qid int64 1 74.7M | question stringlengths 12 33.8k | date stringlengths 10 10 | metadata list | response_j stringlengths 0 115k | response_k stringlengths 2 98.3k |
|---|---|---|---|---|---|
164,206 | EU anti-graft agency OLAF said Thursday it has opened an investigation into alleged misuse of EU funds in Slovakia uncovered by journalist Ján Kuciak, who was killed earlier this year.
Source: <https://www.politico.eu/article/eu-anti-fraud-agency-probes-alleged-misuse-of-funds-in-slovakia-olaf-jan-kuciak/>
Is it all right that before "Thursday" the preposition "on" is missing? | 2018/04/20 | [
"https://ell.stackexchange.com/questions/164206",
"https://ell.stackexchange.com",
"https://ell.stackexchange.com/users/6420/"
] | There is no need for a preposition there at all, and it's not because it's headlinese. These are both idiomatic:
>
> You said Monday that you'd have this finished by Friday.
>
>
> Monday you said you'd have this finished by Friday.
>
>
>
You could also say "on Monday" there as well. | You can omit the 'on' if your intended readership is US (American). If you are aiming for a British readership, it is required. |
51,575 | 
The warning label on each coconut describes how to treat the water within each. The water on the brown coconut is to be discarded; the water on the white coconut is safe for consumption.
**In what ways is the water from the brown coconut different than that of the white coconut?** | 2014/12/14 | [
"https://cooking.stackexchange.com/questions/51575",
"https://cooking.stackexchange.com",
"https://cooking.stackexchange.com/users/30919/"
] | Phew, too long for a comment.
The liquid of (young?) coconuts is sterile and [can even be used for transfusions](http://www.ajemjournal.com/article/S0735-6757%2800%2990062-7/abstract). The abstract of the linked article doesn't say anything about the sterility of older coconuts. I assume that the older coconuts (esperically the peeled ones) are not sterile anymore. [This not very trustworthy looking website](http://www.kokosnussblog.de/kokoswasser-das-besondere-wasser) (this article is written by someone who sells coconut-related products...) says the liquid of older coconuts tastes sour.
If I search for "brown coconut juice sour" (both in English and in German) I get results (of ...well, not very trustworthy websites, too) that say the juice ferments if stored for too long. 1 Then, [the highest rated answer](https://answers.yahoo.com/question/index?qid=20080815125704AANoyfM) on Yahoo Answers makes sense:
>
> The longer a coconut stays and ages on the palm, the stronger and sharper its flavor gets...pick it early at the right age, but store it for an awfully long time, the flavor fades OR if it's a high-sugar variety, then it turns sour fast.
>
>
>
If I search for "brown coconut juice sour ferment" I get (among many, many results about coconut juice-based kefir) an article of the [Cincinnati Herald](http://thecincinnatiherald.com/news/2013/jul/24/health-benefits-coconut-water/) by Timothy Moore:
>
> The fruit itself has a very short life span and likewise must not come into contact with oxygen for too long. Fermentation will begin and the coconut and water will taste sour and smell. It will attract bacteria and complications will arise after it is digested.
>
>
> The commercialization of coconut water prevents it from fermenting and souring as quickly as the natural fruit.[...]
>
>
> If you are shopping for a coconut, don’t be afraid to ask how long the fruit was transported or how old the coconut is? In some cases, ask if you can open the coconut to be on the safe side. You don’t want to leave the store with a bad coconut.
>
>
>
On the one hand I would to take this article with a pinch of salt since the description of the author at the end of the article sound quite sensationalistic - on the other hand this explaination sounds plausible.
---
1 "Bei zu langer Lagerung wird das Kokoswasser sauer oder gärt." Translation: The coconut juice turns sour or even ferments if stored fo too long. [Source](http://www.rohkostwiki.de/wiki/Kokospalme). | If it is unsanitary to drink the water of mature coconuts, it does stand to reason that the meat would also be contaminated. I have yet to find any information suggesting that harm may befall a consumer of said meat, fermented or not. |
51,575 | 
The warning label on each coconut describes how to treat the water within each. The water on the brown coconut is to be discarded; the water on the white coconut is safe for consumption.
**In what ways is the water from the brown coconut different than that of the white coconut?** | 2014/12/14 | [
"https://cooking.stackexchange.com/questions/51575",
"https://cooking.stackexchange.com",
"https://cooking.stackexchange.com/users/30919/"
] | I have been in the coconut export business for over 6 years. Coconuts are either 'young' 7-9 months or 'mature' 11-12 months old at the time of harvest. If you want sweet water, the coconut is harvest young, when the sugar content and volume of water are at their peak.
As the coconut ages, the water is absorbed as the 'meat' in the coconut grows thicker. The water also looses its sugar content. After husking, young coconut shells are white and quickly turn brown UNLESS treated in chemicals to keep them white for an extended period.. The water in an older, browner, coconuts is still safe to drink, just not as sweet.
If the water is rancid, so is the coconut meat. | If it is unsanitary to drink the water of mature coconuts, it does stand to reason that the meat would also be contaminated. I have yet to find any information suggesting that harm may befall a consumer of said meat, fermented or not. |
46,850,592 | Suppose there is a real time feed of stock prices, how do you calculate the average of a subset of it (say over the past week)?
This was an interview question. I can come up with an algorithm to do it in O(n^2), but the interviewer wanted an algorithm that was O(n). | 2017/10/20 | [
"https://Stackoverflow.com/questions/46850592",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/6516383/"
] | A useful approach is to compute the cumulative sum of your array.
This means that each entry in the cumulative sum array is the sum of all previous prices.
This is useful because you can then generate the sum over any particular subarray of your input using a single subtraction.
Note that when a new input arrives, you only need 1 addition to compute the new cumulative sum (because you simply add the new element to the old cumulative sum). | Another approach is akin to computing skew in Genomics.
If you want to compute the average over the past week, create a variable that contains the sum over a moving window. When an entry is created, add the entry to the above sum variable, and subtract the oldest entry in the moving window from it. Since the size of the window is constant, the average over the past week is just the moving sum over the number of entries in the past week. |
112,558 | The original Death Star was blown up by Luke, a relativley untrained pilot flying with the Rebels. Yes, he had the force on his side, but in the end it was pretty easy.
The second (bigger) Death Star also got blown up by the rebels.
>
> In *The Force Awakens*, an even smaller bunch of rebels, with only a handful of ships, manage to blow up an even bigger Deathstar.
>
>
>
Why can't the bad guys manage to make something that's a just a bit harder to destroy? | 2015/12/29 | [
"https://scifi.stackexchange.com/questions/112558",
"https://scifi.stackexchange.com",
"https://scifi.stackexchange.com/users/58150/"
] | Shorter answer.
---------------
If you ask me, the first Death Star and Starkiller Base were well designed to repel assault and attack; Starkiller Base slightly better than the Death Star since it was a full-sized planet and the access to the power storage in Starkiller Base required a ground assault to penetrate; not some dinky thermal exhaust port.
But in the case of Death Star II its destruction was not a design issue but rather an issue of arrogance. By rushing the offensive capabilities of the station to be online in an effort to trap the Rebels, the risk of defensive collapse happening was far greater since the station was partially built. Here is the breakdown.
* [**Death Star I**](http://starwars.wikia.com/wiki/Death_Star): The first Death Star was only destroyed after the Rebels discovered a weakness via the stolen plans and exploited it to their advantage. Without those plans, the Death Star would still be intact and the Rebels would have been destroyed.
* [**Death Star II**](http://starwars.wikia.com/wiki/Death_Star_II): When Death Star II came around, it was partially constructed and mainly functional on an offensive basis in an effort to lure the Rebels into a trap to destroy them. While the shields that protected Death Star II—which were generated from a base on Endor—were markedly stronger than those that protected the first Death Star, the second those shields went down, anyone could attack it with relative ease; which is what the Rebels were hoping for.
* [**Starkiller Base**](http://starwars.wikia.com/wiki/Starkiller_Base): This was a planet converted into a weapon with a headquarters for the First Order. As a planet it had no internal reactor that could have been blown up, so in many ways this is inherently a more stable design than either of the Death Stars. But its main weakness came when it was charging its weapon to fire it and—even with that window of attack possibility in play—to destroy it one would need to engage in a coordinated ground and air assault.
Longer answer.
--------------
### [Death Star I](http://starwars.wikia.com/wiki/Death_Star)
The first Death Star was destroyed by Rebels after stealing plans to the Death Star itself and analyzing it for weaknesses. Without those plans—the key to the whole story of the first *Star Wars* film—the Rebellion had little hope to destroy it. Their conclusion was two-fold:
1. **Not Designed for Small Craft Assault:** The Death Star was designed to repel large craft assaults, but was inadequate to deal with small craft assault. Small craft could easily pass through the station’s protective shields but were not considered a threat that a few random TIE fighters or turbo laser batteries couldn’t deal with.
2. **Weakness Found from Analyzing Plans:** A weakness was found in the overall Death Star design that allowed a small craft assault to successfully deploy a payload that would cause the hypermatter reactor to blow up. This was a weakness even the Death Star staff were unaware off until the Rebel assault was analyzed.
### [Death Star II](http://starwars.wikia.com/wiki/Death_Star_II)
While the second Death Star was more powerful than the first, it was partially constructed. This left huge gaping holes in the infrastructure that the Rebels wanted to take advantage of. But the Emperor’s desire to rush the construction of the station to be offensively capable—so as to lure the Rebels into a trap and destroy them—left the station defensively weak and contributed to it’s destruction:
1. **Offensively Capable:** While the Emperor gloated to Luke about how the station was fully operational, it was really only operational from an offensive point of view. It’s defensive capabilities were minimal to non-existant and they required the Imperial fleet to provide active defense as well as a shield generator planet-side on Endor to provide a passive defense. The Emperor felt that luring the Rebels into a trap and destroying them would be a simple process and didn’t even see those tiny, tribal Ewoks as being capable of aiding the Rebels in their efforts.
2. **Stronger Shields But Defensively Weaker:** Since Death Star II was co-dependent on defensive shields based on a nearby planet, the second the shields on Endor came down, there was no need to fly down a narrow trench and shoot torpedoes down a small exhaust port: The ships could just fly straight into the incomplete infrastructure of the space station, head to the reactor at its core, blow it up and head out before it was blown to bits.
Also worth noting is the shields used on Endor to protect Death Star II were far stronger than the shields on the first Death Star since ships could not get through; remember how Lando had to call off the attack based on Nien Nunb’s observation. So the Empire clearly learned their lesson in some way from the destruction of Death Star I, but screwed up in another way by having the overall strategic impact of a total shield breach being so much worse than it was with Death Star I.
### [Starkiller Base](http://starwars.wikia.com/wiki/Starkiller_Base)
This was a full planet converted into a weapon by the First Order. The defenses here were much better than the first Death Star or Death Star II since the shields were stronger and the setup was more complex.
1. **Only Effective Attack Was To Strike While It’s Charging:** Since it relied on power from a nearby star/sun the only real effective way of attacking/destroying it would be while it was charging its weapon to fire. Without the stored power, the planetary weapon is just an empty, dead weapon on a planet and would require and immense amount of external power to destroy. Meaning the Resistance would need the equivalent of its own Death Star to destroy Starkiller Base if they did not attack it while it was charging its weapon.
2. **Even With Shields Down It Was Not Easy to Take Down:** That said, even with the shields down, the Resistance fighters deployed in the assault could not make a dent in destroying the thermal oscillator that stored the power the weapon was harvesting from a nearby star/sun.
3. **Supportive Efforts of a Ground Assault to Seal the Deal:** The deciding moment was when the small Resistance ground assault team managed to plant explosives in the area around the thermal oscillator that not only significantly damaged it, but provided an opening large enough for the Resistance assault fighters to deliver the final, decisive blow that would release all of that contained energy engulfing the planet itself in a fireball. | The Death Star I wasn't easy to destroy. The thermal exhaust port is only 2 meters wide, and it would take a starfighter near-ideal conditions to successfully fire a proton torpedo in. Add to that the base defences in the form of turbolasers and TIE squadrons, and only one other pilot besides Luke managed to find an opportunity to make a trench run before Darth Vader arrived - his torpedo didn't make it because he was under immense pressure in a race against time. In pretty cliché fashion, you're underestimating the power of the Force (in overcoming battlefield stress).
If Darth Vader wasn't there, if the Rebellion had enough fighters to throw at the Empire, and if there was no time limit, the Death Star may eventually fall without the Force, but the Battle of Yavin was none of that.
The Death Star II was actually indestructible once complete - it's many times bigger, there is no more vulnerability like the thermal exhaust port to exploit, it had proportionately more surface defences (turbolasers) and TIE squadron complements than the first to combat starfighter threats and the superlaser can be calibrated for energy-efficient destruction of enemy capital ships for proportionately lower cooldown instead of the Death Star I's cooldown of 24 hours. It's indestructible when complete, but Palpatine wanted to use its apparent weakness when under construction as bait. That backfired spectacularly, but you can't use that as argument the Empire couldn't build an invincible weapon - it would have been.
As for Starkiller Base, I can't be certain, but it's possible the First Order got overconfident - if something is the size of a planet instead of a moon, surely it'll take so much more to destroy it, right? Still, it took exceptional skill to destroy it.
>
> All the starfighters the Resistance could throw at the First Order wasn't enough to destroy it on their own - they're only scratching the surface and the TIEs that scrambled to defend the base basically picked off half of the Resistance fighters before the situation changed. It was thanks to Han & Chewie's explosives that opened a hole which permitted only one with the skill of Poe (the best pilot of the Resistance and maybe the galaxy) to fly through and destroy it from within.
>
>
>
One should not judge a weapon's ability to stand up to external attack based on its ease of falling to an internal one, I say. |
3,929 | I don't like to use a back pack instead of the side bag supplied by laptop it self. But when I carry the laptop in a back pack for a long distance, I feel a pain on my shoulders. How do I reduce the pain by doing a hack on placing laptop in the bag? | 2015/01/25 | [
"https://lifehacks.stackexchange.com/questions/3929",
"https://lifehacks.stackexchange.com",
"https://lifehacks.stackexchange.com/users/1452/"
] | ***What might help you is:***
* Adding cushions to your backpack or shoulder bag straps. This can be done by getting softer straps or sewing softer materials into the strap, also making the strap broader helps. The broader the strap the more the weight is displaced, there are usually devices for this already on the strap.
* Get a waist belt. A waist belt for your back pack helps it support the weight instead of just lying directly on your shoulders.
***Usually a shoulder bag has more than one handle, it has a longer shoulder strap and a shorter hand strap, by using both of them you lessen the weight directly on you shoulder. Also, switch sides that hold the bag frequently and avoid narrow straps, try broader padded ones.***
---
You may want to invest in a new back pack that is made to hold a appreciable amount of weight. This way the backpack is stronger and more comrfortable.
[Fitting you backpack.](https://www.whitemountain.com.au/fitting_backpacks/index.html)
[hartchirocenter.com](http://www.hartchirocenter.com/backpack-safety-how-to-wear-your-backpack-for-a-healthy-back)
>
> The backpack should fit so the bulk of the weight is transferred from
> the shoulders to the hips and lower back. Shoulder straps that are too
> loose can cause the backpack to dangle uncomfortably, and cause
> misalignment and pain.
>
>
> Choose a backpack with a lumbar support. The lumbar cushion will
> redistribute weight to the lower extremities, creating a fulcrum that
> facilitates an upright standing position and good posture that is
> essential for proper spinal health.
>
>
> | Waist, chest, and hip belts on your backpack take significant stress off your shoulders no matter how much weight you are carrying. |
649,706 | [](https://i.stack.imgur.com/05PS3.png)
**Things I have:**
1. I have an old type **MIDI input device** (Piano Keyboard Panel) which has **a MIDI "OUT" port** (female type jack) that has **5 holes**. (and no USB port). it has a 9V power input (AC to DC adapter)
2. I have a 32 BIT assembled Desktop computer (Intel Microprocessor, Intel Motherboard and an NVIDIA graphics card) running on Windows 7, that has several standard **USB sockets** (female type).
3. In my desktop, I have a plenty of **ROLL piano apps** and **MIDI Editor apps**, which are asking for a MIDI Input device. Some of them also take input from my TEXT keyboard, but it is stressful for me to use the TEXT keyboard or the mouse as an input device, especially for inserting chords, and hinder fluency.
4. I have a **standard USB** (Male type) **plug**, taken out from a Standard USB Mouse, that has with **4 wires** connected to it.
**What I am trying to do:**
>
> I want to use that MIDI piano keyboard **as a** MIDI Input device for my Windows 7 MIDI apps.
>
>
>
**What I have tried already:**
1. I have downloaded various simple schematics from the internet (not displayed here) that **claims** to **directly connect** the MIDI port with an AUX/ RCA/ USB cables. None of the combinations worked. Some of the combination caused horrible buzzing in the built in speaker of the piano keyboard, and bleeping or flashing or flicker of the tiny LCD display screen of the piano input.
2. The apps keep saying **"No MIDI input device found"**.
3. The apps show a dropdown list that I have 2 MIDI output / synth interfaces (**"Microsoft GS Wavetable Synth"** and **"Microsoft MIDI Mapper"**), but it shows **"BLANK"** in the field for **MIDI input Interface** (even after I installed a few drivers from CASIO, ROLLAND, YAMAHA and a few random drivers from web, and restarted the machine for several times).
4. I have tried searching the In the control panel, the **"Device manager"** and the **"Device and printers"** options. They does not show any option for MIDI devices.
**So please suggest me :**
1. **Proper circuit diagram** to connect MIDI to USB port.
2. **Parts list** (Veroboards, IC chips, capacitors, resistors, diodes etc) with part number to do this.
3. How would I make Windows 7 (32 bit) to recognize this incoming USB signal, and to send it to the MIDI applications in an appropriate format?
Note:
1. I Won't be able to put an online order due to some situations. The MIDI TO USB converters are not available in the local radio parts market and musical instrument market. Also not available in computer applience market.
2. The price is too high. (A reliable MIDI to USB converter such as ROLLAND UM-ONE MK2 costing about INR 3000. WIDI Master MIDI to Bluetooth costs INR 10,263).
3. I won't be able to purchase a separate microcontroller. I want to only adjust the voltage of the MIDI signal and ***I want to make the entire conversion and recognition through a custom made driver pack (.EXE, .DLL etc) within my existing system hardware.*** | 2023/01/12 | [
"https://electronics.stackexchange.com/questions/649706",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/107801/"
] | With those specs, you can't build anything that works..
MIDI is not audio, do not connect it to audio input. MIDI is data.
You need a MIDI to USB adapter, or build one from a microcontroller or MCU board that can present itself as USB device, and write firmware that makes it look like USB MIDI adapter.
But you may have damaged the MIDI output already with incorrect cables so it may not work any more.
Cheap USB MIDI adapters do exist, but they often are unreliable. If you pay much less than half the price of the proper known brand adapter you mentioned, they are likely garbage products that almost work but are not built up to MIDI standards. | Just to back up Justme's answer, you cannot make a USB-to-MIDI adapter with these restrictions - it is not possible.
The reason is that your computer software - even the driver - doesn't have direct access to the wires of the USB port. The USB port is connected to a USB Host Controller, a chip that speaks the USB protocol ("USB language") and translates to the CPU. The Host Controller performs functions like detecting whether something is plugged in, and polling each device 8000 times a second, and then telling the driver if it detects something actually happened. And it's hard-wired to use the USB protocol for this. So whatever you plug into a USB port absolutely must speak the USB protocol, not the MIDI protocol.
---
However, I can think of one possible way to make a cheap adapter. MIDI uses a very common framing protocol known as "UART", which is also used by the PC serial port - which is now obsolete, but still used by many people and it is very easy to get a USB UART adapter for just a few dollars (few hundred rupees) - [example](https://www.ebay.de/itm/352982598553?hash=item522f670399:g:dAwAAOSw9b9jicdn) (but you should find one for yourself).
Wikipedia shows an example schematic. The left half shows the wiring inside the keyboard. The right half shows the wiring inside your adapter. As you can see, it's just a few components - and I think the diode isn't even needed, because it's just for safety (protects the opto-isolator from damage if the cable is somehow plugged in backwards).
[](https://en.wikipedia.org/wiki/MIDI#/media/File:MIDI_IN_OUT_schematic.svg)
So where it says "to UART", just wire that to the receive/RX line on your UART; 5V to 5V; ground to ground; then plug the UART's USB plug into your computer. It will show up as a serial port. If you set the baud rate to 31250 baud, you should be able to receive MIDI data.
How to make your computer treat it as a MIDI port, and not a serial port, I'm not sure. Thanks to [Super User](https://superuser.com/questions/910155/can-i-use-an-uart-serial-port-as-a-midi-device-in-linux) I see there is this Linux called [ttyMIDI](https://launchpad.net/ttymidi) but I haven't tried it. It should be much easier on Linux than Windows, just because Linux is a *lot* more customizable, but maybe someone has created a way to do it on Windows.
**The UART adapter must be able to do 31250 baud.** I think most of these adapters are fine with it, but it's not a usual baud rate for PC serial ports, so maybe some adapters (that are designed to work as PC serial ports) don't allow it. I'm not sure if Linux also has a problem with it - I found [this](http://joost.damad.be/2009/05/how-to-set-serial-port-at-midi-speed-in.html) regarding USB adapters on Linux.
---
It should also be possible to use a PC serial port adapter ([example](https://www.reichelt.de/de/de/usb-a-konverter-1x-rs232-9-pol-pl-2303hxd-0-45-m-manhattan-151801-p282842.html)) instead of a UART adapter, but it will be slightly more difficult because the voltage is different, the voltage is reversed, and because you have to attach one of those serial port plugs, assuming that you want to plug your circuit into the adapter, not cut it apart and solder directly to the wires. I think most of these adapters *do* allow non-standard baud rates, just like the UART adapters. |
136,928 | I studied for two years in a university towards a bachelor's degree and then left because my parents and I had to immigrate to another country before I turn 21 (the age limit for an immigrant's child). Now I have to write my resume to apply for jobs. I have decided to include this college experience in my resume.
How can I describe it in my resume? How should I include a brief description of my reason for leaving college?
My draft for this part of the resume goes like this:
>
> Sept 2008 – May 2010 XX University Completed four semesters toward
> Bachelor of Arts in YY (Non-academic leave because of the age
> limit for immigration to the BBB)
>
>
> | 2019/05/20 | [
"https://workplace.stackexchange.com/questions/136928",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/104883/"
] | >
> How can I describe it in my resume? How should I include a brief description of my reason for leaving college?
>
>
>
Don't include your reasons for leaving college. If you are asked about it during an interview, you can explain the reasons at that time. Anything you add to your resume should be relevant to your career and the position that you are applying to. | >
> How can I describe it in my resume? How should I include a brief description of my reason for leaving college?
>
>
>
**I would not attempt to provide a reason in your resume.**
Resumes are intended to spark conversations, to convey the **what**, **where**, and **when** of your work and educational experience. They are not very good for explaining the **why**.
That's what cover letters, introductions, and interviews are for: providing context if necessary.
If a position does not require a college degree, for example, an employer may not care the reason why you left college. Your employer might be happy that you have any college experience at all, and your description would be unnecessary.
In a different situation, an employer might want a more detailed description of why you left college, or would want to have a conversation about it anyway, so your reason in your resume might be unnecessary in that case as well.
By sticking with basic facts in your resume, you keep it short, easy to read, and relevant. It's tempting to add descriptions, but unless something can be conveyed in one or two words, such prose more often than not simply serves to distract. |
388,494 | I've just soldered a SOIC-16 am26c32 on to a breakout board.
The datasheet shows the orientation of the chip by means of a semi circle at one end (<http://www.ti.com/lit/ds/symlink/am26c32.pdf>). However, on the actual device, there is no such semi circle to be found - only the name of the chip.
Am I to assume that pin 1 is the bottom left, when the chip shows the writing the right way up? | 2018/07/30 | [
"https://electronics.stackexchange.com/questions/388494",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/190132/"
] | Some SOIC type of packages don't have a dot or semi-circle. The first row is designated by a bevel,
[](https://i.stack.imgur.com/S9OcD.png) | Look more closely in the *Mechanical Data* section of the datasheet. It shows that the edge is shaved off along the whole row of pins that includes pin 1:
[](https://i.stack.imgur.com/lUs1Y.gif) |
388,494 | I've just soldered a SOIC-16 am26c32 on to a breakout board.
The datasheet shows the orientation of the chip by means of a semi circle at one end (<http://www.ti.com/lit/ds/symlink/am26c32.pdf>). However, on the actual device, there is no such semi circle to be found - only the name of the chip.
Am I to assume that pin 1 is the bottom left, when the chip shows the writing the right way up? | 2018/07/30 | [
"https://electronics.stackexchange.com/questions/388494",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/190132/"
] | Look more closely in the *Mechanical Data* section of the datasheet. It shows that the edge is shaved off along the whole row of pins that includes pin 1:
[](https://i.stack.imgur.com/lUs1Y.gif) | In addition to the features already mentioned, you may also see a white dot on the chip by pin 1. It's more common on DIL packages, but I've seen it on SOIC too. |
388,494 | I've just soldered a SOIC-16 am26c32 on to a breakout board.
The datasheet shows the orientation of the chip by means of a semi circle at one end (<http://www.ti.com/lit/ds/symlink/am26c32.pdf>). However, on the actual device, there is no such semi circle to be found - only the name of the chip.
Am I to assume that pin 1 is the bottom left, when the chip shows the writing the right way up? | 2018/07/30 | [
"https://electronics.stackexchange.com/questions/388494",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/190132/"
] | Some SOIC type of packages don't have a dot or semi-circle. The first row is designated by a bevel,
[](https://i.stack.imgur.com/S9OcD.png) | [](https://i.stack.imgur.com/0Q9nl.png)
As Olin said, and as you can see from the mechanical drawing, one side should have a bevel.
Also, in practice, TI prints/etches/lasers a pretty clear bar onto the left hand side (in above drawing), something like:
[](https://uk.farnell.com/texas-instruments/sn74hc595d/8-bit-sipo-shift-register-soic/dp/2144926) |
388,494 | I've just soldered a SOIC-16 am26c32 on to a breakout board.
