want squad cars parked in front of their premises. After a complete review of all the evidence adduced at trial, we conclude that the defendant knowingly solicited and accepted funds which were unauthorized by law, using his office as chief of police with the intent to influence the performance of an act in his official capacity.
While defendant denied that he received money from either of the motels, the credibility of witnesses and the weight to be given to their testimony are matters for the trier of fact to determine. (People v. Clark, 52 Ill.2d 374, 288 N.E.2d 363.) The evidence adduced at trial was sufficient to establish defendant's guilt on both charges beyond a reasonable doubt.
The judgment is affirmed.
Affirmed.
DRUCKER and LORENZ, JJ., concur.
|
Little Snoring
Little Snoring is a village and a civil parish in Norfolk, England.
The village is approximately east-north-east from the town of Fakenham, west-south-west from Cromer, and to the side of the A148 road.
At the 2011 Census, it had population of 619. For the purposes of local government, the parish falls within the district of North Norfolk.
History
Little Snoring has an entry in the Domesday Book of 1085. In the Great Book (jargon for "The Domesday Book") Little Snoring is recorded by the names "Esnaringa", "Snaringa" and "Snarlinga", named after "Snare", the settlers' leader, and thus the name Little Snoring evolved over the centuries. It was the king's land with the main landholders being William de Warenne and Peter de Valognes and his main tenant is said to be Ralph.
Within the parish of Little Snoring, at a place called Queensgate, was a house of the Order of St Lazarus. It is mentioned in the will of Alexander, Rector of Snoring Parva, in 1380. Nothing further about the house is known.
Landmarks
The church of Little Snoring St Andrew is one of 124 existing round-tower churches in Norfolk. The church and its separate tower are Grade I listed buildings.
The village has an airfield, formerly RAF Little Snoring, part of which is still active as a private airfield, whilst the other part now belongs to a potato-producing company.
Amenities
Village amenities include a school, post office and the Green Man public house.
The nearest railway station is at Sheringham for the Bittern Line which runs between Sheringham, Cromer and Norwich. The nearest airport is Norwich International Airport.
References
External links
"Little Snoring, St. Andrew", Johann Dasch photographs, Roundtowerchurches.de
"The Snoring Villages", The-snorings.co.uk
"Little Snoring Parish Council", Norfolk Rural Community Council
"Little Snoring Primary School", Norfolk County Council
Category:Villages in Norfolk
Category:Civil parishes in Norfolk
Category:North Norfolk |
Q:
How do I check which of my installed modules is unsupported?
When I go to admin/modules, I get this scary warning:
The installed version of at least one of your modules or themes is no
longer supported. Upgrading or disabling is strongly recommended. See
the project homepage for more details. See the available updates page
for more information and to install your missing updates.
That sounds like something I'd like to fix! So I click the "available updates" link, but that page tells me:
All of your projects are up to date.
Hmmm. I tried checking for updates with Drush (drush up) but Drush doesn't find any updates either. How can I find which modules installed in my site are unsupported?
A:
As stated here, you need to go to admin/reports/updates rather than admin/reports/updates/update to see which module is having the problem.
|
Size
Quantity
Details
Our two piece NATO watch straps versatile in style and is a perfect accessory on any watch from the old vintage to todays smart watches. Since it's so easy to change a NATO strap from day to day, you can switch watch straps at any time. |
Q:
Make a trigger to only trigger if a certain column is updated
I'm trying to make a trigger to only trigger if a certain column is updated, and then only if the column is updated to 'Executed'. I can update if the column is changed, but I can't seem to find a way to update if column is updated to 'Executed'
CREATE TRIGGER dbo.NewTrigger
ON dbo.Database
AFTER UPDATE
AS
IF Update(Status) = 'Executed'
BEGIN
--MY insert into statement. This adds data to another table, but I only want the whole process to run if the original table column "Status" is set to "Executed"
END
Could someone assist please?
A:
You'd need to use the inserted and deleted tables in the trigger, see here:
Use Inserted and Deleted Tables
In case of an update:
inserted table: contains new column values of rows that have been updated
deleted table: contains old column values of rows that have been updated
Your trigger could look something like this:
create table t (id int identity, status varchar(100));
create table audit(id int, old_status varchar(100), new_status varchar(100), updated_at datetime);
create trigger StatusUpdate
on t
After UPDATE as
if (update(status)
and exists(select * from inserted i
inner join deleted d on d.id = i.id
where d.status != 'Executed'
and i.status = 'Executed'))
begin
insert into audit (id, old_status, new_status, updated_at)
select i.id, d.status, i.status, getdate()
from inserted i
inner join deleted d on d.id = i.id
where d.status != 'Executed'
and i.status = 'Executed'
end
See Demo on DB Fiddle
Demo 2 - Multiple rows updated together
|
Column liquid chromatography determination of vitamins A and E in powdered milk and local flour: a validation procedure.
A high-performance liquid chromatography method was developed for the simultaneous routine determination of vitamins A and E in powdered milk and flour made from local plants and purchased from open markets in the Ivory Coast. The method involves saponification followed by extraction with a mixture of organic solvents. The vitamins were resolved by reversed-phase HPLC and detection at a single wavelength. The main tests of method validation were applied to the procedure. The results show the reliability of the analytical method for the intended application. |
With Mayor Ed Murray Out, His Donors Head to Jenny Durkan
Jenny Durkan announcing her campaign for mayor last month. Nate Gowdy
Earlier this year, it looked like Mayor Ed Murray was a shoo-in for reelection in this year’s mayoral race. He had netted over $400,000 in contributions, and his only serious challenger was Nikkita Oliver, the lawyer and educator running with the new Peoples Party.
Everything changed when the Seattle Times published three separate accounts from men alleging Murray sexually abused them when they were minors more than three decades ago. The mayor’s race swelled to 21 candidates, six of whom are serious contenders. Murray dropped out, and city politicos wondered where his support would go. If money is any tell, the answer is Jenny Durkan.
Durkan, a former U.S. Attorney who has never held elected office, received $31,133 from 55 donors that had previously donated to Murray’s mayoral bid, according to the latest campaign finance reports. Those Murray donors account for 19.4 percent of all the funds donated to Durkan’s campaign.
Support from Murray's donations helped Durkan jumped to a quick lead in campaign financing, raising more than $160,000 in less than month since she filed her candidacy on May 12. She has been endorsed by the Seattle Metropolitan Chamber of Commerce, former Governor Chris Gregoire and King County Council Member Joe McDermott.
Durkan also has the support of two city council members that were formerly Murray supporters. Council Member Sally Bagshaw donated $500 to Durkan on May 15 after donating $100 to Murray last October. Council Member Tim Burgess gave $500 to Durkan on May 18 after giving $500 to Murray last October. Burgess’s wife, Joleen Burgess, also gave $500 to both Durkan and Murray.
Bagshaw said she is supporting Durkan because of her “thoughtful approach to civil rights and social justice issues.”
“It’s a tragedy around what’s happened around Mayor Murray. Whether it’s true or not, either way it’s a tragedy, and Jenny is someone that I have tremendous amount of respect for and I’m glad she’s willing to run,” Bagshaw said.
Bagshaw said she developed respect for Durkan after working on the state’s marriage equality campaign in 2012 and working in city’s legal community together. Bagshaw was an attorney at the King County Prosecutor’s Office while Durkan was working as a private attorney in Seattle.
“There are some good candidates that are running and I think she just rises to the top,” Bagshaw said.
Burgess was not available for comment for this story.
No other candidate came close to raising as much from former Murray donors as Durkan. Former Mayor Mike McGinn, who is running for mayor again after losing his own reelection fight to Murray in 2013, attracted three former donors of Murray, totally $1,000.
Nikkita Oliver, an attorney and activist, raised $455 from four donors that had previously given to Murray. Cary Moon, an urban designer and prominent critic of the Highway 99 tunnel, raised $350 from two donors that had previously given to Murray’s campaign.
Overall, Oliver is in second place for fundraising as of Monday, pulling in $50,784. McGinn has raised $27,045 and Moon has raised $12,715. The field of 21 candidates will be winnowed down to the top two candidates after a primary vote on August 1. The general election will be held November 7. |
703 So.2d 1011 (1997)
Kevin BARFIELD
v.
STATE.
CR-96-0467.
Court of Criminal Appeals of Alabama.
April 18, 1997.
Kevin Barfield, pro se.
Bill Pryor, atty. gen., and Hense R. Ellis II, asst. atty. gen., for appellee.
BROWN, Judge.
The appellant, Kevin Barfield, purports to appeal from the trial court's denial of his petition for post-conviction relief filed pursuant to Rule 32, Ala.R.Crim.P., in which he attacked his 1995 guilty-plea conviction for first degree rape, and the resulting sentence of 20 years' imprisonment. The attorney general argues that this appeal should be dismissed because, it argues, the appellant's notice of appeal was untimely.
The relevant facts are as follows:
1. On June 26, 1996, the appellant filed a Rule 32 petition in the Circuit Court of Jefferson County. (C.R. 1-10.) In the petition, the appellant maintained that he was entitled to relief because, he argued, a number of his constitutional rights had been violated. The appellant also argued that the trial court was without jurisdiction to impose the sentence, that the sentence exceeded the maximum authorized by law, and that he had failed to appeal his conviction through no fault of his own.
2. On August 26, 1996, the state filed a motion to dismiss the petition on the grounds that the claims were either precluded, were without merit, or were merely bare allegations. (C.R.13-18.)
3. On September 5, 1996, the appellant filed a response to the state's motion to dismiss. (C.R.19-20.)
4. On September 23, 1996, the appellant filed a document entitled "`Motion for Expansion of Records, Introduction of Newly Discovered Evidence," which included an affidavit from a prison inmate who vouched for the appellant's innocence. (C.R. 21-23.)
5. On October 16, 1996, the trial court denied the appellant's petition in the following written order:
"The Court, having reviewed petition for relief from conviction and the state's response, finds that petitioner has not supported any allegations in the petition with fact. Petition for relief from conviction is denied and dismissed."
(C.R.32.)
6. The trial court clerk's record contains a written notice of appeal filed by the appellant in the Jefferson County Circuit Court. (C.R.24.) The notice of appeal does not indicate when the notice was filed in the circuit clerk's office; however, both the notice of appeal and its certificate of service are dated November 11, 1996.
*1012 7. On November 12, 1996, this Court received a copy of the appellant's written notice of appeal, apparently sent by the appellant.
8. The document immediately following the appellant's notice of appeal in the trial court clerk's record is a document entitled "Motion in Objection to Court's Denial of Rule 32 Petition on October 16, 1996." (C.R.25-26.) This document and its certificate of service are also dated November 11, 1996. The stamp on this document indicates that it was filed in the circuit clerk's office on November 18, 1996.
9. The reporter's transcript order and the Court of Criminal Appeals docketing statement, which were executed by the appellant, are both dated November 25, 1996. (C.R.27-30.) The certificate of service attached to these documents is dated November 26, 1996. (C.R.28.) The stamp on these documents indicates that they were not filed in the circuit clerk's office until December 6, 1996.
10. The notice of appeal to this Court by the circuit clerk shows that the appellant filed his notice of appeal on December 2, 1996. (C.R.31.)
11. The circuit clerk's certificate of completion of the record also shows that the appellant filed his notice of appeal on December 2, 1996. (C.R.34.)
The attorney general argues that the appellant's December 2, 1996, notice of appeal was untimely, and that thus, the appeal is due to be dismissed. Symanowski v. State, 606 So.2d 171 (Ala.Crim.App.1992). In Symanowski, we held:
"An appeal must be taken in the manner and within the time prescribed by the Alabama Rules of Appellate Procedure, or it is not taken at all. See Rogers v. Singleton, 286 Ala. 83, 237 So.2d 473 (1970). A.R.App.P. 4(b) provides that in a criminal case the notice of appeal must be filed within 42 days of pronouncement of sentence, provided that the notice of appeal may be orally entered at the sentencing, or it must be filed within 42 days after the denial or overruling of a motion in arrest of judgment, motion for a new trial, or motion for judgment of acquittal filed within 30 days of sentence. `This 42-day period is to be applied uniformly....' Committee Comments, Rule 4. Rule 2(a)(1) provides: `An appeal shall be dismissed if the notice of appeal was not timely filed to invoke the jurisdiction of the appellate court.' This requirement of timely filing of the notice of appeal is `a jurisdictional act'; `[i]t is the only step in the appellate process which is jurisdictional.' Committee Comments, Rule 3. See also Lewis v. State, 463 So.2d 154, 155 (Ala.1985); Woods v. State, 371 So.2d 944, 945 (Ala. 1979); Turner v. State, 365 So.2d 335, 335 (Ala.Cr.App.), cert. denied, 365 So.2d 336 (1978).
"`In the absence of statutory authorization, neither the trial nor appellate courts may extend or shorten the time for appeal... even to relieve against mistake, inadvertence, accident, or misfortune....' Meeks v. State Farm Mut. Auto. Ins. Co., 286 Ala. 513, 515, 243 So.2d 27, 28 (1970) (quoting with approval Hanley v. Hanley, 23 Cal.2d 120, 142 P.2d 423, 149 A.L.R. 1250, 1261-67 (1943)). `In the interest of finality of judgments, the prescribed time within which a notice of appeal must be filed with the trial court cannot be waived nor is it subject to extension of time by agreement of the parties or by order of this Court.' Stewart v. Younger, 375 So.2d 428, 428 (Ala.1979) (emphasis in original). See also Hayden v. Harris, 437 So.2d 1283, 1287 (Ala.1983); State v. Kebe, 399 So. 2d 348 (Ala.1981) (wherein our supreme court noted that a United States District Court could not confer to the court the authority to extend the 42-day period).'"
606 So.2d at 172.
It is undisputed that for the notice of appeal to be timely, it had to be filed with the clerk of the trial court by November 27, 1996, which is 42 days from the trial court's denial of his Rule 32 petition.[1] See Rules 3 and 4, Ala.R.App.P. If, as the state argues, the appellant's notice of appeal was filed on December 2, 1996, then the appeal is due to be *1013 dismissed as untimely; however, from our review of the record, we are unable to tell when the appellant filed his written notice of appeal with the trial court clerk, or how the trial court clerk determined that the date of the filing of the appellant's notice of appeal was December 2, 1996.
As noted above, although the appellant's notice of appeal is dated November 11, 1996, there is no indication in the record as to when that notice of appeal was filed with the trial court clerk. This Court received a copy of the appellant's notice of appeal on November 12, 1996. Moreover, the appellant's "Motion in Objection to Court's Denial of Rule 32 Petition," which was dated the same date as the appellant's notice of appeal and which was filed immediately after the notice of appeal in the clerk's record, was filed in the circuit clerk's office on November 18, well before the November 27 deadline. Although the Court of Criminal Appeals docketing statement was not filed in the trial court clerk's office until December 6, the notice of appeal was not necessarily untimely, because the docketing statement can be filed after the notice of appeal. See Rule 3(e), Ala. R.App.P.
A determination that the appellant's notice of appeal was untimely has significant ramifications, namely, the dismissal of the appeal. Accordingly, we must remand this cause to the trial court with instructions that the trial court determine when the appellant filed his written notice of appeal with the trial court clerk. The trial court shall take the necessary action to ensure that the circuit clerk makes due return to this Court at the earliest possible time within 28 days from the date of this opinion. We pretermit discussion of the other issues raised on appeal pending the return to remand.
REMANDED WITH DIRECTIONS.[*]
All the Judges concur.
NOTES
[1] Although on November 18, 1996, the appellant filed a document entitled, "Motion in Objection to Court's Denial of Rule 32 Petition On October 16, 1996," the filing of the motion did not extend the time for the filing of the appellant's notice of appeal. The Rules of Criminal Procedure do not contemplate the filing of a motion for a new trial or its equivalent following the denial of a Rule 32 petition. See Patterson v. State, 549 So.2d 635, 636 (Ala.Crim.App.1989).
[*] Note from the reporter of decisions: On August 22, 1997, on return to remand, the Court of Criminal Appeals affirmed, without opinion.
|
Well cared for Frank Baines Reflex English jump saddle. Wool Flocked making it possible for a saddle fitter to adjust fit. MW tree and 16 1 2 " seat. The reflex saddle allows for improved rider balance and better weight distribution along the horse s back. It s a comfortable saddle for the horse and rider. See more Pet Supplies, Horses Pet Supplies, Horses in Jacksonville.
Well cared for Frank Baines Reflex English jump saddle. Wool Flocked making it possible for a saddle fitter to adjust fit. MW tree and 16...
This saddle has unique silver tooling. Very light useage. Kept in climate controlled environment. Suede seat is in superb condition. Lighter tone leather. Great show possiblities. Only slight marks on sides which cannot be seen when in saddle. NO other defects.. See more Pet Supplies, Horses Pet Supplies, Horses in Gainesville.
This gorgeous 16" Parker County Cutter/Penner by Billy Cook Saddlerly is a very comfortable ride. It's practically brand new with only six rides on it. Built on a ralide tree with a 7" gullet double dropped stainless dees rough out seat and fenders a 4" cheyenne rolled cantle oversized close contact skirts a 3" horn and natural rawhide hand laced stirrups. It's adorned with antiqued concho's and trimmed with barbed wire tooling. Oiled regularly and stored indoors its been very well maintained. Purchased March of this year for 1375 I'm only asking 1000 obo shipping available at purchasers expense or local pickup in High Springs FL 30 mins NW of Gainesville. Trades for barrel saddle of equal value considered.. See more Pet Supplies, Horses Pet Supplies, Horses in Gainesville.