The datasheet shows the orientation of the chip by means of a semi circle at one end (<http://www.ti.com/lit/ds/symlink/am26c32.pdf>). However, on the actual device, there is no such semi circle to be found - only the name of the chip.
Am I to assume that pin 1 is the bottom left, when the chip shows the writing the right way up? | 2018/07/30 | [
"https://electronics.stackexchange.com/questions/388494",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/190132/"
] | Some SOIC type of packages don't have a dot or semi-circle. The first row is designated by a bevel,
[](https://i.stack.imgur.com/S9OcD.png) | In addition to the features already mentioned, you may also see a white dot on the chip by pin 1. It's more common on DIL packages, but I've seen it on SOIC too. |
388,494 | I've just soldered a SOIC-16 am26c32 on to a breakout board.
The datasheet shows the orientation of the chip by means of a semi circle at one end (<http://www.ti.com/lit/ds/symlink/am26c32.pdf>). However, on the actual device, there is no such semi circle to be found - only the name of the chip.
Am I to assume that pin 1 is the bottom left, when the chip shows the writing the right way up? | 2018/07/30 | [
"https://electronics.stackexchange.com/questions/388494",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/190132/"
] | [](https://i.stack.imgur.com/0Q9nl.png)
As Olin said, and as you can see from the mechanical drawing, one side should have a bevel.
Also, in practice, TI prints/etches/lasers a pretty clear bar onto the left hand side (in above drawing), something like:
[](https://uk.farnell.com/texas-instruments/sn74hc595d/8-bit-sipo-shift-register-soic/dp/2144926) | In addition to the features already mentioned, you may also see a white dot on the chip by pin 1. It's more common on DIL packages, but I've seen it on SOIC too. |
14,294 | I live in Sunset zone 17. I purchased a pair of blueberries in February. I transplanted them into five gallon pots, and used a mix of potting soil and coconut coir that drains pretty well. I also amended with sulfur to make the soil more acidic, and have given them an E.B. Stone fruit tree/bush fertilizer once every couple of months.
One, a sunshine blue, has been doing great the whole time. It is still producing nice healthy growth. The other is an earliblue, and it has been getting steadily sicker. Initially, the leaves curled down, but were mostly still a nice green. They started browning (and becoming brittle) from the center of the leaves. Some of the newer growth is also looking chlorotic.
I thought the bush might be getting root bound, so I transplanted both blueberries into a single large oak barrel with roughly the same soil mix. The sunshine blue is still looking happy, but the earliblue symptoms continue to get worse. The picture shows some of the worst leaves on the bush (I've been removed leaves as I notice them).
I think my watering is not the problem. I have not let the soil dry out completely, but it is never soggy either. I would appreciate any help figuring out what is going on.
 | 2014/10/05 | [
"https://gardening.stackexchange.com/questions/14294",
"https://gardening.stackexchange.com",
"https://gardening.stackexchange.com/users/3288/"
] | ### Two suggestions:
* As it seems you are aware blueberries like acid soil, it may be a good idea to check the soil ph and make sure it's correct, the incorrect ph may be causing a nutrient deficiency.
* It could also be a fungus (which may result indirectly from the incorrect ph).
I think the best place to start is by testing the soils ph.
**These may be useful to read:**
* <http://msue.anr.msu.edu/news/soil_test_before_you_plant_blueberries>
* <http://vric.ucdavis.edu/pdf/Soil/ChangingpHinSoil.pdf>
If your Blueberries are potted, the easiest way would be to buy acid potting soil from the nursery. | [](https://i.stack.imgur.com/SaC61.jpg)Looks like sunburn to me. My leaves looked the same and it was from full sun all day long in high heat dry climate we have in NM.i had to move to afternoon shade |
14,294 | I live in Sunset zone 17. I purchased a pair of blueberries in February. I transplanted them into five gallon pots, and used a mix of potting soil and coconut coir that drains pretty well. I also amended with sulfur to make the soil more acidic, and have given them an E.B. Stone fruit tree/bush fertilizer once every couple of months.
One, a sunshine blue, has been doing great the whole time. It is still producing nice healthy growth. The other is an earliblue, and it has been getting steadily sicker. Initially, the leaves curled down, but were mostly still a nice green. They started browning (and becoming brittle) from the center of the leaves. Some of the newer growth is also looking chlorotic.
I thought the bush might be getting root bound, so I transplanted both blueberries into a single large oak barrel with roughly the same soil mix. The sunshine blue is still looking happy, but the earliblue symptoms continue to get worse. The picture shows some of the worst leaves on the bush (I've been removed leaves as I notice them).
I think my watering is not the problem. I have not let the soil dry out completely, but it is never soggy either. I would appreciate any help figuring out what is going on.
 | 2014/10/05 | [
"https://gardening.stackexchange.com/questions/14294",
"https://gardening.stackexchange.com",
"https://gardening.stackexchange.com/users/3288/"
] | A couple things to check for, aside from the pH levels, would be the Nitrogen and the Nitrates. The Nitrogen content should be fairly high, and the Nitrate content low. It's possible that high nitrates could be causing the burn.
Another thing to watch out for, is when you water, don't water the leaves of the plant, as it increases the likeliness of fungus growth. | [](https://i.stack.imgur.com/SaC61.jpg)Looks like sunburn to me. My leaves looked the same and it was from full sun all day long in high heat dry climate we have in NM.i had to move to afternoon shade |
411,765 | There have been numerous discussions over the years on the acceptability of partial answers:
| Year | Q&A | Theme |
| --- | --- | --- |
| 2010 | [Should I answer the question with just a partial solution?](https://meta.stackexchange.com/questions/72392/should-i-answer-the-question-with-just-a-partial-solution) | General acceptability of posting partial *solutions* to the problem |
| 2012 | [Should I downvote partial answer?](https://meta.stackexchange.com/questions/118640/should-i-downvote-partial-answer) | Self-evident: whether partial answers should be considered useful |
| 2012 | [Is it okay to put partial answers?](https://meta.stackexchange.com/questions/144452/is-it-okay-to-put-partial-answers) | Also self-evident: is this acceptable to post partial answers? |
| 2014 | [Is a partial answer OK?](https://meta.stackoverflow.com/questions/267126/is-a-partial-answer-ok) | Also concerned with providing a partial *solution* |
| 2015 | [Partial self-answering etiquette](https://meta.stackexchange.com/questions/250666/partial-self-answering-etiquette) | Ethical considerations of accepting partial self-answers |
| 2017 | [Best practices for partial answers](https://meta.stackexchange.com/questions/301807/best-practices-for-partial-answers) | Advice for writing acceptable partial answers |
| 2019 | [Why was my answer to my own question deleted?](https://meta.stackoverflow.com/q/385841/11407695) | "Partial answers/workarounds are considered answers." |
| 2021 | [What's the etiquette for improving answers that only address part of a question?](https://meta.stackoverflow.com/questions/411553) | Guidance on how to improve partial answers |
None of the those (except for maybe the one from 2017), however, address the following part in the help center (the highlighted part) on [how to answer](https://stackoverflow.com/help/how-to-answer):
>
> Still no answer to the question, and you have the same problem? Help us find a solution by researching the problem, then **contribute the results of your research and anything additional you’ve tried as a partial answer**. That way, even if we can’t figure it out, the next person has more to go on. You can also vote up the question or set a bounty on it so the question gets more attention.
>
>
>
This part seems to be in direct contradiction as to what is considered to be a partial answer as per community consensus (if you examine posts listed above, the general idea is that it is ok to post partial answers as long as they actually *solve* a part of the original problem).
This clause creates a loophole that allows "answers" that would be better off as comments, edits to the question, or potential new questions that could be used as duplicate targets if answered to be posted in a forum-like manner as a "me too" response. This also makes NAA flagging such responses, *by the letter of the law*, incorrect too.
The concern is not theoretical. It is already used (see [relevant SOCVR discussion](https://chat.stackoverflow.com/transcript/message/53053161#53053161)) to justify contesting the removal of such content. Please note that I am not saying that the information provided by such responses is not valuable, just that they are using the wrong medium and thus should not be officially endorsed.
This discussion has been brought up at least once before in 2016 as a clarification request that lead nowhere (given the discussion there, I accept the possibility of duplicate linking to it as the target, but I would like to make this a part of the [faq](/questions/tagged/faq "show questions tagged 'faq'") instead should the discussion be in favor of the proposal without invalidating the old Q&A):
[Why does the Help center seem to indicate "Me too!" is a valid answer?](https://meta.stackoverflow.com/q/324340/11407695)
I propose **removing** the clause from the help center to stop confusing *both* conscientious users that do read the help center as we ask them to (and thus justifiably think this is an acceptable way to answer) and curators that find themselves in the predicament of acting against the literal interpretation of the rules.
Shamelessly piggybacking on the [famous Shog9](https://meta.stackexchange.com/q/225370/786798)'s image on what is and what is not an NAA, here is a proposed revised rendition clarifying the abovementioned ambiguity:
[](https://i.stack.imgur.com/ugTvb.png) | 2021/09/26 | [
"https://meta.stackoverflow.com/questions/411765",
"https://meta.stackoverflow.com",
"https://meta.stackoverflow.com/users/11407695/"
] | No, "me too" answers with additional details are not answers (even partial). An answer needs to actually solve at least part of the OP's problem to be considered an answer. The consensus in the community is clear on that, as far as I'm aware, and based on the discussions you've linked to.
I absolutely agree that the wording in the Help Center is misleading in this regard. I don't imagine many users read the help pages thoroughly, and even fewer who try to follow it strictly, but we shouldn't be penalizing users who do so. So, yes, the help pages could do with some editorial fixes. Something as simple as the following might be sufficient.
>
> Help us find a solution by researching the problem, then contribute the results of your research and anything additional you’ve tried as a partial answer, **but make sure that the answer actually solves at least part of the question**.
>
>
>
I wouldn't say it's worth making an edit if the page needs to be consistent network-wide, but a quick search reveals at least one [site](https://codereview.stackexchange.com/help/how-to-answer) with different wording, so it's certainly possible.
---
Also, if the page is going to be edited, the next bullet on that page **Answer the question** says,
>
> Any answer that gets the asker going in the right direction is helpful, [...]
>
>
>
which is just not true on this site, if taken literally. An answer that just says "Learn how to use a debugger." is extremely helpful advice, and definitely gets the asker going in the right direction, but such an answer would (and should) be summarily deleted, or converted to a comment at best. Similar wording as above might work here as well. | We've [updated the copy on this Help Center article](https://stackoverflow.com/help/how-to-answer). I've also taken the liberty to shuffle it around a bit for clarity, and to put the more important sections first. (I do know this is a bit subjective, but the old order did seem very strange to me.)
It now clearly specifies that any 'me too'-style answer needs to fully answer at least part of the question:
>
> So long as you fully answer at least a part of the original question, then you can contribute the results of your research and anything additional you’ve tried.
>
>
>
At this point in time I'm open to making another revision only if it's especially important, ex. if I've inadvertently made the guidance much more opaque, or there's a painful typo. Or if y'all absolutely can't stand the new section order and want the old one back.
Otherwise, I'm markin' this one [status-completed](/questions/tagged/status-completed "show questions tagged 'status-completed'"). |
206,138 | so i'm going to use IRF530 as a LED driver for my project, but as soon as i open datasheet, i can't find K for calculating current Id. so how to calculate Id to make sure i'm supplying enough current for 10-15 LED @ 1 watt (350mA) and not overrated it? i've heard from other source that you just don't use K. i mean if that was true...how to precisely control current that passed MOSFET?
thanks in advance
[](https://i.stack.imgur.com/aPV0u.png) | 2015/12/15 | [
"https://electronics.stackexchange.com/questions/206138",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/50252/"
] | This is a bad way to control LEDs. For a start you have them all in parallel - have you done no research at all about LEDs? LEDs need individual current limiting/sharing resistors when placed in parallel or one of them hogs all the current and blows up after a short while then, the current that was to be shared among 4 LEDs is now inflicted on three and, the one that has a slightly lower forward volt drop, takes most of that current and blows.
You can see where this is going - basically after a few minutes or seconds you have no more working LEDs.
You have a 12 V supply and four LEDs which I assume to be standard 2V, 20mA types. So, put them all in series and calculate a current limiting resistor to go in series with them.
If you want to control LED current to control brightness then try googling current sources like this one: -
[](https://i.stack.imgur.com/JqWup.png)
Varying R2 can give you some control on brightness. Of course you could consider a more elegant solution involving a proper LED control chip: -
[](https://i.stack.imgur.com/Zxa2J.png) | What's 'k'? I assume from the context you mean transconductance, that is the Ids per Vgs ratio.
This parameter varies from FET to FET, and with temperature, and with voltage, so it's not good to assume it's stable and to try to program the LED current by dead reckoning from Vgs, much as you wouldn't take a BJT's \$\beta\$ to be constant and try to do the same thing by base current.
What you *should* do is to add a source resistor Rs to sample the source current. Sense the voltage across it with an opamp or a transistor to control Vgs, or to really cheapskate it, keep the present circuit diagram you have, and the voltage developed across Rs effectively applies negative feedback onto Vgs, to provide some stabilisation against transconductance variations.
**BEWARE** IRF530, and almost every other FET in the world, are **switching** FETs, not linear FETs. In the circuit you have drawn, you will be using the FET in a linear mode to dissipate power. These FETs are built from multiple cells, that share dissipation OK if they are fully on, or fully off, and you can use them to their rated power. If in the linear region, they become thermally unstable after a short time (seconds, milliseconds, not microseconds which is why they can switch fast), and fail to share the dissipation, causing local thermal runaway and device failure. If you stick to <10% of its rated dissipation, and keep it cool, so well heatsunk, you will probably be OK. BJTs do not have this problem. You can buy linear rated FETs for audio amplifiers, but they are mucho denarios and difficult to find. |
1,424,179 | For any matrix the column rank and row rank are equal. As I understand it rank means the number of linearly independent vectors, where vectors is either the rows or columns of the matrix.
This seems to mean that the number of linearly independent rows in a matrix is equal to the number of linearly independent columns? But my gut tells me this shouldn't be the case, at least intuitively I can't see why it would be the case. | 2015/09/06 | [
"https://math.stackexchange.com/questions/1424179",
"https://math.stackexchange.com",
"https://math.stackexchange.com/users/2075/"
] | It is true.
There are many ways to show this.
One of the most constructive ones is to transform the matrix to its "echelon form", using elementary transformations which do not change the number of linearly independent rows or columns.
What you obtain in the end is a diagonal matrix, with ones followed by zero in the diagonal. Clearly, in such a matrix the number of linearly independent rows is the same with the number of linearly independent columns. | For instance the rank of the matrix is the largest dimension of an invertible square submatrix. This criterion is independenty of whether you work with rows or with columns.
You also can say it is the size of the largest non-zero minor of the associated determinant. |
42,371 | I'm learning how to use `sox` and i encountered this explanation:
>
> We can also use sox to convert a raw file to another format. In this
> case, we have to supply some information about the raw file:
>
>
> sox -r 44100 -s -w foo.raw foo.wav
>
>
> The three flags preceding the input file name tell sox that the input
> file has a sampling rate of 44,100 samples per second, that **the data
> is signed**, and that each sample consists of a two byte word ...
>
>
>
What **signed** means in this context? The paragraph above is [from here](http://billposer.org/Linguistics/Computation/SoxTutorial.html#play). | 2017/11/11 | [
"https://sound.stackexchange.com/questions/42371",
"https://sound.stackexchange.com",
"https://sound.stackexchange.com/users/23357/"
] | Signed and Unsigned refer to how numbers are stored for the samples. A sample corresponds to voltage above or below 0.
Signed data is a format that can store negative numbers, so 0 is no voltage and positive or negative numbers can be used to describe the direction of the voltage.
In unsigned data, all numbers are positive and the middle of the range is used for 0 volts.
This refers to the same kind of signed/unsigned used in computer programing when dealing with integer values. | Signed data in audio just means that a value of 0 corresponds to a voltage of 0V and the number being interpreted in two's complement.
Unsigned data means that the lowest (negative) and highest (positive) voltage are represented by all zeros and all ones, with 0V being in the middle of the range (what would be the most negative number in two's complement). You can switch from one representation to the other by flipping the most significant bit in a word. |
1,475,168 | Hey all, I have a linq app using C# express2008 and sqlserver express 2005 (mdf file connection)
I followed the regular dml generation and vanilla datacontext. However i created a repository class to manage the Linq stuff.
In using the functions, selecting data works fine, updating data works in the app.But when i check the data in the tables, nothing has changed.Needless to say, when i close the app, also no change.
I used SQL profiler to see what was being sent to sqlserver express, nothing showed up.
What could be my issues? | 2009/09/25 | [
"https://Stackoverflow.com/questions/1475168",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/178812/"
] | Are you calling SubmitChanges() on the DataContext? | No primary key, or no column(s) in the L2S model marked as primary key member(s)..? |
1,475,168 | Hey all, I have a linq app using C# express2008 and sqlserver express 2005 (mdf file connection)
I followed the regular dml generation and vanilla datacontext. However i created a repository class to manage the Linq stuff.
In using the functions, selecting data works fine, updating data works in the app.But when i check the data in the tables, nothing has changed.Needless to say, when i close the app, also no change.
I used SQL profiler to see what was being sent to sqlserver express, nothing showed up.
What could be my issues? | 2009/09/25 | [
"https://Stackoverflow.com/questions/1475168",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/178812/"
] | Are you calling SubmitChanges() on the DataContext? | Ok... found 'a' soln.
1. changed (forced) default connection string from looking at a file to an instance of sql server (using database instead of file)
2. Attached file to sql server express..and renamed the db.
Ran app again and everything works. Only thing is..the dml still uses the old connection string..so any mods to tables have to be done in sqlserver.
This is just my quicky patch, anyone care to provide a more elaborate view? |
1,475,168 | Hey all, I have a linq app using C# express2008 and sqlserver express 2005 (mdf file connection)
I followed the regular dml generation and vanilla datacontext. However i created a repository class to manage the Linq stuff.
In using the functions, selecting data works fine, updating data works in the app.But when i check the data in the tables, nothing has changed.Needless to say, when i close the app, also no change.
I used SQL profiler to see what was being sent to sqlserver express, nothing showed up.
What could be my issues? | 2009/09/25 | [
"https://Stackoverflow.com/questions/1475168",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/178812/"
] | No primary key, or no column(s) in the L2S model marked as primary key member(s)..? | Ok... found 'a' soln.
1. changed (forced) default connection string from looking at a file to an instance of sql server (using database instead of file)
2. Attached file to sql server express..and renamed the db.
Ran app again and everything works. Only thing is..the dml still uses the old connection string..so any mods to tables have to be done in sqlserver.
This is just my quicky patch, anyone care to provide a more elaborate view? |
307,392 | I'm a co-op student working on a hardware design for an industry client, and I'm trying to use a [TPS63000](http://www.ti.com/lit/ds/symlink/tps63000-q1.pdf) to get a Vout of 3.45V. Vin is 5V. When I power this circuit up, I read 5V on Vin but only 0.7V from Vout. [](https://i.stack.imgur.com/vQ9VR.png)
Here's the PCB design:
[](https://i.stack.imgur.com/gjGLL.png)
I'm not sure what approach to take in troubleshooting this. Any help would be appreciated.
EDIT: Unfortunately I don't have easy access to an oscilloscope, but when I do, I will definitely try to confirm that there's a pulse across the inductor.
Yesterday I assembled another PCB, and I'm getting a very similar result. Vout actually stabilizes at 322mV, and does so with a high degree to accuracy. During startup, I do see Vout shift drastically, but always settles around 322mV. Sorry that that's not very much to go on, an oscilloscope will definitely help give a better picture.
So it seems to me possible causes are: A. lack of R3/C3 on the Vin pin, B. narrow traces on the PCB, or C. bad chip. Thanks for all the responses so far. | 2017/05/25 | [
"https://electronics.stackexchange.com/questions/307392",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/150224/"
] | You have no delay on the VINA, EN and PS/SYNC pins. It is unclear what mode the thing will go into when connected like that.
The [device specs](http://www.ti.com/lit/ds/symlink/tps63000.pdf) shows a suitable delay circuit.
[](https://i.stack.imgur.com/u3P2M.png) | A few guesses/pointers: -
* Try fitting link W1
* Check R1 isn't 1.8 kohm (or a wrong value)
* Is exposed thermal pad properly earthed?
* Have you got too much output load?
* Check that oscillations are running about 1350 kHz
* Is your meter working correctly?
Good luck. |
307,392 | I'm a co-op student working on a hardware design for an industry client, and I'm trying to use a [TPS63000](http://www.ti.com/lit/ds/symlink/tps63000-q1.pdf) to get a Vout of 3.45V. Vin is 5V. When I power this circuit up, I read 5V on Vin but only 0.7V from Vout. [](https://i.stack.imgur.com/vQ9VR.png)
Here's the PCB design:
[](https://i.stack.imgur.com/gjGLL.png)
I'm not sure what approach to take in troubleshooting this. Any help would be appreciated.
EDIT: Unfortunately I don't have easy access to an oscilloscope, but when I do, I will definitely try to confirm that there's a pulse across the inductor.
Yesterday I assembled another PCB, and I'm getting a very similar result. Vout actually stabilizes at 322mV, and does so with a high degree to accuracy. During startup, I do see Vout shift drastically, but always settles around 322mV. Sorry that that's not very much to go on, an oscilloscope will definitely help give a better picture.
So it seems to me possible causes are: A. lack of R3/C3 on the Vin pin, B. narrow traces on the PCB, or C. bad chip. Thanks for all the responses so far. | 2017/05/25 | [
"https://electronics.stackexchange.com/questions/307392",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/150224/"
] | A few guesses/pointers: -
* Try fitting link W1
* Check R1 isn't 1.8 kohm (or a wrong value)
* Is exposed thermal pad properly earthed?
* Have you got too much output load?
* Check that oscillations are running about 1350 kHz
* Is your meter working correctly?
Good luck. | I know this question is old, but Google brought me here on this exact topic and I see no concrete answer, I solved the issue myself, the error is on the feedback, the TPS63000 is a 3.3V (fixed regulator) and no set point is required, only direct feedback, however the circuit and pcb layout shown can accommodate both TPS63000 (fixed) and TPS63001 (adjust)
Solution for your circuit using a TPS63000 option, requires a zero ohm resistor for R1, so feedback is taken directly from the output, and omit R2. |
307,392 | I'm a co-op student working on a hardware design for an industry client, and I'm trying to use a [TPS63000](http://www.ti.com/lit/ds/symlink/tps63000-q1.pdf) to get a Vout of 3.45V. Vin is 5V. When I power this circuit up, I read 5V on Vin but only 0.7V from Vout. [](https://i.stack.imgur.com/vQ9VR.png)
Here's the PCB design:
[](https://i.stack.imgur.com/gjGLL.png)
I'm not sure what approach to take in troubleshooting this. Any help would be appreciated.
EDIT: Unfortunately I don't have easy access to an oscilloscope, but when I do, I will definitely try to confirm that there's a pulse across the inductor.
Yesterday I assembled another PCB, and I'm getting a very similar result. Vout actually stabilizes at 322mV, and does so with a high degree to accuracy. During startup, I do see Vout shift drastically, but always settles around 322mV. Sorry that that's not very much to go on, an oscilloscope will definitely help give a better picture.
So it seems to me possible causes are: A. lack of R3/C3 on the Vin pin, B. narrow traces on the PCB, or C. bad chip. Thanks for all the responses so far. | 2017/05/25 | [
"https://electronics.stackexchange.com/questions/307392",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/150224/"
] | You have no delay on the VINA, EN and PS/SYNC pins. It is unclear what mode the thing will go into when connected like that.
The [device specs](http://www.ti.com/lit/ds/symlink/tps63000.pdf) shows a suitable delay circuit.
[](https://i.stack.imgur.com/u3P2M.png) | My guess is that you don't have a proper power supply bypassing. You have omitted the separate bypass cap C3 from the typical application schematics, where R3 & C3 form a low-pass filter to separate the power supply of the power part from the controller part of the IC. You might be able to succeed without R3 (but you shouldn't), but you definitely need C3, and the controller side power supply should be separated enough (with at least a thin wire) so that the highly fluctuating current between C1 & the power side of the IC (pin 5) does not affect the controller side power supply of the IC (pin 8).
You should also look at page 16 in the data sheet and read the layout recommendations (note that the layout recommendation picture shows a differing schematics, with a different C3). |
307,392 | I'm a co-op student working on a hardware design for an industry client, and I'm trying to use a [TPS63000](http://www.ti.com/lit/ds/symlink/tps63000-q1.pdf) to get a Vout of 3.45V. Vin is 5V. When I power this circuit up, I read 5V on Vin but only 0.7V from Vout. [](https://i.stack.imgur.com/vQ9VR.png)
Here's the PCB design:
[](https://i.stack.imgur.com/gjGLL.png)
I'm not sure what approach to take in troubleshooting this. Any help would be appreciated.
EDIT: Unfortunately I don't have easy access to an oscilloscope, but when I do, I will definitely try to confirm that there's a pulse across the inductor.
Yesterday I assembled another PCB, and I'm getting a very similar result. Vout actually stabilizes at 322mV, and does so with a high degree to accuracy. During startup, I do see Vout shift drastically, but always settles around 322mV. Sorry that that's not very much to go on, an oscilloscope will definitely help give a better picture.
So it seems to me possible causes are: A. lack of R3/C3 on the Vin pin, B. narrow traces on the PCB, or C. bad chip. Thanks for all the responses so far. | 2017/05/25 | [
"https://electronics.stackexchange.com/questions/307392",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/150224/"
] | You have no delay on the VINA, EN and PS/SYNC pins. It is unclear what mode the thing will go into when connected like that.
The [device specs](http://www.ti.com/lit/ds/symlink/tps63000.pdf) shows a suitable delay circuit.
[](https://i.stack.imgur.com/u3P2M.png) | I know this question is old, but Google brought me here on this exact topic and I see no concrete answer, I solved the issue myself, the error is on the feedback, the TPS63000 is a 3.3V (fixed regulator) and no set point is required, only direct feedback, however the circuit and pcb layout shown can accommodate both TPS63000 (fixed) and TPS63001 (adjust)
Solution for your circuit using a TPS63000 option, requires a zero ohm resistor for R1, so feedback is taken directly from the output, and omit R2. |
307,392 | I'm a co-op student working on a hardware design for an industry client, and I'm trying to use a [TPS63000](http://www.ti.com/lit/ds/symlink/tps63000-q1.pdf) to get a Vout of 3.45V. Vin is 5V. When I power this circuit up, I read 5V on Vin but only 0.7V from Vout. [](https://i.stack.imgur.com/vQ9VR.png)
Here's the PCB design:
[](https://i.stack.imgur.com/gjGLL.png)
I'm not sure what approach to take in troubleshooting this. Any help would be appreciated.
EDIT: Unfortunately I don't have easy access to an oscilloscope, but when I do, I will definitely try to confirm that there's a pulse across the inductor.
Yesterday I assembled another PCB, and I'm getting a very similar result. Vout actually stabilizes at 322mV, and does so with a high degree to accuracy. During startup, I do see Vout shift drastically, but always settles around 322mV. Sorry that that's not very much to go on, an oscilloscope will definitely help give a better picture.
So it seems to me possible causes are: A. lack of R3/C3 on the Vin pin, B. narrow traces on the PCB, or C. bad chip. Thanks for all the responses so far. | 2017/05/25 | [
"https://electronics.stackexchange.com/questions/307392",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/150224/"
] | My guess is that you don't have a proper power supply bypassing. You have omitted the separate bypass cap C3 from the typical application schematics, where R3 & C3 form a low-pass filter to separate the power supply of the power part from the controller part of the IC. You might be able to succeed without R3 (but you shouldn't), but you definitely need C3, and the controller side power supply should be separated enough (with at least a thin wire) so that the highly fluctuating current between C1 & the power side of the IC (pin 5) does not affect the controller side power supply of the IC (pin 8).
You should also look at page 16 in the data sheet and read the layout recommendations (note that the layout recommendation picture shows a differing schematics, with a different C3). | I know this question is old, but Google brought me here on this exact topic and I see no concrete answer, I solved the issue myself, the error is on the feedback, the TPS63000 is a 3.3V (fixed regulator) and no set point is required, only direct feedback, however the circuit and pcb layout shown can accommodate both TPS63000 (fixed) and TPS63001 (adjust)
Solution for your circuit using a TPS63000 option, requires a zero ohm resistor for R1, so feedback is taken directly from the output, and omit R2. |
4,521,457 | I have a .NET solution with 4 library projects resulting in 4 assemblies.
There are classes in these assemblies that generate some partial information in a linear workflow.
By information=generated I mean some business logic information inclusive of timestamp, # of records etc.
The final table into which this information has to be inserted contains 16 columns...each of these assemblies have loggable information for just 4 columns.
My objective is to aggregate all this information and log it to the database in a single DB call...instead of doing it 4 times.
To be more clear, a class in namespace A which (namespace/assembly A imports/uses B and C)... instantiatesa class inside B and C respectively .
I do not have access to information generated in methods inside B and C classes in the class in namespaceA. But I want to aggregate all of the information and push it to the database in a single insert.
How should I approach this problem?
I want to persist the data across all these different assemblies and log the data to the databse in a single call.How can I persist the data? | 2010/12/23 | [
"https://Stackoverflow.com/questions/4521457",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/336489/"
] | If you want to insert data produced by classes A, B, and C all at once, then *something* needs to be able to access all of that data at the insertion point.
If, as you say, B and C aren't available to A, have you considered a separate class (in its own assembly, for example) that is concerned only with persistence?
Such a class could reference A, B, and C, pull the data from each (assuming that they expose their data, somehow... if not, I've no idea how you'd accomplish what you seek), aggregate that data and insert it.
This way, A, B, and C remain ignorant of each other, and of the class that persists their data. | As far as I can interpret the question, I suspect you might benefit from an aggregator class (a [Collecting Parameter](http://www.industriallogic.com/xp/refactoring/accumulationToCollection.html) or even a more complicated [Visitor](http://www.industriallogic.com/xp/refactoring/accumulationToVisitor.html)) that visits each information generating class, compiles data, and can then eventually be asked to publish the compiled data somewhere else. |
4,521,457 | I have a .NET solution with 4 library projects resulting in 4 assemblies.
There are classes in these assemblies that generate some partial information in a linear workflow.
By information=generated I mean some business logic information inclusive of timestamp, # of records etc.
The final table into which this information has to be inserted contains 16 columns...each of these assemblies have loggable information for just 4 columns.
My objective is to aggregate all this information and log it to the database in a single DB call...instead of doing it 4 times.
To be more clear, a class in namespace A which (namespace/assembly A imports/uses B and C)... instantiatesa class inside B and C respectively .
I do not have access to information generated in methods inside B and C classes in the class in namespaceA. But I want to aggregate all of the information and push it to the database in a single insert.
How should I approach this problem?
I want to persist the data across all these different assemblies and log the data to the databse in a single call.How can I persist the data? | 2010/12/23 | [
"https://Stackoverflow.com/questions/4521457",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/336489/"
] | If I understand what you're asking, then it doesn't make sense.
If you have individual classes generating data that should be logged, then these individual classes should log the data. No other class should have to know which data the individual classes find interesting.
Let your logging framework deal with bundling the logging information into an appropriate number of database calls. That's an implementation detail of the logging framework. | As far as I can interpret the question, I suspect you might benefit from an aggregator class (a [Collecting Parameter](http://www.industriallogic.com/xp/refactoring/accumulationToCollection.html) or even a more complicated [Visitor](http://www.industriallogic.com/xp/refactoring/accumulationToVisitor.html)) that visits each information generating class, compiles data, and can then eventually be asked to publish the compiled data somewhere else. |
1,242 | I have some information to enter in a text box. The data is structured and only some sequences are valid. For a phone number, for instance, a letter such as `A` is invalid.
So should I intercept keyboard events and prevent typed letters from reaching the text box, or should I draw some red squiggly lines (or another visual cue) to let my user know the letter is not acceptable? Which is a better user experience? | 2010/09/02 | [
"https://ux.stackexchange.com/questions/1242",
"https://ux.stackexchange.com",
"https://ux.stackexchange.com/users/202/"
] | I would suggest a hybrid of both approaches, but only for strictly-validated fields like phone numbers (SSN, zipcodes, etc ... anything that **must** be numeric by definition).
Track each keystroke and, when the user types an invalid character, show it ... but immediately signal that something's wrong. Maybe turn the input field red and display a "Please enter numeric characters only" to the right of the field.
Blocking input is a no-no because it can be misinterpreted by the user. They might assume a problem with their machine or input device, not that they weren't supposed to type a letter in a numeric entry box.
Validating *after* the input is completed can be annoying. Particularly with fields like those used in password creation. Some sites require 6-character passwords with only alphanumeric characters. Others require 9-character passwords and allow any character (including !@#$% and such). Typing a password and only finding out *after* I've typed it is the most annoying form type I've found and has actually turned me away from several services.
Providing dynamic feedback as the user enters data helps them
1. Know which field had the error while they're still in the field
2. Proactively correct input errors before submitting data
3. Learn from the experience as they go
Blocking input fails on all 3 accounts. Highlighting validation errors after the textbox is filled in fails on all 3 as well. | As with most things..depends on context. I have a more specialized take on the topic. Most of the applications I, or my group, work on are technical in nature and are only used by the reasonably trained. As such we can place more responsibility and freedom in the hands of the users than you could with the general public. Briefly, some of the guidelines we use...
* Disregard any invalid entry....block invalid keystrokes.
* Indicate invalid fields, a state that
should be all but impossible, by
contrast and color\*.
* Focus is always under user control.
Do not trap the user or prevent
a change of focus.
* Never throw a modal warning/error
box. Ever.
* Never ask the user questions. Software is a tool not a person. Yes I am sure.
* The Enter/Return keys always mean:
Accept changes and move on. Do not
change this by using focus on a
cancel/exit button as the default on
Enter.
* The Escape key always means: Discard
and exit.
* Provide auto-completion whenever
possible.
* Provide short-cut entry for numerics
that have a min/max limit. e.g. If a
valid value is between 0.12 and 100.387 then
entries greater that 100.387, like 999, are taken as 100.387, etc.
* Establish very clear black and white
rules on behavior..rules the user can
understand. If this becomes overly complex...re-think it.
Almost all of the examples are user requested and, yes...implementation is sometimes a real bear/nightmare/killer but rarely impossible.
Caveat: *Again, these examples are given in the context of a vertical system with a technical user base....not a web general public form, etc. Your mileage my vary.*
\*Except red. In our business red means: Danger you could be killed...or worse. |
1,242 | I have some information to enter in a text box. The data is structured and only some sequences are valid. For a phone number, for instance, a letter such as `A` is invalid.
So should I intercept keyboard events and prevent typed letters from reaching the text box, or should I draw some red squiggly lines (or another visual cue) to let my user know the letter is not acceptable? Which is a better user experience? | 2010/09/02 | [
"https://ux.stackexchange.com/questions/1242",
"https://ux.stackexchange.com",
"https://ux.stackexchange.com/users/202/"
] | Overriding a user's control of his or her computer will always be an interruption, and often be perceived as an invasion, of that user's workflow. The web site "isn't supposed to" have control of our computers, so exerting such control will be at least a little alarming.
That said, you broach a really important topic, being the importance of letting a user know their entry is invalid before they fill out an entire set of fields and submit them.
The approach I've found to be the least-intrusive but provide the quickest turn-around response is the flow below.
Essentially, wait for them to fill out their first field, then as they click/tab around a new field (or a field that has been invalidated), check the field that was just completed. If the field that was just completed is valid, print an unobtrusive "good" icon next to it (green, checkmark, happy face, you get the idea). If it fails the validation rules, discretely print a "bad" icon (red, "X", etc.) and a short line about what failed.
But remember, do not interrupt their workflow:
* No pop-ups
* no overlays
When they're finished with the field they moved to, they'll either go back to the invalid field or move on and go back later. Either way, they'll see that they have an edit to make but can do it on their own time.
In the end, they'll have a completely valid form before ever submitting, which means
1. They'll only have to submit once
2. They don't spend time getting mad or frustrated when their submission "fails"
3. They'll like your site more
4. Everyone will have cake | It depends on the case. If possible use Auto-suggest. If you intercepts user's input, an immediate feedback is crucial. One of the possible solutions is "clue box" that pops up immediately after key is intercepted. You can see it on Windows login screen, it pops up if you press $ sign for instance. |
1,242 | I have some information to enter in a text box. The data is structured and only some sequences are valid. For a phone number, for instance, a letter such as `A` is invalid.
So should I intercept keyboard events and prevent typed letters from reaching the text box, or should I draw some red squiggly lines (or another visual cue) to let my user know the letter is not acceptable? Which is a better user experience? | 2010/09/02 | [
"https://ux.stackexchange.com/questions/1242",
"https://ux.stackexchange.com",
"https://ux.stackexchange.com/users/202/"
] | Overriding a user's control of his or her computer will always be an interruption, and often be perceived as an invasion, of that user's workflow. The web site "isn't supposed to" have control of our computers, so exerting such control will be at least a little alarming.
That said, you broach a really important topic, being the importance of letting a user know their entry is invalid before they fill out an entire set of fields and submit them.
The approach I've found to be the least-intrusive but provide the quickest turn-around response is the flow below.
Essentially, wait for them to fill out their first field, then as they click/tab around a new field (or a field that has been invalidated), check the field that was just completed. If the field that was just completed is valid, print an unobtrusive "good" icon next to it (green, checkmark, happy face, you get the idea). If it fails the validation rules, discretely print a "bad" icon (red, "X", etc.) and a short line about what failed.
But remember, do not interrupt their workflow:
* No pop-ups
* no overlays
When they're finished with the field they moved to, they'll either go back to the invalid field or move on and go back later. Either way, they'll see that they have an edit to make but can do it on their own time.
In the end, they'll have a completely valid form before ever submitting, which means
1. They'll only have to submit once
2. They don't spend time getting mad or frustrated when their submission "fails"
3. They'll like your site more
4. Everyone will have cake | I would suggest a hybrid of both approaches, but only for strictly-validated fields like phone numbers (SSN, zipcodes, etc ... anything that **must** be numeric by definition).
Track each keystroke and, when the user types an invalid character, show it ... but immediately signal that something's wrong. Maybe turn the input field red and display a "Please enter numeric characters only" to the right of the field.
Blocking input is a no-no because it can be misinterpreted by the user. They might assume a problem with their machine or input device, not that they weren't supposed to type a letter in a numeric entry box.
Validating *after* the input is completed can be annoying. Particularly with fields like those used in password creation. Some sites require 6-character passwords with only alphanumeric characters. Others require 9-character passwords and allow any character (including !@#$% and such). Typing a password and only finding out *after* I've typed it is the most annoying form type I've found and has actually turned me away from several services.
Providing dynamic feedback as the user enters data helps them
1. Know which field had the error while they're still in the field
2. Proactively correct input errors before submitting data
3. Learn from the experience as they go
Blocking input fails on all 3 accounts. Highlighting validation errors after the textbox is filled in fails on all 3 as well. |
1,242 | I have some information to enter in a text box. The data is structured and only some sequences are valid. For a phone number, for instance, a letter such as `A` is invalid.
So should I intercept keyboard events and prevent typed letters from reaching the text box, or should I draw some red squiggly lines (or another visual cue) to let my user know the letter is not acceptable? Which is a better user experience? | 2010/09/02 | [
"https://ux.stackexchange.com/questions/1242",
"https://ux.stackexchange.com",
"https://ux.stackexchange.com/users/202/"
] | Overriding a user's control of his or her computer will always be an interruption, and often be perceived as an invasion, of that user's workflow. The web site "isn't supposed to" have control of our computers, so exerting such control will be at least a little alarming.
That said, you broach a really important topic, being the importance of letting a user know their entry is invalid before they fill out an entire set of fields and submit them.
The approach I've found to be the least-intrusive but provide the quickest turn-around response is the flow below.
Essentially, wait for them to fill out their first field, then as they click/tab around a new field (or a field that has been invalidated), check the field that was just completed. If the field that was just completed is valid, print an unobtrusive "good" icon next to it (green, checkmark, happy face, you get the idea). If it fails the validation rules, discretely print a "bad" icon (red, "X", etc.) and a short line about what failed.
But remember, do not interrupt their workflow:
* No pop-ups
* no overlays
When they're finished with the field they moved to, they'll either go back to the invalid field or move on and go back later. Either way, they'll see that they have an edit to make but can do it on their own time.
In the end, they'll have a completely valid form before ever submitting, which means
1. They'll only have to submit once
2. They don't spend time getting mad or frustrated when their submission "fails"
3. They'll like your site more
4. Everyone will have cake | Being on both ends of using a form, here are some guidelines I prefer:
* User inputs should hardly ever be prevented. Meaning if I type the letter 'A' in a phone number field, I would expect to see that 'A'.
* I begin my text field validation on a `onBlur()` event, and no sooner than that. This gives the user a chance to fix that mis-typed 'A' before I show an error message. Allowing the user to correct an error while still in the input field provides one less chance for an obtrusive "hey, you messed up, dude" kind of message. |
1,242 | I have some information to enter in a text box. The data is structured and only some sequences are valid. For a phone number, for instance, a letter such as `A` is invalid.
So should I intercept keyboard events and prevent typed letters from reaching the text box, or should I draw some red squiggly lines (or another visual cue) to let my user know the letter is not acceptable? Which is a better user experience? | 2010/09/02 | [
"https://ux.stackexchange.com/questions/1242",
"https://ux.stackexchange.com",
"https://ux.stackexchange.com/users/202/"
] | Overriding a user's control of his or her computer will always be an interruption, and often be perceived as an invasion, of that user's workflow. The web site "isn't supposed to" have control of our computers, so exerting such control will be at least a little alarming.
That said, you broach a really important topic, being the importance of letting a user know their entry is invalid before they fill out an entire set of fields and submit them.
The approach I've found to be the least-intrusive but provide the quickest turn-around response is the flow below.
Essentially, wait for them to fill out their first field, then as they click/tab around a new field (or a field that has been invalidated), check the field that was just completed. If the field that was just completed is valid, print an unobtrusive "good" icon next to it (green, checkmark, happy face, you get the idea). If it fails the validation rules, discretely print a "bad" icon (red, "X", etc.) and a short line about what failed.
But remember, do not interrupt their workflow:
* No pop-ups
* no overlays
When they're finished with the field they moved to, they'll either go back to the invalid field or move on and go back later. Either way, they'll see that they have an edit to make but can do it on their own time.
In the end, they'll have a completely valid form before ever submitting, which means
1. They'll only have to submit once
2. They don't spend time getting mad or frustrated when their submission "fails"
3. They'll like your site more
4. Everyone will have cake | As with most things..depends on context. I have a more specialized take on the topic. Most of the applications I, or my group, work on are technical in nature and are only used by the reasonably trained. As such we can place more responsibility and freedom in the hands of the users than you could with the general public. Briefly, some of the guidelines we use...
* Disregard any invalid entry....block invalid keystrokes.
* Indicate invalid fields, a state that
should be all but impossible, by
contrast and color\*.
* Focus is always under user control.
Do not trap the user or prevent
a change of focus.
* Never throw a modal warning/error
box. Ever.
* Never ask the user questions. Software is a tool not a person. Yes I am sure.
* The Enter/Return keys always mean:
Accept changes and move on. Do not
change this by using focus on a
cancel/exit button as the default on
Enter.
* The Escape key always means: Discard
and exit.
* Provide auto-completion whenever
possible.
* Provide short-cut entry for numerics
that have a min/max limit. e.g. If a
valid value is between 0.12 and 100.387 then
entries greater that 100.387, like 999, are taken as 100.387, etc.
* Establish very clear black and white
rules on behavior..rules the user can
understand. If this becomes overly complex...re-think it.
Almost all of the examples are user requested and, yes...implementation is sometimes a real bear/nightmare/killer but rarely impossible.
Caveat: *Again, these examples are given in the context of a vertical system with a technical user base....not a web general public form, etc. Your mileage my vary.*
\*Except red. In our business red means: Danger you could be killed...or worse. |
1,242 | I have some information to enter in a text box. The data is structured and only some sequences are valid. For a phone number, for instance, a letter such as `A` is invalid.
So should I intercept keyboard events and prevent typed letters from reaching the text box, or should I draw some red squiggly lines (or another visual cue) to let my user know the letter is not acceptable? Which is a better user experience? | 2010/09/02 | [
"https://ux.stackexchange.com/questions/1242",
"https://ux.stackexchange.com",
"https://ux.stackexchange.com/users/202/"
] | Being on both ends of using a form, here are some guidelines I prefer:
* User inputs should hardly ever be prevented. Meaning if I type the letter 'A' in a phone number field, I would expect to see that 'A'.
* I begin my text field validation on a `onBlur()` event, and no sooner than that. This gives the user a chance to fix that mis-typed 'A' before I show an error message. Allowing the user to correct an error while still in the input field provides one less chance for an obtrusive "hey, you messed up, dude" kind of message. | I would suggest a hybrid of both approaches, but only for strictly-validated fields like phone numbers (SSN, zipcodes, etc ... anything that **must** be numeric by definition).
Track each keystroke and, when the user types an invalid character, show it ... but immediately signal that something's wrong. Maybe turn the input field red and display a "Please enter numeric characters only" to the right of the field.
Blocking input is a no-no because it can be misinterpreted by the user. They might assume a problem with their machine or input device, not that they weren't supposed to type a letter in a numeric entry box.
Validating *after* the input is completed can be annoying. Particularly with fields like those used in password creation. Some sites require 6-character passwords with only alphanumeric characters. Others require 9-character passwords and allow any character (including !@#$% and such). Typing a password and only finding out *after* I've typed it is the most annoying form type I've found and has actually turned me away from several services.
Providing dynamic feedback as the user enters data helps them
1. Know which field had the error while they're still in the field
2. Proactively correct input errors before submitting data
3. Learn from the experience as they go
Blocking input fails on all 3 accounts. Highlighting validation errors after the textbox is filled in fails on all 3 as well. |
1,242 | I have some information to enter in a text box. The data is structured and only some sequences are valid. For a phone number, for instance, a letter such as `A` is invalid.
So should I intercept keyboard events and prevent typed letters from reaching the text box, or should I draw some red squiggly lines (or another visual cue) to let my user know the letter is not acceptable? Which is a better user experience? | 2010/09/02 | [
"https://ux.stackexchange.com/questions/1242",
"https://ux.stackexchange.com",
"https://ux.stackexchange.com/users/202/"
] | If you prevent invalid characters while they type, they could potentially think that their keyboard is malfunctioning. User's like things to work the way they expect them to. If you are going to prevent certain characters on key press, at the very least you should also display an error message near the input at the same time. This way they know why it isn't typing.
Another thing that can get annoying is when you just finished filling out a long form and the site uses server side validation so you have to go back through the form to find what's wrong. I like to use client side validation primarily but fall back on server side. I will display an error either as they are typing or when the input loses focus. That way the user knows right away that there is an issue and knows exactly what the problem is. | I would suggest a hybrid of both approaches, but only for strictly-validated fields like phone numbers (SSN, zipcodes, etc ... anything that **must** be numeric by definition).
Track each keystroke and, when the user types an invalid character, show it ... but immediately signal that something's wrong. Maybe turn the input field red and display a "Please enter numeric characters only" to the right of the field.
Blocking input is a no-no because it can be misinterpreted by the user. They might assume a problem with their machine or input device, not that they weren't supposed to type a letter in a numeric entry box.
Validating *after* the input is completed can be annoying. Particularly with fields like those used in password creation. Some sites require 6-character passwords with only alphanumeric characters. Others require 9-character passwords and allow any character (including !@#$% and such). Typing a password and only finding out *after* I've typed it is the most annoying form type I've found and has actually turned me away from several services.
Providing dynamic feedback as the user enters data helps them
1. Know which field had the error while they're still in the field
2. Proactively correct input errors before submitting data
3. Learn from the experience as they go
Blocking input fails on all 3 accounts. Highlighting validation errors after the textbox is filled in fails on all 3 as well. |
1,242 | I have some information to enter in a text box. The data is structured and only some sequences are valid. For a phone number, for instance, a letter such as `A` is invalid.
So should I intercept keyboard events and prevent typed letters from reaching the text box, or should I draw some red squiggly lines (or another visual cue) to let my user know the letter is not acceptable? Which is a better user experience? | 2010/09/02 | [
"https://ux.stackexchange.com/questions/1242",
"https://ux.stackexchange.com",
"https://ux.stackexchange.com/users/202/"
] | Being on both ends of using a form, here are some guidelines I prefer:
* User inputs should hardly ever be prevented. Meaning if I type the letter 'A' in a phone number field, I would expect to see that 'A'.
* I begin my text field validation on a `onBlur()` event, and no sooner than that. This gives the user a chance to fix that mis-typed 'A' before I show an error message. Allowing the user to correct an error while still in the input field provides one less chance for an obtrusive "hey, you messed up, dude" kind of message. | As with most things..depends on context. I have a more specialized take on the topic. Most of the applications I, or my group, work on are technical in nature and are only used by the reasonably trained. As such we can place more responsibility and freedom in the hands of the users than you could with the general public. Briefly, some of the guidelines we use...
* Disregard any invalid entry....block invalid keystrokes.
* Indicate invalid fields, a state that
should be all but impossible, by
contrast and color\*.
* Focus is always under user control.
Do not trap the user or prevent
a change of focus.
* Never throw a modal warning/error
box. Ever.
* Never ask the user questions. Software is a tool not a person. Yes I am sure.
* The Enter/Return keys always mean:
Accept changes and move on. Do not
change this by using focus on a
cancel/exit button as the default on
Enter.
* The Escape key always means: Discard
and exit.
* Provide auto-completion whenever
possible.
* Provide short-cut entry for numerics
that have a min/max limit. e.g. If a
valid value is between 0.12 and 100.387 then
entries greater that 100.387, like 999, are taken as 100.387, etc.
* Establish very clear black and white
rules on behavior..rules the user can
understand. If this becomes overly complex...re-think it.
Almost all of the examples are user requested and, yes...implementation is sometimes a real bear/nightmare/killer but rarely impossible.
Caveat: *Again, these examples are given in the context of a vertical system with a technical user base....not a web general public form, etc. Your mileage my vary.*
\*Except red. In our business red means: Danger you could be killed...or worse. |
1,242 | I have some information to enter in a text box. The data is structured and only some sequences are valid. For a phone number, for instance, a letter such as `A` is invalid.
So should I intercept keyboard events and prevent typed letters from reaching the text box, or should I draw some red squiggly lines (or another visual cue) to let my user know the letter is not acceptable? Which is a better user experience? | 2010/09/02 | [
"https://ux.stackexchange.com/questions/1242",
"https://ux.stackexchange.com",
"https://ux.stackexchange.com/users/202/"
] | Being on both ends of using a form, here are some guidelines I prefer:
* User inputs should hardly ever be prevented. Meaning if I type the letter 'A' in a phone number field, I would expect to see that 'A'.
* I begin my text field validation on a `onBlur()` event, and no sooner than that. This gives the user a chance to fix that mis-typed 'A' before I show an error message. Allowing the user to correct an error while still in the input field provides one less chance for an obtrusive "hey, you messed up, dude" kind of message. | It depends on the case. If possible use Auto-suggest. If you intercepts user's input, an immediate feedback is crucial. One of the possible solutions is "clue box" that pops up immediately after key is intercepted. You can see it on Windows login screen, it pops up if you press $ sign for instance. |
1,242 | I have some information to enter in a text box. The data is structured and only some sequences are valid. For a phone number, for instance, a letter such as `A` is invalid.
So should I intercept keyboard events and prevent typed letters from reaching the text box, or should I draw some red squiggly lines (or another visual cue) to let my user know the letter is not acceptable? Which is a better user experience? | 2010/09/02 | [
"https://ux.stackexchange.com/questions/1242",
"https://ux.stackexchange.com",
"https://ux.stackexchange.com/users/202/"
] | If you prevent invalid characters while they type, they could potentially think that their keyboard is malfunctioning. User's like things to work the way they expect them to. If you are going to prevent certain characters on key press, at the very least you should also display an error message near the input at the same time. This way they know why it isn't typing.