This gorgeous 16" Parker County Cutter/Penner by Billy Cook Saddlerly is a very comfortable ride. It's practically brand new with only si... |
The peptidoglycan biosynthesis genes MurA and MraY are related to chloroplast division in the moss Physcomitrella patens.
In the moss Physcomitrella patens, 10 Mur genes involved in peptidoglycan biosynthesis were found, and the MurE and Pbp genes are related to plastid division. Although the MraY and MurG genes were missing in our previous expressed sequence tag screening, they were discovered in the P. patens genome in this study, indicating that P. patens has a full set of genes capable of synthesizing peptidoglycan. In addition, a second MurA gene (PpMurA2) was found. Whereas Northern analyses indicated that PpMurA1, PpMurG and PpMraY were expressed, transcripts of PpMurA2 were detected only when RT-PCR was employed. Whereas GFP fusion proteins with either PpMurA1 or PpMraY were detected in chloroplasts, the PpMurA2 fusion proteins were located in the cytoplasm. Protonema cells in the wild-type plants had an average of 46 chloroplasts. PpMurA1 gene-disrupted lines had <10 chloroplasts, whereas approximately 30 chloroplasts existed in the PpMurA2 knockout lines. The PpMurA1/A2 double-knockout lines had only a few macrochloroplasts, suggesting a redundant function for these two genes. Disruption of the PpMraY gene in P. patens resulted in the appearance of macrochloroplasts. Anabaena MraY fused to the N-terminal region of PpMraY and A. thaliana MraY could complement the macrochloroplast phenotype in the PpMraY knockout line. Electron microscopic observations showed no obvious differences in the shape or stacking of thylakoid membranes between all knockout transformants and wild-type plants, suggesting that these Mur genes are related only to plastid division in moss. |
let mockPath = "/app/designer/zones/1";
jest.mock("../../history", () => ({
getPathArray: jest.fn(() => mockPath.split("/")),
push: jest.fn(),
}));
jest.mock("../../api/crud", () => ({
edit: jest.fn(),
save: jest.fn(),
}));
import * as React from "react";
import { mount, shallow } from "enzyme";
import {
RawEditZone as EditZone, EditZoneProps, mapStateToProps,
} from "../edit_zone";
import { fakeState } from "../../__test_support__/fake_state";
import { fakePointGroup } from "../../__test_support__/fake_state/resources";
import {
buildResourceIndex,
} from "../../__test_support__/resource_index_builder";
import { save, edit } from "../../api/crud";
import { push } from "../../history";
describe("<EditZone />", () => {
const fakeProps = (): EditZoneProps => ({
dispatch: jest.fn(),
findZone: () => undefined,
botSize: {
x: { value: 3000, isDefault: true },
y: { value: 1500, isDefault: true },
},
});
it("redirects", () => {
mockPath = "/app/designer/zones/nope";
const wrapper = mount(<EditZone {...fakeProps()} />);
expect(wrapper.text()).toContain("Redirecting...");
expect(push).toHaveBeenCalledWith("/app/designer/zones");
});
it("doesn't redirect", () => {
mockPath = "/app/logs";
const wrapper = mount(<EditZone {...fakeProps()} />);
expect(wrapper.text()).toContain("Redirecting...");
expect(push).not.toHaveBeenCalled();
});
it("renders", () => {
mockPath = "/app/designer/zones/1";
const p = fakeProps();
p.findZone = () => fakePointGroup();
const wrapper = mount(<EditZone {...p} />);
expect(wrapper.text().toLowerCase()).toContain("edit");
});
it("changes name", () => {
mockPath = "/app/designer/zones/1";
const p = fakeProps();
const group = fakePointGroup();
p.findZone = () => group;
const wrapper = shallow(<EditZone {...p} />);
wrapper.find("input").first().simulate("blur", {
currentTarget: { value: "new name" }
});
expect(edit).toHaveBeenCalledWith(group, { name: "new name" });
expect(save).toHaveBeenCalledWith(group.uuid);
});
});
describe("mapStateToProps()", () => {
it("returns props", () => {
const state = fakeState();
state.resources = buildResourceIndex([fakePointGroup()]);
const props = mapStateToProps(state);
expect(props.findZone(1)).toEqual(undefined);
});
});
|
A highly unsettling phenomenon has been sweeping through our television news programmes of late. The phenomenon may have been inspired by the behaviour of that rude, horse-faced dreep, Jeremy Paxman. Or its origins may go back even further to the bad-tempered, dismissive approach of the singularly curmudgeonly Robin Day, he of the ridiculously oversized, spotted bowtie.I’m referring, of course, to the aggressive interview, particularly of politicians. The Paxo style of interview – attack like a Rottweiler from the off, never let up, and then when your victim is bleeding and defeated smile sweetly and say thank you – seems to have spread to all the news presenters. Even that cloying, homely Fiona Bruce – she of the Blue Peter school of presentation, with her constant hand gestures and undulating voice – is at it.Although tucked away on its News Channel, the BBC actually devotes a whole programme to the phenomenon. HARDtalk (the ominous uppercase letters are the BBC’s, not mine) is billed as: In-depth interviews with hard-hitting questions and sensitive topics being covered. Recently, I caught the tail-end of an interview on that programme with 98-year-old Ben Ferencz, the last surviving prosecutor at the Nuremberg trials. I ask you, hard-hitting questions directed at a man who has more gravitas in his pinkie than is possessed collectively by a studio full of self-important news presenters?Not long after that travesty, I also caught the tail-end of another broadcast on the programme. It was a cosy, self-congratulatory chat over lunch between the main presenters of HARDtalk, who were reminiscing on memorable interviews they had conducted over the years. One smug fellow recalled the time he interviewed the late Mo Mowlam when she was Secretary of State for Northern Ireland. (He spoke her name so disdainfully that he might as well have added the word “Oik” after it.) It seemed that the bold, brave (but clearly uncouth) Mo actually kicked him under the table when he pressed her to answer a particularly sensitive question. Oh, how they all laughed!I’m not against aggressive interviewing per se. What I object to are the aggressors. They are increasingly rude and surly to their interviewees. They have the arrogance to believe that it’s perfectly all right to impart their personal views, their personal values, with the rudeness and surliness. And they have the audacity to presume that they are representing the same views and values of their audience. Given that by and large the interviewers are ex-public school, ex-Oxbridge members of the upper middle-class, I hardly think they could ever represent my particular views and values. Yet I think they do believe their role is to act as the public’s moral guardians.So what can we do to combat this phenomenon of the aggro television interview? Perhaps if enough of us complain to Ofcom or sign a petition that’s debated in Parliament, the current pack of interviewers will receive a message from the public they haughtily presume to represent. You are not our moral guardians. We no longer wish to know what you think or what makes you angry and contemptuous. We wish to hear what your interviewees answer in response to your robust, probing questions. But we demand civility at all times in your approach. Above all, we demand objectivity.(While we’re at it, we could also get a message to BBC News. Get rid of Fiona Bruce – the woman is useless.) |
Q:
Defining a topology on the middle third Cantor set $C$
Is there some sort of a topology defined on the ternary Cantor set $C$ if we consider that $(a,b) \cap C$ is considered to be an open set? Is this the right approach to be taken if wanting to establish a topology on $C$? Hopefully, I will learn more about this in the Topology course I will be taking at my university this semester.
A:
That is exactly the approach taken to define the subspace topology on $C$ as a subspace of the real line with its usual topology. Start with the fact that the open intervals form a base for the usual topology on $\mathbb{R}$: a subset of $\mathbb{R}$ is open if and only if it’s a union of sets of the form $(a,b)$. Now transfer this ‘down’ to $C$: declare the sets of the form $(a,b) \cap C$ to be the basic open sets in $C$, and say that a subset of $C$ is open in the subspace topology on $C$ if and only if it’s a union of basic open sets, i.e., of sets of the form $(a,b) \cap C$.
However, you can just as well get all of the open sets in the subspace topology on $C$ at once, without using open intervals. We’ve just said that if $V \subseteq C$ is open in the subspace topology on $C$, there is some family $\mathscr{I}$ of open intervals in $\mathbb{R}$ such that $$V = \bigcup\{I \cap C:I \in \mathscr{I}\} = C \cap \bigcup\{I:I \in \mathscr{I}\} = C \cap \bigcup\mathscr{I}.$$ $\bigcup\mathscr{I}$ is a union of open intervals in $\mathbb{R}$, so it’s just some open set $W$ in $\mathbb{R}$. Thus, every subset $V$ of $C$ that is open in the subspace topology on $C$ is of the form $W \cap C$ for some open set $W$ in $\mathbb{R}$. Conversely, if you start with some open set $W$ in $\mathbb{R}$, you can write it as $W = \bigcup\mathscr{I}$ for some family $\mathscr{I}$ of open intervals, and carry out the same calculation in reverse to see that $W \cap C$ is a union of sets of the form $(a,b) \cap C$.
The sets of the form $(a,b) \cap C$ are the nicest open sets in the subspace topology on $C$, and the easiest to visualize, and they’re enough to give you all of the open sets in the subspace topology on $C$ by taking unions, but you can get all of the open sets at once by taking the sets of the form $W \cap C$ where $W$ is any open set in $\mathbb{R}$, not just an open interval.
As Davide Giraudo said in his comment, one can do essentially the same thing with any subset of any topological space, and you should indeed learn more about this in your topology course.
|
SPECIALIST ROOF REFURBISHMENT AND CONVERSION FOR INDUSTRIAL AND COMMERCIAL PREMISES
A total solution for roof repair, refurbishment, conversion, renovation and construction requirements across the South West and South Wales.
The simple answer is no. Surfaces need to be dry for work to commence – for this reason working between showers is also difficult. We are also unable to work in freezing temperatures and in wind speeds that exceed 30 knots.
Can I walk on a non-fragile roof light?
Only roof lights specified as “non fragility for the life of the roof” may be walked on. We recommend that you never walk on translucent, coloured GRP or polycarbonate roof light.
What is meant by ‘non fragile’?
Non fragility is determined under a test method detailed in ACR[M]001:2005. Please refer to the link to the non fragile specification table on page 8 supplied by FILON.
Why do I need to erect fall protection netting under my fragile roof when you are working on it?
It is a legal requirement in accordance with ‘HSE HSG33 Health and Safety in Roof Work’ to supply a safe working platform and fall arrest protection for anybody working on a fragile roof. Scaffolding and safety netting are the safest and most cost-effective way of providing this.
Do you offer insurance-backed guarantees for your work?
Insurance-backed guarantees are available via the manufacturers of some of the systems we install. Work is inspected by a third party and the guarantee is issued via the manufacturer. Guarantee periods can range from 10-25 years.
I am overlaying my flat roof; do I need to upgrade the insulation values to meet current Building Regulation requirements?
No you do not – the definition of renovation in the building regulations has been amended to read: “the repair of a flat roof by the addition of a layer to the existing layers of over 25% of the area will not require an upgrade to the thermal properties”.
I am renewing my roof covering; do I need to upgrade the insulation values to meet current Building Regulation requirements?
If you are renewing over 25% of the roof area you will need to raise insulation values and seek building regulation approval.
What size contracts do you undertake?
From small patch repairs through to major refurbishment projects. Project values range from £300 – £1 million. |
---
author:
- |
Hamid Rezatofighi$^{1,2}$Nathan Tsoi$^1$JunYoung Gwak$^1$Amir Sadeghian$^1$Ian Reid$^2$Silvio Savarese$^1$\
\
$^1$Computer Science Department, Stanford University, United States\
$^2$School of Computer Science, The University of Adelaide, Australia\
[hamidrt@stanford.edu]{}
bibliography:
- 'egbib.bib'
title: 'Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression'
---
|
The capture proteasome assay: A method to measure proteasome activity in vitro.
Because of its crucial role in various cellular processes, the proteasome is the focus of intensive research for the development of proteasome inhibitors to treat cancer and autoimmune diseases. Here, we describe a new and easy assay to measure the different proteasome activities in vitro (chymotrypsin-like, caspase-like, and trypsin-like) based on proteasome capture on antibody-coated plates, namely the capture proteasome assay (CAPA). Applying the CAPA to lysates from cells expressing standard proteasome, immunoproteasome, or intermediate proteasomes β5i or β1i-β5i, we can monitor the activity of the four proteasome subtypes. The CAPA provided similar results as the standard whole-cell proteasome-Glo assay without the problem of contaminating proteases requiring inhibitors. However, the profile of trypsin-like activity differed between the two assays. This could be partly explained by the presence of MgSO4 in the proteasome-Glo buffer, which inhibits the trypsin-like activity of the proteasome. The CAPA does not need MgSO4 and, therefore, provides a more precise measurement of the trypsin-like activity. The CAPA provides a quick and accurate method to measure proteasome activity in vitro in a very specific manner and should be useful for the development of proteasome inhibitors. |
Crypto Should Try Self-Regulation, CFTC Commissioner Says
That was a common thread in the conversations Wednesday at Yahoo Finance’s cryptocurrency event in New York.
Whether that’s a good or bad thing is a matter of opinion. But even for those wary of government oversight, it’s perhaps a less unpleasant thought in light of the previous day’s hearing before the U.S. Senate Banking Committee, where the chairmen of the SEC and the CFTC expressed a generally open attitude toward crypto.
Against that backdrop, Brian Quintenz, a member of the Commodity Futures Trading Commission, pointed out that the Tuesday hearing was free of the scaremongering pronunciations that some in the ecosystem perhaps expected.
“One of the other takeaways from yesterday was you didn’t hear either chairman say ‘no, absolutely not, this is not safe, we must stop this at all costs.’ No one said that,” Quintenz said at the Yahoo event during an onstage interview with CoinDesk managing editor Marc Hochstein.
Noting that any congressional action to fill in jurisdictional gaps identified by the SEC and CFTC will likely take a long time, the commissioner called on the industry to consider forming a self-regulatory organization.
Successful examples of this model include the Financial Regulatory Industry Authority and the National Futures Association, he said. Such an organization would provide necessary oversight of the spot market and exchanges “between now and however long Congress chooses to act,” he said.
Quintenz went on to add:
“We don’t want to be saying no to innovators, or to advancing technology.”
Adam White, general manager of the Coinbase-operated digital asset exchange GDAX, said that government oversight was something the company welcomed.
“I think we embrace regulation at Coinbase,” he said. “We recognize that regulations are a complementary part of the financial system in many ways.”
Ripple’s ‘revolution’
Similarly, Brad Garlinghouse, the CEO of Ripple, stressed his company’s longtime friendly posture toward regulators (which made it stand out from the libertarian early adopters of bitcoin), saying:
“I don’t think the US dollar’s going away in my lifetime. I don’t think the US government’s going away in my lifetime. The revolution isn’t happening outside the system. The revolution’s going to happen inside the system.”
For others, including BitPesa CEO Elizabeth Rossiello, regulation is an everyday reality, especially for businesses that serve as bridges between cryptocurrencies and government-issued money.
“In an ideal world, we wouldn’t have regulation. But in the real world, my customers ask me: ‘Do we work with companies which [comply with investor rules]?” she said.
On the investor side, speaker Alex Sunnarborg, founding partner of hedge fund Tetras Capital, described how lack of regulatory oversight could have a significant impact, for example in Simple Agreements for Future Tokens. What if those future tokens are never delivered?
“The big fear there is you enter into a SAFT which gives you a discount or allocation into a sale in the future … but the sale [doesn’t happen],” he said.
Image of Brian Quintenz (right) by Nikhilesh De for CoinDesk
Disclosure: CoinDesk is a subsidiary of Digital Currency Group, which has ownership stakes in BitPesa, Coinbase and Ripple.
The leader in blockchain news, CoinDesk is a media outlet that strives for the highest journalistic standards and abides by a strict set of editorial policies. CoinDesk is an independent operating subsidiary of Digital Currency Group, which invests in cryptocurrencies and blockchain startups. |
GuidedStepSupportFragment
A GuidedStepSupportFragment is used to guide the user through a decision or series of decisions.
It is composed of a guidance view on the left and a view on the right containing a list of
possible actions.
Basic Usage
Clients of GuidedStepSupportFragment must create a custom subclass to attach to their Activities.
This custom subclass provides the information necessary to construct the user interface and
respond to user actions. At a minimum, subclasses should override:
If app chooses not to use the helper function, it is the app's responsibility to call
setUiStyle(int) to select fragment transition and remember the stack entry where it
need pops to.
Theming and Stylists
GuidedStepSupportFragment delegates its visual styling to classes called stylists. The GuidanceStylist is responsible for the left guidance view, while the GuidedActionsStylist is responsible for the right actions view. The stylists use theme
attributes to derive values associated with the presentation, such as colors, animations, etc.
Most simple visual aspects of GuidanceStylist and GuidedActionsStylist can be customized
via theming; see their documentation for more information.
GuidedStepSupportFragments must have access to an appropriate theme in order for the stylists to
function properly. Specifically, the fragment must receive Theme_Leanback_GuidedStep, or a theme whose parent is
is set to that theme. Themes can be provided in one of three ways:
The simplest way is to set the theme for the host Activity to the GuidedStep theme or a
theme that derives from it.
If the Activity already has a theme and setting its parent theme is inconvenient, the
existing Activity theme can have an entry added for the attribute LeanbackGuidedStepTheme_guidedStepTheme. If present,
this theme will be used by GuidedStepSupportFragment as an overlay to the Activity's theme.