Another thing that can get annoying is when you just finished filling out a long form and the site uses server side validation so you have to go back through the form to find what's wrong. I like to use client side validation primarily but fall back on server side. I will display an error either as they are typing or when the input loses focus. That way the user knows right away that there is an issue and knows exactly what the problem is. | As with most things..depends on context. I have a more specialized take on the topic. Most of the applications I, or my group, work on are technical in nature and are only used by the reasonably trained. As such we can place more responsibility and freedom in the hands of the users than you could with the general public. Briefly, some of the guidelines we use...
* Disregard any invalid entry....block invalid keystrokes.
* Indicate invalid fields, a state that
should be all but impossible, by
contrast and color\*.
* Focus is always under user control.
Do not trap the user or prevent
a change of focus.
* Never throw a modal warning/error
box. Ever.
* Never ask the user questions. Software is a tool not a person. Yes I am sure.
* The Enter/Return keys always mean:
Accept changes and move on. Do not
change this by using focus on a
cancel/exit button as the default on
Enter.
* The Escape key always means: Discard
and exit.
* Provide auto-completion whenever
possible.
* Provide short-cut entry for numerics
that have a min/max limit. e.g. If a
valid value is between 0.12 and 100.387 then
entries greater that 100.387, like 999, are taken as 100.387, etc.
* Establish very clear black and white
rules on behavior..rules the user can
understand. If this becomes overly complex...re-think it.
Almost all of the examples are user requested and, yes...implementation is sometimes a real bear/nightmare/killer but rarely impossible.
Caveat: *Again, these examples are given in the context of a vertical system with a technical user base....not a web general public form, etc. Your mileage my vary.*
\*Except red. In our business red means: Danger you could be killed...or worse. |
1,124 | In Bonn, we've been have a discussion on the topic in the title:
>
> Suppose that A and B is are classes and that there are injections from A to B and fom B to A. Does it follow that there is a bijection between A and B?
>
>
>
Example: Let A the class of sets of cardinality one and let B be the class of sets of cardinality two. There is an injection
A -> B sending a to {a, emptyset},
B-> A sending b to {{b}}.
Does it follow that there is a bijection between A and B? | 2009/10/19 | [
"https://mathoverflow.net/questions/1124",
"https://mathoverflow.net",
"https://mathoverflow.net/users/296/"
] | Ignoring set-theoretic technicalities of formulating the question properly, I see no reason that the usual proof of Schroder-Bernstein wouldn't work.
(Set-theoretic technicalities: In the standard language of set theory, you can't quantify over classes, so you can't quite state this. However, you can prove a metatheorem saying that whenever you exhibit two such injections, you can prove there is also a bijection. Alternatively, you could work in [set theory with classes](http://en.wikipedia.org/wiki/Von_Neumann%E2%80%93Bernays%E2%80%93G%C3%B6del_set_theory), in which the statement can be made properly and you ought to be able to prove it just like ordinary Schroder-Bernstein. Alternatively, it is a trivial corollary of the "global" axiom of choice (which implies, in particular, that all proper classes have the same size), though this is kind of applying a sledgehammer.) | To ask this question, you have to first be clear what you mean by a "class". Do you mean a finite formula in ZFC language with one free variable, P(x)? (Writing P(x) roughly means "x has property P".)
Second, what do you mean by a map from a class P to a class Q? Do you mean a class of ordered pairs?
If "yes" to both of the above, then I think the answer to your question is also "yes", because the Schroeder-Bernstein argument will allow you to explicitly write down a formula F(x,y) from the formulas P( ) and Q( ) that is a "bijection" in the sense that for all x such that P(x) there is a unique y such that Q(y) and F(x,y).
If, however, you mean class in some other sense, like the undefined notion of "class" used in NBG set theory, or something more vague, a different answer will be required.
I recommend reading the wikipedia article on [Zermelo-Frenkel Set Theory](http://en.wikipedia.org/wiki/Zermelo%E2%80%93Fraenkel_set_theory)
and browsing related articles until you feel comfortable the precise meanings of the basic terminology :) |
137,477 | Searching for a story I read many years ago. I think the premise was that one man was treated as though he was a genius, but later discovered the rest of humanity had evolved far beyond him. He was treated specially so he would not feel bad about himself.
Might have been a short story in a magazine or anthology. Thanks for any help! | 2016/08/08 | [
"https://scifi.stackexchange.com/questions/137477",
"https://scifi.stackexchange.com",
"https://scifi.stackexchange.com/users/69941/"
] | ["Kindness"](https://www.isfdb.org/cgi-bin/title.cgi?47571) by [Lester del Rey](https://en.wikipedia.org/wiki/Lester_del_Rey), first published in [*Astounding Science Fiction*, October 1944](https://www.isfdb.org/cgi-bin/pl.cgi?57642+c), available at the [Internet Archive](https://archive.org/stream/Astounding_v34n02_1944-10_dtsg0318-LennyS#page/n79/mode/2up). Does any of [these covers](https://www.isfdb.org/cgi-bin/titlecovers.cgi?47571) look familiar?
In del Rey's story, Danny Black is the last surviving *Homo sapiens* in a world of supermen:
>
> *Homo sapiens!* The type of man who had come out of the caves and built a world of atomic power, electronics and other old-time wonders—thinking man, as it translated from the Latin. In the dim past, when his ancestors had owned the world, they had made a joke of it, shortening it to homo sap, and laughing, because there had been no other species to rival them. Now it was no longer a joke.
>
>
> Normal man had been only a "sap" to *homo intelligens*—intelligent man—who was now the master of the world. Danny was only a left-over, the last normal man in a world of supermen, hating the fact that he had been born, and that his mother had died at his birth to leave him only loneliness as his heritage.
>
>
>
Danny does not think he is a genius; he is painfully aware of his inferiority, although the supermen go to great lengths to treat him kindly and give him a make-work job. In the end, he thinks he has outwitted them by stealing an old rocket ship and escaping into space:
>
> The note might be ten years or half a dozen centuries old—but his people had been here, fighting on and managing to live, after Earth had been lost to them. If they could, so could he!
>
>
>
>
Meanwhile, back on Earth:
>
> "Well, he swallowed our lies and ran off with the ship we built him. I hope he's happy, for a while at least."
>
>
> | Perhaps you're looking for [Caliban](https://www.isfdb.org/cgi-bin/title.cgi?58602) by Robert Silverberg, first published in 1972. Here's a short summary:
>
> It's 2103 and people can change their bodily appearance practically at will. Everyone has chosen (this month) to be sleek-skinned, athletic, blue-eyed blondes. Into this world comes a man born in 1967, with all his bodily flaws, lumpy nose, dark hair, and quaint attitudes. He's a renowned curiosity, the world's greatest novelty, and gets to have sex with many women. But popularity in this world has a price ...
>
> [Source](http://www.majipoor.com/work.php?id=126)
>
>
> |
2,155 | Over the last few months I've started meditating late on a night (as well as my usual morning practice). I'm finding that I am sleeping less but not feeling tired because of it. Is there any information around about how mediation may or may not affect sleep patterns?
For instance [Dipa Ma](http://en.wikipedia.org/wiki/Dipa_Ma), the Indian housewife and meditation teacher was up at 4.00 am every morning to meditate for a few hours presumably finding that perfectly OK. That said [Margaret Thatcher](http://en.wikipedia.org/wiki/Margaret_Thatcher) only slept for 4 hours a night and I would hesitate to call her a Buddhist meditation master. | 2014/07/16 | [
"https://buddhism.stackexchange.com/questions/2155",
"https://buddhism.stackexchange.com",
"https://buddhism.stackexchange.com/users/157/"
] | The reason the body do not need as much sleep when one do a lot of meditation is because while we meditate, we are getting some rest that the body needs. Meditation increases the production of melatonin in our bodies. Melatonin is a hormone known for creating restful sleep.
[This article might help](http://eocinstitute.org/meditation/require-less-sleep-with-meditation-460/). | More you meditate less sleep you need. The need for sleep is based on how energetic you are. Meditation stirs up more energy and you become energetic.
>
> Similarly, when you go to bed at night, close your eyes and feel sensation anywhere within the body. If you fall asleep with this awareness, naturally as soon as you wake up in the morning, you will be aware of sensation. Perhaps you may not sleep soundly, or you may even remain fully awake throughout the night. This is wonderful, provided you stay lying in bed and maintain awareness and equanimity. The body will receive the rest it needs, and there is no greater rest for the mind than to remain aware and equanimous. However, if you start worrying that you are developing insomnia, then you will generate tensions, and will feel exhausted the next day. Nor should you forcefully try to stay awake, remaining in a seated posture all night; that would be going to an extreme. If sleep comes, very good; sleep. If sleep does not come, allow the body to rest by remaining in a recumbent position, and allow the mind to rest by remaining aware and equanimous.
>
>
>
...
>
> At first it may seem a heavy burden to devote two hours a day to meditation, but you will soon find that much time will be saved that was wasted in the past. Firstly, you will need less time for sleep. Secondly, you will be able to complete your work more quickly, because your capacity for work will increase. When a problem arises you will remain balanced, and will be able immediately to find the correct solution. As you become established in the technique, you will find that having meditated in the morning, you are full of energy throughout the day, without any agitation.
>
>
>
Source: [The Discourse Summaries](http://www.vridhamma.org/The-Discourse-Summaries) |
2,155 | Over the last few months I've started meditating late on a night (as well as my usual morning practice). I'm finding that I am sleeping less but not feeling tired because of it. Is there any information around about how mediation may or may not affect sleep patterns?
For instance [Dipa Ma](http://en.wikipedia.org/wiki/Dipa_Ma), the Indian housewife and meditation teacher was up at 4.00 am every morning to meditate for a few hours presumably finding that perfectly OK. That said [Margaret Thatcher](http://en.wikipedia.org/wiki/Margaret_Thatcher) only slept for 4 hours a night and I would hesitate to call her a Buddhist meditation master. | 2014/07/16 | [
"https://buddhism.stackexchange.com/questions/2155",
"https://buddhism.stackexchange.com",
"https://buddhism.stackexchange.com/users/157/"
] | The reason the body do not need as much sleep when one do a lot of meditation is because while we meditate, we are getting some rest that the body needs. Meditation increases the production of melatonin in our bodies. Melatonin is a hormone known for creating restful sleep.
[This article might help](http://eocinstitute.org/meditation/require-less-sleep-with-meditation-460/). | In Dvedhavitakka Sutta it is said that thinking tires the body. So I believe that meditation helps the body be more energetic by taking away some of the tiring thinking.
[...]
>
> "And as I remained thus heedful, ardent, & resolute, thinking imbued with renunciation arose in me. I discerned that 'Thinking imbued with renunciation has arisen in me; and that leads neither to my own affliction, nor to the affliction of others, nor to the affliction of both. It fosters discernment, promotes lack of vexation, & leads to Unbinding. If I were to think & ponder in line with that even for a night... even for a day... even for a day & night, I do not envision any danger that would come from it, **except that thinking & pondering a long time would tire the body. When the body is tired, the mind is disturbed; and a disturbed mind is far from concentration.' So I steadied my mind right within, settled, unified, & concentrated it. Why is that? So that my mind would not be disturbed.**
>
>
>
[...] |
2,155 | Over the last few months I've started meditating late on a night (as well as my usual morning practice). I'm finding that I am sleeping less but not feeling tired because of it. Is there any information around about how mediation may or may not affect sleep patterns?
For instance [Dipa Ma](http://en.wikipedia.org/wiki/Dipa_Ma), the Indian housewife and meditation teacher was up at 4.00 am every morning to meditate for a few hours presumably finding that perfectly OK. That said [Margaret Thatcher](http://en.wikipedia.org/wiki/Margaret_Thatcher) only slept for 4 hours a night and I would hesitate to call her a Buddhist meditation master. | 2014/07/16 | [
"https://buddhism.stackexchange.com/questions/2155",
"https://buddhism.stackexchange.com",
"https://buddhism.stackexchange.com/users/157/"
] | The reason the body do not need as much sleep when one do a lot of meditation is because while we meditate, we are getting some rest that the body needs. Meditation increases the production of melatonin in our bodies. Melatonin is a hormone known for creating restful sleep.
[This article might help](http://eocinstitute.org/meditation/require-less-sleep-with-meditation-460/). | Meditation can definitely cause you to sleep less. When meditating one is naturally entering a calmer state of mind.
When doing Samatha meditation one is temporarily suspending the hindrances thereby limiting the effect of *worries, excessive thinking, restlessness, cravings, aversions, anger etc.* This naturally leads to a more simple and unified mind. Such a mind will affect the physical body and calm it down.
The human brain uses more energy than both arms and legs together on e.g. cognitive processes. All the stress from the hindrances will sooner or later manifest itself in the physical body as stress, i.e. an increased secretion of stress hormones such as *cortisol and adrenalin*. This will create a lot extra tension and load on the internal organs which in turn will increase the demand for energy.
The result of that is the need for more sleep.
Now if one removes the above things then one naturally needs less sleep.
If we take a look at Vipassana Meditation then here we are directly working with the hindrances, i.e. by weakening them to the point where they can be eradicated completely. In vipassana meditation we learn to see reality for what it is, meaning that the mind do not extrapolate on phenomena. The mind do no follow after phenomena thereby launching a process of ideation and planning on how to posess the object.
A non-meditator, i.e. the untrained mind will not have this ability to see clearly. The untrained mind will see an object for a brief moment whereafter it will launch the process of ideation, extrapolation, judging, imagining, liking, disliking etc. Imagine how much energy these cognitive processes demand.
If one trains in clear comprehension one gradually builds the ability to stop the mind from launching these cognitive processes by applying mindfulness to mental and physical phenomena as they arise and cease. With time and training the mind will become deeply rooted in the present. It will become calm, peaceful and equanimous no matter what arises.
A reduction in mental stress levels results in the physiological processes of secretion of stress hormones will be stopped and reversed resulting in a calmer body and a lot less load on internal organs.
At night the body regrows tissue, builds muscle and bone and strengthens the immune system. When there is less load on the body then it also needs less time for restitution, i.e. less time to sleep. |
9,847,519 | I've a workflow that use a Business layer (in a separated dll)
This Business Layer check permissions using the IPrincipal roles, so workflow activities must set the Principal on current thread before calling a method on the business layer.
I've a problem when a wrokflow is resumed after a delay activity: the roles / IIdentity are lost (or worse: are wrong).
Does someone have an idea on how I can handle this case and be sure that, when resuming, it uses the IPrincipal set before the delay ?
Or do you have any idea on how to manage roles in workflows ?
Thank you! | 2012/03/23 | [
"https://Stackoverflow.com/questions/9847519",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/667449/"
] | My way around this has been to store the original principal as a workflow variable (IClaimsPrincipal in my case).
This has two advantages.
Firstly, it was persisted so that if the workflow was persisted then resumed, the original principal was still there. This was also important because the original context from which the principal was obtained may no longer be available when the workflow resumes.
Secondly (specifically for workflow services), it allowed me to check whether the principal continuing the workflow by calling another service operation (essentially the same logical session via correlation) is the same user that started the session (the same principal as the variable). | The authentication information may still be available via OperationContext.Current.ServiceSecurityContext- the following articles might be of help:
<http://zamd.net/2010/07/04/using-wif-with-workflow-services/>
http://msmvps.com/blogs/theproblemsolver/archive/2010/09/21/using-the-wcf-operationcontext-from-a-receive-activity.aspx |
46,337 | The picture below depicts one way to go around the rosary beads when praying the [Holy Rosary](https://christianity.stackexchange.com/questions/1155/what-is-the-purpose-of-a-rosary).
[](https://i.stack.imgur.com/oJQ3x.jpg)
If a child were to ask, as they typical do, as to why Catholics when praying the Holy Rosary customarily go around the rosary beads one way rather than the other - if that's what they do, how would one answer? Is there a customary way to go around rosary beads? If yes, what is the reason for that order?
---
A question in the "same series" as: [Is there a particular order in lighting candles on an advent wreath?](https://christianity.stackexchange.com/questions/35537/is-there-a-particular-order-in-lighting-candles-on-an-advent-wreath) | 2016/01/12 | [
"https://christianity.stackexchange.com/questions/46337",
"https://christianity.stackexchange.com",
"https://christianity.stackexchange.com/users/-1/"
] | Either direction is acceptable
------------------------------
This might be semi-authoritative since it comes straight from the only approved Marian apparition site in the United States. Up near Green Bay, Wisconsin there is an outdoor rosary walk with the mysteries visible for people who traverse in a right-hand or left-hand direction.
For instance, when you walk around clockwise, you see 1st glorious mystery, 1st luminous. Etc... On a big banner, but if you look at the backside of the banner, its got 5th glorious, 5th luminous, etc... So no matter whether you walk around clockwise or counter-clockwise, you will see the mysteries in their proper order. Which I remarked to my wife, "that's mighty clever, those Catholics are always thinking..." | The spacing on the beads are the same in both directions. You do not even need beads, I use an app on my phone, I'm techno savy. You can use your fingers and keep track of the mysteries that way. I have prayed the Rosary many times before Mass, hardly anyone has the beads as they pray together. There are many ways to pray the Rosary. Many things to contemplate when doing it. It is a method of Prayer not a system of Prayer. |
46,337 | The picture below depicts one way to go around the rosary beads when praying the [Holy Rosary](https://christianity.stackexchange.com/questions/1155/what-is-the-purpose-of-a-rosary).
[](https://i.stack.imgur.com/oJQ3x.jpg)
If a child were to ask, as they typical do, as to why Catholics when praying the Holy Rosary customarily go around the rosary beads one way rather than the other - if that's what they do, how would one answer? Is there a customary way to go around rosary beads? If yes, what is the reason for that order?
---
A question in the "same series" as: [Is there a particular order in lighting candles on an advent wreath?](https://christianity.stackexchange.com/questions/35537/is-there-a-particular-order-in-lighting-candles-on-an-advent-wreath) | 2016/01/12 | [
"https://christianity.stackexchange.com/questions/46337",
"https://christianity.stackexchange.com",
"https://christianity.stackexchange.com/users/-1/"
] | The spacing on the beads are the same in both directions. You do not even need beads, I use an app on my phone, I'm techno savy. You can use your fingers and keep track of the mysteries that way. I have prayed the Rosary many times before Mass, hardly anyone has the beads as they pray together. There are many ways to pray the Rosary. Many things to contemplate when doing it. It is a method of Prayer not a system of Prayer. | The question is not clear. If you hold the beads flat, there's a "direction" ( e.g. clockwise), but if they hang down e.g. while standing to pray, how would you tell which way is which? There's never an order marked on the beads. It's not even particularly strict that one must say the 3 Aves, Paternoster and Creed before the decades, as opposed to after. |
46,337 | The picture below depicts one way to go around the rosary beads when praying the [Holy Rosary](https://christianity.stackexchange.com/questions/1155/what-is-the-purpose-of-a-rosary).
[](https://i.stack.imgur.com/oJQ3x.jpg)
If a child were to ask, as they typical do, as to why Catholics when praying the Holy Rosary customarily go around the rosary beads one way rather than the other - if that's what they do, how would one answer? Is there a customary way to go around rosary beads? If yes, what is the reason for that order?
---
A question in the "same series" as: [Is there a particular order in lighting candles on an advent wreath?](https://christianity.stackexchange.com/questions/35537/is-there-a-particular-order-in-lighting-candles-on-an-advent-wreath) | 2016/01/12 | [
"https://christianity.stackexchange.com/questions/46337",
"https://christianity.stackexchange.com",
"https://christianity.stackexchange.com/users/-1/"
] | Either direction is acceptable
------------------------------
This might be semi-authoritative since it comes straight from the only approved Marian apparition site in the United States. Up near Green Bay, Wisconsin there is an outdoor rosary walk with the mysteries visible for people who traverse in a right-hand or left-hand direction.
For instance, when you walk around clockwise, you see 1st glorious mystery, 1st luminous. Etc... On a big banner, but if you look at the backside of the banner, its got 5th glorious, 5th luminous, etc... So no matter whether you walk around clockwise or counter-clockwise, you will see the mysteries in their proper order. Which I remarked to my wife, "that's mighty clever, those Catholics are always thinking..." | The question is not clear. If you hold the beads flat, there's a "direction" ( e.g. clockwise), but if they hang down e.g. while standing to pray, how would you tell which way is which? There's never an order marked on the beads. It's not even particularly strict that one must say the 3 Aves, Paternoster and Creed before the decades, as opposed to after. |
62,802 | >
> Not Counting the OS And the requirements to run the Development software. The storage space required.
>
>
>
Strictly speaking from work perspective(company setup and not freelancers).A individual developer ( not considering a build system) unless into areas of video,audio processing( huge raw files) 3d/graphics development.
How much storage space would be required.
* Even if we account for the software trials to download or reading material.is it Right or Safe to assume 20GB would most suffice and any thing more would be a waste and or would be improperly utilized?.
* What is the typical hard disk space allotted per developer in an office setup.This may differ per role or specific requirement and on what type of work the company is into. but on an average for a developer/programmer how much space is normally allotted.
Edit:
**To Clarify Intent**
These are questions i had faced by business/management people.I only wish to understand more in this regard to give an answer(or better answer)the next time i come across them. I am neither making assumptions or intend to give offense to any one in this regard.It would be helpful if some links to data online were provided in this regard.
Edit 2:
* The issue as i understand was restricting the storage space to only the saving of work files to discourage extraneous usage...
* Not about scrimping/cost saving on hardware. | 2011/03/28 | [
"https://softwareengineering.stackexchange.com/questions/62802",
"https://softwareengineering.stackexchange.com",
"https://softwareengineering.stackexchange.com/users/19554/"
] | My main work machine has a 160GB internal drive, a 160GB external drive, and a 250GB external drive.
All but the 250GB drive are almost full (which is why I added (at my own expense, btw...) the 250GB drive).
We're using VMWare images a lot, and those tend to get rather large.
3 projects, each with a 20GB image plus a backup image, is 120GB.
That doesn't include the space needed to install Oracle, WebLogic, 5 JBoss instances, IDEs, office software, operating system, etc. etc. which together take up another 100GB or so.
Then there's documents, eBooks, and whatever else you need. For productivity that'll likely include a few GB of music for iTunes (most anyone I know works better to music, and everyone prefers different music). And yes, I therefore do consider iTunes (or another music library/player) a productivity tool. | >
> These are questions i had faced by
> business/management people
>
>
>
You do realize that modern internal harddisks cost something like ¢3-5/GB? So you're suggesting that ¢60 should be enough for developer. That's ridiculous, a sounds very dilbertian to make management issue out of it. |
62,802 | >
> Not Counting the OS And the requirements to run the Development software. The storage space required.
>
>
>
Strictly speaking from work perspective(company setup and not freelancers).A individual developer ( not considering a build system) unless into areas of video,audio processing( huge raw files) 3d/graphics development.
How much storage space would be required.
* Even if we account for the software trials to download or reading material.is it Right or Safe to assume 20GB would most suffice and any thing more would be a waste and or would be improperly utilized?.
* What is the typical hard disk space allotted per developer in an office setup.This may differ per role or specific requirement and on what type of work the company is into. but on an average for a developer/programmer how much space is normally allotted.
Edit:
**To Clarify Intent**
These are questions i had faced by business/management people.I only wish to understand more in this regard to give an answer(or better answer)the next time i come across them. I am neither making assumptions or intend to give offense to any one in this regard.It would be helpful if some links to data online were provided in this regard.
Edit 2:
* The issue as i understand was restricting the storage space to only the saving of work files to discourage extraneous usage...
* Not about scrimping/cost saving on hardware. | 2011/03/28 | [
"https://softwareengineering.stackexchange.com/questions/62802",
"https://softwareengineering.stackexchange.com",
"https://softwareengineering.stackexchange.com/users/19554/"
] | Actual space needed? It depends. Despite your edit I'm still not really sure I get the point. You couldn't even buy a 20gb hard drive right now if you tried. You would almost always use whatever size hard drive has best GB/$ ratio. (Or a small fast drive (SSD) and a large slower storage drive) Thats about a 1TB right now (last time I looked anyway...). If you wanted to drop to 500gb, you'd be paying 80% of the cost for 50% of the space.
I would argue that somebody asking you to put a number on how much space you "need", doesn't even really understand what they are asking. | >
> These are questions i had faced by
> business/management people
>
>
>
You do realize that modern internal harddisks cost something like ¢3-5/GB? So you're suggesting that ¢60 should be enough for developer. That's ridiculous, a sounds very dilbertian to make management issue out of it. |
62,802 | >
> Not Counting the OS And the requirements to run the Development software. The storage space required.
>
>
>
Strictly speaking from work perspective(company setup and not freelancers).A individual developer ( not considering a build system) unless into areas of video,audio processing( huge raw files) 3d/graphics development.
How much storage space would be required.
* Even if we account for the software trials to download or reading material.is it Right or Safe to assume 20GB would most suffice and any thing more would be a waste and or would be improperly utilized?.
* What is the typical hard disk space allotted per developer in an office setup.This may differ per role or specific requirement and on what type of work the company is into. but on an average for a developer/programmer how much space is normally allotted.
Edit:
**To Clarify Intent**
These are questions i had faced by business/management people.I only wish to understand more in this regard to give an answer(or better answer)the next time i come across them. I am neither making assumptions or intend to give offense to any one in this regard.It would be helpful if some links to data online were provided in this regard.
Edit 2:
* The issue as i understand was restricting the storage space to only the saving of work files to discourage extraneous usage...
* Not about scrimping/cost saving on hardware. | 2011/03/28 | [
"https://softwareengineering.stackexchange.com/questions/62802",
"https://softwareengineering.stackexchange.com",
"https://softwareengineering.stackexchange.com/users/19554/"
] | Actual space needed? It depends. Despite your edit I'm still not really sure I get the point. You couldn't even buy a 20gb hard drive right now if you tried. You would almost always use whatever size hard drive has best GB/$ ratio. (Or a small fast drive (SSD) and a large slower storage drive) Thats about a 1TB right now (last time I looked anyway...). If you wanted to drop to 500gb, you'd be paying 80% of the cost for 50% of the space.
I would argue that somebody asking you to put a number on how much space you "need", doesn't even really understand what they are asking. | My main work machine has a 160GB internal drive, a 160GB external drive, and a 250GB external drive.
All but the 250GB drive are almost full (which is why I added (at my own expense, btw...) the 250GB drive).
We're using VMWare images a lot, and those tend to get rather large.
3 projects, each with a 20GB image plus a backup image, is 120GB.
That doesn't include the space needed to install Oracle, WebLogic, 5 JBoss instances, IDEs, office software, operating system, etc. etc. which together take up another 100GB or so.
Then there's documents, eBooks, and whatever else you need. For productivity that'll likely include a few GB of music for iTunes (most anyone I know works better to music, and everyone prefers different music). And yes, I therefore do consider iTunes (or another music library/player) a productivity tool. |
62,802 | >
> Not Counting the OS And the requirements to run the Development software. The storage space required.
>
>
>
Strictly speaking from work perspective(company setup and not freelancers).A individual developer ( not considering a build system) unless into areas of video,audio processing( huge raw files) 3d/graphics development.
How much storage space would be required.
* Even if we account for the software trials to download or reading material.is it Right or Safe to assume 20GB would most suffice and any thing more would be a waste and or would be improperly utilized?.
* What is the typical hard disk space allotted per developer in an office setup.This may differ per role or specific requirement and on what type of work the company is into. but on an average for a developer/programmer how much space is normally allotted.
Edit:
**To Clarify Intent**
These are questions i had faced by business/management people.I only wish to understand more in this regard to give an answer(or better answer)the next time i come across them. I am neither making assumptions or intend to give offense to any one in this regard.It would be helpful if some links to data online were provided in this regard.
Edit 2:
* The issue as i understand was restricting the storage space to only the saving of work files to discourage extraneous usage...
* Not about scrimping/cost saving on hardware. | 2011/03/28 | [
"https://softwareengineering.stackexchange.com/questions/62802",
"https://softwareengineering.stackexchange.com",
"https://softwareengineering.stackexchange.com/users/19554/"
] | Hard disk space is **incredibly cheap**. Why would you not give each developer a terabyte harddisk? Saves them from having to waste time cleaning up their disks every week.
That being said, I'm using about 150 gb of that terabyte, our code base alone is around 9 gb. | I usually go for very fast drive for the important files (programs and code) and a large drive for everything else.
That usually means a small SSD (128GB) in this configurarion:
80GB C: for the OS, Office, Visual Studio, SQL server etc.
30GB D: (the rest of the SSD) where i put all my code and some VM's
1.5TB E: here i put all the rest, isos, my music, videos, etc. |
62,802 | >
> Not Counting the OS And the requirements to run the Development software. The storage space required.
>
>
>
Strictly speaking from work perspective(company setup and not freelancers).A individual developer ( not considering a build system) unless into areas of video,audio processing( huge raw files) 3d/graphics development.
How much storage space would be required.
* Even if we account for the software trials to download or reading material.is it Right or Safe to assume 20GB would most suffice and any thing more would be a waste and or would be improperly utilized?.
* What is the typical hard disk space allotted per developer in an office setup.This may differ per role or specific requirement and on what type of work the company is into. but on an average for a developer/programmer how much space is normally allotted.
Edit:
**To Clarify Intent**
These are questions i had faced by business/management people.I only wish to understand more in this regard to give an answer(or better answer)the next time i come across them. I am neither making assumptions or intend to give offense to any one in this regard.It would be helpful if some links to data online were provided in this regard.
Edit 2:
* The issue as i understand was restricting the storage space to only the saving of work files to discourage extraneous usage...
* Not about scrimping/cost saving on hardware. | 2011/03/28 | [
"https://softwareengineering.stackexchange.com/questions/62802",
"https://softwareengineering.stackexchange.com",
"https://softwareengineering.stackexchange.com/users/19554/"
] | My main work machine has a 160GB internal drive, a 160GB external drive, and a 250GB external drive.
All but the 250GB drive are almost full (which is why I added (at my own expense, btw...) the 250GB drive).
We're using VMWare images a lot, and those tend to get rather large.
3 projects, each with a 20GB image plus a backup image, is 120GB.
That doesn't include the space needed to install Oracle, WebLogic, 5 JBoss instances, IDEs, office software, operating system, etc. etc. which together take up another 100GB or so.
Then there's documents, eBooks, and whatever else you need. For productivity that'll likely include a few GB of music for iTunes (most anyone I know works better to music, and everyone prefers different music). And yes, I therefore do consider iTunes (or another music library/player) a productivity tool. | Sufficient that your developers don't have to worry about running out of space whilst working. In my case that's room for big checkouts, (sanitised) copies of databases, VMs and "other stuff" - although some of the other stuff ought to be on shared space (of which I'd need a lot) to avoid duplication in a team environment.
Use (abuse) of space on a work PC for extraneous purposes should be a matter of policy and one not of enforced physical limitions on their principal dev boxes - not least because developers are the kind of people who will be able to find creative ways round restrictions quite possibly to the detriment of your network as a whole.
Note also that this is one of those areas where developers and other users do actually need to receive different treatment. |
62,802 | >
> Not Counting the OS And the requirements to run the Development software. The storage space required.
>
>
>
Strictly speaking from work perspective(company setup and not freelancers).A individual developer ( not considering a build system) unless into areas of video,audio processing( huge raw files) 3d/graphics development.
How much storage space would be required.
* Even if we account for the software trials to download or reading material.is it Right or Safe to assume 20GB would most suffice and any thing more would be a waste and or would be improperly utilized?.
* What is the typical hard disk space allotted per developer in an office setup.This may differ per role or specific requirement and on what type of work the company is into. but on an average for a developer/programmer how much space is normally allotted.
Edit:
**To Clarify Intent**
These are questions i had faced by business/management people.I only wish to understand more in this regard to give an answer(or better answer)the next time i come across them. I am neither making assumptions or intend to give offense to any one in this regard.It would be helpful if some links to data online were provided in this regard.
Edit 2:
* The issue as i understand was restricting the storage space to only the saving of work files to discourage extraneous usage...
* Not about scrimping/cost saving on hardware. | 2011/03/28 | [
"https://softwareengineering.stackexchange.com/questions/62802",
"https://softwareengineering.stackexchange.com",
"https://softwareengineering.stackexchange.com/users/19554/"
] | My main work machine has a 160GB internal drive, a 160GB external drive, and a 250GB external drive.
All but the 250GB drive are almost full (which is why I added (at my own expense, btw...) the 250GB drive).
We're using VMWare images a lot, and those tend to get rather large.
3 projects, each with a 20GB image plus a backup image, is 120GB.
That doesn't include the space needed to install Oracle, WebLogic, 5 JBoss instances, IDEs, office software, operating system, etc. etc. which together take up another 100GB or so.
Then there's documents, eBooks, and whatever else you need. For productivity that'll likely include a few GB of music for iTunes (most anyone I know works better to music, and everyone prefers different music). And yes, I therefore do consider iTunes (or another music library/player) a productivity tool. | I usually go for very fast drive for the important files (programs and code) and a large drive for everything else.
That usually means a small SSD (128GB) in this configurarion:
80GB C: for the OS, Office, Visual Studio, SQL server etc.
30GB D: (the rest of the SSD) where i put all my code and some VM's
1.5TB E: here i put all the rest, isos, my music, videos, etc. |
62,802 | >
> Not Counting the OS And the requirements to run the Development software. The storage space required.
>
>
>
Strictly speaking from work perspective(company setup and not freelancers).A individual developer ( not considering a build system) unless into areas of video,audio processing( huge raw files) 3d/graphics development.
How much storage space would be required.
* Even if we account for the software trials to download or reading material.is it Right or Safe to assume 20GB would most suffice and any thing more would be a waste and or would be improperly utilized?.
* What is the typical hard disk space allotted per developer in an office setup.This may differ per role or specific requirement and on what type of work the company is into. but on an average for a developer/programmer how much space is normally allotted.
Edit:
**To Clarify Intent**
These are questions i had faced by business/management people.I only wish to understand more in this regard to give an answer(or better answer)the next time i come across them. I am neither making assumptions or intend to give offense to any one in this regard.It would be helpful if some links to data online were provided in this regard.
Edit 2:
* The issue as i understand was restricting the storage space to only the saving of work files to discourage extraneous usage...
* Not about scrimping/cost saving on hardware. | 2011/03/28 | [
"https://softwareengineering.stackexchange.com/questions/62802",
"https://softwareengineering.stackexchange.com",
"https://softwareengineering.stackexchange.com/users/19554/"
] | Development needs a LOT of space.
We use VM images as units of configuration management for developer setups.
Once you've copied the VM onto your machine you start it, update the source code from the VCS and you're running. No futzing with developer setups.
Each VM image is about 20Gb. 4-5 of those. i.e. we generally need 100 to 120Gb Gb
Mind you, they are not "per user" on the machine things, only need one. | The only reasonable answer to this is "**enough**".
It doesn't take many developer hours to pay for a new harddrive... |
62,802 | >
> Not Counting the OS And the requirements to run the Development software. The storage space required.
>
>
>
Strictly speaking from work perspective(company setup and not freelancers).A individual developer ( not considering a build system) unless into areas of video,audio processing( huge raw files) 3d/graphics development.
How much storage space would be required.
* Even if we account for the software trials to download or reading material.is it Right or Safe to assume 20GB would most suffice and any thing more would be a waste and or would be improperly utilized?.
* What is the typical hard disk space allotted per developer in an office setup.This may differ per role or specific requirement and on what type of work the company is into. but on an average for a developer/programmer how much space is normally allotted.
Edit:
**To Clarify Intent**
These are questions i had faced by business/management people.I only wish to understand more in this regard to give an answer(or better answer)the next time i come across them. I am neither making assumptions or intend to give offense to any one in this regard.It would be helpful if some links to data online were provided in this regard.
Edit 2:
* The issue as i understand was restricting the storage space to only the saving of work files to discourage extraneous usage...
* Not about scrimping/cost saving on hardware. | 2011/03/28 | [
"https://softwareengineering.stackexchange.com/questions/62802",
"https://softwareengineering.stackexchange.com",
"https://softwareengineering.stackexchange.com/users/19554/"
] | Development needs a LOT of space.
We use VM images as units of configuration management for developer setups.
Once you've copied the VM onto your machine you start it, update the source code from the VCS and you're running. No futzing with developer setups.
Each VM image is about 20Gb. 4-5 of those. i.e. we generally need 100 to 120Gb Gb
Mind you, they are not "per user" on the machine things, only need one. | Actual space needed? It depends. Despite your edit I'm still not really sure I get the point. You couldn't even buy a 20gb hard drive right now if you tried. You would almost always use whatever size hard drive has best GB/$ ratio. (Or a small fast drive (SSD) and a large slower storage drive) Thats about a 1TB right now (last time I looked anyway...). If you wanted to drop to 500gb, you'd be paying 80% of the cost for 50% of the space.
I would argue that somebody asking you to put a number on how much space you "need", doesn't even really understand what they are asking. |
62,802 | >
> Not Counting the OS And the requirements to run the Development software. The storage space required.
>
>
>
Strictly speaking from work perspective(company setup and not freelancers).A individual developer ( not considering a build system) unless into areas of video,audio processing( huge raw files) 3d/graphics development.
How much storage space would be required.
* Even if we account for the software trials to download or reading material.is it Right or Safe to assume 20GB would most suffice and any thing more would be a waste and or would be improperly utilized?.
* What is the typical hard disk space allotted per developer in an office setup.This may differ per role or specific requirement and on what type of work the company is into. but on an average for a developer/programmer how much space is normally allotted.
Edit:
**To Clarify Intent**
These are questions i had faced by business/management people.I only wish to understand more in this regard to give an answer(or better answer)the next time i come across them. I am neither making assumptions or intend to give offense to any one in this regard.It would be helpful if some links to data online were provided in this regard.
Edit 2:
* The issue as i understand was restricting the storage space to only the saving of work files to discourage extraneous usage...
* Not about scrimping/cost saving on hardware. | 2011/03/28 | [
"https://softwareengineering.stackexchange.com/questions/62802",
"https://softwareengineering.stackexchange.com",
"https://softwareengineering.stackexchange.com/users/19554/"
] | Development needs a LOT of space.
We use VM images as units of configuration management for developer setups.
Once you've copied the VM onto your machine you start it, update the source code from the VCS and you're running. No futzing with developer setups.
Each VM image is about 20Gb. 4-5 of those. i.e. we generally need 100 to 120Gb Gb
Mind you, they are not "per user" on the machine things, only need one. | Sufficient that your developers don't have to worry about running out of space whilst working. In my case that's room for big checkouts, (sanitised) copies of databases, VMs and "other stuff" - although some of the other stuff ought to be on shared space (of which I'd need a lot) to avoid duplication in a team environment.
Use (abuse) of space on a work PC for extraneous purposes should be a matter of policy and one not of enforced physical limitions on their principal dev boxes - not least because developers are the kind of people who will be able to find creative ways round restrictions quite possibly to the detriment of your network as a whole.
Note also that this is one of those areas where developers and other users do actually need to receive different treatment. |
64,146 | Should I go out of my way to write certain female archetypes out of my stories? I think some archetypes for female characters are offensive. For example, there's the femme fatale archetype. The femme fatale generally reduces women to just their sexuality, and results in one-dimensional female characters that appeal to the male fantasies.
Should I write them out of my stories, or are there ways to make the femme fatale archetype more women friendly by modifying some of the characteristics of the archetype? | 2023/01/11 | [
"https://writers.stackexchange.com/questions/64146",
"https://writers.stackexchange.com",
"https://writers.stackexchange.com/users/49648/"
] | The main problem with archetypical characters (regardless of gender) is, as you mentioned, their one-dimensionality.
You can avoid one-dimensional characters by giving them:
1. A motivation which is if not ethically justifiable then at least understandable for the audience
2. A more fleshed out personality with quirks and flaws
3. A more nuanced and adaptive approach to how they act in different situations and what methods they use to reach their objectives
4. An inner conflict about what they do and how they do it | A character archetype is just another trope used in fiction that is specific to a character role in the story. Like all tropes, the mere existence is not an inherit flaw and it's important to know why the femme fatal is a widely used archtype for a female character. The character is normally not initially motivated to assist the hero, though will typically come around to the heroes side by the climax (but not necessarily for the same reason as the hero). In truth, the femme fatal in it's most basic version is a very pro-feminist character: While they typically aren't the stories main villain, or even aligned with them, they are almost universally the character that threatens the hero more than the villains, because they are the female foil of the male hero. Both are jaded to life, motivated out of self-interests instead of noble intent, and exude confidence.
In many stories, the male protaganist is written to appeal to the audience in the same way: Guys want to be him, girls want to be with him. The Femme Fatal was introduced not only to make the female lead something more than the prize for the hero to win... but to appeal to women who want to be more than just a prize. In effect, women want to be her, men want to be with her.
One of the other reasons the femme fatal is more pro-woman is that she doesn't sacrafice female beauty but is able to use it to her advantage. Afterall, if James Bond can be a badass in a tux or a spy suit, why can't a woman be a badass both in evening wear and a cat suit. Also, while she normally will fall for the hero in the story, the hero's attraction is typically not because she is attractive. The typical hero of noir and spy thriller settings, when they encounter a woman who throws herself to his feat, will step over them. The reason for the fatal's attraction isn't that she's attractive... but that she's actually playing his own game just as well as he plays it, if not better.
Now, that's the archtype in theory. In practice... well... I mean... some of the early Bond girls had some names that made the subtext rapidly become text (Looking at you, Pussy Galore). But just as James Bond is not a realistic depiction of espionage, but based on a glorification of real life espionage, the femme fatal Bond girl is not the most realistic depection of women in espionage, it's not without a basis in real life.
Because there were some pretty badass female spies who did their jobs and looked good doing it. The fallacy of this, however, is if a spy does their job right, then nobody would know they did it at all. Most real life "famous" spies are the ones that got exposed... aka... the bad one. The others are usually confined to document releases decades after the fact. Marlene Dietrich, a well known icon of beauty and considered one of the most attractive women in Hollywood, worked with the OSS to demoralize the troops in her Native Germany by incorperating her iconic status with her knowledge of German to create recordings that were intended to be heard by German troops on the front lines. Julia Child was also famously employed by the OSS at the same time as a chemist and devised a shark repellent that is still in use to this day (the repellent was used to keep sharks from getting too close underwater mines meant to take out U-Boats... which, lets face it, if you had a nickel for every famous person who was associated with a combination of the words "exploding", "shark", "repellant" and "submarine filled with bad guys" you'd have two nickels, which is not a lot, but it's weird that it happened twice... and she's the one who actually did it.).
More consistently, on of the most consistent tools used for espionage has been the "honey trap" which entices a potential informant by having a person who they find sexually attractive start a relationship with the potential informant, and then using the compromise of sleeping with someone who is an enemy spy to blackmail them into slipping classified documents to her and her handler. Although this is not a game played just by women, as men can perform the role just as well (as espionage is an industry that benefits from being an equal opportunity employer, and not all men are into women.). The honey trap is typically run by a woman who is using her physical beauty against an informant is a thing that definitely happens in real life, and during the time where the character was rising to prominence, the relationship between the U.S. and U.S.S.R. had deteriorated into a cold war where the major battles were fought in intelligence gathering. By their own admission, Russia was a notorious user of the honey trap (though they insisted East Germany was better at running same sex honey traps than they were).
But even then, there are feminine fatals in fiction that are plenty wholesome and latch on to the archtype because of their skill sets. The modern poster girl for the archtype is the Black Widow from the MCU movies, who's deception skills put her in positions where she's not suspected to be an action hero who is capable of having an upper hand (in Iron Man II, she poses as an assistent to Tony Stark and makes it clear that Tony may have a reputation as a man whore, but she want's no part in that (given she was trying to discreetly watch Tony, the surest way to get around with out Tony's notice is to specifically tell him she will not be another notch in his belt. In Avenger's she hints that her capture by the thugs in her intro scene was all part of the plan and that she's been manipulating them through out "their" interrogation and has all she wants to know and is presently humoring them. We next see her in action, she's trying to take in Bruce Banner and is being completely straight forward with him because she wants him specifically to know he's being brought in to be Bruce, not the Hulk and she has no interest in meeting the big guy.
Additionally, the only abuse she suffered was at the hands of the Red Room training she was given as part of the Black Widow program and she is not confident in her ability to sexually satisfy a man, given she's infertile, but she's pefectly capable of playing that in the short term, but it prevents her personal relationships.
Another example of a femme fatal with a positive outlook in the MCU is Peggy Carter in the Agent Carter series, who again never resorts to sex and also tells Tony's father that they are never getting together despite his womanizing reputation and how uncomfortable it makes her feel... but she does treat him as a platonic friend and he returns the relationship. Much of the show revolves around her using her position as the one female agent in the SSO office to gather to her advantage by playing to expectations. This results in her actually making some progress in her cases that elude her male co-workers, but because of the off the books nature, also gets her in some trouble for her actions that she has to take.
Turning to the noir and away from superheroes, Jessica Rabbit, actually comes out and tells the person she's manipulating that it's an act while she's doing that ("I'm not bad, I'm just drawn that way" It turns out that she's 100% telling the truth. Not only is she not the guilty party, she is in fact a victim in the matter and the line has multiple layers to it beyond that.). It's implied that for Toons like Jessica, physical attraction is not as important as the ability to make someone laugh. Jessica is conventionally attractive to humans but any toon who's asked about the matter strongly imply that Rodger is settling for Jessica and not the other way around as the humans believe. Much of her Femme Fatal coding is an act on her part because she knows humans find her very attractive, even if toons do not, as they are the only men other than Rodger who show her any kind of attention. The only reason her Femme Fatal character is played up is that, from her point of view, she can't trust Eddy and for all she knows, Eddy's in on the conspiracy. It's only after each other confirm that they are on the same side does she drop the Femme Fatal act around Eddy and talk more like a normal person (she figured it out first, but was still aware of Eddy not trusting her.). Up until the end of the Toon Town sequence, the audience is only shown half of Jessica's side, which makes her come off as someone different the the wife in the rare Hollywood marriage where both partners are deeply truly in love, faithful, and committed to each other. Hell, both show knowledge of the other that would only come from years of devotion. All of her appearances are motivated for us to assume the worst of her. Her first scene, her mannerisms are part of her act, which is a major draw to toon run club that strictly caters to humans. She knows her audience and is doing her part to put butts in seats. The Patty Cake scene was manipulated to deliberatly look like she was cheating on Rodger. As she explains, had she not done it, Rodger's employer would have fired Rodger and use his clout to blacklist him from working for antother studio, and it was only because she knew Rodger took his career seriously (as serious as one can when your a comedian at least.) and never mentions maintaining the perks of being the wife of a A-List cartoon star. We next seeing her slap Eddy at the crime scene and yelling at him, but this is given a new light in the context of her motives. She's genuinely pissed off that Eddy would assume of her the worst motives and the evil that Eddy brought into her and Rodger's life because of his pictures. And when we see her in Eddy's office, in her most riskee interaction, she isn't hiding the ball, but she isn't giving it away either. Again, she's trying to feel out Eddy's role in the mystery to see if he's someone who she should work with or work against. It's only in the Toon Town sequence do we learn that she's a good person when she saves Eddy, who would have been shot in the back by Doom had she not been there. Again, nothing she does is inconsistent with her motives (Knocking Rodger out with a frying pan makes sense to her because, well, no one ever accused her of not being a toon.). Her femme fatal persona, for the audience to hide in plain site and deliberately hide some obvious hints at her real role in the noir story.
Going back to the layers of her iconic line "I'm not bad, I'm just drawn that way" there is a double meaning. The more obvious one is a subtle nod that as a toon, she was delibertaly drawn (as in animated) to look the way she does for a certain purpose but that isn't who she is. This works because the central premise of the film is that all cartoons are made by toon actors playing a part on a set and like in real life, their public and private persona may be different. For whatever reason Jessica looks the way she does, there is more too her than what we see. The second meaning is one of two hints to her true role in the story... drawn here is a play on how people can "Draw conclusions". That is, Jessica is telling Eddy that she's not involved or a part of the grand conspiracy... but she understands why Eddy might have her on his suspect list. There is a character achtype that covers it and by now, it should be as obvious as her second clue, her iconic dress: Jessica isn't the femme fatal, but the Red Herring... a character who's purpose is to look suspicious in a mystery story, but be totally innocent of the crime. The love between Jessica and Rodger is real and she will do anything to protect her husband. But she also knows how people sees her, and is not ashamed to use her sex appeal if it was to keep Rodger safe, but at the end of the day, she knows that what she is doing looks wrong, but she doesn't care because she knows why she is doing it and it's for nothing but the purest of motives. |
64,146 | Should I go out of my way to write certain female archetypes out of my stories? I think some archetypes for female characters are offensive. For example, there's the femme fatale archetype. The femme fatale generally reduces women to just their sexuality, and results in one-dimensional female characters that appeal to the male fantasies.
Should I write them out of my stories, or are there ways to make the femme fatale archetype more women friendly by modifying some of the characteristics of the archetype? | 2023/01/11 | [
"https://writers.stackexchange.com/questions/64146",
"https://writers.stackexchange.com",
"https://writers.stackexchange.com/users/49648/"
] | The main problem with archetypical characters (regardless of gender) is, as you mentioned, their one-dimensionality.
You can avoid one-dimensional characters by giving them:
1. A motivation which is if not ethically justifiable then at least understandable for the audience
2. A more fleshed out personality with quirks and flaws
3. A more nuanced and adaptive approach to how they act in different situations and what methods they use to reach their objectives
4. An inner conflict about what they do and how they do it | The issue when writing archetypical characters into your story is making sure they aren't **just** that. hszmv's answer points out a whole bunch of characters and real people who possess all the traits of a femme fatale but crucially are all far more than just a list of traits.
You're right that a trap people fall into when using archetypes is that they "reduce" people into just being just a hollow shell wholly defined by whatever role they've been given in the story, and it's often that that puts people's backs up (particularly if from a section of the population that exist infrequently in certain genres of fiction *not* all possessing some prescribed attributes with little else to those characters). So instead, why not expand on their character instead. Why do they do what they do? Where do they come from? What do they want out of life? What are their thoughts and feelings? Their hobbies? What do they do when they're on their own or in a social setting that doesn't allow them to fill that role? All these questions (and many more) can give you a greater insight into who the character is than simply the general role their archetypical traits fulfil. |
64,146 | Should I go out of my way to write certain female archetypes out of my stories? I think some archetypes for female characters are offensive. For example, there's the femme fatale archetype. The femme fatale generally reduces women to just their sexuality, and results in one-dimensional female characters that appeal to the male fantasies.
Should I write them out of my stories, or are there ways to make the femme fatale archetype more women friendly by modifying some of the characteristics of the archetype? | 2023/01/11 | [
"https://writers.stackexchange.com/questions/64146",
"https://writers.stackexchange.com",
"https://writers.stackexchange.com/users/49648/"
] | A character archetype is just another trope used in fiction that is specific to a character role in the story. Like all tropes, the mere existence is not an inherit flaw and it's important to know why the femme fatal is a widely used archtype for a female character. The character is normally not initially motivated to assist the hero, though will typically come around to the heroes side by the climax (but not necessarily for the same reason as the hero). In truth, the femme fatal in it's most basic version is a very pro-feminist character: While they typically aren't the stories main villain, or even aligned with them, they are almost universally the character that threatens the hero more than the villains, because they are the female foil of the male hero. Both are jaded to life, motivated out of self-interests instead of noble intent, and exude confidence.
In many stories, the male protaganist is written to appeal to the audience in the same way: Guys want to be him, girls want to be with him. The Femme Fatal was introduced not only to make the female lead something more than the prize for the hero to win... but to appeal to women who want to be more than just a prize. In effect, women want to be her, men want to be with her.
One of the other reasons the femme fatal is more pro-woman is that she doesn't sacrafice female beauty but is able to use it to her advantage. Afterall, if James Bond can be a badass in a tux or a spy suit, why can't a woman be a badass both in evening wear and a cat suit. Also, while she normally will fall for the hero in the story, the hero's attraction is typically not because she is attractive. The typical hero of noir and spy thriller settings, when they encounter a woman who throws herself to his feat, will step over them. The reason for the fatal's attraction isn't that she's attractive... but that she's actually playing his own game just as well as he plays it, if not better.