Finally, custom subclasses of GuidedStepSupportFragment may provide a theme through the onProvideTheme() method. This can be useful if a subclass is used across multiple
Activities.
If the theme is provided in multiple ways, the onProvideTheme override has priority, followed by
the Activity's theme. (Themes whose parent theme is already set to the guided step theme do not
need to set the guidedStepTheme attribute; if set, it will be ignored.)
If themes do not provide enough customizability, the stylists themselves may be subclassed and
provided to the GuidedStepSupportFragment through the onCreateGuidanceStylist() and onCreateActionsStylist() methods. The stylists have simple hooks so that subclasses
may override layout files; subclasses may also have more complex logic to determine styling.
Guided sequences
GuidedStepSupportFragments can be grouped together to provide a guided sequence. GuidedStepSupportFragments
grouped as a sequence use custom animations provided by GuidanceStylist and
GuidedActionsStylist (or subclasses) during transitions between steps. Clients
should use add(FragmentManager, GuidedStepSupportFragment) to place subsequent GuidedFragments onto the fragment stack so that
custom animations are properly configured. (Custom animations are triggered automatically when
the fragment stack is subsequently popped by any normal mechanism.)
Note: Currently GuidedStepSupportFragments grouped in this way must all be defined programmatically,
rather than in XML. This restriction may be removed in the future.
LeanbackGuidedStepTheme_guidedActionsBackgroundDark
LeanbackGuidedStepTheme_guidedActionsElevation
LeanbackGuidedStepTheme_guidedStepBackground
LeanbackGuidedStepTheme_guidedStepTheme
Constants
EXTRA_UI_STYLE
Fragment argument name for UI style. The argument value is persisted in fragment state and
used to select fragment transition. The value is initially UI_STYLE_ENTRANCE and
might be changed in one of the three helper functions:
UI_STYLE_ACTIVITY_ROOT
One possible value of argument EXTRA_UI_STYLE. This is the case that we show first
GuidedStepSupportFragment in a separate activity. The default behavior of this style:
Enter transition is assigned null (will rely on activity transition), exit transition is
same as UI_STYLE_ENTRANCE. Note: Changing exit transition by UI style is not working
because fragment transition asks for exit transition before UI style is restored in
Fragment.onCreate().
UI_STYLE_ENTRANCE
Default value for argument EXTRA_UI_STYLE. The default value is assigned in
GuidedStepSupportFragment constructor. This is the case that we show GuidedStepSupportFragment on top of
other content. The default behavior of this style:
Enter transition slides in from two sides, exit transition slide out to START(left).
Background will be faded in. Note: Changing exit transition by UI style is not working
because fragment transition asks for exit transition before UI style is restored in Fragment
.onCreate().
When popping multiple GuidedStepSupportFragment, finishGuidedStepSupportFragments() also changes
the top GuidedStepSupportFragment to UI_STYLE_ENTRANCE in order to run the return transition
(reverse of enter transition) of UI_STYLE_ENTRANCE.
GuidedStepSupportFragment
Public methods
add
Adds the specified GuidedStepSupportFragment to the fragment stack, replacing any existing
GuidedStepSupportFragments in the stack, and configuring the fragment-to-fragment custom
transitions. A backstack entry is added, so the fragment will be dismissed when BACK key
is pressed.
add
Adds the specified GuidedStepSupportFragment to the fragment stack, replacing any existing
GuidedStepSupportFragments in the stack, and configuring the fragment-to-fragment custom
transitions. A backstack entry is added, so the fragment will be dismissed when BACK key
is pressed.
If current fragment on stack is GuidedStepSupportFragment: assign UI_STYLE_REPLACE
If current fragment on stack is not GuidedStepSupportFragment: assign UI_STYLE_ENTRANCE
Note: currently fragments added using this method must be created programmatically rather
than via XML.
Parameters
fragmentManager
FragmentManager: The FragmentManager to be used in the transaction.
fragment
GuidedStepSupportFragment: The GuidedStepSupportFragment to be inserted into the fragment stack.
Returns
int
The ID returned by the call FragmentTransaction.commit.
addAsRoot
Adds the specified GuidedStepSupportFragment as content of Activity; no backstack entry is added so
the activity will be dismissed when BACK key is pressed. The method is typically called in
Activity.onCreate() when savedInstanceState is null. When savedInstanceState is not null,
the Activity is being restored, do not call addAsRoot() to duplicate the Fragment restored
by FragmentManager.
UI_STYLE_ACTIVITY_ROOT is assigned.
Note: currently fragments added using this method must be created programmatically rather
than via XML.
Parameters
activity
FragmentActivity: The Activity to be used to insert GuidedstepFragment.
fragment
GuidedStepSupportFragment: The GuidedStepSupportFragment to be inserted into the fragment stack.
id
int: The id of container to add GuidedStepSupportFragment, can be android.R.id.content.
Returns
int
The ID returned by the call FragmentTransaction.commit, or -1 there is already
GuidedStepSupportFragment.
finishGuidedStepSupportFragments
Convenient method to close GuidedStepSupportFragments on top of other content or finish Activity if
GuidedStepSupportFragments were started in a separate activity. Pops all stack entries including
UI_STYLE_ENTRANCE; if UI_STYLE_ENTRANCE is not found, finish the activity.
Note that this method must be paired with add(FragmentManager, GuidedStepSupportFragment, int) which sets up the stack entry name for finding which fragment we need to pop back to.
getUiStyle
Read UI style from fragment arguments. Default value is UI_STYLE_ENTRANCE when
fragment is first initialized. UI style is used to choose different fragment transition
animations and determine if this is the first GuidedStepSupportFragment on backstack.
isFocusOutEndAllowed
Returns true if allows focus out of end edge of GuidedStepSupportFragment, false otherwise.
Default value is false, the reason is to disable FocusFinder to find focusable views
beneath content of GuidedStepSupportFragment. Subclass may override.
Returns
boolean
True if allows focus out of end edge of GuidedStepSupportFragment.
isFocusOutStartAllowed
Returns true if allows focus out of start edge of GuidedStepSupportFragment, false otherwise.
Default value is false, the reason is to disable FocusFinder to find focusable views
beneath content of GuidedStepSupportFragment. Subclass may override.
Note that this can be called while the fragment's activity is
still in the process of being created. As such, you can not rely
on things like the activity's content view hierarchy being initialized
at this point. If you want to do work once the activity itself is
created, see onActivityCreated(Bundle).
Any restored child fragments will be created before the base
Fragment.onCreate method returns.
Parameters
savedInstanceState
Bundle: If the fragment is being re-created from
a previous saved state, this is the state.
onCreateView
Called to have the fragment instantiate its user interface view.
This is optional, and non-graphical fragments can return null (which
is the default implementation). This will be called between
onCreate(Bundle) and onActivityCreated(Bundle).
If you return a View from here, you will later be called in
onDestroyView() when the view is being released.
Parameters
inflater
LayoutInflater: The LayoutInflater object that can be used to inflate
any views in the fragment,
container
ViewGroup: If non-null, this is the parent view that the fragment's
UI should be attached to. The fragment should not add the view itself,
but this can be used to generate the LayoutParams of the view.
savedInstanceState
Bundle: If non-null, this fragment is being re-constructed
from a previous saved state as given here.
This corresponds to Activity.onSaveInstanceState(Bundle) and most of the discussion there
applies here as well. Note however: this method may be called
at any time before onDestroy(). There are many situations
where a fragment may be mostly torn down (such as when placed on the
back stack with no UI showing), but its state will not be saved until
its owning activity actually needs to save its state.
setUiStyle
Set UI style to fragment arguments. Default value is UI_STYLE_ENTRANCE when fragment
is first initialized. UI style is used to choose different fragment transition animations and
determine if this is the first GuidedStepSupportFragment on backstack. In most cases app does not
directly call this method, app calls helper function
add(FragmentManager, GuidedStepSupportFragment, int). However if the app creates Fragment
transaction and controls backstack by itself, it would need call setUiStyle() to select the
fragment transition to use.
Protected methods
onAddSharedElementTransition
Called when this fragment is added to FragmentTransaction with UI_STYLE_REPLACE (aka
when the GuidedStepSupportFragment replacing an existing GuidedStepSupportFragment). Default implementation
establishes connections between action background views to morph action background bounds
change from disappearing GuidedStepSupportFragment into this GuidedStepSupportFragment. The default
implementation heavily relies on GuidedActionsStylist's layout, app may override this
method when modifying the default layout of GuidedActionsStylist.
TIP: because the fragment view is removed during fragment transition, in general app cannot
use two Visibility transition together. Workaround is to create your own Visibility
transition that controls multiple animators (e.g. slide and fade animation in one Transition
class). |
Mambray Creek
Mambray Creek is part
of the Mount Remarkable National Park.
Access is from the Princes Highway near Port Pirie and is sign
posted.
It's a great area for camping and walking.
The camp ground is very good with natural bush camps well apart
from other campers with flushing toilets, drinking water and
Hot showers. The walks available start at a couple of hours to
multi day walks. |
Actigraphic monitoring during sleep of children with ADHD on methylphenidate and placebo.
Sleep disturbances appear as a comorbid condition in children with attention-deficit/hyperactivity disorder. The aim of this study was to investigate the relationship of activity levels during sleep and therapeutic response to methylphenidate (MPH). Nightly sleep actigraphic recordings during a double-blind, placebo-controlled, crossover clinical study (1-week of 0.5 mg/kg MPH; 1-week of placebo) were obtained on 44 children, 6 to 12 years old, diagnosed with attention-deficit/hyperactivity disorder (DSM-IV). Significant (p <.005) differences between the conditions were found in several software-computed parameters: sleep onset latency (MPH, 39.3 minutes; placebo, 28.2 minutes), sleep efficiency (MPH, 78.0%; placebo, 80.4%), total sleep time (MPH, 7 hours; 57 minutes; placebo, 8 hours, 16 minutes). No significant differences on any of these measures were found among the 26 subjects who showed a moderate or large global improvement on MPH over placebo compared with 18 subjects who showed mild or no clinical improvement. MPH, given twice daily, induces a slight but significant sleep disturbance. Motor activity levels during sleep did not differentiate children who responded to MPH from those who did not respond. This suggests that responders to MPH treatment do not experience greater sleep disturbances than nonresponders, at least at the dose studied. |
Robert R. Provine
Laughter
American Scientist 84. 1 (Jan-Feb, 1996): 38-47.
© Sigma Xi, The Scientific Research Society 1996
The study of laughter provides a novel approach to the mechanisms and evolution of vocal production, perception and social behavior
Consider the bizarre events of the 1962 outbreak of contagious laughter in Tanganyika. What began as an isolated fit of laughter (and sometimes crying) in a group of 12- to 18-year-old schoolgirls rapidly rose to epidemic proportions. Contagious laughter propagated from one individual to the next, eventually infecting adjacent communities. The epidemic was so severe that it required the closing of schools. It lasted for six months.
The Tanganyikan laughter epidemic is a dramatic example of the infectious power of laughter - something that many of us may have experienced in our own lives. Many readers will be familiar with the laugh tracks of television situation comedies - attempts to stimulate contagious laughter in viewers -and the difficulty of extinguishing their own "laugh jags," fits of nearly uncontrollable laughter. Have you ever been overcome by a comparable urge to chant "hello-hello-hello?" Rather than dismissing contagious laughter as a behavioral curiosity, we should recognize it and other laugh-related phenomena as clues to broader and deeper issues.
Clearly, laughter is a powerful and pervasive part of our lives - an important component of that biobehavioral bedrock of our species known as human nature. Laughter's significance has been recognized at various times and in various ways by such scientific and philosophical dignitaries as Aristotle, Kant, Darwin, Bergson and Freud. Yet aside from a general appreciation that laughter is good for us -"the best medicine" - and is somehow associated with humor, we know little about laughter itself.
My approach to understanding laughter is one that a visiting extraterrestrial might take were it to encounter a group of laughing human beings. What would the visitor make of the large bipedal animals emitting paroxysms of sound from a toothy vent in their faces? A reasonable approach would be to describe the simplest and most obvious aspects of the noisy behavior: its physical characteristics, the rules that govern its expression, characteristics of the animals emitting the sounds (such as gender), the mechanism of sound production, and whether similar sounds are made by related species. To Earthlings this naturalistic approach is known as ethology - a biologically oriented scientific discipline devoted to understanding what animals do and how and why they do it. Ethologists treat behavior as an evolutionary adaptation. The species-wide distribution of laughter and its stereotypical (and simple) structure suggests that the behavior has strong genetic and neurophysiological bases - qualities attractive to those who wish to understand the mechanisms and natural history of behavior.
During the past eight years I have been observing human laughter in various natural habitats - shopping malls, classrooms, sidewalks, offices and cocktail parties - with the investigative spirit of our hypothetical alien. Observing everyday behavior in these settings has provided an opportunity to appreciate laughter as a social vocalization of the human animal. These studies have produced some unexpected insights into the phenomenon of human laughter - its social nature, the lawful relationship between laughter and speech, gender differences and the biological basis of contagion.
Laugh Structure
One of my first goals was to describe the sonic structure of human laughter. This proved to be more difficult than ! expected. Like other spontaneous acts, laughter often disappears when one attempts to observe it, especially in the laboratory. Some unconventional approaches were called for. Although I could occasionally elicit laughter from friends and colleagues during playful conversations, I was often forced to engage in shameless hamming (something that graduate school did not prepare me for). One of the most productive approaches was to encounter people in public places and simply ask them to laugh. The request was usually answered with a burst of laughter. About half of the laughing subjects reported that they could not laugh on command. Indeed, we have much less conscious control over laughter than over speech. It is easy to say "ha-ha-ha," but difficult to laugh on cue. We do not "speak" laughter.
In collaboration with an undergraduate assistant, Yvonne Yong, I took the recordings to the Sound Laboratory of the National Zoo in Washington, D.C. Here the laughs were analyzed with a sound spectrograph, a device that translates a sound into an image that reveals the changes in frequency and intensity of the sound over time. Giggles, shrieks and belly laughs replaced the laboratory's usual sonic fare of indigo bunting songs and the calls of golden lion tamarins. Laboratory workers gave us quizzical looks but politely refrained from asking about the origins of the sounds.
The sound spectra revealed the distinct signature of laughter. A laugh is characterized by a series of short vowel-like notes (syllables), each about 75 milliseconds long, that are repeated at regular intervals about 210 milliseconds apart. A specific vowel sound does not define laughter, but similar vowel sounds are typically used for the notes of a given laugh. For example, laughs have the structure of "ha-ha-ha" or "ho- ho-ho," but not "ha-ho-ha-ho." There are intrinsic constraints against producing such laughs. Try to simulate a "ha-ho-ha-ho" laugh - it should feel quite unnatural. When there are variations in the notes, they most often involve the first or last note in a sequence. Thus, "cha-ha-ha" or "ha-ha-ho" laughs are possible variants.
The explosively voiced blasts of a laugh have a strong harmonic structure, with each harmonic being a multiple of a low (fundamental) frequency. The harmonic structure is revealed in a sound spectrogram by the evenly spaced stacks of short horizontal lines in the spectrum, the lowest of which is the fundamental frequency. Given their higher-pitched voices, it is not surprising that the laughter of females has a higher fundamental frequency (about 502 hertz) than male laughter (about 276 hertz). Whether it is a deep belly laugh or a high-pitched titter, however, all human laughter is a variation of this basic form. It is this structure that allows us to recognize laughter in spite of individual differences.
The notes and internote intervals carry most of the information that allows us to identify a sound as laughter. If the sounds between laugh notes are edited out of a tape recording - leaving the notes separated by intervals of silence - a laugh still sounds normal. The internote time interval carries information, but the internote expiratory sounds do not. If the notes are removed from a recording and the gaps between intervals are closed, all that remains of laughter is a long, breathy sigh.
The stereotypic structure of a laugh is, at least in part, a result of the limitations of our vocal apparatus. It is difficult to laugh with abnormally long note durations, such as "haaa-haaa-haaa," or abnormally short durations (much less than 75 milliseconds in length). Likewise, normal note durations with abnormally long or short internote intervals do not occur. Try to produce a natural laugh with a long internote interval, such as "ha------ha------ha." As with the natural rhythms of walking or running, there are only so many ways to laugh.
The structural simplicity of a laugh is also suggested by its reversibility. A short segment of laughter -" ha-ha-ha" - played backward on a tape recorder still sounds rather like "ha-ha-ha." Indeed the sound spectrum of a laugh is similar whether scanned from left to right or from right to left -a laugh note has a high degree of temporal symmetry. Yet one aspect of a laugh that is not symmetrical is its loudness. Laughter is characterized by a decrescendo in which the laugh notes that are late in a sequence are usually lower in amplitude than earlier notes (presumably because we run out of air). Recordings of laughter played backward produce a bizarre-sounding crescendo.
Chimpanzee Laughter
There is a common misperception that laughter is exclusive to human beings. From at least the time of Darwin, however, it has been known that chimpanzees and other great apes perform a laugh-like vocalization when tickled or during play To pursue the details of this primate laughter, I teamed up with Kim Bard, who is nursery director and caregiver for young chimpanzees at the Yerkes Regional Primate Center in Atlanta. It is a pleasure to be able to play with young chimpanzees in the pursuit of one's science.