Now, that's the archtype in theory. In practice... well... I mean... some of the early Bond girls had some names that made the subtext rapidly become text (Looking at you, Pussy Galore). But just as James Bond is not a realistic depiction of espionage, but based on a glorification of real life espionage, the femme fatal Bond girl is not the most realistic depection of women in espionage, it's not without a basis in real life.
Because there were some pretty badass female spies who did their jobs and looked good doing it. The fallacy of this, however, is if a spy does their job right, then nobody would know they did it at all. Most real life "famous" spies are the ones that got exposed... aka... the bad one. The others are usually confined to document releases decades after the fact. Marlene Dietrich, a well known icon of beauty and considered one of the most attractive women in Hollywood, worked with the OSS to demoralize the troops in her Native Germany by incorperating her iconic status with her knowledge of German to create recordings that were intended to be heard by German troops on the front lines. Julia Child was also famously employed by the OSS at the same time as a chemist and devised a shark repellent that is still in use to this day (the repellent was used to keep sharks from getting too close underwater mines meant to take out U-Boats... which, lets face it, if you had a nickel for every famous person who was associated with a combination of the words "exploding", "shark", "repellant" and "submarine filled with bad guys" you'd have two nickels, which is not a lot, but it's weird that it happened twice... and she's the one who actually did it.).
More consistently, on of the most consistent tools used for espionage has been the "honey trap" which entices a potential informant by having a person who they find sexually attractive start a relationship with the potential informant, and then using the compromise of sleeping with someone who is an enemy spy to blackmail them into slipping classified documents to her and her handler. Although this is not a game played just by women, as men can perform the role just as well (as espionage is an industry that benefits from being an equal opportunity employer, and not all men are into women.). The honey trap is typically run by a woman who is using her physical beauty against an informant is a thing that definitely happens in real life, and during the time where the character was rising to prominence, the relationship between the U.S. and U.S.S.R. had deteriorated into a cold war where the major battles were fought in intelligence gathering. By their own admission, Russia was a notorious user of the honey trap (though they insisted East Germany was better at running same sex honey traps than they were).
But even then, there are feminine fatals in fiction that are plenty wholesome and latch on to the archtype because of their skill sets. The modern poster girl for the archtype is the Black Widow from the MCU movies, who's deception skills put her in positions where she's not suspected to be an action hero who is capable of having an upper hand (in Iron Man II, she poses as an assistent to Tony Stark and makes it clear that Tony may have a reputation as a man whore, but she want's no part in that (given she was trying to discreetly watch Tony, the surest way to get around with out Tony's notice is to specifically tell him she will not be another notch in his belt. In Avenger's she hints that her capture by the thugs in her intro scene was all part of the plan and that she's been manipulating them through out "their" interrogation and has all she wants to know and is presently humoring them. We next see her in action, she's trying to take in Bruce Banner and is being completely straight forward with him because she wants him specifically to know he's being brought in to be Bruce, not the Hulk and she has no interest in meeting the big guy.
Additionally, the only abuse she suffered was at the hands of the Red Room training she was given as part of the Black Widow program and she is not confident in her ability to sexually satisfy a man, given she's infertile, but she's pefectly capable of playing that in the short term, but it prevents her personal relationships.
Another example of a femme fatal with a positive outlook in the MCU is Peggy Carter in the Agent Carter series, who again never resorts to sex and also tells Tony's father that they are never getting together despite his womanizing reputation and how uncomfortable it makes her feel... but she does treat him as a platonic friend and he returns the relationship. Much of the show revolves around her using her position as the one female agent in the SSO office to gather to her advantage by playing to expectations. This results in her actually making some progress in her cases that elude her male co-workers, but because of the off the books nature, also gets her in some trouble for her actions that she has to take.
Turning to the noir and away from superheroes, Jessica Rabbit, actually comes out and tells the person she's manipulating that it's an act while she's doing that ("I'm not bad, I'm just drawn that way" It turns out that she's 100% telling the truth. Not only is she not the guilty party, she is in fact a victim in the matter and the line has multiple layers to it beyond that.). It's implied that for Toons like Jessica, physical attraction is not as important as the ability to make someone laugh. Jessica is conventionally attractive to humans but any toon who's asked about the matter strongly imply that Rodger is settling for Jessica and not the other way around as the humans believe. Much of her Femme Fatal coding is an act on her part because she knows humans find her very attractive, even if toons do not, as they are the only men other than Rodger who show her any kind of attention. The only reason her Femme Fatal character is played up is that, from her point of view, she can't trust Eddy and for all she knows, Eddy's in on the conspiracy. It's only after each other confirm that they are on the same side does she drop the Femme Fatal act around Eddy and talk more like a normal person (she figured it out first, but was still aware of Eddy not trusting her.). Up until the end of the Toon Town sequence, the audience is only shown half of Jessica's side, which makes her come off as someone different the the wife in the rare Hollywood marriage where both partners are deeply truly in love, faithful, and committed to each other. Hell, both show knowledge of the other that would only come from years of devotion. All of her appearances are motivated for us to assume the worst of her. Her first scene, her mannerisms are part of her act, which is a major draw to toon run club that strictly caters to humans. She knows her audience and is doing her part to put butts in seats. The Patty Cake scene was manipulated to deliberatly look like she was cheating on Rodger. As she explains, had she not done it, Rodger's employer would have fired Rodger and use his clout to blacklist him from working for antother studio, and it was only because she knew Rodger took his career seriously (as serious as one can when your a comedian at least.) and never mentions maintaining the perks of being the wife of a A-List cartoon star. We next seeing her slap Eddy at the crime scene and yelling at him, but this is given a new light in the context of her motives. She's genuinely pissed off that Eddy would assume of her the worst motives and the evil that Eddy brought into her and Rodger's life because of his pictures. And when we see her in Eddy's office, in her most riskee interaction, she isn't hiding the ball, but she isn't giving it away either. Again, she's trying to feel out Eddy's role in the mystery to see if he's someone who she should work with or work against. It's only in the Toon Town sequence do we learn that she's a good person when she saves Eddy, who would have been shot in the back by Doom had she not been there. Again, nothing she does is inconsistent with her motives (Knocking Rodger out with a frying pan makes sense to her because, well, no one ever accused her of not being a toon.). Her femme fatal persona, for the audience to hide in plain site and deliberately hide some obvious hints at her real role in the noir story.
Going back to the layers of her iconic line "I'm not bad, I'm just drawn that way" there is a double meaning. The more obvious one is a subtle nod that as a toon, she was delibertaly drawn (as in animated) to look the way she does for a certain purpose but that isn't who she is. This works because the central premise of the film is that all cartoons are made by toon actors playing a part on a set and like in real life, their public and private persona may be different. For whatever reason Jessica looks the way she does, there is more too her than what we see. The second meaning is one of two hints to her true role in the story... drawn here is a play on how people can "Draw conclusions". That is, Jessica is telling Eddy that she's not involved or a part of the grand conspiracy... but she understands why Eddy might have her on his suspect list. There is a character achtype that covers it and by now, it should be as obvious as her second clue, her iconic dress: Jessica isn't the femme fatal, but the Red Herring... a character who's purpose is to look suspicious in a mystery story, but be totally innocent of the crime. The love between Jessica and Rodger is real and she will do anything to protect her husband. But she also knows how people sees her, and is not ashamed to use her sex appeal if it was to keep Rodger safe, but at the end of the day, she knows that what she is doing looks wrong, but she doesn't care because she knows why she is doing it and it's for nothing but the purest of motives. | The issue when writing archetypical characters into your story is making sure they aren't **just** that. hszmv's answer points out a whole bunch of characters and real people who possess all the traits of a femme fatale but crucially are all far more than just a list of traits.
You're right that a trap people fall into when using archetypes is that they "reduce" people into just being just a hollow shell wholly defined by whatever role they've been given in the story, and it's often that that puts people's backs up (particularly if from a section of the population that exist infrequently in certain genres of fiction *not* all possessing some prescribed attributes with little else to those characters). So instead, why not expand on their character instead. Why do they do what they do? Where do they come from? What do they want out of life? What are their thoughts and feelings? Their hobbies? What do they do when they're on their own or in a social setting that doesn't allow them to fill that role? All these questions (and many more) can give you a greater insight into who the character is than simply the general role their archetypical traits fulfil. |
71,569,007 | I'm trying to count the number of factory operators used to manufacture parts during a shift but I am double counting them as this example illustrates:
**Machine Groups**
* Group A : Machines 1 and 2, employing 3 operators per shift
* Group B : Machines 3 and 4, employing 2 operators per shift
**Shift Output**
| Group | Machine | Operators | Item | Quantity |
| --- | --- | --- | --- | --- |
| Grp A | Mach 1 | 3 | Nuts | 1000 |
| Grp A | Mach 2 | 3 | Bolts | 500 |
| Grp B | Mach 3 | 2 | Washers | 2000 |
| Grp B | Mach 4 | 2 | Springs | 1500 |
| **Total** | | **10** | | **5000** |
So the total quantity of parts is correct but the total number of operators is incorrect as it should only be 5. Operators are being double-counted because they make 2 different parts.
I have tried using an implicit sum on the operators column and also a DAX sum
* Sum Operators = SUM(Production(Operators))
I have also tried with a matrix rather than a simple table but get the same result.
(There will not always be 2 items per shift. There could sometimes be 3 or 4) | 2022/03/22 | [
"https://Stackoverflow.com/questions/71569007",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/9871468/"
] | If I understand you correctly, you can
* click the down arrow on the Operators Values line.
* Select to create a new quick measure
* Average (or Min or Max since they would all be the same) by Group
Then you can delete the original Operators entry and rename this new one.
[](https://i.stack.imgur.com/KVzSD.png)
*Note: I am very new with this so there may be more efficient methods to do this* | I've had a read and play and come up with an alternative which also works:
[](https://i.stack.imgur.com/WsIHX.png) |
11,230,067 | I am looking at taking the inverse of a large matrix, common size of 1000 x 1000, but sometimes exceeds 100000 x 100000 (which is currently failing due to time and memory). I know that the normal sentiment is 'don't take the inverse, find some other way to do it', but that is not possible at the moment. The reason for this is due to the usage of software that is already made that expects to get the matrix inverse. (Note: I am looking into ways of changing this, but that will take a long time)
At the moment we are using an LU decomposition method from numerical recopies, and I am currently in the process of testing the eigen library. The eigen library seems to be more stable and a bit faster, but I am still in testing phase for accuracy. I have taken a quick look at other libraries such as ATLAS and LAPACK but have not done any substantial testing with these yet.
It seems as though the eigen library does not use concurrent methods to compute the inverse (though does for LU factorization part of the inverse) and as far as I can tell ATLAS and LAPACK are similar in this limitation. (I am currently testing the speed difference for eigen with openMP and without.)
First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization. I found an article [here](http://www.bhrc.ac.ir/Portal/LinkClick.aspx?fileticket=oLt3rF1LlTo=&tabid=584) that talks about matrix inversion parallel algorithms, but I did not understand. It seems [this](http://world-comp.org/p2011/PDP5100.pdf) article talks about another method? I am also not sure if scaLAPACK or PETSc are useful?
Second question, I read [this](http://www.hpca.uji.es/ficheros/remon/pdp11.pdf) article of using the GPUs to increase performance, but I have never coded for GPUs and so have no idea what is trying to convey, but the charts at the bottom looked rather alarming. How is this even possible, and how where do I start to go about implementing something like this if it is to be true.
I also found [this](http://lara.inist.fr/bitstream/handle/2332/784/RR2002-04.pdf?sequence=1) article, have yet had the time to read through it to understand, but it seems promising, as memory is a current issue with our software.
Any information about these articles or the problems in general would be of great help. And again I apologize if this question seems vague, I will try to expand more if necessary. | 2012/06/27 | [
"https://Stackoverflow.com/questions/11230067",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1444573/"
] | *First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization.*
I'd hazard a guess that this, and related topics in linear algebra, is one of the most studied topics in parallel computing. If you're stuck looking for somewhere to start reading, well good old [Golub and Van Loan](http://www.amazon.co.uk/Computations-Hopkins-Studies-Mathematical-Sciences/dp/0801854148/ref=sr_1_1?s=books&ie=UTF8&qid=1340812109&sr=1-1) have a chapter on the topic. As to whether Scalapack and Petsc are likely to be useful, certainly the former, probably the latter. Of course, they both depend on MPI but that's kind of taken for granted in this field.
*Second question ...*
Use GPUs if you've got them and you can afford to translate your code into the programming model supported by your GPUs. If you've never coded for GPUs and have access to a cluster of commodity-type CPUs you'll get up to speed quicker by using the cluster than by wrestling with a novel technology.
As for the last article you refer to, it's now 10 years old in a field that changes very quickly (try finding a 10-year old research paper on using GPUs for matrix inversion). I can't comment on its excellence or other attributes, but the problem sizes you mention seem to me to be well within the capabilities of modern clusters for in-core (to use an old term) computation. If your matrices are very big, are they also sparse ?
Finally, I strongly support your apparent intention to use existing off-the-shelf codes rather than to try to develop your own. | 100000 x 100000 is 80GB at double precision. You need a library that supports memory-mapped matrices on disk. I can't recommend a particular library and I didn't find anything with quick Google searches. But code from Numerical Recipes certainly isn't going to be adequate. |
11,230,067 | I am looking at taking the inverse of a large matrix, common size of 1000 x 1000, but sometimes exceeds 100000 x 100000 (which is currently failing due to time and memory). I know that the normal sentiment is 'don't take the inverse, find some other way to do it', but that is not possible at the moment. The reason for this is due to the usage of software that is already made that expects to get the matrix inverse. (Note: I am looking into ways of changing this, but that will take a long time)
At the moment we are using an LU decomposition method from numerical recopies, and I am currently in the process of testing the eigen library. The eigen library seems to be more stable and a bit faster, but I am still in testing phase for accuracy. I have taken a quick look at other libraries such as ATLAS and LAPACK but have not done any substantial testing with these yet.
It seems as though the eigen library does not use concurrent methods to compute the inverse (though does for LU factorization part of the inverse) and as far as I can tell ATLAS and LAPACK are similar in this limitation. (I am currently testing the speed difference for eigen with openMP and without.)
First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization. I found an article [here](http://www.bhrc.ac.ir/Portal/LinkClick.aspx?fileticket=oLt3rF1LlTo=&tabid=584) that talks about matrix inversion parallel algorithms, but I did not understand. It seems [this](http://world-comp.org/p2011/PDP5100.pdf) article talks about another method? I am also not sure if scaLAPACK or PETSc are useful?
Second question, I read [this](http://www.hpca.uji.es/ficheros/remon/pdp11.pdf) article of using the GPUs to increase performance, but I have never coded for GPUs and so have no idea what is trying to convey, but the charts at the bottom looked rather alarming. How is this even possible, and how where do I start to go about implementing something like this if it is to be true.
I also found [this](http://lara.inist.fr/bitstream/handle/2332/784/RR2002-04.pdf?sequence=1) article, have yet had the time to read through it to understand, but it seems promising, as memory is a current issue with our software.
Any information about these articles or the problems in general would be of great help. And again I apologize if this question seems vague, I will try to expand more if necessary. | 2012/06/27 | [
"https://Stackoverflow.com/questions/11230067",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1444573/"
] | 100000 x 100000 is 80GB at double precision. You need a library that supports memory-mapped matrices on disk. I can't recommend a particular library and I didn't find anything with quick Google searches. But code from Numerical Recipes certainly isn't going to be adequate. | An LU decomp on a GPU can be ~10x faster than on a CPU. Although this is now changing, GPU's have traditionally been designed around single precision arithmetic, and so on older hardware single precision arithmetic is generally much faster than double precision arithmetic. Also, storage requirements and performance will be greatly impacted by the structure of your matrices. A sparse 100,000 x 100,000 matrix LU decomp is a reasonable problem to solve and will not require much memory.
Unless you want to become a specialist and spend *a lot* of time tuning for hardware updates, I would strongly recommend using a commercial library. I would suggest [CULA tools](http://www.culatools.com/). They have both sparse and dense GPU libraries and in fact their [free library](http://www.culatools.com/cula_dense_programmers_guide/) offers SGETRF - a single precision (dense) LU decomp routine. You'll have to pay for their double precision libraries. |
11,230,067 | I am looking at taking the inverse of a large matrix, common size of 1000 x 1000, but sometimes exceeds 100000 x 100000 (which is currently failing due to time and memory). I know that the normal sentiment is 'don't take the inverse, find some other way to do it', but that is not possible at the moment. The reason for this is due to the usage of software that is already made that expects to get the matrix inverse. (Note: I am looking into ways of changing this, but that will take a long time)
At the moment we are using an LU decomposition method from numerical recopies, and I am currently in the process of testing the eigen library. The eigen library seems to be more stable and a bit faster, but I am still in testing phase for accuracy. I have taken a quick look at other libraries such as ATLAS and LAPACK but have not done any substantial testing with these yet.
It seems as though the eigen library does not use concurrent methods to compute the inverse (though does for LU factorization part of the inverse) and as far as I can tell ATLAS and LAPACK are similar in this limitation. (I am currently testing the speed difference for eigen with openMP and without.)
First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization. I found an article [here](http://www.bhrc.ac.ir/Portal/LinkClick.aspx?fileticket=oLt3rF1LlTo=&tabid=584) that talks about matrix inversion parallel algorithms, but I did not understand. It seems [this](http://world-comp.org/p2011/PDP5100.pdf) article talks about another method? I am also not sure if scaLAPACK or PETSc are useful?
Second question, I read [this](http://www.hpca.uji.es/ficheros/remon/pdp11.pdf) article of using the GPUs to increase performance, but I have never coded for GPUs and so have no idea what is trying to convey, but the charts at the bottom looked rather alarming. How is this even possible, and how where do I start to go about implementing something like this if it is to be true.
I also found [this](http://lara.inist.fr/bitstream/handle/2332/784/RR2002-04.pdf?sequence=1) article, have yet had the time to read through it to understand, but it seems promising, as memory is a current issue with our software.
Any information about these articles or the problems in general would be of great help. And again I apologize if this question seems vague, I will try to expand more if necessary. | 2012/06/27 | [
"https://Stackoverflow.com/questions/11230067",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1444573/"
] | 100000 x 100000 is 80GB at double precision. You need a library that supports memory-mapped matrices on disk. I can't recommend a particular library and I didn't find anything with quick Google searches. But code from Numerical Recipes certainly isn't going to be adequate. | I know it's old post - but really - OpenCL (you download the relevant one based on your graphics card) + OpenMP + Vectorization (not in that order) is the way to go.
Anyhow - for me my experience with matrix anything is really to do with overheads from copying double double arrays in and out the system and also to pad up or initialize matrices with 0s before any commencement of computation - especially when I am working with creating .xll for Excel usage.
If I were to reprioritize the top -
1. try to vectorize the code (Visual Studio 2012 and Intel C++ has autovectorization - I'm not sure about MinGW or GCC, but I think there are flags for the compiler to analyse your for loops to generate the right assembly codes to use instead of the normal registers to hold your data, to populate your processor's vector registers. I think Excel is doing that because when I monitored Excel's threads while running their MINVERSE(), I notice only 1 thread is used.
I don't know much assembly language - so I don't know how to vectorize manually... (haven't had time to go learn this yet but sooooo wanna do it!)
2. Parallelize with OpenMP (omp pragma) or MPI or pthreads library (parallel\_for) - very simple - but... here's the catch - I realise that if your matrix class is completely single threaded in the first place - then parallelizing the operation like mat multiply or inverse is scrappable - cuz parallelizing will deteriorate the speed due to initializing or copying to or just accessing the non-parallelized matrix class.
But... where parallelization helps is - if you're designing your own matrix class and you parallelize its constructor operation (padding with 0s etc), then your computation of LU(A^-1) = I will also be faster.
It's also mathematically straightforward to also optimize your LU decomposition, and also optimizing ur forward backward substitution for the special case of identity. (I.e. don't waste time creating any identity matrix - analyse where your for (row = col) and evaluate to be a function with 1 and the rest with 0.)
3. Once it's been parallelized (on the outer layers) - the matrix operations requiring element by element can be mapped to be computed by GPU(SSSSSS) - hundreds of processors to compute elements - beat that!. There is now sample Monte Carlo code available on ATI's website - using ATI's OpenCL - don't worry about porting code to something that uses GeForce - all u gotta do is recompile there.
For 2 and 3 though - remember that overheads are incurred so no point unless you're handling F\**K*\*G HUGE matrices - but I see 100k^2? wow...
Gene |
11,230,067 | I am looking at taking the inverse of a large matrix, common size of 1000 x 1000, but sometimes exceeds 100000 x 100000 (which is currently failing due to time and memory). I know that the normal sentiment is 'don't take the inverse, find some other way to do it', but that is not possible at the moment. The reason for this is due to the usage of software that is already made that expects to get the matrix inverse. (Note: I am looking into ways of changing this, but that will take a long time)
At the moment we are using an LU decomposition method from numerical recopies, and I am currently in the process of testing the eigen library. The eigen library seems to be more stable and a bit faster, but I am still in testing phase for accuracy. I have taken a quick look at other libraries such as ATLAS and LAPACK but have not done any substantial testing with these yet.
It seems as though the eigen library does not use concurrent methods to compute the inverse (though does for LU factorization part of the inverse) and as far as I can tell ATLAS and LAPACK are similar in this limitation. (I am currently testing the speed difference for eigen with openMP and without.)
First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization. I found an article [here](http://www.bhrc.ac.ir/Portal/LinkClick.aspx?fileticket=oLt3rF1LlTo=&tabid=584) that talks about matrix inversion parallel algorithms, but I did not understand. It seems [this](http://world-comp.org/p2011/PDP5100.pdf) article talks about another method? I am also not sure if scaLAPACK or PETSc are useful?
Second question, I read [this](http://www.hpca.uji.es/ficheros/remon/pdp11.pdf) article of using the GPUs to increase performance, but I have never coded for GPUs and so have no idea what is trying to convey, but the charts at the bottom looked rather alarming. How is this even possible, and how where do I start to go about implementing something like this if it is to be true.
I also found [this](http://lara.inist.fr/bitstream/handle/2332/784/RR2002-04.pdf?sequence=1) article, have yet had the time to read through it to understand, but it seems promising, as memory is a current issue with our software.
Any information about these articles or the problems in general would be of great help. And again I apologize if this question seems vague, I will try to expand more if necessary. | 2012/06/27 | [
"https://Stackoverflow.com/questions/11230067",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1444573/"
] | *First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization.*
I'd hazard a guess that this, and related topics in linear algebra, is one of the most studied topics in parallel computing. If you're stuck looking for somewhere to start reading, well good old [Golub and Van Loan](http://www.amazon.co.uk/Computations-Hopkins-Studies-Mathematical-Sciences/dp/0801854148/ref=sr_1_1?s=books&ie=UTF8&qid=1340812109&sr=1-1) have a chapter on the topic. As to whether Scalapack and Petsc are likely to be useful, certainly the former, probably the latter. Of course, they both depend on MPI but that's kind of taken for granted in this field.
*Second question ...*
Use GPUs if you've got them and you can afford to translate your code into the programming model supported by your GPUs. If you've never coded for GPUs and have access to a cluster of commodity-type CPUs you'll get up to speed quicker by using the cluster than by wrestling with a novel technology.
As for the last article you refer to, it's now 10 years old in a field that changes very quickly (try finding a 10-year old research paper on using GPUs for matrix inversion). I can't comment on its excellence or other attributes, but the problem sizes you mention seem to me to be well within the capabilities of modern clusters for in-core (to use an old term) computation. If your matrices are very big, are they also sparse ?
Finally, I strongly support your apparent intention to use existing off-the-shelf codes rather than to try to develop your own. | Regarding the first question (how to parallellize computing the inverse):
I assume you are computing the inverse by doing an LU decomposition of your matrix and then using the decomposition to solve A\*B = I where A is your original matrix, B is the matrix you solve for, and I is the identity matrix. Then B is the inverse.
The last step is easy to parallellize. Divide your identity matrix along the columns. If you have p CPUs and your matrix is n-by-n, then every part has n/p columns and n rows. Lets call the parts I1, I2, etc. On every CPU, solve a system of the form A\*B1 = I1, this gives you the parts B1, B2, etc., and you can combine them to form B which is the inverse. |
11,230,067 | I am looking at taking the inverse of a large matrix, common size of 1000 x 1000, but sometimes exceeds 100000 x 100000 (which is currently failing due to time and memory). I know that the normal sentiment is 'don't take the inverse, find some other way to do it', but that is not possible at the moment. The reason for this is due to the usage of software that is already made that expects to get the matrix inverse. (Note: I am looking into ways of changing this, but that will take a long time)
At the moment we are using an LU decomposition method from numerical recopies, and I am currently in the process of testing the eigen library. The eigen library seems to be more stable and a bit faster, but I am still in testing phase for accuracy. I have taken a quick look at other libraries such as ATLAS and LAPACK but have not done any substantial testing with these yet.
It seems as though the eigen library does not use concurrent methods to compute the inverse (though does for LU factorization part of the inverse) and as far as I can tell ATLAS and LAPACK are similar in this limitation. (I am currently testing the speed difference for eigen with openMP and without.)
First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization. I found an article [here](http://www.bhrc.ac.ir/Portal/LinkClick.aspx?fileticket=oLt3rF1LlTo=&tabid=584) that talks about matrix inversion parallel algorithms, but I did not understand. It seems [this](http://world-comp.org/p2011/PDP5100.pdf) article talks about another method? I am also not sure if scaLAPACK or PETSc are useful?
Second question, I read [this](http://www.hpca.uji.es/ficheros/remon/pdp11.pdf) article of using the GPUs to increase performance, but I have never coded for GPUs and so have no idea what is trying to convey, but the charts at the bottom looked rather alarming. How is this even possible, and how where do I start to go about implementing something like this if it is to be true.