Chimpanzee (Pan troglodytes) laughter differs in many ways from its human counterpart. The vowel-like notes of human laughter are performed by chopping a single expiration, whereas chimpanzee laughter is a breathy panting vocalization that is produced during each brief expiration and inspiration. Unlike human laughter, the laughter of a chimpanzee lacks discrete, vowel-like notes that have sharp leading and trailing edges on sound spectra. Chimpanzee laughter has the sound and cadence of a handsaw cutting wood. The sounds of chimpanzee and human laughter are sufficiently different that without viewing the characteristic "play face" and source of stimulation (such as play and tickle), naive human beings may be unable to identify the chimpanzee vocalization as laughter. You can experience the difference in production between the two forms of laughter by placing a hand on your abdomen and comparing the abdominal pulsations of chimpanzee-like panting with the smoother act of speaking "ha-ha-ha" during a single expiration.
People laugh as we speak. If chimpanzees laugh as they speak, by producing one laugh sound per expiration and inspiration, we have identified an important and previously unrecognized constraint on the evolution of speech and language in chimpanzees and presumably other great apes. The close coupling of laughter to breathing in chimpanzees may be evidence of a more general limitation on these animals to speak. (In contrast to the success of teaching hundreds of signs to chimpanzees, efforts to teach them to speak English have produced meager results.) Indeed, the inability to modulate expiratory airflow may be at least as limiting to speech as the structure of the vocal tracts of nonhuman primates.
Breathy, panting laughter is probably the primal form that dates back to the common ancestor of all great apes and people. Human beings evolved their characteristic laughter after branching from an ancestor in common with chimpanzees (estimated to be around six million years ago, according to DNA hybridization data).
It is noteworthy that chimpanzee laughter occurs almost exclusively during physical contact, or during the threat of such contact, during chasing games, wrestling or tickling. (The individual being chased laughs the most.) Although people laugh when tickled, most adult human laughter occurs during conversation, typically in the absence of physical contact.
Social and Linguistic Context
Laughter is a decidedly social signal, not an egocentric expression of emotion. In the absence of stimulating media (television, radio or books), people are about 30 times more likely to laugh when they are in a social situation than when they are alone. Indeed people are more likely to smile or talk to themselves than they are to laugh when they are alone. Aside from the obvious implication that sociality can enhance laughter and perhaps one's mood, these observations indicate that laughter has a social function. What can we say about laughter as communication?
In an attempt to gather some clues, my colleagues and I have collected observations on 1,200 instances of naturally occurring human laughter. Three undergraduate assistants (Lisa Greisman, Tina Runyan, Michelle Bowers) and I wandered various public gathering places where we eavesdropped on groups of laughing people. We carefully took note of the principals engaged in the behavior - the gender of the speaker and the audience, whether the speaker or the audience laughed and what was said immediately before the laughter.
Contrary to our expectations we found that most conversational laughter is not a response to structured attempts at humor, such as jokes or stories. Less than 20 percent of the laughter in our sample was a response to anything resembling a formal effort at humor. Most of the laughter seemed to follow rather banal remarks, such as "Look, it's Andre," "Are you sure?" and "It was nice meeting you too." Even our "greatest hits," the funniest of the 1,200 pre-laugh comments were not necessarily howlers: "You don't have to drink, just buy us drinks," "She's got a sex disorder - she doesn't like sex," and "Do you date within your species?" Mutual playfulness, in-roup feeling and positive emotional tone - not comedy - mark the social settings of most naturally occurring laughter. Research that focuses only on the response of an audience to jokes (a common laboratory scenario) targets only a small subset of laughter.
One of the key features of natural laughter is its placement in speech. Laughter is not randomly scattered throughout the speech stream.
The speaker and the audience seldom interrupt the phrase structure of speech with laughter. In our sample of 1,200 laughs there were only eight interruptions of speech by laughter, all of them by the speaker. Thus a speaker may say "You are going where?... ha-ha," but rarely "You are going... ha-ha... where?" The occurrence of laughter during pauses at the end of phrases suggests that a lawful and probably neurologically based process governs the placement of laughter in speech - a process in which speech has priority access to the single vocalization channel. The strong and orderly relationship between laughter and speech is akin to punctuation in written communication (and is called the punctuation effect).
Our field study revealed other clues about laughter in human communication. A counterintuitive finding was that the average speaker laughs about 46 percent more often than the audience. This finding reveals the limits of analyses that report only audience behavior - the typical approach of humor research - and neglect the social nature of the laughing relationship.
The gender of the principals involved plays a large role in determining the amount of speaker laughter. Whether they are speakers or audiences (in mixed-sex groups), females laugh more often than males. Female speakers laugh 127 percent more than their male audience. In contrast, male speakers laugh about 7 percent less than their female audience. Neither males nor females laugh as much to female speakers as they do to male speakers. (The lot of the female comedian is not an easy one - whether her audience is male or female.)
These gender differences in the pattern of laughter are at least as strong as those noted for speech by the linguist Deborah Tannen of Georgetown University. The limited cross-cultural evidence suggests that males are the leading humor producers and that females are the leading laughers. These differences are already present by the time that joking first appears around six years of age.
What message is being conveyed by a laughing speaker or a laughing audience? In some respects laughter may be a signal of dominance/submission or acceptance/rejection. Consider the distinction between laughing with and laughing at someone. Valuable insights about laughter's social function will come from studies of laughter in groups of people who differ in social rank and gender.
A response of laughter by the audience may affirm or negate the spirit of the speaker's message. "Polite" laughter, for example, may be a forced effort on the part of the audience to signal their accord with the speaker, quite the opposite of the indignant "ha!" A speaker, in other cases, may buffer an aggressive comment with laughter or deliver a remark using "laugh-speak," a consciously controlled hybrid of laughter and speech. Talk-show hosts, who are experts at shaping the course of a conversation, commonly use laugh-speak. In this sense laughter may modify the behavior of others by shaping the emotional tone of a conversation.
Laugh Tracks and Contagion
The use of laughter to evoke laughter or a positive mood is familiar to viewers of situation comedy shows on television. "Laugh tracks" (dubbed-in sounds of laughter) have accompanied most "sitcoms" since 7:00 p.m. (Eastern Standard Time) on September 9, 1950. On that evening the Hank McCune Show -comedy about "a likeable blunderer, a devilish fellow who tries to cut corners only to find himself the sucker" - first used a laugh track to compensate for the absence of a live audience. Despite the fact that the show was short-lived, the television industry discovered the power of laughter to evoke audience laughter. The recording Industry recognized the seductive power of laughter shortly after World War I with the distribution of the OKeh Laugh Record, which consisted of trumpet playing that was intermittently interrupted by laughter It remains one of the most successful novelty records of all time. Acknowledging the commercial potential of this novelty market, Louis Armstrong, Sidney Bechet, Woody Herman and Spike Jones all attempted to cash in with laugh records of their own.
In the intervening years social scientists have confirmed that laugh tracks do indeed increase audience laughter and the audience's rating of the humorousness of the comedy material. However, scientists did not consider that, in the absence of a joke or a remark, laughter by itself can evoke laughter. This is a key element in the propagation of contagious laughter.
I recently performed some investigations of the phenomenon of contagious laughter in an undergraduate psychology classroom. The stimulus was a "laugh box" - a small battery-operated record player from a novelty store -that emitted an 18-second span of laughter. The "canned" laughter was played 10 times, with the beginning of each segment separated by a one-minute interval.
On the first stimulus nearly half of the students reported that they responded with laughter themselves. (More than 90 percent reported smiling on the first stimulus.) However, the effectiveness of the stimulus declined with each repetition until only 3 of the 128 students laughed on the tenth trial. By that point about 75 percent of the students rated the laugh stimulus as "obnoxious."
The negative effect of the repeated stimulus seems to go beyond the response expected from the recurrent exposure to a generic auditory stimulus, such as "Hello, my name is George." The reaction may reflect the deep biological significance of laughter, which in this case may be perceived as jeering or ridicule. (Colleagues whose offices adjoin my own can attest to the aversiveness of periodic canned laughter. Personally, find myself wincing every time one of the laugh boxes in my office is accidently activated.) Certainly it is pleasurable to laugh at or with people, but it is quite unpleasant to be laughed at, or to be the recipient of a scornful "ha." Court fools and presidential aides learn early in their careers that it is safer to laugh with the boss than at him or her.
The efficacy of laughter alone to elicit laughter raises the intriguing possibility that human beings have auditory "feature detectors" - neural circuits that respond exclusively to this species-typical vocalization. In turn, the feature detector triggers the neural circuits that generate the stereotyped action pattern of laughter. This mechanism, involving a laugh detector that drives a laugh generator, may be the foundation of contagious laughter. (Contagious yawning appears to involve a similar process in the visual domain.) Those who attempt to explain away their laugh-evoked (contagious) laughter as nothing more than a response to a "funny" stimulus are saying that they laughed in response to a stimulus that made them laugh, a circular argument.
The structural simplicity and species-typical character of laughter makes it a prime candidate for the evolution of such a laugh detection and releasing process. Future psychophysical studies must determine which of laughter's parameters - note structure, note duration, internote interval and amplitude dynamics - are necessary for the perception of laughter and the activation of the hypothetical laugh detector or releasing mechanism. Similar detectors may have evolved for universal phonemic features of speech but the variability and complexity of language and the absence of a contagious response to assay the activation of the detectors will make their discovery more difficult.
Future Directions
Now that the critical dimensions of laughter as a social stimulus and motor act have been identified, we can pursue a variety of promising issues. Consider "pathological laughter," a frequent and often vaguely described medical symptom. Damage to a wide variety of brain regions produces abnormal laughter, a result consistent with the diverse emotional, respiratory, motor, cognitive and communicative aspects of the act. The most common cases of pathological laughter are found in pseudobulbar palsy, gelastic epilepsy and psychiatric illness. However, pathological laughter has also been reported in multiple sclerosis, amyotrophic lateral sclerosis (Lou Gehrig's disease), and cases of tumors and lesions (especially in the limbic system and the brain stem). Particularly mystifying to both patient and clinician are sudden bursts of laughter that are not associated with a feeling of mirth or an environmental stimulus. Here we have a segregation of the emotional, cognitive and motor mechanisms of laughter. Other cases are more subtle.
Some people with forebrain damage have their readjustment to society impeded by a tendency to laugh at almost anything - breaches in laugh etiquette have more serious consequences than one might think. Using our improved descriptive tools, we can now specify more precisely what is "abnormal," "pathological" or "inappropriate" about these cases (whether it is sonic structure, placement in speech, social context, contagion sensitivity, perception or relation to humor). We may even discover new laugh-related syndromes.
The next time that you or a friend have one beer too many, you may research the age-old question of alcohol effects - while taking careful notes on a cocktail napkin, of course. Do alcohol, "laughing gas" and other drugs known to increase laughter simply lower the threshold for laughter, or do they alter its pattern or quality? In aphasia (a disorder of language production or perception) is there sparing of laughter and, if so, which of laughter's several dimensions are spared? Does vocal laughter punctuate the signed speech of the congenitally deaf, in whom there is not a shared organ of expression? The left cerebral hemisphere has a specialized role in language - is this also true of the production or perception of laughter?
Many developmental issues remain open. Laughter typically appears in human babies around 3-1/2 to 4 months of age, but we know little about the details of the developmental process. Must babies hear their own laughter or the laughter of others for laughter to mature? If so, is there a critical period during which such laughter must be experienced? The report of laughter in a few congenitally deaf-blind children suggests that at least some features of laughter develop without benefit of auditory and visual stimulation, evidence of a strong maturational and genetic basis.
For a more satisfying account of laugh acquisition, we must conduct high- resolution studies that contrast the development of normal and hearing- impaired children.
All of us have encountered people with bizarre-sounding laughter. What is different about such laughter and what does this tell us about the mechanism of normal laugh production? Do these odd types of laughter run in families? If so, what is the nature of its development and heritability? In my otherwise forgettable high-school physics class there was a kid who brayed like a donkey when he laughed. Where is Roger now that I need him?
Comparative studies may provide clues about both the evolution and social function of laughter. Does the low level of conscious control that we have over our own laughter reflect the typical level of control that non-uman animals have over their own species-typical vocalizations? Do the great apes show the sexually dimorphic or contagious laughter described in human beings? Does the pattern of laughter vary with rank within a troop? Aside from the great apes, do other animals produce laugh-like vocalizations? How do the neurobehavioral mechanisms of laugh production vary between species? Tickle may be a kind of Rosetta Stone for such comparative laugh research because it triggers laugh-like vocalizations in all of the great apes and perhaps other species. Can you tickle your pet dog or cat? How can you tell? Is a laugh-evoking stimulus that works equally well in a variety of species the ultimate example of "low" humor?
Laughter research is still in its infancy, an exciting time when the frontiers are near at hand and accessible with modest resources. Certainly much of the research described in this article can be replicated or extended by almost anyone, making it suitable for college or even high school research projects. Laughter research is a reminder that not all science concerns arcane or narrow problems. We should resist neglecting or trivializing the commonplace. There are rewards for approaching nature with a naive curiosity and attempting to see the familiar in new ways.
Bibliography
Apte, M. L. 1985. Humor and Laughter: An Anthropological Approach. Ithaca, N.Y.: Cornell University Press.
Black, D. W. 1982. Pathological laughter. Journal of Nervous and Mental Diseases 170:67-71.
Chapman, A. J., and H. C. Foot. 1976. Humor and Laughter: Theory, Research and Applications. New York: Wiley.
Chapman, A. J., and H. C. Foot. 1977. It's a Funny Thing, Humour. Oxford: Pergamon Press.
Demento, Dr. 1985. The Okeh laughing record. Dr. Demento Presents the Greatest Novelty Records of All Time: Volume 1. The 1940's (and before) (Rhino Records R4 70820). Santa Monica, Calif.: Rhino Records.
Fry, W. F., Jr. 1963. Sweet Madness: A Study of Humor. Palo Alto, Calif.: Pacific.
Marler, P., and R. Tenaza. 1977. Signalling behavior of apes with special reference to vocalization. In How Animals Communicate, pp. 965- 1033, ed. T. A. Seboek. Bloomington: Indiana University Press.
Provine, R. R. 1986. Yawning as a stereotyped action pattern and releasing stimulus. Ethology 72:109-122.
Provine, R. R. 1992. Contagious laughter: Laughter is a sufficient stimulus for laughs and smiles. Bulletin of the Psychonomic Society 30:1- 4.
Provine, R. R. 1993. Laughter punctuates speech: Linguistic, social and gender contexts of laughter. Ethology 95:291-298.
Provine, R. R. In press. Contagious yawning and laughter: Significance for sensory feature detection, motor pattern generation, imitation, and the evolution of social behavior. In Social Learning in Animals: The Roots of Culture, ed. C. M. Heyes and B. G. Galef. New York: Academic Press.
Provine, R. R., and K. R. Fischer. 1989. Laughing, smiling, and talking: Relation to sleeping and social context in humans. Ethology 83:295-305.
Provine, R. R., and Y. L. Yong. 1991. Laughter: A stereotyped human vocalization. Ethology 89:115-124.
|
Q:
Uncaught SyntaxError: Unexpected token <
I am running a server on go. When I access the localhost:8888/static/ajax.html,
I get no errors. But when I just access localhost:8888 I get
an error saying:
"Uncaught SyntaxError: Unexpected
token <"
By default "/" serves ajax.html file, but doing so I don't get the
expected result. On the other hand on calling /static/ajax.html I am
getting the expected result without any errors.
server.go contains the following :
package main
import (
"http"
"flag"
)
//var path = flag.String("storage", "/home/chinmay/work/jstree/", "Set storage directory, use absolute path")
var root = flag.String("root", "/home/chinmay/work/ajax/", "Set root directory, use absolute path")
func temp(w http.ResponseWriter, r *http.Request){
http.ServeFile(w,r,*root +"ajax.html")
}
func main(){
http.HandleFunc("/", temp)
http.Handle("/static/", http.FileServer(*root, "/static/"))
http.ListenAndServe(":8888", nil)
}
ajax.html contains the following:
<html>
<head>
<script type="text/javascript" src="/static/jquery.js"></script>
<script type="text/javascript" src="/static/three/Three.js"></script>
<script type="text/javascript" src="/static/js/Detector.js"></script>
<script type="text/javascript" src="/static/js/RequestAnimationFrame.js"></script>
<script type="text/javascript" src="/static/js/Stats.js"></script>
<script type="text/javascript">
$.ajaxSetup ({
cache:false;
})
$("#container").ready(function(){
$("button").click(function(){
$.getScript("model.js");
});
});
</script>
</head>
<body>
<button>Use Ajax to get and then run a JavaScript</button>
<div id="container">
</div>
</body>
</html>
socket.js : http://www.grideads.net/redmine/attachments/download/113/socket.js
model.js : http://www.grideads.net/redmine/attachments/download/112/model.js
A:
I did not decode the question, but a possible issue is the way you call your script
$.getScript("model.js")
http://localhost:8888/ will call http://localhost:8888/model.js
http://localhost:8888/static/ajax.html will call http://localhost:8888/static/model.js
EDIT
also your model.js has an error on line 133
for( j = 0, jl = meshes.length; j < jl; j++ )
the proper for loop format is
for (variable=startvalue;variable<=endvalue;variable=variable+increment)
j < jl; is extra, resulting in the "Unexpected token <" error message
|
Diaper Junction Review
Diaper Junction is a great company offering cloth diapers and accessories. Here is my review and a Diaper Junction coupon code for all of you savvy parents out there. They are based out of Virginia Beach, Virginia, and ship out online orders.
First, let me begin with my positive experience and then I will chat more about the products.
1. Easy shopping experience on their site and in store
2. All different kinds of cloth diapers are available
3. Necessary accessories are available
4. Potty training pants are available for order
5. Breastfeeding supplies
6. They sell gently used items as well
7. Very fair pricing!
Diaper Junction
Here is a list of what to expect to see available:
1. Prefolds
2. Fitted
3. Hybrids
4. Contour
5. Newborn
6. Nighttime Heavy Wetters
7. Organic
8. Swim
9. Gently Used items
Another great thing I love about this company is the ability to try out cloth diapers risk-free for 30 days! Yes, that’s right, you are able to take them for a test run to learn which type of diaper is best for your baby’s needs! Another thing to take note of is how you earn rewards for every purchase you make online. To grab your risk-free 30 day trial, click here: DiaperJunction.com.
Suggested Reading: Kawaii Diapers Review and Kissed By The Moon Review
Online Coupons:
Search Coupon Codes:
Order Clipped and Whole Coupon Inserts Here:
Who’s Behind Coupons and More?
Hi, I'm Debbie and welcome to my frugal living blog! I search the net looking for the best online coupon codes, printable coupons, and sales to share with my readers. I hope to help others save their hard earned money with coupons like I have.
Please check out my other blog:Coupon Inserts for Sunday coupon preview's and store coupon matchups. |
<?xml version="1.0" encoding="utf-8"?><!-- Copyright (C) 2014 The Android Open Source Project
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->
<selector xmlns:android="http://schemas.android.com/apk/res/android">
<item
android:state_enabled="true"
android:state_pressed="true"
>
<set>
<objectAnimator
android:duration="100"
android:propertyName="translationZ"
android:valueTo="4dp"
android:valueType="floatType"
/>
<objectAnimator
android:duration="0"
android:propertyName="elevation"
android:valueTo="2dp"
android:valueType="floatType"
/>
</set>
</item>
<!-- base state -->
<item android:state_enabled="true">
<set>
<objectAnimator
android:duration="100"
android:propertyName="translationZ"
android:startDelay="100"
android:valueTo="0"
android:valueType="floatType"
/>
<objectAnimator
android:duration="0"
android:propertyName="elevation"
android:valueTo="2dp"
android:valueType="floatType"
/>
</set>
</item>
<item>
<set>
<objectAnimator
android:duration="0"
android:propertyName="translationZ"
android:valueTo="0"
android:valueType="floatType"
/>
<objectAnimator
android:duration="0"
android:propertyName="elevation"
android:valueTo="0"
android:valueType="floatType"
/>
</set>
</item>
</selector>
|
What is this?
HALIFAX – Saint Mary’s University says after a year of working behind the scenes it’s ready to roll out a plan that will close the door on a dark year in the school’s history.
The Welcome Week revamp was sparked by the controversial chant that came to light during last year’s introduction week events. At a now cancelled event called Turf Burn students and frosh leaders sang a sexually explicit chant condoning non-consensual sex and under age sex.
The chant was widely condemned, and in the aftermath the student association president resigned and the university took over the management of all Welcome Week activities.
“There’s a great deal happening and most of it is going to roll out and become visible at the start of the academic year,” said Margaret Murphy, Associate Vice President of External Affairs at St. Mary’s. “So it starts with Welcome Week but we’re not going to accomplish cultural change in one week. It’s the start of activities that will roll out through the month of September, through this entire academic year and continues years beyond.”
Welcome Week will be the start of what she says she hopes will be a much bigger cultural shift at the university.
A cultural shift is exactly what Lewis Rendell says the university needs. She was the media contact for the campus Women’s Centre when the controversy was sparked in September 2013. She says she felt pressured by people at her own school to stop speaking out, ultimately she said the stress forced her to take a break from the university. She won’t be back until January but she says she hopes to see a difference.
“I’m hoping for a bigger focus for safety for women and LGBT people on campus because historically at St Mary’s it hasn’t been the most welcoming place,” said Rendell.
SMU administration said this year’s introduction week will have a stronger focus on academics, and faculty will be more involved in the activities. Murphy said the activities will also involve learning modules and more training around sexual health and campus safety.
“Talking to them about alcohol prevention, bystander training, talking about consent. We’re going to be teaching bystander training, rolling it out peer to peer, starting with the new students, starting with the student leaders, its also going to touch faculty and staff,” said Murphy.
But Murphy says university administration recognizes they still have a ways to go before everyone feels safe again.
“I’m really hopeful that in the year ahead that those students that if they felt marginalized in the past, that they see the difference,” said Murphy. “We want to stand for the campus that is welcoming to the LGBTQ community as well as all of the other diversity on campus.”
She knows it will take more than just one introductory week to make that change a reality.
“If people see something going on, they know that they have the support of the St. Mary’s community, whether it’s their fellow students, their professors, there is support. If you see something happening you can step in and be that positive bystander and they’re not going to be alone,” said Murphy. “That’s not all going to happen in week one, that takes time, and it takes people seeing it starting with student leaders and other leaders on campus.” |
Gävleborg County
Gävleborg County () is a county or län on the Baltic Sea coast of Sweden. It borders to the counties of Uppsala, Västmanland, Dalarna, Jämtland and Västernorrland. The capital is Gävle.
Provinces
Gävleborg County encompasses the provinces of Gästrikland and Hälsingland, except for the northwestern part of the latter which is located in Jämtland County, most notably Ytterhogdal.
Administration
Gävleborg County was established in 1762 when it was separated from Västernorrland County. For the list of Governors see main article.
The main aim of the County Administrative Board is to fulfil the goals set by the national policy by the Riksdag and the Government, to coordinate the interests and promote the development of the county, to establish regional goals and safeguard the due process of law in the handling of each case. The County Administrative Board is a Government Agency headed by a Governor. See List of Gävleborg Governors.
Politics
The county council of Gävleborg or Region Gävleborg.
After the county council election in 2018, the following political parties are represented in the Gävleborg county council:
Governors
Municipalities
In Gästrikland Province:
Gävle
Hofors
Ockelbo
Sandviken
In Hälsingland Province:
Bollnäs
Hudiksvall
Ljusdal
Nordanstig
Ovanåker
Söderhamn
Heraldry
The arms for Gävleborg County is a combination of the arms of Gästrikland and Hälsingland. When it is shown with a royal crown it represents the County Administrative Board. Blazon: "Quartered, the arms of Gästrikland and the arms of Hälsingland."
See also
Duke of Gästrikland & Duke of Hälsingland, titles for members of the royal family (see Duchies in Sweden)The tiles have been created for the first time for present Princess Madeleine, Duchess of Hälsingland and Gästrikland
References and notes
External links
Gävleborg County Administrative Board
Gävleborg Regional Development Council
Gävleborg County Council
Category:Counties of Sweden
Category:Gästrikland
Category:Hälsingland
Category:1762 establishments in Sweden
Category:States and territories established in 1762 |
A few years ago, Wim approached a Nijmegen-based research team at the Radboud Medical Center and challenged them to scientifically investigate his purported ability to influence his autonomic nervous system – a skill commonly deemed impossible by the medical community.
The first two studies were conducted by Dutch physicians Dr. Matthijs Kox and Prof. Peter Pickkers who work on studying the human immune response, mainly in ICU patients, to help reduce the mortality of critically ill patients. They were able to publish their work in high-ranking scientific journals and continue to investigate the effects the WHM has in various patient groups.
Enjoy this week’s episode where we bring you a fascinating, rich, and highly articulate interview with Matthijs Kox, the man who has investigated the Iceman! |
Why disease persists: an evolutionary nosology.
Although natural selection might be expected to reduce the incidence and severity of disease, disease persists. Natural selection leads to increases in the mean fitness of populations and so will reduce the frequency of disease-associated alleles, but other evolutionary processes, such as mutation and gene flow, may introduce or increase the frequency of these deleterious alleles. The pleiotropic actions of genes and the epistatic interactions between them complicate the relationship between genotype and phenotype, and may result in the preservation of disease-associated alleles. Deleterious alleles may also be maintained because of linkage to beneficial alleles. The inability of natural selection to eliminate diseases of aging is a reminder that fitness -- success in producing progeny, or in contributing genes to the population gene pool -- is not equivalent to the absence of disease. Nutritional or psychosocial cues may lead to life history strategies that maximize survival to reproductive maturity at the expense of disease later in life. Natural selection acts on genes, cells, and groups, as well as on organisms; the outcome of evolution reflects selection at different levels of biological organization. Finally, the human environment is constantly changing, largely because of the evolution of our parasites and because of changes in cultural beliefs and practices; genetic evolution is comparatively slow and lags behind environmental change. An evolutionary nosology complements traditional medical nosologies and enhances our understanding of the persistence of disease and the meaning of human variation. |
CheatCC’s PlayStation Meeting Predictions
The PlayStation meeting is upon us! It's time to stop speculating and guessing what Sony might do in response to Microsoft announcing Project Scorpio, and see what they have up their sleeves at last. There are a few things that we know and a few things that we think we know. At this point, I think we can piece together a pretty clear picture of what Sony will be presenting. Let's take a moment to recap. What are we talking about again? Sony is holding a special event, simply called "PlayStation Meeting," on Wednesday, September 7. The event will begin at 3PM EST at the PlayStation Theater in New York, but you can watch the livestream from the comfort of your home (or cubicle). Invitations were sent to the media to attend the event where Sony will "share details about the PlayStation business," which means they'll likely discuss multiple initiatives and products. It could be really exciting, because French publication Gameblog, who correctly predicted the date of the PlayStation Meeting, stated that the main point of the meeting would be to reveal the PS4 Neo.
What can we count on? There are few new products that we know we can count on seeing soon. Since this meeting is being held to discuss all things PlayStation, we can pretty much guarantee that a few new items will share the limelight. PlayStation on PC The PlayStation Now service very recently began rolling out on PCs worldwide. Sony also revealed a new DualShock 4 USB wireless adapter that you can use with your PC to play games wirelessly as well. I'd be surprised if they don't take a little time to promote these things. Along with PS4 remote play, these features can turn any capable PC or laptop into the semi-portable PlayStation hub. The PlayStation Meeting could coincide with the launch of PS Now on North American PCs. PS4 Slim Well thanks to some goober who goes by the name "shortman82" on Twitter, we already know quite a bit about the PS4 Slim. He's shared photos and details about the console, the new controller, and different wifi options. His photos indicate that he received his PS4 Slim in what looks like finalized retail packaging, so it'll probably hit shelves pretty soon. Expect Sony to officially reveal the Slim, and announce a release date in the very near future. |
Q:
Add rows dynamically to a table with dynamically recieved number of columns
I need to add rows to a table dynamically when the command was given by the user to add them to the end of a printed rows. But the number of columns must add for a particular row vary by table to table.
The number of columns should add is getting dynamically to each table. So the printed number of columns must vary time to time. This is the code I used.
$(document).ready(function(){
$("#anc_add").click(function(){ //This is the ID of the button where user gives the command to add rows.
var Total=$("#NumberOfColumns").val(); //This is the number of input boxes I must add to that row......
$('#tbl1 tr').last().after(
'<tr><td>Static Content ['+cnt+']</td><td><input type="text" name="txtbx'+cnt+'" value="'+cnt+'">;
); // I need to print the given(Total) number of input boxes instead of printing textboxes like this.
cnt++;
});
});
Could anyone please suggest me a way to do this. Thank you so much.
A:
This will loop the total number of inputs you want to add.
$("#anc_add").click(function(){ //This is the ID of the button where user gives the command to add rows.
var Total=$("#NumberOfColumns").val();
var rowElem = $('<tr>').append(
$('<td>').html('Static Content ['+cnt+']')
);
for(var i = 0; i < Total; i++) {
rowElem.append(
$('<td>').html('<input type="text" name="txtbx'+cnt+'" value="'+cnt+'">')
)
cnt++;
}
$('#tbl1 tr').last().after(
rowElem
);
});
});
|
This calculator works out how much you can pay for a house. It figures out how large a mortgage your deposit and your payment capacity will support. You should be conservative when adding those two numbers because you need to allow for some sensible headroom in case unexpected things happen. And also don't forget the bank will apply a 'serviceability test' of their own, usually by checking you could still make the payments if the interest rate rose by a full +2%. You can do that in this calculator too.
It is here as part of our partnership with Calculate.co.nz. |
"""
用于将本地天软分钟数据转换为 QA 格式
有条件的用户可自行转换
"""
import datetime
import concurrent.futures
from concurrent.futures import ProcessPoolExecutor, ThreadPoolExecutor
import pandas as pd
import pymongo
import os
import QUANTAXIS as QA
from QUANTAXIS.QAFetch.QATdx import QA_fetch_get_stock_list
from QUANTAXIS.QAUtil import (
DATABASE, QA_util_date_stamp, QA_util_get_real_date, QA_util_log_info,
QA_util_time_stamp, QA_util_to_json_from_pandas, trade_date_sse)
def QA_SU_trans_stock_min(client=DATABASE, ui_log=None, ui_progress=None,
data_path: str = "D:\\skysoft\\", type_="1min"):
"""
将天软本地数据导入 QA 数据库
:param client:
:param ui_log:
:param ui_progress:
:param data_path: 存放天软数据的路径，默认文件名格式为类似 "SH600000.csv" 格式
"""
code_list = list(map(lambda x: x[2:8], os.listdir(data_path)))
coll = client.stock_min
coll.create_index([
("code", pymongo.ASCENDING),
("time_stamp", pymongo.ASCENDING),
("date_stamp", pymongo.ASCENDING),
])
err = []
def __transform_ss_to_qa(file_path: str = None, end_time: str = None, type_="1min"):
"""
导入相应 csv 文件，并处理格式
1. 这里默认为天软数据格式:
time symbol open high low close volume amount
0 2013-08-01 09:31:00 SH600000 7.92 7.92 7.87 7.91 518700 4105381
...
2. 与 QUANTAXIS.QAFetch.QATdx.QA_fetch_get_stock_min 获取数据进行匹配，具体处理详见相应源码
open close high low vol amount ...
datetime
2018-12-03 09:31:00 10.99 10.90 10.99 10.90 2.211700e+06 2.425626e+07 ...
"""
if file_path is None:
raise ValueError("输入文件地址")
df_local = pd.read_csv(file_path)
# 列名处理
df_local = df_local.rename(
columns={"time": "datetime", "volume": "vol"})
# 格式处理
df_local = df_local.assign(
code=df_local.symbol.map(str).str.slice(2),
date=df_local.datetime.map(str).str.slice(0, 10),
).drop(
"symbol", axis=1)
df_local = df_local.assign(
datetime=pd.to_datetime(df_local.datetime),
date_stamp=df_local.date.apply(lambda x: QA_util_date_stamp(x)),
time_stamp=df_local.datetime.apply(
lambda x: QA_util_time_stamp(x)),
type="1min",
).set_index(
"datetime", drop=False)
df_local = df_local.loc[slice(None, end_time)]
df_local["datetime"] = df_local["datetime"].map(str)
df_local["type"] = type_
return df_local[[
"open",
"close",
"high",
"low",
"vol",
"amount",
"datetime",
"code",
"date",
"date_stamp",
"time_stamp",
"type",
]]
def __saving_work(code, coll):
QA_util_log_info(
"##JOB03 Now Saving STOCK_MIN ==== {}".format(code), ui_log=ui_log)
try:
col_filter = {"code": code, "type": type_}
ref_ = coll.find(col_filter)
end_time = ref_[0]['datetime'] # 本地存储分钟数据最早的时间
filename = "SH"+code+".csv" if code[0] == '6' else "SZ"+code+".csv"
__data = __transform_ss_to_qa(
data_path+filename, end_time, type_) # 加入 end_time， 避免出现数据重复
QA_util_log_info(
"##JOB03.{} Now Saving {} from {} to {} == {}".format(
type_,
code,
__data['datetime'].iloc[0],
__data['datetime'].iloc[-1],
type_,
),
ui_log=ui_log,
)
if len(__data) > 1:
coll.insert_many(
QA_util_to_json_from_pandas(__data)[1::])
except Exception as e:
QA_util_log_info(e, ui_log=ui_log)
err.append(code)
QA_util_log_info(err, ui_log=ui_log)
executor = ThreadPoolExecutor(max_workers=4)
res = {
executor.submit(__saving_work, code_list[i_], coll)
for i_ in range(len(code_list))
}
count = 0
for i_ in concurrent.futures.as_completed(res):
strProgress = "TRANSFORM PROGRESS {} ".format(
str(float(count / len(code_list) * 100))[0:4] + "%")
intProgress = int(count / len(code_list) * 10000.0)
count = count + 1
if len(err) < 1:
QA_util_log_info("SUCCESS", ui_log=ui_log)
else:
QA_util_log_info(" ERROR CODE \n ", ui_log=ui_log)
QA_util_log_info(err, ui_log=ui_log)
if len(err) < 1:
QA_util_log_info("SUCCESS", ui_log=ui_log)
else:
QA_util_log_info(" ERROR CODE \n ", ui_log=ui_log)
QA_util_log_info(err, ui_log=ui_log)
if __name__ == "__main__":
QA_SU_trans_stock_min()()
|
<?xml version="1.0" encoding="UTF-8"?>
<beans xmlns="http://www.springframework.org/schema/beans" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns:context="http://www.springframework.org/schema/context" xmlns:tx="http://www.springframework.org/schema/tx"
xsi:schemaLocation="http://www.springframework.org/schema/beans
http://www.springframework.org/schema/beans/spring-beans-3.0.xsd
http://www.springframework.org/schema/tx
http://www.springframework.org/schema/tx/spring-tx-3.0.xsd
http://www.springframework.org/schema/context
http://www.springframework.org/schema/context/spring-context-3.0.xsd">
<context:annotation-config />
<context:component-scan base-package="fm.last.musicbrainz.data" />
<bean id="musicBrainzDataSource" class="com.mchange.v2.c3p0.ComboPooledDataSource" init-method="getConnection"
destroy-method="close" lazy-init="false">
<property name="driverClass" value="${db.musicbrainz.driver.class}" />
<property name="jdbcUrl" value="${db.musicbrainz.url}" />
<property name="user" value="${db.musicbrainz.user}" />
<property name="password" value="${db.musicbrainz.password}" />
<property name="preferredTestQuery" value="SELECT version();" />
<property name="initialPoolSize" value="${db.musicbrainz.initial.pool.size:1}" />
<property name="maxPoolSize" value="${db.musicbrainz.max.pool.size:1}" />
<property name="acquireIncrement" value="${db.musicbrainz.acquire.increment:1}" />
<property name="maxIdleTime" value="${db.musicbrainz.max.idle.time:0}" />
<property name="idleConnectionTestPeriod" value="${db.musicbrainz.idle.connection.test.period:0}" />
</bean>
<bean id="musicBrainzSessionFactory" class="org.springframework.orm.hibernate3.annotation.AnnotationSessionFactoryBean">
<property name="dataSource" ref="musicBrainzDataSource" />
<property name="hibernateProperties">
<props>
<prop key="hibernate.dialect">org.hibernate.dialect.PostgreSQLDialect</prop>
<prop key="hibernate.hbm2ddl.auto">validate</prop>
<prop key="hibernate.show_sql">${hibernate.show_sql:false}</prop>
<prop key="hibernate.cache.use_second_level_cache">false</prop>
<prop key="hibernate.cache.use_query_cache">false</prop>
</props>
</property>
<property name="packagesToScan">
<list>
<value>fm.last.musicbrainz.data.model</value>
</list>
</property>
</bean>
<bean id="musicBrainzTransactionManager" class="org.springframework.orm.hibernate3.HibernateTransactionManager">
<property name="sessionFactory" ref="musicBrainzSessionFactory" />
</bean>
<tx:annotation-driven transaction-manager="musicBrainzTransactionManager" />
</beans>
|
PHR sent teams to the U.S.-Mexico border. Here are eight things you should know and two things you can do.