I also found [this](http://lara.inist.fr/bitstream/handle/2332/784/RR2002-04.pdf?sequence=1) article, have yet had the time to read through it to understand, but it seems promising, as memory is a current issue with our software.
Any information about these articles or the problems in general would be of great help. And again I apologize if this question seems vague, I will try to expand more if necessary. | 2012/06/27 | [
"https://Stackoverflow.com/questions/11230067",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1444573/"
] | *First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization.*
I'd hazard a guess that this, and related topics in linear algebra, is one of the most studied topics in parallel computing. If you're stuck looking for somewhere to start reading, well good old [Golub and Van Loan](http://www.amazon.co.uk/Computations-Hopkins-Studies-Mathematical-Sciences/dp/0801854148/ref=sr_1_1?s=books&ie=UTF8&qid=1340812109&sr=1-1) have a chapter on the topic. As to whether Scalapack and Petsc are likely to be useful, certainly the former, probably the latter. Of course, they both depend on MPI but that's kind of taken for granted in this field.
*Second question ...*
Use GPUs if you've got them and you can afford to translate your code into the programming model supported by your GPUs. If you've never coded for GPUs and have access to a cluster of commodity-type CPUs you'll get up to speed quicker by using the cluster than by wrestling with a novel technology.
As for the last article you refer to, it's now 10 years old in a field that changes very quickly (try finding a 10-year old research paper on using GPUs for matrix inversion). I can't comment on its excellence or other attributes, but the problem sizes you mention seem to me to be well within the capabilities of modern clusters for in-core (to use an old term) computation. If your matrices are very big, are they also sparse ?
Finally, I strongly support your apparent intention to use existing off-the-shelf codes rather than to try to develop your own. | An LU decomp on a GPU can be ~10x faster than on a CPU. Although this is now changing, GPU's have traditionally been designed around single precision arithmetic, and so on older hardware single precision arithmetic is generally much faster than double precision arithmetic. Also, storage requirements and performance will be greatly impacted by the structure of your matrices. A sparse 100,000 x 100,000 matrix LU decomp is a reasonable problem to solve and will not require much memory.
Unless you want to become a specialist and spend *a lot* of time tuning for hardware updates, I would strongly recommend using a commercial library. I would suggest [CULA tools](http://www.culatools.com/). They have both sparse and dense GPU libraries and in fact their [free library](http://www.culatools.com/cula_dense_programmers_guide/) offers SGETRF - a single precision (dense) LU decomp routine. You'll have to pay for their double precision libraries. |
11,230,067 | I am looking at taking the inverse of a large matrix, common size of 1000 x 1000, but sometimes exceeds 100000 x 100000 (which is currently failing due to time and memory). I know that the normal sentiment is 'don't take the inverse, find some other way to do it', but that is not possible at the moment. The reason for this is due to the usage of software that is already made that expects to get the matrix inverse. (Note: I am looking into ways of changing this, but that will take a long time)
At the moment we are using an LU decomposition method from numerical recopies, and I am currently in the process of testing the eigen library. The eigen library seems to be more stable and a bit faster, but I am still in testing phase for accuracy. I have taken a quick look at other libraries such as ATLAS and LAPACK but have not done any substantial testing with these yet.
It seems as though the eigen library does not use concurrent methods to compute the inverse (though does for LU factorization part of the inverse) and as far as I can tell ATLAS and LAPACK are similar in this limitation. (I am currently testing the speed difference for eigen with openMP and without.)
First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization. I found an article [here](http://www.bhrc.ac.ir/Portal/LinkClick.aspx?fileticket=oLt3rF1LlTo=&tabid=584) that talks about matrix inversion parallel algorithms, but I did not understand. It seems [this](http://world-comp.org/p2011/PDP5100.pdf) article talks about another method? I am also not sure if scaLAPACK or PETSc are useful?
Second question, I read [this](http://www.hpca.uji.es/ficheros/remon/pdp11.pdf) article of using the GPUs to increase performance, but I have never coded for GPUs and so have no idea what is trying to convey, but the charts at the bottom looked rather alarming. How is this even possible, and how where do I start to go about implementing something like this if it is to be true.
I also found [this](http://lara.inist.fr/bitstream/handle/2332/784/RR2002-04.pdf?sequence=1) article, have yet had the time to read through it to understand, but it seems promising, as memory is a current issue with our software.
Any information about these articles or the problems in general would be of great help. And again I apologize if this question seems vague, I will try to expand more if necessary. | 2012/06/27 | [
"https://Stackoverflow.com/questions/11230067",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1444573/"
] | *First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization.*
I'd hazard a guess that this, and related topics in linear algebra, is one of the most studied topics in parallel computing. If you're stuck looking for somewhere to start reading, well good old [Golub and Van Loan](http://www.amazon.co.uk/Computations-Hopkins-Studies-Mathematical-Sciences/dp/0801854148/ref=sr_1_1?s=books&ie=UTF8&qid=1340812109&sr=1-1) have a chapter on the topic. As to whether Scalapack and Petsc are likely to be useful, certainly the former, probably the latter. Of course, they both depend on MPI but that's kind of taken for granted in this field.
*Second question ...*
Use GPUs if you've got them and you can afford to translate your code into the programming model supported by your GPUs. If you've never coded for GPUs and have access to a cluster of commodity-type CPUs you'll get up to speed quicker by using the cluster than by wrestling with a novel technology.
As for the last article you refer to, it's now 10 years old in a field that changes very quickly (try finding a 10-year old research paper on using GPUs for matrix inversion). I can't comment on its excellence or other attributes, but the problem sizes you mention seem to me to be well within the capabilities of modern clusters for in-core (to use an old term) computation. If your matrices are very big, are they also sparse ?
Finally, I strongly support your apparent intention to use existing off-the-shelf codes rather than to try to develop your own. | I know it's old post - but really - OpenCL (you download the relevant one based on your graphics card) + OpenMP + Vectorization (not in that order) is the way to go.
Anyhow - for me my experience with matrix anything is really to do with overheads from copying double double arrays in and out the system and also to pad up or initialize matrices with 0s before any commencement of computation - especially when I am working with creating .xll for Excel usage.
If I were to reprioritize the top -
1. try to vectorize the code (Visual Studio 2012 and Intel C++ has autovectorization - I'm not sure about MinGW or GCC, but I think there are flags for the compiler to analyse your for loops to generate the right assembly codes to use instead of the normal registers to hold your data, to populate your processor's vector registers. I think Excel is doing that because when I monitored Excel's threads while running their MINVERSE(), I notice only 1 thread is used.
I don't know much assembly language - so I don't know how to vectorize manually... (haven't had time to go learn this yet but sooooo wanna do it!)
2. Parallelize with OpenMP (omp pragma) or MPI or pthreads library (parallel\_for) - very simple - but... here's the catch - I realise that if your matrix class is completely single threaded in the first place - then parallelizing the operation like mat multiply or inverse is scrappable - cuz parallelizing will deteriorate the speed due to initializing or copying to or just accessing the non-parallelized matrix class.
But... where parallelization helps is - if you're designing your own matrix class and you parallelize its constructor operation (padding with 0s etc), then your computation of LU(A^-1) = I will also be faster.
It's also mathematically straightforward to also optimize your LU decomposition, and also optimizing ur forward backward substitution for the special case of identity. (I.e. don't waste time creating any identity matrix - analyse where your for (row = col) and evaluate to be a function with 1 and the rest with 0.)
3. Once it's been parallelized (on the outer layers) - the matrix operations requiring element by element can be mapped to be computed by GPU(SSSSSS) - hundreds of processors to compute elements - beat that!. There is now sample Monte Carlo code available on ATI's website - using ATI's OpenCL - don't worry about porting code to something that uses GeForce - all u gotta do is recompile there.
For 2 and 3 though - remember that overheads are incurred so no point unless you're handling F\**K*\*G HUGE matrices - but I see 100k^2? wow...
Gene |
11,230,067 | I am looking at taking the inverse of a large matrix, common size of 1000 x 1000, but sometimes exceeds 100000 x 100000 (which is currently failing due to time and memory). I know that the normal sentiment is 'don't take the inverse, find some other way to do it', but that is not possible at the moment. The reason for this is due to the usage of software that is already made that expects to get the matrix inverse. (Note: I am looking into ways of changing this, but that will take a long time)
At the moment we are using an LU decomposition method from numerical recopies, and I am currently in the process of testing the eigen library. The eigen library seems to be more stable and a bit faster, but I am still in testing phase for accuracy. I have taken a quick look at other libraries such as ATLAS and LAPACK but have not done any substantial testing with these yet.
It seems as though the eigen library does not use concurrent methods to compute the inverse (though does for LU factorization part of the inverse) and as far as I can tell ATLAS and LAPACK are similar in this limitation. (I am currently testing the speed difference for eigen with openMP and without.)
First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization. I found an article [here](http://www.bhrc.ac.ir/Portal/LinkClick.aspx?fileticket=oLt3rF1LlTo=&tabid=584) that talks about matrix inversion parallel algorithms, but I did not understand. It seems [this](http://world-comp.org/p2011/PDP5100.pdf) article talks about another method? I am also not sure if scaLAPACK or PETSc are useful?
Second question, I read [this](http://www.hpca.uji.es/ficheros/remon/pdp11.pdf) article of using the GPUs to increase performance, but I have never coded for GPUs and so have no idea what is trying to convey, but the charts at the bottom looked rather alarming. How is this even possible, and how where do I start to go about implementing something like this if it is to be true.
I also found [this](http://lara.inist.fr/bitstream/handle/2332/784/RR2002-04.pdf?sequence=1) article, have yet had the time to read through it to understand, but it seems promising, as memory is a current issue with our software.
Any information about these articles or the problems in general would be of great help. And again I apologize if this question seems vague, I will try to expand more if necessary. | 2012/06/27 | [
"https://Stackoverflow.com/questions/11230067",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1444573/"
] | Regarding the first question (how to parallellize computing the inverse):
I assume you are computing the inverse by doing an LU decomposition of your matrix and then using the decomposition to solve A\*B = I where A is your original matrix, B is the matrix you solve for, and I is the identity matrix. Then B is the inverse.
The last step is easy to parallellize. Divide your identity matrix along the columns. If you have p CPUs and your matrix is n-by-n, then every part has n/p columns and n rows. Lets call the parts I1, I2, etc. On every CPU, solve a system of the form A\*B1 = I1, this gives you the parts B1, B2, etc., and you can combine them to form B which is the inverse. | An LU decomp on a GPU can be ~10x faster than on a CPU. Although this is now changing, GPU's have traditionally been designed around single precision arithmetic, and so on older hardware single precision arithmetic is generally much faster than double precision arithmetic. Also, storage requirements and performance will be greatly impacted by the structure of your matrices. A sparse 100,000 x 100,000 matrix LU decomp is a reasonable problem to solve and will not require much memory.
Unless you want to become a specialist and spend *a lot* of time tuning for hardware updates, I would strongly recommend using a commercial library. I would suggest [CULA tools](http://www.culatools.com/). They have both sparse and dense GPU libraries and in fact their [free library](http://www.culatools.com/cula_dense_programmers_guide/) offers SGETRF - a single precision (dense) LU decomp routine. You'll have to pay for their double precision libraries. |
11,230,067 | I am looking at taking the inverse of a large matrix, common size of 1000 x 1000, but sometimes exceeds 100000 x 100000 (which is currently failing due to time and memory). I know that the normal sentiment is 'don't take the inverse, find some other way to do it', but that is not possible at the moment. The reason for this is due to the usage of software that is already made that expects to get the matrix inverse. (Note: I am looking into ways of changing this, but that will take a long time)
At the moment we are using an LU decomposition method from numerical recopies, and I am currently in the process of testing the eigen library. The eigen library seems to be more stable and a bit faster, but I am still in testing phase for accuracy. I have taken a quick look at other libraries such as ATLAS and LAPACK but have not done any substantial testing with these yet.
It seems as though the eigen library does not use concurrent methods to compute the inverse (though does for LU factorization part of the inverse) and as far as I can tell ATLAS and LAPACK are similar in this limitation. (I am currently testing the speed difference for eigen with openMP and without.)
First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization. I found an article [here](http://www.bhrc.ac.ir/Portal/LinkClick.aspx?fileticket=oLt3rF1LlTo=&tabid=584) that talks about matrix inversion parallel algorithms, but I did not understand. It seems [this](http://world-comp.org/p2011/PDP5100.pdf) article talks about another method? I am also not sure if scaLAPACK or PETSc are useful?
Second question, I read [this](http://www.hpca.uji.es/ficheros/remon/pdp11.pdf) article of using the GPUs to increase performance, but I have never coded for GPUs and so have no idea what is trying to convey, but the charts at the bottom looked rather alarming. How is this even possible, and how where do I start to go about implementing something like this if it is to be true.
I also found [this](http://lara.inist.fr/bitstream/handle/2332/784/RR2002-04.pdf?sequence=1) article, have yet had the time to read through it to understand, but it seems promising, as memory is a current issue with our software.
Any information about these articles or the problems in general would be of great help. And again I apologize if this question seems vague, I will try to expand more if necessary. | 2012/06/27 | [
"https://Stackoverflow.com/questions/11230067",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1444573/"
] | Regarding the first question (how to parallellize computing the inverse):
I assume you are computing the inverse by doing an LU decomposition of your matrix and then using the decomposition to solve A\*B = I where A is your original matrix, B is the matrix you solve for, and I is the identity matrix. Then B is the inverse.
The last step is easy to parallellize. Divide your identity matrix along the columns. If you have p CPUs and your matrix is n-by-n, then every part has n/p columns and n rows. Lets call the parts I1, I2, etc. On every CPU, solve a system of the form A\*B1 = I1, this gives you the parts B1, B2, etc., and you can combine them to form B which is the inverse. | I know it's old post - but really - OpenCL (you download the relevant one based on your graphics card) + OpenMP + Vectorization (not in that order) is the way to go.
Anyhow - for me my experience with matrix anything is really to do with overheads from copying double double arrays in and out the system and also to pad up or initialize matrices with 0s before any commencement of computation - especially when I am working with creating .xll for Excel usage.
If I were to reprioritize the top -
1. try to vectorize the code (Visual Studio 2012 and Intel C++ has autovectorization - I'm not sure about MinGW or GCC, but I think there are flags for the compiler to analyse your for loops to generate the right assembly codes to use instead of the normal registers to hold your data, to populate your processor's vector registers. I think Excel is doing that because when I monitored Excel's threads while running their MINVERSE(), I notice only 1 thread is used.
I don't know much assembly language - so I don't know how to vectorize manually... (haven't had time to go learn this yet but sooooo wanna do it!)
2. Parallelize with OpenMP (omp pragma) or MPI or pthreads library (parallel\_for) - very simple - but... here's the catch - I realise that if your matrix class is completely single threaded in the first place - then parallelizing the operation like mat multiply or inverse is scrappable - cuz parallelizing will deteriorate the speed due to initializing or copying to or just accessing the non-parallelized matrix class.
But... where parallelization helps is - if you're designing your own matrix class and you parallelize its constructor operation (padding with 0s etc), then your computation of LU(A^-1) = I will also be faster.
It's also mathematically straightforward to also optimize your LU decomposition, and also optimizing ur forward backward substitution for the special case of identity. (I.e. don't waste time creating any identity matrix - analyse where your for (row = col) and evaluate to be a function with 1 and the rest with 0.)
3. Once it's been parallelized (on the outer layers) - the matrix operations requiring element by element can be mapped to be computed by GPU(SSSSSS) - hundreds of processors to compute elements - beat that!. There is now sample Monte Carlo code available on ATI's website - using ATI's OpenCL - don't worry about porting code to something that uses GeForce - all u gotta do is recompile there.
For 2 and 3 though - remember that overheads are incurred so no point unless you're handling F\**K*\*G HUGE matrices - but I see 100k^2? wow...
Gene |
11,230,067 | I am looking at taking the inverse of a large matrix, common size of 1000 x 1000, but sometimes exceeds 100000 x 100000 (which is currently failing due to time and memory). I know that the normal sentiment is 'don't take the inverse, find some other way to do it', but that is not possible at the moment. The reason for this is due to the usage of software that is already made that expects to get the matrix inverse. (Note: I am looking into ways of changing this, but that will take a long time)
At the moment we are using an LU decomposition method from numerical recopies, and I am currently in the process of testing the eigen library. The eigen library seems to be more stable and a bit faster, but I am still in testing phase for accuracy. I have taken a quick look at other libraries such as ATLAS and LAPACK but have not done any substantial testing with these yet.
It seems as though the eigen library does not use concurrent methods to compute the inverse (though does for LU factorization part of the inverse) and as far as I can tell ATLAS and LAPACK are similar in this limitation. (I am currently testing the speed difference for eigen with openMP and without.)
First question is can anyone explain how it would be possible to optimize matrix inversion by parallelization. I found an article [here](http://www.bhrc.ac.ir/Portal/LinkClick.aspx?fileticket=oLt3rF1LlTo=&tabid=584) that talks about matrix inversion parallel algorithms, but I did not understand. It seems [this](http://world-comp.org/p2011/PDP5100.pdf) article talks about another method? I am also not sure if scaLAPACK or PETSc are useful?
Second question, I read [this](http://www.hpca.uji.es/ficheros/remon/pdp11.pdf) article of using the GPUs to increase performance, but I have never coded for GPUs and so have no idea what is trying to convey, but the charts at the bottom looked rather alarming. How is this even possible, and how where do I start to go about implementing something like this if it is to be true.
I also found [this](http://lara.inist.fr/bitstream/handle/2332/784/RR2002-04.pdf?sequence=1) article, have yet had the time to read through it to understand, but it seems promising, as memory is a current issue with our software.
Any information about these articles or the problems in general would be of great help. And again I apologize if this question seems vague, I will try to expand more if necessary. | 2012/06/27 | [
"https://Stackoverflow.com/questions/11230067",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1444573/"
] | An LU decomp on a GPU can be ~10x faster than on a CPU. Although this is now changing, GPU's have traditionally been designed around single precision arithmetic, and so on older hardware single precision arithmetic is generally much faster than double precision arithmetic. Also, storage requirements and performance will be greatly impacted by the structure of your matrices. A sparse 100,000 x 100,000 matrix LU decomp is a reasonable problem to solve and will not require much memory.
Unless you want to become a specialist and spend *a lot* of time tuning for hardware updates, I would strongly recommend using a commercial library. I would suggest [CULA tools](http://www.culatools.com/). They have both sparse and dense GPU libraries and in fact their [free library](http://www.culatools.com/cula_dense_programmers_guide/) offers SGETRF - a single precision (dense) LU decomp routine. You'll have to pay for their double precision libraries. | I know it's old post - but really - OpenCL (you download the relevant one based on your graphics card) + OpenMP + Vectorization (not in that order) is the way to go.
Anyhow - for me my experience with matrix anything is really to do with overheads from copying double double arrays in and out the system and also to pad up or initialize matrices with 0s before any commencement of computation - especially when I am working with creating .xll for Excel usage.
If I were to reprioritize the top -
1. try to vectorize the code (Visual Studio 2012 and Intel C++ has autovectorization - I'm not sure about MinGW or GCC, but I think there are flags for the compiler to analyse your for loops to generate the right assembly codes to use instead of the normal registers to hold your data, to populate your processor's vector registers. I think Excel is doing that because when I monitored Excel's threads while running their MINVERSE(), I notice only 1 thread is used.
I don't know much assembly language - so I don't know how to vectorize manually... (haven't had time to go learn this yet but sooooo wanna do it!)
2. Parallelize with OpenMP (omp pragma) or MPI or pthreads library (parallel\_for) - very simple - but... here's the catch - I realise that if your matrix class is completely single threaded in the first place - then parallelizing the operation like mat multiply or inverse is scrappable - cuz parallelizing will deteriorate the speed due to initializing or copying to or just accessing the non-parallelized matrix class.
But... where parallelization helps is - if you're designing your own matrix class and you parallelize its constructor operation (padding with 0s etc), then your computation of LU(A^-1) = I will also be faster.
It's also mathematically straightforward to also optimize your LU decomposition, and also optimizing ur forward backward substitution for the special case of identity. (I.e. don't waste time creating any identity matrix - analyse where your for (row = col) and evaluate to be a function with 1 and the rest with 0.)
3. Once it's been parallelized (on the outer layers) - the matrix operations requiring element by element can be mapped to be computed by GPU(SSSSSS) - hundreds of processors to compute elements - beat that!. There is now sample Monte Carlo code available on ATI's website - using ATI's OpenCL - don't worry about porting code to something that uses GeForce - all u gotta do is recompile there.
For 2 and 3 though - remember that overheads are incurred so no point unless you're handling F\**K*\*G HUGE matrices - but I see 100k^2? wow...
Gene |
198,523 | This weekend I placed an order for a part from an online retailer ($300-400). Shortly thereafter I received an email from their sales team saying something similar to the following:
>
> Hi, we have received your order #12345 and it has been triggered by
> our fraud alert system. Please send us a copy of your drivers license
> and credit card used in the transaction. You my black out any personal
> details like the DL# but please leave the last 4 digits of the credit
> card, name, and address.
>
>
> This is no different than presenting your ID in a brick and mortar
> store. Thank you for your cooperation, blah blah blah.
>
>
>
My response to them was basically:
>
> Hi, I would be happy to perform any verification through the bank
> issuing the card but unfortunately do not feel comfortable sending
> copies of my DL and credit card via email to an online retailer.
>
>
>
Their response basically reiterated their first email, but insisting that they already had all this information anyways and just needed to verify it. So I ended up cancelling the order and buying the same part from a different retailer. They are indeed legitimate retailers. Their name bounces around the forums without any issues and I found them, they didn't find me.
My question is, **did I make an issue out of something that was reasonable for me to comply with?**
A few points that crossed my mind:
1) I've had "fraud alerts" on my credit card before when making a larger than usual purchase. They come in the form of an email from my bank asking me to confirm whether I made a given purchase or not. Sometimes they even require that I call in to verify. This "fraud alert" was only from the retailer. My bank did not raise this issue.
2) As far as I'm concerned, I should never take a picture of my credit card. Even if I block out the important details, the original photo probably got synced to the cloud which means I have to go make sure it was deleted everywhere and trust the our google/apple/amazon overlords did indeed delete it. I'm sure there's ways to mitigate this, but it every hoop I jump through is an opportunity for mistakes.
3) If I was a fraudster, I'm pretty sure I could fake a credit card and corresponding ID via Photoshop that I could take a crappy photo of with my phone and satisfy them. I don't think there's much security in their process. | 2018/11/27 | [
"https://security.stackexchange.com/questions/198523",
"https://security.stackexchange.com",
"https://security.stackexchange.com/users/185906/"
] | This is rational response. Better be safe than sorry.
Any goods purchase that contains the delivery address (software purchase may not have the info) that similar to your credit card is verifiable by the issuer.
Point 1 is not always true because it depends on the level of escalation, i.e. the middleman like Visa/Master/AMEX may have a delayed clearance issues with the bank. So the logical way to deal with this is calling the bank support.
Point 2 is valid, you should never send a copy of your credit card/ID to through email. At minimum, the merchant should at least provide an encrypted web interface to let you upload it. If they cannot do it due to cost and commission issue (some payment gateways are pretty expensive), then they should ask you to contact your bank than performing such shady process, even it is legitimate request. | Yeah this sounds really sketchy to me. Good call on trusting your gut here. Fraud detection like that would obviously usually happen on the bank side. If for some reason the retailer were doing heavy fraud detection (which, for a store selling items of presumably smaller value, would surprise me), then asking for a copy of your identification without telling you what triggered the fraud alert smells weird. It may have been legit - after all, they did instruct you to obfuscate the valuable parts of the image - but not giving details is weird. |
198,523 | This weekend I placed an order for a part from an online retailer ($300-400). Shortly thereafter I received an email from their sales team saying something similar to the following:
>
> Hi, we have received your order #12345 and it has been triggered by
> our fraud alert system. Please send us a copy of your drivers license
> and credit card used in the transaction. You my black out any personal
> details like the DL# but please leave the last 4 digits of the credit
> card, name, and address.
>
>
> This is no different than presenting your ID in a brick and mortar
> store. Thank you for your cooperation, blah blah blah.
>
>
>
My response to them was basically:
>
> Hi, I would be happy to perform any verification through the bank
> issuing the card but unfortunately do not feel comfortable sending
> copies of my DL and credit card via email to an online retailer.
>
>
>
Their response basically reiterated their first email, but insisting that they already had all this information anyways and just needed to verify it. So I ended up cancelling the order and buying the same part from a different retailer. They are indeed legitimate retailers. Their name bounces around the forums without any issues and I found them, they didn't find me.
My question is, **did I make an issue out of something that was reasonable for me to comply with?**
A few points that crossed my mind:
1) I've had "fraud alerts" on my credit card before when making a larger than usual purchase. They come in the form of an email from my bank asking me to confirm whether I made a given purchase or not. Sometimes they even require that I call in to verify. This "fraud alert" was only from the retailer. My bank did not raise this issue.
2) As far as I'm concerned, I should never take a picture of my credit card. Even if I block out the important details, the original photo probably got synced to the cloud which means I have to go make sure it was deleted everywhere and trust the our google/apple/amazon overlords did indeed delete it. I'm sure there's ways to mitigate this, but it every hoop I jump through is an opportunity for mistakes.
3) If I was a fraudster, I'm pretty sure I could fake a credit card and corresponding ID via Photoshop that I could take a crappy photo of with my phone and satisfy them. I don't think there's much security in their process. | 2018/11/27 | [
"https://security.stackexchange.com/questions/198523",
"https://security.stackexchange.com",
"https://security.stackexchange.com/users/185906/"
] | Those responding here, probably haven't sold anything online and have no idea of the amount of credit card fraud retailers have to deal with.
It is very common to have some fraudster use someone else's credit card to make purchases. Merchants have zero protection in these cases and not only pay for the charge processed as well as a dispute fee. Also, their credit card processor will blacklist them and basically throw them out of business if the dispute rate increases.
So, almost any retailer that processes $1M+ in volume a month will use some kind fraud early warning software. There are plenty in market and they are making a killing as SaaS companies. Once a transaction is triggered as potential fraud, the merchant will attempt to verify via multiple methods.