October 5, 2018October 24, 2018
1. There are now more than 12,800 migrant children detained in the United States.
The number of migrant children reported in federal shelters increased
fivefold in the last year and a half, reaching a total of 12,800 in September
2018. The staggering increase is largely
a result of sponsors, to whom unaccompanied children are released, being scared
away.
“The U.S. administration is using children as bait,” explains PHR’s Asylum Network program officer, Kathryn Hampton. “Sponsors, many of them family members of the children who may be in the United States undocumented, come forward to claim the children and are then arrested on the spot. So it’s no wonder that fewer and fewer people are willing to step forward and sponsor these children, and that the numbers of people detained are growing. It’s essentially a system of mass detention where you’re transferring children to unlicensed facilities, indefinitely, and then detaining their sponsors,” Hampton adds.
2. Some 1,600 migrant children were transferred to a desert tent city at the beginning of October.
In
a clandestine, nighttime operation, some 1,600 migrant children were
transferred from detention centers around the United States to a tent city in
the desert in Tornillo, Texas, at the start of October. The center, which looks
like a prison from the outside, lacks adequate standards and oversight
mechanisms. This means there are no official systems in place to ensure that
the children’s best interests are being looked after.
“It
is simply unacceptable that children be held in these inhumane conditions, in
unregulated camps which have no mechanisms for accountability,” said Dr. Ranit
Mishori, PHR Asylum Network member and medical consultant.
“As health professionals, we wonder what goes on behind
these bars, jail walls, behind the fences, and in the tents in the desert. We
worry about the basic needs of these children – whether and how they are being
met,” Mishori adds.
3. Clinical evidence shows that detention of any kind has negative health effects on children.
PHR’s Asylum Network is made up of more than 1,200
health professionals who do pro bono work throughout the United States – many
with migrant children along the U.S.-Mexico border. Our work shows us time and
time again that detention causes intense psychological distress in children, often
resulting in developmental delays and other life-long symptoms.
Hajar Habbach, PHR’s Asylum Network program
associate, visited a detention center in Texas with Dr. Mishori as part of a
trip by human rights organizations and faith groups to deliver aid to detained
children. “What really struck me was the contradictory setting in which we
found ourselves,” she said. “This detention facility, the Ursula facility in
McAllen, was across the street from a community baseball field. So, while we
were delivering aid to detained migrant children who don’t have adequate access
to water, food, clothing, or a proper education, the local children were
playing little league baseball across the street,” Habbach added.
Said Dr. Mishori: “In addition to feelings of isolation, detention can cause or exacerbate trauma and contribute to ‘adverse childhood experiences’ or ACEs. Such experiences can disrupt actual brain development, alter the very architecture of the brain, which in turn can result in social, emotional, and cognitive impairment – which can last decades. Extreme and repetitive stress has a name – it is called toxic stress – and it is linked to an increased risk of developing chronic mental health conditions: depression and post-traumatic stress disorder and even physical conditions such as cancer, stroke, diabetes, and heart disease.
4.The U.S. administration is attempting to dismantle protections so that children can be detained indefinitely.
The Trump administration wants to revoke the Flores
settlement that prevents the detention of minors for more than 20 days.
Revoking this would essentially mean that children can be detained
indefinitely.
“Not only do they want to extend detentions
indefinitely, they are also attempting to legalize tent cities by eliminating the
requirement for state licensing for detention facilities,” explains Hampton. “This
would mean that children could be held practically anywhere – even in places
where there are no proper mechanisms for accountability or for ensuring that
conditions are suitable for these children. And they could be held there
without any time limit being placed on their detention.”
Dr. Mishori adds: “The Flores settlement states that the
children in custody should be ‘treated with dignity, respect and special concern for their
particular vulnerability as minors’, but I cannot see how dignity and respect can be honored
when one is confined to a long-term detention facility. The decision to allow
indefinite detention goes against the Flores settlement, whose sole
purpose is to protect children from harm.”
5. The journey across the U.S.-Mexico border is a treacherous one, and most migrants don’t embark on the journey because they want to – they do it because they have to.
“Many of
those who decide to make the journey are fleeing abuse back home, including
gang violence and domestic violence. They don’t want to leave their homes, they
have to, in order to save themselves from violence and potential death,”
explains Hampton.
Then, along the treacherous journey, they face
similar threats. There are snakes and scorpions to contend with, and the water
which collects in ravines is contaminated by the feces of livestock who roam
the area. Those who drink the water are likely to experience vomiting and
diarrhea, which dangerously dehydrates them further.
“Exposure to the elements is the major cause of fatalities, according to forensic reports of local medical examiners – but harsh border patrol apprehension practices, such as high speed nighttime chases with dogs and helicopters, also increase the risk of injury and death. Smugglers also may abandon their charges or deceive them about the difficulties of crossing, leaving them vulnerable and unprepared for the dangers they face,” Hampton added.
PHR Executive Director Donna McKay, who traveled to Texas in August, added: “Despite these threats, people are still deciding to make the brutal journey and the overall number of asylum seekers has not declined. More families are arriving with children than in previous years. The people I spoke with in Texas explained it this way: ‘If your house is burning, you flee. No matter what.’”
6.Hundreds of people die each year trying to cross the border.
In 2013, U.S. Customs and Border Protection counted
more than 2,300 people, in just that year, who had to be rescued along the
border. Each year before that, and every year since, hundreds of migrants have
died while trying to make the treacherous journey in wildly variable temperatures which can
exceed 115 degrees Fahrenheit in summer months.
“There are hundreds of fatalities every year,” says Hampton. “And some doctors, nurses, and EMTs are being arrested and prosecuted as ‘smugglers,’ just for providing urgent medical care and water to the migrants. Whenever migrants decide to embark on the journey, they know that they might not make it to the other side alive.”
7.There are still about 400 children who are separated from their parents.
It’s
been more than three months since U.S. President Trump signed an Executive
Order ending family separations, but today, some 400 children remain separated
from their parents. In most of these cases, the parents were deported.
“What we learned was that many deportations take place under duress or through manipulation. In addition, more than 900 parents have been deemed ‘ineligible’ for reunification and stripped of their parental rights, despite the fact that the government has failed to demonstrate in any way that they were unfit, nor that they had committed any crime apart from irregular border crossing,” McKay explains. “These deportations result in hundreds of immigrant children, who have loving and caring parents, being forcibly orphaned in the United States. What’s even more alarming is that, as a result of the U.S. government’s failure to properly document separated families, many of these children are essentially lost in the system, which means they may never be reunited with their parents, ever,” she adds.
An internal report by the U.S. Department of Homeland Security stated that Border Patrol took no measures to ensure that pre-verbal children could be correctly identified after separation.
8. Most detainees are afraid to speak out about the conditions in detention.
“When we walked through the West Texas Detention Facility,
the conditions which we observed looked very un-therapeutic. It was clear that
many migrants were being treated as criminals. Their crime: seeking asylum,”
Hampton says.
“We met a woman who didn’t speak a word of
English. She was actually from India, but had entered the United States via
Mexico. She spoke only Gujarati when she arrived, but after a full year of
detention, she had learned some basic Spanish from her fellow detainees. It was
unclear which, if any, accommodations had been made to ensure that she would be
able to access information about medical and mental health care in the facility,
or to report complaints,” Hampton adds.
“We also met a counselor who didn’t speak a word
of Spanish. Imagine trying to counsel someone for their trauma, including
sexual violence, and not being able to communicate with them in their own
language. What’s worse is that the counselor told us that correctional officers
are the only translators available for the counseling sessions and are present
when these sensitive doctor-patient issues, including sexual abuse, are being
discussed.
“We saw a man who was locked up in a cell, under
solitary confinement, for having mental health issues. I can’t imagine a less
therapeutic environment, completely isolated in a small concrete room, with a
trap door for food trays. We heard these stories time and time again, but
people there told us that they are too afraid to speak up or to tell us more
details. They kept looking fearfully at the door where the guard was standing
outside. They’re afraid of being treated even worse as a result of complaining,
or of having their asylum claims rejected if they speak out about anything.
It’s a heartbreaking situation,” Hampton says.
Here are two things you can do, today:
1. Tell U.S. Secretary of Homeland Security Kirstjen Nielsen to stop the expansion of family detention at the U.S. border.Click HERE to send a message and join our unified call to respect children’s constitutional right to protection and abide by the universal human rights principle of the best interests of the child
2.Tell Congress to investigation the harmful effects of family detention.As health professionals, we know the severe trauma experienced by children in detention and we know the long-term physical and psychological effects. Click HERE to join us today in demanding that Congress immediately hold oversight hearings on the harmful and life-threatening detention facilities at the U.S. southern border. With thousands of children at risk, there is no time to lose. |
// Jest Snapshot v1, https://goo.gl/fbAQLP
exports[`TooltipIcon should allow the user to specify the direction 1`] = `
<TooltipIcon
align="center"
className="custom-class"
direction="top"
tooltipText="tooltip text"
>
<button
aria-describedby="icon-tooltip-4"
className="bx--tooltip__trigger bx--tooltip--a11y custom-class bx--tooltip--top bx--tooltip--align-center"
onFocus={[Function]}
onMouseEnter={[Function]}
type="button"
>
<span
className="bx--assistive-text"
id="icon-tooltip-4"
>
tooltip text
</span>
<svg />
</button>
</TooltipIcon>
`;
exports[`TooltipIcon should render 1`] = `
<TooltipIcon
align="center"
className="custom-class"
direction="bottom"
tooltipText="tooltip text"
>
<button
aria-describedby="icon-tooltip-1"
className="bx--tooltip__trigger bx--tooltip--a11y custom-class bx--tooltip--bottom bx--tooltip--align-center"
onFocus={[Function]}
onMouseEnter={[Function]}
type="button"
>
<span
className="bx--assistive-text"
id="icon-tooltip-1"
>
tooltip text
</span>
<svg />
</button>
</TooltipIcon>
`;
|
Celeb Dirty Laundry reported today about John Travolta’s former gay lover, Doug Gotterba, breaking his silence to reveal the intimate details of their six-year sexual relationship. According to the National Enquirer, these two lovebirds already started dating back in 1981, one year after Travolta starred in “Urban Cowboy”.
The most shocking factor behind these revelations is a rumoured sex tape doing the rounds. A what? A sex tape of Travolta and Gotterba? Could this be any worse? Well, yes, dear reader, the only event that would be worse than this sex tape would be an extinction level event, when the planet is overrun by aliens in Justin Bieber suits and bitch sized guns incinerating all you love with sparkle poop. We imagine the sex video to be very early 80’s vintage soft core porn, with a horribly synthesized soundtrack and John Travolta dancing a pre-historic Macarena with a polio gait. There’s nothing wrong with a well produced celebrity sex tape ever so often, but honestly now, how well produced could this sex tape be?
More importantly, what will this mean to a rather prominent Scientologist? Will Travolta be cleansed with the lady oils of Xenu, or will he be lambasted by his ex-BFF Tom Cruise? This scandal has been one of the most important and most well documented scandals of the past decade.
We have already washed our eyes with the strongest detergent we could find and after scraping our corneas and dabbing our eyelids in acid; we think we might be ready to move on from our wild imaginations. This tape will even scar Bruce Jenner and as we all know, nothing ever scars Brucey. He might consider moving into Khloe Kardashian’s (the lesser insane bitch) house for a day or two, because this tape might be a little bit too much for his gentle demeanor.
Bitches, why were you so stupid? Why make a sex tape back in 1981? You could have waited for 3D and HD. |
Saturday, June 30, 2007
As humans we naturally see in 3D. However, most people have only ever seen visual playback on a 2D television. Most have come to accept it. Recently though, there have been advances in 3D film.
Getting to the point, today I came across this new 3D display being developed. I urge you to at least look at the images in the paper as it is really fascinating, and if you have the bandwidth the video shows off the technology very well. It is basically works by spinning a mirror and shining light onto it to produce a 3D image viewable from all angles. Since the mirror spins at a high speed, it creates the illusion of a 3D object.
As you move around the image, so you get a different view. This allows for "view-independent effects such as occlusion." It recreates the object in real 3D space, so there's no need for any special glasses such as IMAX 3D. It also allows for a number of people to gather around the display and watch from different angles. The rendering is so efficient that they can display fully interactive scenes.
If you're interested, there is plenty more to read up about it in the paper. It goes into a fair amount of detail on how the display functions and the rendering behind it. |
Q:
mysql subquery limit is also limiting parent query
SELECT * FROM forum_posts AS post
INNER JOIN( SELECT parent AS rparent, author AS rauthor, MAX(created_at( AS rdate FROM forum_replies) AS reply
ON post.id = reply.rparent
I want to retrieve all records from forum with 1 latest reply on that thread.
Problem is the limit also effects the parent query resulting in only 1 thread being returned.
Help will be greatly appreciated, thanks in advance.
A:
As @AgRizzo commented above, you probably need to group the subquery.
Furthermore, an inner join will only result in a record where the join criterion is matched in both tables: that is, posts for which there are no replies will be excluded.
If you wish to keep records from one table even where the join criterion is not matched, you will need an outer join; in this case, a left outer join (so that records from the left operand of the join are always included in the resultset).
See A Visual Explanation of SQL Joins for more information.
Therefore:
SELECT * FROM forum_posts AS post LEFT JOIN (
SELECT parent AS rparent
, author AS rauthor
, MAX(created_at) AS rdate
FROM forum_replies
GROUP BY parent
) AS reply ON post.id = reply.rparent
Alternatively, perform the grouping after the join:
SELECT post.*
, reply.parent AS rparent
, reply.author AS rauthor
, MAX(reply.created_at) AS rdate
FROM forum_posts AS post
LEFT JOIN forum_replies AS reply ON reply.parent = post.id
GROUP BY post.id
|
tag:blogger.com,1999:blog-8416170281378297037.post1990420884184922695..comments2018-04-22T10:47:04.775-07:00Comments on Vanilla Clouds and Lemon Drops: Tea Party SandwichesLyndsey Fleenorhttps://plus.google.com/101946818146623356160noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-8416170281378297037.post-30155886872343622862011-03-24T10:08:17.916-07:002011-03-24T10:08:17.916-07:00Thanks everyone! A delicious treat for a great cau...Thanks everyone! A delicious treat for a great cause! :)Lyndshttps://www.blogger.com/profile/12797353530520663668noreply@blogger.comtag:blogger.com,1999:blog-8416170281378297037.post-25979970989538563672011-03-24T10:03:41.145-07:002011-03-24T10:03:41.145-07:00Delicous!! You make the best cucumber sandwiches ;...Delicous!! You make the best cucumber sandwiches ;) and the salmon ones were so yummy too!! xxxLaurennoreply@blogger.comtag:blogger.com,1999:blog-8416170281378297037.post-85371363300270229522011-03-23T13:28:11.895-07:002011-03-23T13:28:11.895-07:00Oh lovely
Carol
www.dormouseandtheteapot.blogsp...Oh lovely<br /><br />Carol <br /><br />www.dormouseandtheteapot.blogspot.com<br />www.dormouseandtheteapot.comcarolhttps://www.blogger.com/profile/01731986853819172058noreply@blogger.comtag:blogger.com,1999:blog-8416170281378297037.post-86642041765932424202011-03-23T06:27:17.557-07:002011-03-23T06:27:17.557-07:00The only thing better than the worthy cause for wh...The only thing better than the worthy cause for which this post is inspired are these delectable little sandwiches! The wasabi and lime cream cheese paired with the cucumber has me over the moon. Another example of how smaller portions, when executed well and packed with flavor can be as satisfying as a full-sized mealBrooks Walkerhttps://www.blogger.com/profile/13058269329581501665noreply@blogger.comtag:blogger.com,1999:blog-8416170281378297037.post-11905932387386489132011-03-22T22:17:50.192-07:002011-03-22T22:17:50.192-07:00These looks so delicious! Beautiful photos too!
N...These looks so delicious! Beautiful photos too! <br />Nicely done and great cause!Sandrahttps://www.blogger.com/profile/06183466872546715233noreply@blogger.com |
Cognos
COGN
shares surged $1.60 to close at $25.32 after the company on Wednesday posted a profit of $29.6 million, or 33 cents a share, on sales of $164 million for the period ended Feb. 28. Analysts surveyed by Thomson First Call forecast Cognos would earn 27 cents a share on sales of $161 million.
Cognos' profits nearly tripled from the $10 million, or 11 cents a share, it earned during the same period a year ago when its sales were $143 million.
The company reported operating cash flow of $56 million, bringing its total cash and cash equivalents to $242 million.
Rob Ashe, Cognos' president and chief operating officer, attributed part of the performance to completing the purchase of Adaytum, which makes planning software for large enterprises. "We think our strategy has taken a huge step since we added Adaytum," Ashe said.
Going forward, Cognos expects first-quarter revenue of between $146 million and $150 million, with profits in a range of 12 cents to 14 cents a share. Wall Street analysts have forecast Cognos to report first-quarter sales of $144.5 million and earnings of 14 cents a share.
Cognos' report gave a lift to other business-software companies. Its main rival, Business Objects
BOBJ
rose $1.56, or 9 percent, to $18.74, while Hyperion Solutions
HYSL
rose $1.44 to $27.44.
Intraday Data provided by SIX Financial Information and subject to terms of use. Historical and current end-of-day data provided by SIX Financial Information. All quotes are in local exchange time. Real-time last sale data for U.S. stock quotes reflect trades reported through Nasdaq only. Intraday data delayed at least 15 minutes or per exchange requirements. |
Heating oil help offered
Fed up with angering customers when bad weather or strife in oil-producing countries drive petroleum prices up, a growing number of Maryland heating oil dealers are offering customers protection against dramatic price increases.
By playing the futures markets and, essentially, buying price insurance, some oil dealers are able to guarantee that they won't charge more than, say, $1.05 a gallon during the upcoming winter.
The companies that offer the guarantees, however, may end up charging slightly more than those that don't, since the price insurance can cost the companies about 5 cents a gallon.
Though oil prices jumped briefly during the attempted Soviet coup, heating oil was selling recently at anywhere from 80 cents to $1 a gallon, area dealers said.
Last winter, as U.S. troops massed in Saudi Arabia, home heating oil prices reached $1.30 a gallon, but they fell more than a dime after the fighting started, according to the U.S. Department of Energy.
The year before that, Maryland heating oil prices peaked at $1.20 a gallon after a record-breaking cold spell in December 1989. Prices fell about 20 cents a gallon after temperatures warmed up in February 1990.
Les Hubbard, general manager of Southern Maryland Oil, said that his company and a sister company, Delmarva Oil, will promise customers on automatic delivery plans that they won't be charged more than $1.049 per gallon this winter.
If prices fall, the two oil companies will pass through the savings, Mr. Hubbard said. The only catch is that customers have to agree to take delivery on his company's schedule.
Mr. Hubbard said that the two oil companies, which are owned by a La Plata-based company called the Wills Group, decided to offer the guarantees this year after trying to calm customers who were upset last fall when home heating oil prices rocketed because of Iraq's invasion of Kuwait.
He said that the companies are able to offer the guarantees by buying futures contracts guaranteeing delivery of oil at set prices. The companies won't lose money unless the bottom falls out of the market and prices drop to about half their current $22-a-barrel level, he said.
Mr. Hubbard said that he thinks oil customers are tired of price fluctuations and that such price guarantees "are the future."
Mark Londer, a manager for Wilhelm Oil in Randallstown, said that his company has offered such guarantees for more than five years. Recently, he has noticed competitors starting to offer similar programs.
Mr. Londer said that his company is guaranteeing its customers that its price won't go above 99 cents a gallon this winter. His company, too, promises to pass through savings if prices fall.
Alan Kaufman, president of a firm that sells price insurance policies, said that he charges dealers about 5 cents a gallon to guarantee that they will be able to keep their prices about $1 a gallon this winter.
He says his company, Mount Lucas Associates of Princeton, N.J., is able to sell the insurance policies by buying and selling contracts for future oil deliveries at different prices. Winter delivery oil is now selling for about 65 cents a gallon on the wholesale markets.
Mr. Kaufman said that when he first started offering price insurance to oil dealers, "people were skeptical" and thought the nickel-a-gallon charge was too high. But after the wild price fluctuations of the past two winters, "business has grown dramatically," he said.
Not all local dealers like the idea, though.
Rick Phelps, president of Carroll Independent Fuel Co. of Baltimore, said that he has considered buying price insurance, especially after last year's Persian Gulf crisis, but "didn't like the risk" associated with hedging through futures contracts.
Though it is painful, customers who do have to order oil during high price periods usually end up paying only about $30 more for the entire heating season, Mr. Phelps said.
"To be honest, I haven't heard a big clamor from people wanting fixed prices," he said. |
/*
Copyright 2017 The Kubernetes Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package apiclient
import (
"net"
"strings"
"github.com/pkg/errors"
"k8s.io/api/core/v1"
apierrors "k8s.io/apimachinery/pkg/api/errors"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
"k8s.io/apimachinery/pkg/runtime"
"k8s.io/apimachinery/pkg/util/intstr"
core "k8s.io/client-go/testing"
"k8s.io/kubernetes/cmd/kubeadm/app/constants"
"k8s.io/kubernetes/pkg/registry/core/service/ipallocator"
)
// InitDryRunGetter implements the DryRunGetter interface and can be used to GET/LIST values in the dryrun fake clientset
// Need to handle these routes in a special manner:
// - GET /default/services/kubernetes -- must return a valid Service
// - GET /clusterrolebindings/system:nodes -- can safely return a NotFound error
// - GET /kube-system/secrets/bootstrap-token-* -- can safely return a NotFound error
// - GET /nodes/<node-name> -- must return a valid Node
// - ...all other, unknown GETs/LISTs will be logged
type InitDryRunGetter struct {
controlPlaneName string
serviceSubnet string
}
// InitDryRunGetter should implement the DryRunGetter interface
var _ DryRunGetter = &InitDryRunGetter{}
// NewInitDryRunGetter creates a new instance of the InitDryRunGetter struct
func NewInitDryRunGetter(controlPlaneName string, serviceSubnet string) *InitDryRunGetter {
return &InitDryRunGetter{
controlPlaneName: controlPlaneName,
serviceSubnet: serviceSubnet,
}
}
// HandleGetAction handles GET actions to the dryrun clientset this interface supports
func (idr *InitDryRunGetter) HandleGetAction(action core.GetAction) (bool, runtime.Object, error) {
funcs := []func(core.GetAction) (bool, runtime.Object, error){
idr.handleKubernetesService,
idr.handleGetNode,
idr.handleSystemNodesClusterRoleBinding,
idr.handleGetBootstrapToken,
}
for _, f := range funcs {
handled, obj, err := f(action)
if handled {
return handled, obj, err
}
}
return false, nil, nil
}
// HandleListAction handles GET actions to the dryrun clientset this interface supports.
// Currently there are no known LIST calls during kubeadm init this code has to take care of.
func (idr *InitDryRunGetter) HandleListAction(action core.ListAction) (bool, runtime.Object, error) {
return false, nil, nil
}
// handleKubernetesService returns a faked Kubernetes service in order to be able to continue running kubeadm init.
// The kube-dns addon code GETs the Kubernetes service in order to extract the service subnet
func (idr *InitDryRunGetter) handleKubernetesService(action core.GetAction) (bool, runtime.Object, error) {
if action.GetName() != "kubernetes" || action.GetNamespace() != metav1.NamespaceDefault || action.GetResource().Resource != "services" {
// We can't handle this event
return false, nil, nil
}
_, svcSubnet, err := net.ParseCIDR(idr.serviceSubnet)
if err != nil {
return true, nil, errors.Wrapf(err, "error parsing CIDR %q", idr.serviceSubnet)
}
internalAPIServerVirtualIP, err := ipallocator.GetIndexedIP(svcSubnet, 1)
if err != nil {
return true, nil, errors.Wrapf(err, "unable to get first IP address from the given CIDR (%s)", svcSubnet.String())
}
// The only used field of this Service object is the ClusterIP, which kube-dns uses to calculate its own IP
return true, &v1.Service{
ObjectMeta: metav1.ObjectMeta{
Name: "kubernetes",
Namespace: metav1.NamespaceDefault,
Labels: map[string]string{
"component": "apiserver",
"provider": "kubernetes",
},
},
Spec: v1.ServiceSpec{
ClusterIP: internalAPIServerVirtualIP.String(),
Ports: []v1.ServicePort{
{
Name: "https",
Port: 443,
TargetPort: intstr.FromInt(6443),
},
},
},
}, nil
}
// handleGetNode returns a fake node object for the purpose of moving kubeadm init forwards.
func (idr *InitDryRunGetter) handleGetNode(action core.GetAction) (bool, runtime.Object, error) {
if action.GetName() != idr.controlPlaneName || action.GetResource().Resource != "nodes" {
// We can't handle this event
return false, nil, nil
}
return true, &v1.Node{
ObjectMeta: metav1.ObjectMeta{
Name: idr.controlPlaneName,
Labels: map[string]string{
"kubernetes.io/hostname": idr.controlPlaneName,
},
Annotations: map[string]string{},
},
}, nil
}
// handleSystemNodesClusterRoleBinding handles the GET call to the system:nodes clusterrolebinding
func (idr *InitDryRunGetter) handleSystemNodesClusterRoleBinding(action core.GetAction) (bool, runtime.Object, error) {
if action.GetName() != constants.NodesClusterRoleBinding || action.GetResource().Resource != "clusterrolebindings" {
// We can't handle this event
return false, nil, nil
}
// We can safely return a NotFound error here as the code will just proceed normally and don't care about modifying this clusterrolebinding
// This can only happen on an upgrade; and in that case the ClientBackedDryRunGetter impl will be used
return true, nil, apierrors.NewNotFound(action.GetResource().GroupResource(), "clusterrolebinding not found")
}
// handleGetBootstrapToken handles the case where kubeadm init creates the default token; and the token code GETs the
// bootstrap token secret first in order to check if it already exists
func (idr *InitDryRunGetter) handleGetBootstrapToken(action core.GetAction) (bool, runtime.Object, error) {
if !strings.HasPrefix(action.GetName(), "bootstrap-token-") || action.GetNamespace() != metav1.NamespaceSystem || action.GetResource().Resource != "secrets" {
// We can't handle this event
return false, nil, nil
}
// We can safely return a NotFound error here as the code will just proceed normally and create the Bootstrap Token
return true, nil, apierrors.NewNotFound(action.GetResource().GroupResource(), "secret not found")
}
|
Q:
HTML Onlcick event only works with # Anchor Tag
When I attempt to call a function my this page using the below code. I just seems to refresh the page and not call the script.
<form role="search" name="locationForm">
<div class="form-group">
<input id="locationInput" type="text" class="form-control" placeholder="Search">
</div>
<button class="btn btn-primary btn-lg" type="submit" onclick="start();">Submit</button>
</form>
If I add a '#' to the end of the url, reload the page, then the onlcick event works as it is suppose to.
As far as I knew these were Anchor tags and I have no idea why they would be required in the calling of a function.
How do I correct this? As I don't want to have to use the #.
A:
You are using a button element, whose default behavior, when clicked, submits its parent form. return false will stop the form from submitting:
<button class="btn btn-primary btn-lg" type="submit" onclick="start(); return false;">Submit</button>
|
VS-APPLE: A Virtual Screening Algorithm Using Promiscuous Protein-Ligand Complexes.
As the number of structurally resolved protein-ligand complexes increases, the ligand-binding pockets of many proteins have been found to accommodate multiple different compounds. Effective use of these structural data is important for developing virtual screening (VS) methods that identify bioactive compounds. Here, we introduce a VS method, VS-APPLE (Virtual Screening Algorithm using Promiscuous Protein-Ligand complExes), based on promiscuous protein-ligand binding structures. In VS-APPLE, multiple ligands bound to a pocket are combined into a query template for screening. Both the structural match between a test compound and the multiple-ligand template and the possible collisions between the test compound and the target protein are evaluated by an efficient geometric hashing method. The performance of VS-APPLE was examined on a filtered, clustered version of the Directory of Useful Decoys data set. In Area Under the Curve analyses of this data set, VS-APPLE outperformed several popular screening programs. Judging from the performance of VS-APPLE, the structural data of promiscuous protein-ligand bindings could be further analyzed and exploited for developing VS methods. |
Review: DEXTER Seasons I-V
If you haven't heard of Showtime's series DEXTER, I'd be surprised. Since 2006, the show has been a mainstay of horror and genre fiction in general. Since it involves blood, murder, more blood, serial killers and then a little blood, it's the ideal show for me to review.
You may ask yourself, "why the hell is he reviewing this show now?" Well, two reasons.
First, Dexter: The Sixth Season hits on DVD on August 14, plenty of time for you to lose hundreds of irreplacable hours of your life watching two-dimensional people act out made-up stories. Not that any of us would ever do that, of course, but you know how people are. If you're made of money, you can buy Dexter: Seasons 1-5 and veg out to the whole thing if you're not a NetFlix kinda person.
The second reason is a promotional company sent me Season Five over a year ago, because I told them I would review it in this blog. Then, I proceeded to not review it for a long, long time. It's been in my to-dos for, like, forever. They only gave Season Five out to a limited number of reviewers, I took one then failed to do what I promised to do, which means they were (possibly) robbed of another review. Some of these promotional companies get paid based on how many reviews they generate, which means I suck.
So now it's too late to help them, but that damn to-do is still in my RememberTheMilk app and will be there for always mocking me and calling me a loser so I must purge this demon from my life! I still failed to keep my promise, but this puts a little positive karma back my way. I'll feel a touch better about it, so back off!
Ahhh, I feel better!
Without a doubt, the series is some of the best televsion ever. Like, in history. And for all genres, not just scifi/horror/fantasy. The series stars Michael C. Hall, also known for his whoop-ass work in SIX FEET UNDER. The DEXTER series is based on the novels of Jeff Lindsay, who hit the f-ing literary lottery when he came up with the concept of a serial killer that kills other serial killers. Seriously, folks, this is the kind of high-concept, franchise-character story that pop-culture authors fantasize about. The idea is so interesting and compelling it damn near doesn't matter if the books are any good (which they are, I've read the first two), because some TV exec is going to hear it and hallucinate dollar signs. It's called "the slugline," that one sentence that instantly tells everyone what your book or series is about. Jeff Lindsay? Best slugline evah.
The character of Dexter loses his parents to violence. He is adopted by a police officer. As Dexter hits his pre-teens and teens and starts to show the classic indicators of a serial killer, the cop recognizes it immediately. Here's where the real brilliance of the series comes in: the cop loves his son so much, he knows he has to protect him, but he also knows that you can't "cure" serial-killer tendencies. Dexter has to act out, so what does the Cop Daddy do? He teaches Dexter how to find other serial killers, allowing his son to both satisfy that "dark passenger" while simultaneously making the world a better place.
I'm not going to review all five seasons here, so I'll try to summarize for you. First, Hall is fantastic in the role. He plays a man trying to fly under society's radar, just another face in the crowd. He slow-plays the role so much that when he finally shows serial-killer rage, I found myself leaning away from the TV. Dude blows me away, particularly in the first three seasons. After that he's still fantastic, but there's only so much "new" he can bring to the role to surprise you.
I also love the supporting cast. I have a huge crush on Jennifer Carpenter, who plays Dexter's adoptive sister Deb. She's a detective in the same police department where Dex is a blood-splatter analyst. Deb is a foul-mouthed tomboy cop who is always looking for recognition and validation. She didn't get enough attention from her father, because her father was spending all of his time trying to mold the monster that was Dexter. David Zayas plays Angel Batista, a smiling, kind-hearted detective that is an excellent counter-balance to the channeled insanity of Dexter. C.S Lee plays Vince Masuka, who could easily slide into the role of Pookie Chang should NOCTURNAL ever be made into a movie.
For me, the first three seasons of DEXTER are the best television ever. That's all-inclusive, hands-down, even better than the first season of the re-imagined BATTLESTAR GALACTICA, which is high up on my Shelf of Awesome. Each season is a twelve-hour-long movie, and the writers know exactly how each season will end (translation: you don't get the bullshit endings of GALACTICA or LOST). These thriller-based plots are tight and build to satisfying conclusions. I'm not only a fan of the acting, I think the writers and show-runners are some of the best in the biz.
Season IV brings a lights-out performance by a serial-killing John Lithgow, and that alone makes it worth your time. Season V brings us Julia Stiles as a femme fatal (in this case pronounced FAY-tal, not fah-TAL). I wasn't a fan of hers and wasn't expecting much from her performance, but she delivered the acting equivalent of a pimp-slap to my face and really rocked it. The bad guys were a little weak, but still an excellent season full of drama and heartache.
As for Season VI, I hope to get to that in a few weeks. Many people didn't like it, and in the final episode six years of careful character development goes completely off the rails, but Commander Adama -- I mean, Edward James Olmos -- is in it, so it can't be all bad.
11 Comments
Say what you will about it going off the reservation at the end, but I think this is going to launch Dexter into some new territory. New territory they have to take or risk becoming dull or jumping the shark. And we have to figure the sharks around Dexter's place have to be well fed little bastards. ;)
I have LOVED Dexter since the first one i came across. Someone told me i NEED to see it, and i was like, yeah, whatever. I love the narration, it makes the show. You get to hear his thoughts and they are amazing. when he is in certain scenes and he is acting all nice, and he starts saying how he is going to cut them up, etc in the narration, its just awesome.
As for Vince Masuka, he IS Pookie Chang. he is what i thought of when i was listening to Nocturnal. Rude, Crude, and just 100% politically incorrect.
Hall is amazing in this show, he has that calm manner, but damn when he goes serial-killer on someone, look out.
I agree with pretty much everything you said, seasons 1-3 are off the chain. John Lithgow is just perfect in season 4. BUTT did we need to see his BUTT? LOL
This is my "guilty pleasure", my 16 year old son wants to watch it so bad its killing him, but i keep telling him he is not ready for it. Maybe this summer.. LOL
Hall does the voice over for Dodge, and every time i hear his voice i keep waiting to see him with the knives..
CANT wait for Season 7 (and hate it that its the end of the show). And agree with the WTF ending of Season 6, hoping they work that out the way they have worked every other crazy thing.
I would love to see Showtime take up Nocturnal and do with with the same team that does Dexter, or even the GFL books. Even thought i think the GFL books would be better as a animated series to make it better for the different species.
Dexter is the only reason I still even have a TV. I thought book three was complete crap, and I lost a little love for the books, but the show is so well done, and is story-separated from the books, that I can dislike where the books veered off without it hurting my love of the show.
I think Season One was the best written show on TV ... ever. I was hesitant about preordering Season Six because a fair amount of Amazon reviews say it sucked, but a sucky Dexter is probably better than 90% of the other shows out there. You, sir, have made a sale. I will preorder Season Six and help ease your guilty conscience.
I will say all those who disliked Season 6- probably have never read the books. The writers are going back to an original story line. I actually though Season 6 was getting back to Season 1 style, I liked it.
Dexter is the only reason I still even have a TV. I thought book three was complete crap, and I lost a little love for the books, but the show is so well done, and is story-separated from the books, that I can dislike where the books veered off without it hurting my love of the show.
It. F'ing. Rocks.
Period.
If that was the "dark passenger" book- I completely agree. Hated that book- but the Dexter Delicious book- made up for all others he wrote.
I think Season One was the best written show on TV ... ever. I was hesitant about preordering Season Six because a fair amount of Amazon reviews say it sucked, but a sucky Dexter is probably better than 90% of the other shows out there. You, sir, have made a sale. I will preorder Season Six and help ease your guilty conscience.
Season VI gets a bad rap, but the big reveal in it blew me away. Many people hated it, I thought it was brilliant and very consistent with the show's history. To each their own. Also? Commander Fucking Adama.
I saw the twist in 6 coming, but thought it was good just the same. I also liked the main plot line as I love Revelations/Apocalyptic stories. The new character development with Deb felt strange and rushed, but I know it's there in the books, too. I just think they could've rolled it out better.
I saw the twist in 6 coming, but thought it was good just the same. I also liked the main plot line as I love Revelations/Apocalyptic stories. The new character development with Deb felt strange and rushed, but I know it's there in the books, too. I just think they could've rolled it out better.
I know as a writer I should have seen the Season 6 plot twist coming, but I didn't. Not a bit. And when it hit, I was all, like, "Daaaayyyummmm!" |
Ab initio theory for femtosecond spin dynamics, angle-resolved fidelity analysis, and the magneto-optical Kerr effect in the Ni3(CH3OH) and Co3(+)(CH3OH) clusters.
We present a systematic analysis of the ab initio controlled femtosecond spin dynamics in Ni3(CH3OH) and Co3(+)(CH3OH) clusters achieved by a spin-orbit-coupling enabled Λ process. The distortion caused by the attachment of CH3OH to one of the active magnetic centers of the Ni3 and the Co3(+) clusters induces asymmetric geometries which result in well localized spin densities on the magnetic centers. With the use of high-level quantum chemistry methods, successful spin-flip scenarios are demonstrated for both clusters. In order to assess the experimental accessibility of those effects, we compute their tolerance with respect to two laser pulse parameters, i.e., the energy detuning as well as the deviation of the polar angle ϕ from its optimized value. Finally, we calculate the magneto-optical Kerr effect in order to connect to the susceptibility tensor χ as an experimentally measurable quantity. |
'Somos todos racistas?': o teste de Harvard que promete revelar preconceito implícito
Crédito, Getty Images Legenda da foto, O Teste de Associação Implícita (IAT) é uma forma de identificar o preconceito implícito das pessoas
Poucas pessoas admitiriam seu racismo abertamente, mas muitos psicólogos dizem que somos racistas mesmo sem a intenção de sê-lo. Nós teríamos o que eles chamam de "preconceito implícito". O que é isso, como é medido e o que pode ser feito a respeito?
"Seus dados sugerem que você tem uma leve preferência automática por pessoas brancas em relação a pessoas negras".
Esse não era o resultado que eu esperava. Então eu sou racista? Um intolerante? Isso seria bem diferente de como eu me vejo. Mas talvez seja algo que eu deva admitir e encarar de frente?
É que eu acabei de fazer o Teste de Associação Implícita (IAT), uma forma de identificar o preconceito implícito das pessoas. A prova não analisa apenas o preconceito de cor, mas também em relação a orientação sexual, deficiências e obesidade.
Segue uma breve descrição de como é um teste IAT, com base no de raça: aparecem palavras e rostos. As palavras podem ser positivas (como "maravilha", "amizade", "feliz" e "comemore") ou negativas (como "dor", "desprezo", "sujo" e "desastre"). Em uma parte do processo, é preciso pressionar uma tecla toda vez em que você vir um rosto de uma pessoa negra ou uma palavra ruim e pressionar outra tecla quando aparecer um rosto de uma pessoa branca ou uma palavra boa.
Em seguida, o jogo vira: uma tecla para rostos negros e palavras boas e outra para rostos brancos e palavras ruins. É muita coisa para guardar na cabeça. E aí está a questão. Você precisa pressionar a tecla apropriada o mais rápido possível. O computador mede a sua velocidade.
Legenda da foto, Teclas associam palavras boas e ruins a rostos com traços negros e brancos
Você pode fazer o teste aqui.
A ideia por trás do IAT é que alguns conceitos e categorias podem estar mais conectados nas nossas mentes do que outros. Podemos achar mais fácil, e portanto mais rápido, ligar rostos de pessoas negras a palavras ruins do que de pessoas brancas.
Legenda da foto, Imagem tirada do teste mostra série de rostos brancos e negros e palavras que devem ser associadas a eles
Nas últimas décadas, as estatísticas sobre preconceito explícito têm mudado rapidamente. Por exemplo, na Inglaterra dos anos 1980, 50% da população se dizia contra casamentos interraciais. Essa taxa caiu para 15% em 2011. Os Estados Unidos experimentaram uma mudança parecida. Em 1958, 94% dos americanos disseram desaprovar casamentos entre pessoas brancas e negras. Esse número caiu para 11% em 2013.
O Brasil, apesar da fama de miscigenação, tem poucos casamentos interraciais, segundo o Censo Demográfico de 2010 do IBGE. Cerca de 75% das pessoas que se identificam como brancas casam com outras brancas, 69% dos pardos vivem com pardos e 50% dos negros se casam com outras pessoas negras. Os homens que se identificam de "cor amarela" são os que mais se unem com mulheres de outra cor - 38% se casam com asiáticas, 29,2% com pardas, 22% com brancas e 9,8% com mulheres negras.
Mas o preconceito implícito - visões que alimentamos sem intenção - é muito mais aderente e difícil de erradicar. Ao menos esse é o preceito do teste.
O IAT, introduzido há duas décadas como uma maneira de medir o preconceito implícito, agora é usado em laboratórios no mundo inteiro. No site do Projeto Implícito de Harvard, o teste foi realizado quase 18 milhões de vezes. E há um padrão. O meu resultado não é único. No teste de raça, a maioria das pessoas demonstrou alguma inclinação a favor de brancos e contra negros. Elas conectam mais rapidamente rostos de pessoas negras a conceitos ruins do que rostos brancos - e mesmo as pessoas negras não são imunes a esse fenômeno.
O preconceito implícito foi usado para explicar, ao menos parcialmente, tudo, desde a eleição do presidente Donald Trump (preconceito implícito contra sua oponente mulher) até o número desproporcional de homens negros que são baleados pela polícia sem estar armados nos Estados Unidos.
A ideia de que a maioria de nós temos várias formas de preconceito implícito se tornou tão comum que foi até mencionada por Hillary Clinton durante um debate na campanha presidencial com Trump. "Eu acho que o preconceito implícito é um problema para todos", disse ela.
Crédito, Getty Images Legenda da foto, Hillary Clinton e Donald Trump durante debate da campanha presidencial em 2016
Trump respondeu à frase uma semana depois. "No nosso debate na semana passada, ela acusou o país inteiro, sugerindo que absolutamente todo mundo, inclusive a polícia, é racista e tem preconceito. É algo ruim de se dizer".
Poucos experimentos na psicologia social tiveram um impacto tão grande. Jules Holroyd, especialista no tema na Universidade de Sheffield, vê uma conexão entre o sucesso do teste e a sua explicação dos "motivos pelos quais várias formas de discriminação e exclusão persistem".
Na última década, usar testes de associação implícita em treinamentos de diversidade virou rotina em grandes empresas. O objetivo é mostar à equipe, especialmente aqueles com o poder de contratar e promover, que mesmo sem saber, e apesar de suas melhores intenções, eles podem ter preconceitos.
Há um interesse óbvio por parte das empresas para eliminar o preconceito. Se você consegue impedir sua equipe de se comportar com preconceito e irracionalmente, você consegue contratar e promover o melhor talento. Menos preconceito implícito = mais lucros.
No entanto, praticamente tudo que sabemos sobre preconceito implícito é incerto, inclusive questões bastante fundamentais. Por exemplo, há um dissenso sobre quão inconscientes são esses estados da mente. Alguns psicólogos acreditam que temos consciência de nossos próprios preconceitos em alguma medida.
E há o próprio IAT. Há dois problemas com ele. O primeiro é o que os cientistas chamam de replicabilidade. Idealmente, um estudo que produz um certo resultado em uma segunda-feira deveria produzir o mesmo na terça. Mas, segundo Greg Mitchell, professor de Direito da Universidade de Virgínia, a replicabilidade do IAT é extremamente baixa.
Se o teste sugere que você tem um preconceito forte contra pessoas negras e "você o repete uma hora depois, você vai ter uma pontuação bem diferente", diz ela. Um dos motivos para isso é que o seu resultado é sensível às circunstâncias em que você se encontra quando o faz. É possível que seu resultado varie dependendo se você o fez antes ou depois de um almoço saudável, por exemplo.
Também parece existir uma relação entre IAT e comportamento. Por exemplo, se o seu colega, Pessoa A, vai pior no IAT que outro colega, Pessoa B, seria imprudente concluir que a Pessoa A demonstrará um comportamento mais discriminatório no local de trabalho. Se há uma linha entre IAT e compotamento em geral, ela é tênue.
Defensores dizem que há uma correlação entre IAT e questões de comportamento. Naturalmente, o comportamento humano é bastante complexo, e somos influenciados por vários fatores, incluindo clima, humor e nível de açúcar no sangue. Mas todo dia milhões de decisões relacionadas a emprego, questões judiciais e qualificações de alunos são tomadas. Mesmo que poucas dessas decisões tenham sido afetadas por preconceito implícito, os números em geral seriam significativos.
Crédito, Getty Images Legenda da foto, Testes de associação implícita em treinamentos de diversidade viraram rotina em grandes empresas
Nos últimos anos, especialistas contra e a favor do IAT competiram entre si com várias pesquisas. Um estudo observou pacientes brancos e negros com sintomas cardiovasculares: a questão era por que médicos frequentemente tratavam esses pacientes de maneira diferente. Pesquisadores descobriram uma relação entre o número de vezes em que médicos tratam pacientes diferentemente em relação a raça e sua pontução no IAT.
Outro estudo mostrou aos participantes imagens de rostos brancos e negros e em seguida imagens de armas ou outros objetos. A tarefa era identificar o mais rapidamente possível se o objeto mostrado era uma arma. E no fim das contas, as pessoas tendiam a identificar um objeto qualquer como uma arma quando haviam visto um rosto negro antes. E isso estava ligado ao teste: paticipantes com IAT baixo tendiam a cometer mais erros.
Mas os céticos em relação à ligação entre comportamento e o IAT citam outros estudos. Um experimento mostrou que policiais com performance ruim no IAT não tinham uma tendência maior a atirar contra suspeitos negros desarmados em relação a brancos, por exemplo.
Além disso, não há consenso quanto a se a solução é ter plena consciência dos seus preconceitos - ou se isso seria ainda pior.
É um quadro confuso. Mas muitos acadêmicos dizem que há outras formas de identificar preconceito implícito.
"O IAT não é o único paradigma em que observamos preconceitos não intencionais, então mesmo que achemos que o IAT é uma medição ruim do preconceito implícito, isso não significa que o fenômeno não seja real", diz Sophie Stammers, professora da Universidade de Birmingham.
Há estudos que mostram, por exemplo, que empregadores responderam mais positivamente a currículos com nomes masculinos do que femininos ou com nomes tipicamente brancos em relação a negros.
Enquanto isso, Sarah Jane Leslie, professora de Filosofia da Universidade de Princeton, fez algumas pesquisas fascinantes sobre por que algumas disciplinas acadêmicas - como matemática, física e sua própria área (filosofia) têm tão poucas mulheres.
A sua explicação é que algumas disciplinas - como a filosofia - enfatizam a necessidade de genialidade bruta para ser bem-sucedido, enquanto outras - como história - tendem a dar mais importância a dedicação e trabalho duro.
"Já que há estereótipos culturais que ligam mais homens que mulheres à genialidade bruta, previmos e de fato descobrimos que essas disciplinas que enfatizam a genialidade tinham muito menos mulheres", diz Leslie. Na cultura popular, é difícil pensar em uma equivalente a Sherlock Holmes, por exemplo, já que as deduções impressionantes do detetitve seriam um produto de sua genialidade singular.
Quão cedo em nossas vidas absorvemos esses estereótipos culturais? Sarah Jane Leslie conduziu um estudo fascinante com crianças.
Os pesquisadores comentam, com meninos e meninas, a respeito de uma pessoa muito inteligente sem dar nenhuma pista do gênero dela. Em seguida, quatro fotos são apresentadas, de dois homens e duas mulheres, e as crianças têm que adivinhar quem é a pessoa inteligente. Aos cinco anos, tanto meninos quanto meninas demonstraram uma tendência maior a escolher um homem do que uma mulher.
Crédito, Getty Images Legenda da foto, Em estudo em escolas, crianças precisavam adivinhar o gênero de uma pessoa descrita como "muito inteligente"
"Com seis anos, as garotas têm uma tendência bem menor em relação aos garotos a pensar que uma pessoa do seu gênero pode ser alguém tão, tão inteligente", diz Leslie.
A afirmação de que a maioria de nós temos vários preconceitos implícitos é parte de uma explosão de pesquisas sobre a irracionalidade nos nossos pensamentos, decisões e crenças. Nós não somos as criaturas lógicas e sistemáticas que imaginamos.
As descobertas do IAT, e de pesquisas em preconceito implícito em geral, são sem sombra de dúvidas importantes. Mas ainda há muitas coisas que não sabemos. Ainda precisamos de muitas pesquisas rigorosas sobre como preconceitos implícitos interagem com comportamentos e como eles podem ser superados. |
cmake_minimum_required(VERSION 3.5)
project(status_ground)
set(CMAKE_C_STANDARD 11)
IF (NOT CMAKE_BUILD_TYPE)
SET(CMAKE_BUILD_TYPE Release ... FORCE)
ENDIF ()
IF (CMAKE_BUILD_TYPE MATCHES Release)
SET(CMAKE_C_FLAGS "-O3") ## Optimize
message(STATUS "${PROJECT_NAME} module: Release configuration")
ELSE ()
message(STATUS "${PROJECT_NAME} module: Debug configuration")
ENDIF ()
add_subdirectory(../common db_common)
set(SOURCE_FILES status_main.c)
add_executable(db_status ${SOURCE_FILES})
target_link_libraries(db_status db_common) |
33.725530 -0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 1.000000
33.829280 -0.009501 -0.013836 0.804392 -0.004697 -0.000869 -0.000404 0.999989
33.933050 -0.021543 -0.029555 1.604878 -0.006722 -0.001959 -0.001536 0.999974
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.