One of the most common methods employed, is asking for Driving Licence/ID and a copy of credit card with numbers hidden. There is almost nothing anyone can do with this piece of info. You are giving away, nothing confidential. It just proves that you are in possession of the card and your name is on the card. If the charge is disputed, these images can be used to prove that the merchant took an effort to verify the transaction and shift liability. This method is employed by many famous companies: AirBnB, Agoda/Priceline, Lyft used it at one point.
It is not foolproof e.g. someone stealing your wallet with the ID can also complete this verification. Many merchants are willing to take that risk. Wish there was a better way. | This is rational response. Better be safe than sorry.
Any goods purchase that contains the delivery address (software purchase may not have the info) that similar to your credit card is verifiable by the issuer.
Point 1 is not always true because it depends on the level of escalation, i.e. the middleman like Visa/Master/AMEX may have a delayed clearance issues with the bank. So the logical way to deal with this is calling the bank support.
Point 2 is valid, you should never send a copy of your credit card/ID to through email. At minimum, the merchant should at least provide an encrypted web interface to let you upload it. If they cannot do it due to cost and commission issue (some payment gateways are pretty expensive), then they should ask you to contact your bank than performing such shady process, even it is legitimate request. |
198,523 | This weekend I placed an order for a part from an online retailer ($300-400). Shortly thereafter I received an email from their sales team saying something similar to the following:
>
> Hi, we have received your order #12345 and it has been triggered by
> our fraud alert system. Please send us a copy of your drivers license
> and credit card used in the transaction. You my black out any personal
> details like the DL# but please leave the last 4 digits of the credit
> card, name, and address.
>
>
> This is no different than presenting your ID in a brick and mortar
> store. Thank you for your cooperation, blah blah blah.
>
>
>
My response to them was basically:
>
> Hi, I would be happy to perform any verification through the bank
> issuing the card but unfortunately do not feel comfortable sending
> copies of my DL and credit card via email to an online retailer.
>
>
>
Their response basically reiterated their first email, but insisting that they already had all this information anyways and just needed to verify it. So I ended up cancelling the order and buying the same part from a different retailer. They are indeed legitimate retailers. Their name bounces around the forums without any issues and I found them, they didn't find me.
My question is, **did I make an issue out of something that was reasonable for me to comply with?**
A few points that crossed my mind:
1) I've had "fraud alerts" on my credit card before when making a larger than usual purchase. They come in the form of an email from my bank asking me to confirm whether I made a given purchase or not. Sometimes they even require that I call in to verify. This "fraud alert" was only from the retailer. My bank did not raise this issue.
2) As far as I'm concerned, I should never take a picture of my credit card. Even if I block out the important details, the original photo probably got synced to the cloud which means I have to go make sure it was deleted everywhere and trust the our google/apple/amazon overlords did indeed delete it. I'm sure there's ways to mitigate this, but it every hoop I jump through is an opportunity for mistakes.
3) If I was a fraudster, I'm pretty sure I could fake a credit card and corresponding ID via Photoshop that I could take a crappy photo of with my phone and satisfy them. I don't think there's much security in their process. | 2018/11/27 | [
"https://security.stackexchange.com/questions/198523",
"https://security.stackexchange.com",
"https://security.stackexchange.com/users/185906/"
] | As someone who owns a small online business this is something that we are told to do by banks and our payment processors to prevent fraud. If someone is using a stolen card and you report it, it's not the bank refunding you, it's the business. The business loses the product and the money and often even get a chargeback.
It can be very damaging to businesses and so we need to take all the precautions necessary. If you are really concerned you can send a photo with your finger or a piece of paper blocking the personal info so it can't be stolen.
All we really need is just to see that the names match. | This is rational response. Better be safe than sorry.
Any goods purchase that contains the delivery address (software purchase may not have the info) that similar to your credit card is verifiable by the issuer.
Point 1 is not always true because it depends on the level of escalation, i.e. the middleman like Visa/Master/AMEX may have a delayed clearance issues with the bank. So the logical way to deal with this is calling the bank support.
Point 2 is valid, you should never send a copy of your credit card/ID to through email. At minimum, the merchant should at least provide an encrypted web interface to let you upload it. If they cannot do it due to cost and commission issue (some payment gateways are pretty expensive), then they should ask you to contact your bank than performing such shady process, even it is legitimate request. |
198,523 | This weekend I placed an order for a part from an online retailer ($300-400). Shortly thereafter I received an email from their sales team saying something similar to the following:
>
> Hi, we have received your order #12345 and it has been triggered by
> our fraud alert system. Please send us a copy of your drivers license
> and credit card used in the transaction. You my black out any personal
> details like the DL# but please leave the last 4 digits of the credit
> card, name, and address.
>
>
> This is no different than presenting your ID in a brick and mortar
> store. Thank you for your cooperation, blah blah blah.
>
>
>
My response to them was basically:
>
> Hi, I would be happy to perform any verification through the bank
> issuing the card but unfortunately do not feel comfortable sending
> copies of my DL and credit card via email to an online retailer.
>
>
>
Their response basically reiterated their first email, but insisting that they already had all this information anyways and just needed to verify it. So I ended up cancelling the order and buying the same part from a different retailer. They are indeed legitimate retailers. Their name bounces around the forums without any issues and I found them, they didn't find me.
My question is, **did I make an issue out of something that was reasonable for me to comply with?**
A few points that crossed my mind:
1) I've had "fraud alerts" on my credit card before when making a larger than usual purchase. They come in the form of an email from my bank asking me to confirm whether I made a given purchase or not. Sometimes they even require that I call in to verify. This "fraud alert" was only from the retailer. My bank did not raise this issue.
2) As far as I'm concerned, I should never take a picture of my credit card. Even if I block out the important details, the original photo probably got synced to the cloud which means I have to go make sure it was deleted everywhere and trust the our google/apple/amazon overlords did indeed delete it. I'm sure there's ways to mitigate this, but it every hoop I jump through is an opportunity for mistakes.
3) If I was a fraudster, I'm pretty sure I could fake a credit card and corresponding ID via Photoshop that I could take a crappy photo of with my phone and satisfy them. I don't think there's much security in their process. | 2018/11/27 | [
"https://security.stackexchange.com/questions/198523",
"https://security.stackexchange.com",
"https://security.stackexchange.com/users/185906/"
] | Those responding here, probably haven't sold anything online and have no idea of the amount of credit card fraud retailers have to deal with.
It is very common to have some fraudster use someone else's credit card to make purchases. Merchants have zero protection in these cases and not only pay for the charge processed as well as a dispute fee. Also, their credit card processor will blacklist them and basically throw them out of business if the dispute rate increases.
So, almost any retailer that processes $1M+ in volume a month will use some kind fraud early warning software. There are plenty in market and they are making a killing as SaaS companies. Once a transaction is triggered as potential fraud, the merchant will attempt to verify via multiple methods.
One of the most common methods employed, is asking for Driving Licence/ID and a copy of credit card with numbers hidden. There is almost nothing anyone can do with this piece of info. You are giving away, nothing confidential. It just proves that you are in possession of the card and your name is on the card. If the charge is disputed, these images can be used to prove that the merchant took an effort to verify the transaction and shift liability. This method is employed by many famous companies: AirBnB, Agoda/Priceline, Lyft used it at one point.
It is not foolproof e.g. someone stealing your wallet with the ID can also complete this verification. Many merchants are willing to take that risk. Wish there was a better way. | Yeah this sounds really sketchy to me. Good call on trusting your gut here. Fraud detection like that would obviously usually happen on the bank side. If for some reason the retailer were doing heavy fraud detection (which, for a store selling items of presumably smaller value, would surprise me), then asking for a copy of your identification without telling you what triggered the fraud alert smells weird. It may have been legit - after all, they did instruct you to obfuscate the valuable parts of the image - but not giving details is weird. |
198,523 | This weekend I placed an order for a part from an online retailer ($300-400). Shortly thereafter I received an email from their sales team saying something similar to the following:
>
> Hi, we have received your order #12345 and it has been triggered by
> our fraud alert system. Please send us a copy of your drivers license
> and credit card used in the transaction. You my black out any personal
> details like the DL# but please leave the last 4 digits of the credit
> card, name, and address.
>
>
> This is no different than presenting your ID in a brick and mortar
> store. Thank you for your cooperation, blah blah blah.
>
>
>
My response to them was basically:
>
> Hi, I would be happy to perform any verification through the bank
> issuing the card but unfortunately do not feel comfortable sending
> copies of my DL and credit card via email to an online retailer.
>
>
>
Their response basically reiterated their first email, but insisting that they already had all this information anyways and just needed to verify it. So I ended up cancelling the order and buying the same part from a different retailer. They are indeed legitimate retailers. Their name bounces around the forums without any issues and I found them, they didn't find me.
My question is, **did I make an issue out of something that was reasonable for me to comply with?**
A few points that crossed my mind:
1) I've had "fraud alerts" on my credit card before when making a larger than usual purchase. They come in the form of an email from my bank asking me to confirm whether I made a given purchase or not. Sometimes they even require that I call in to verify. This "fraud alert" was only from the retailer. My bank did not raise this issue.
2) As far as I'm concerned, I should never take a picture of my credit card. Even if I block out the important details, the original photo probably got synced to the cloud which means I have to go make sure it was deleted everywhere and trust the our google/apple/amazon overlords did indeed delete it. I'm sure there's ways to mitigate this, but it every hoop I jump through is an opportunity for mistakes.
3) If I was a fraudster, I'm pretty sure I could fake a credit card and corresponding ID via Photoshop that I could take a crappy photo of with my phone and satisfy them. I don't think there's much security in their process. | 2018/11/27 | [
"https://security.stackexchange.com/questions/198523",
"https://security.stackexchange.com",
"https://security.stackexchange.com/users/185906/"
] | As someone who owns a small online business this is something that we are told to do by banks and our payment processors to prevent fraud. If someone is using a stolen card and you report it, it's not the bank refunding you, it's the business. The business loses the product and the money and often even get a chargeback.
It can be very damaging to businesses and so we need to take all the precautions necessary. If you are really concerned you can send a photo with your finger or a piece of paper blocking the personal info so it can't be stolen.
All we really need is just to see that the names match. | Yeah this sounds really sketchy to me. Good call on trusting your gut here. Fraud detection like that would obviously usually happen on the bank side. If for some reason the retailer were doing heavy fraud detection (which, for a store selling items of presumably smaller value, would surprise me), then asking for a copy of your identification without telling you what triggered the fraud alert smells weird. It may have been legit - after all, they did instruct you to obfuscate the valuable parts of the image - but not giving details is weird. |
198,523 | This weekend I placed an order for a part from an online retailer ($300-400). Shortly thereafter I received an email from their sales team saying something similar to the following:
>
> Hi, we have received your order #12345 and it has been triggered by
> our fraud alert system. Please send us a copy of your drivers license
> and credit card used in the transaction. You my black out any personal
> details like the DL# but please leave the last 4 digits of the credit
> card, name, and address.
>
>
> This is no different than presenting your ID in a brick and mortar
> store. Thank you for your cooperation, blah blah blah.
>
>
>
My response to them was basically:
>
> Hi, I would be happy to perform any verification through the bank
> issuing the card but unfortunately do not feel comfortable sending
> copies of my DL and credit card via email to an online retailer.
>
>
>
Their response basically reiterated their first email, but insisting that they already had all this information anyways and just needed to verify it. So I ended up cancelling the order and buying the same part from a different retailer. They are indeed legitimate retailers. Their name bounces around the forums without any issues and I found them, they didn't find me.
My question is, **did I make an issue out of something that was reasonable for me to comply with?**
A few points that crossed my mind:
1) I've had "fraud alerts" on my credit card before when making a larger than usual purchase. They come in the form of an email from my bank asking me to confirm whether I made a given purchase or not. Sometimes they even require that I call in to verify. This "fraud alert" was only from the retailer. My bank did not raise this issue.
2) As far as I'm concerned, I should never take a picture of my credit card. Even if I block out the important details, the original photo probably got synced to the cloud which means I have to go make sure it was deleted everywhere and trust the our google/apple/amazon overlords did indeed delete it. I'm sure there's ways to mitigate this, but it every hoop I jump through is an opportunity for mistakes.
3) If I was a fraudster, I'm pretty sure I could fake a credit card and corresponding ID via Photoshop that I could take a crappy photo of with my phone and satisfy them. I don't think there's much security in their process. | 2018/11/27 | [
"https://security.stackexchange.com/questions/198523",
"https://security.stackexchange.com",
"https://security.stackexchange.com/users/185906/"
] | Those responding here, probably haven't sold anything online and have no idea of the amount of credit card fraud retailers have to deal with.
It is very common to have some fraudster use someone else's credit card to make purchases. Merchants have zero protection in these cases and not only pay for the charge processed as well as a dispute fee. Also, their credit card processor will blacklist them and basically throw them out of business if the dispute rate increases.
So, almost any retailer that processes $1M+ in volume a month will use some kind fraud early warning software. There are plenty in market and they are making a killing as SaaS companies. Once a transaction is triggered as potential fraud, the merchant will attempt to verify via multiple methods.
One of the most common methods employed, is asking for Driving Licence/ID and a copy of credit card with numbers hidden. There is almost nothing anyone can do with this piece of info. You are giving away, nothing confidential. It just proves that you are in possession of the card and your name is on the card. If the charge is disputed, these images can be used to prove that the merchant took an effort to verify the transaction and shift liability. This method is employed by many famous companies: AirBnB, Agoda/Priceline, Lyft used it at one point.
It is not foolproof e.g. someone stealing your wallet with the ID can also complete this verification. Many merchants are willing to take that risk. Wish there was a better way. | As someone who owns a small online business this is something that we are told to do by banks and our payment processors to prevent fraud. If someone is using a stolen card and you report it, it's not the bank refunding you, it's the business. The business loses the product and the money and often even get a chargeback.
It can be very damaging to businesses and so we need to take all the precautions necessary. If you are really concerned you can send a photo with your finger or a piece of paper blocking the personal info so it can't be stolen.
All we really need is just to see that the names match. |
138,536 | A PC in my campaign that is tailing an NPC. She is basically posing as one of his body guards.
At some point I want him to suspect that maybe something is up, and I'm thinking she would either roll a deception check or a stealth check. I just don't know which. | 2019/01/05 | [
"https://rpg.stackexchange.com/questions/138536",
"https://rpg.stackexchange.com",
"https://rpg.stackexchange.com/users/51209/"
] | ### Posing as someone else typically uses Deception checks.
In the situation you describe, the PC is posing as a bodyguard and doesn't want to be exposed. Whether this means maintaining a disguise, or hiding her motives, or keeping up a convincing guard persona, the [Charisma (Deception)](https://roll20.net/compendium/dnd5e/Ability%20Scores#toc_33) skill is probably most appropriate.
>
> **Deception**. Your Charisma (Deception) check determines whether you can convincingly hide the truth, either verbally or through your actions. This Deception can encompass everything from misleading others through ambiguity to telling outright lies. Typical situations include trying to fast- talk a guard, con a merchant, earn money through gambling, pass yourself off in a disguise, dull someone’s suspicions with false assurances, or maintain a straight face while telling a blatant lie.
>
>
>
To determine the PC's success at posing as a guard while the NPC is observing her, you would probably have the PC and NPC roll contested skill checks. Typically the NPC would roll a Wisdom (Insight) check versus the PC's Charisma (Deception) check, although as the DM you have the liberty to choose which skill the NPC rolls, depending on the circumstances. | It's a fine line, but it comes down to "Is the player trying to be unseen right now?" or "Is the player trying to fool someone into thinking they're a guard right now?"
If the player's trying to sneak into the guard detail unseen, it's a stealth check against perception. If she is showing up and saluting and pretending to be a guard, it's a deception check against insight. If the NPC suspects something, the player can attempt to hide (stealth), or convince the NPC she's really a guard (deception)
Also keep in mind the difference between passive stats and checks.
-The guards will be actively looking for trouble, but the NPC will probably be distracted with business or his own thoughts and only passively paying attention.
-A guard who is keeping file and chatting with his neighbor may not actively pay attention to how she's marching or who she claims to be, but guard captain who's suspicious of her will be actively looking for incorrect information when she talks and studying her appearance. |
461,867 | I thought about posting this in woodworking, but everything there seems to be related to actual problems in carpentry.
I'm trying to describe a kitten’s behaviour: **Frisky's so desperate for an opponent to wrestle with, even the [wall corner] will do.** I want to know if there's a technical term for this kind of corner. I already tried searching here and also googling ***convex corner***, ***corner facing out***, and ***wall corner facing out***, all without finding the sought after term.
Here's a picture of the actual thing:
[](https://i.stack.imgur.com/qHfJU.jpg) | 2018/08/26 | [
"https://english.stackexchange.com/questions/461867",
"https://english.stackexchange.com",
"https://english.stackexchange.com/users/167518/"
] | There is a word *quoin* or *coign*, which were originally variants of *coin* but relate to corner. [Per the OED](http://www.oed.com/view/Entry/156875?rskey=8OELwH&result=1&isAdvanced=false#eid):
>
> 1. *Building*.
>
>
> * a. Originally: an external angle of a wall; an outer corner of a building. Subsequently also: any of the stones or bricks serving to form this angle; a cornerstone. Cf. *coin* *n*. 1.
>
>
>
> >
> > Esp. used of cornerstones of a material or decoration differing from that of the rest of the wall.
> >
> >
> >
> * b. An internal angle or corner of a wall, esp. of a room. Also: a recess in a wall, as for a gate; = *hollow quoin* n. at *hollow* adj. and adv. Special uses 3b.
>
>
>
The *coign* version is less often seen these days. The OED says that when it means the same as *quoin*, then *quoin* is the ordinary modern spelling.
Various other specialized senses of *quoin* exist, but this is the oldest of them all. | Angular - having angles or sharp corners <https://dictionary.cambridge.org/dictionary/english/angular> |
281,766 | Having read a style guide recently, I was under the impression that you don't add commas after the ellipsis when the ellipsis indications pauses or thoughts trailing off, followed by "he says" or the like.
E.g.
"If only I had known..." she whispered.
"If I'd known, I would have never..." he says, his voice going quiet.
However, I've recently seen several novels which stick a comma after the ellipsis.
E.g.
"If only I had known...," she whispered.
"If I'd known, I would have never...," he says, his voice going quiet.
Does anyone have an opinion about which is correct?
I also have a question about spaces before the ellipsis. I see books with and without the space and have heard different opinions on the subject.
Is this correct:
"If only I had known..."
Or should it have been:
"If only I had known ..."
Thank you for any advice. | 2015/10/22 | [
"https://english.stackexchange.com/questions/281766",
"https://english.stackexchange.com",
"https://english.stackexchange.com/users/139887/"
] | Punctuation is a matter of style, and you should be guided by your manual of style. There isn't any absolute right or wrong. I use *The Chicago Manual of Style*, which recommends using the comma to separate the speech from the speaker. It also recommends space (3-to-em) around each of the points used in *quoted* text, but the applies that typography to its examples of faltering speech. | It is a matter of personal preference. Nothing's written in stone. Styles change, and those who change them are called humans.
That said, an ellipsis immediately followed by a comma looks silly. |
281,766 | Having read a style guide recently, I was under the impression that you don't add commas after the ellipsis when the ellipsis indications pauses or thoughts trailing off, followed by "he says" or the like.
E.g.
"If only I had known..." she whispered.
"If I'd known, I would have never..." he says, his voice going quiet.
However, I've recently seen several novels which stick a comma after the ellipsis.
E.g.
"If only I had known...," she whispered.
"If I'd known, I would have never...," he says, his voice going quiet.
Does anyone have an opinion about which is correct?
I also have a question about spaces before the ellipsis. I see books with and without the space and have heard different opinions on the subject.
Is this correct:
"If only I had known..."
Or should it have been:
"If only I had known ..."
Thank you for any advice. | 2015/10/22 | [
"https://english.stackexchange.com/questions/281766",
"https://english.stackexchange.com",
"https://english.stackexchange.com/users/139887/"
] | The ellipsis **is** punctuation, representing a kind of pause, so it's redundant to add another punctuation mark. Think of other similar situations, such as where quoted speech ends with a **?** or **!**: a comma after either of those would be out of place. | It is a matter of personal preference. Nothing's written in stone. Styles change, and those who change them are called humans.
That said, an ellipsis immediately followed by a comma looks silly. |
1,004 | I came here to seek truth and I assume others also look for the same, which means I should expect beliefs against mine and rejection of my basics in religion. All good, but I've seen recently where some posts link to sites in which explicitly hate-speech and insulting to other beliefs is common.
In the interest of providing particulars, the incident which prompted this discussion was seeing an answer reference material from <http://shia.bs> (which is clearly anti-Shi'a). I don't want to link to that answer since I'm not interested in calling it (or the author who posted it) out in particular, I just want a general discussion on how and when such sites should be used, if at all.
For me this is very disappointing since this goes against rational reasoning and Q&A norms. And we are helping such sites spread hate and hurt the unity and peace between Muslims, which is for sure tricks of our enemies.
Should we forbid such sites? | 2014/04/15 | [
"https://islam.meta.stackexchange.com/questions/1004",
"https://islam.meta.stackexchange.com",
"https://islam.meta.stackexchange.com/users/-1/"
] | First, we should recognize this is in the gray area between censorship and not hosting offensive material, and people will reasonably disagree with one another. I therefore expect that no policy will be a perfect solution. Thus...
**Summary**: I recommend:
1. respecting the original author's decision to link or not link,
2. optionally adding an unobtrusive and impartial-sounding adjective such as "...the Shia-critical site [link]..." if relevant, and
3. adding a pertinent snippet from the site (to reduce the need to access the site).
[Monica Cellio](https://meta.stackexchange.com/a/291082/351283), diamond moderator at Mi Yodeya (i.e., Judaism.SE), proposed removing links to actively hostile sites, which seems reasonable to me.
---
**Should we forbid linking to anti-any-religious groups sites?**
Forbidding these links goes against the sectarian nature of the site (this is a question and answer site; it's neither pro-XYZ nor anti-XYZ). While it's against StackExchange policy to host offensive material (see: [Stack Exchange Network Terms of Service](http://stackexchange.com/legal)), this is counterbalanced by StackExchange's general attitude towards censorship. In this situation, Islam.SE doesn't host the content of those links.
*These links provide context*. It may be that visiting the site is needed to understand a post, and not providing the link is an obstacle. Providing a pertinent snippet from the site, thereby giving a self-contained answer/question, reduces the need for someone to click on the link. (And many anti-[anything] sites are prone to link rot.)
It's going to be a judgement call (i.e., a matter of opinion). In these cases, we should respect the original author's decision; they best understand whether or not a link is needed to understand their post. Otherwise, we risk arguments, edit wars, and antipathy towards the site. (There's going to be exceptions, e.g. if the link encourages violence.)
We can optionally add some adjective e.g. "...the Shia-critical site [link]...", something that's both unobtrusive and is not itself offensive. Then reader can make an informed decision as to whether or not they want to click the link.
---
A similar question was asked on meta.Judaism.SE [Censoring comments](https://judaism.meta.stackexchange.com/questions/1632/), and it was argued against for the following reasons: (a) an implementation nightmare, (b) a bureaucratic nightmare, (c) unnecessary, (d) bad practice for this site.
I asked the question at meta.SE: [Should we edit out useful links to likely offensive websites?](https://meta.stackexchange.com/questions/291046/should-we-edit-out-useful-links-to-likely-offensive-websites) Hopefully, we can get a better perspective from more experienced sites and people more familiarity with StackExchange policy. The question got this response from [Monica Cellio](https://meta.stackexchange.com/a/291082/351283):
>
> We almost always do this in the case of a site that is actively hostile, abusive, or fraudulent (as opposed to just wrong). We feel no obligation to provide visibility and promotion for hate sites, for example.
>
>
> ...
>
>
> For sites that aren't like that but are more mildly problematic, often we keep links but make sure the nature of the link is clear without clicking.
>
>
>
This is somewhat consistent with goldPseudo's [answer](https://islam.meta.stackexchange.com/a/1863/17163). | If the site is marked by a reputable organization as a hate group, then I think we should not link to them. Instead, leave the link in footnotes without the http:// like I have done in [this question](https://islam.stackexchange.com/questions/33364/answering-the-protagonists-of-the-misinterperted-islamic-jihad).
linking to them helps their site by driving more traffic to them. We are here to promote rational prespectives of Islam not drive traffic to irrational sites. |
1,004 | I came here to seek truth and I assume others also look for the same, which means I should expect beliefs against mine and rejection of my basics in religion. All good, but I've seen recently where some posts link to sites in which explicitly hate-speech and insulting to other beliefs is common.
In the interest of providing particulars, the incident which prompted this discussion was seeing an answer reference material from <http://shia.bs> (which is clearly anti-Shi'a). I don't want to link to that answer since I'm not interested in calling it (or the author who posted it) out in particular, I just want a general discussion on how and when such sites should be used, if at all.
For me this is very disappointing since this goes against rational reasoning and Q&A norms. And we are helping such sites spread hate and hurt the unity and peace between Muslims, which is for sure tricks of our enemies.
Should we forbid such sites? | 2014/04/15 | [
"https://islam.meta.stackexchange.com/questions/1004",
"https://islam.meta.stackexchange.com",
"https://islam.meta.stackexchange.com/users/-1/"
] | I think the real question here is, **why is the link there in the first place**?
In my experience, this sort of issue typically shows up in one of three use cases:
1. A question seeking the context and/or veracity of a claim made on the site:
>
> According to **badsite.foo**, Muslims eat babies for lunch. Does this actually happen?
>
>
>
2. A poster is actively trying to drive traffic to the site:
>
> To find out what Muslims *really* believe about babies, check out **badsite.foo**.
>
>
>
3. A poster is actually using it as a reference:
>
> According to one hadith (e.g. as posted on **badsite.foo**), babies are particularly delicious. While Muslim scholars have unanimously declared this particular narration as *mawdu*, it is still commonly accepted by certain fringe groups.
>
>
>
In the first case, I don't see anything particularly wrong with that. Such a link provides valuable context which is useful for (a) making sure that the OP is accurately understanding and representing the claim and (b) determining exactly what arguments were presented to *make* this claim so they can be verified/refuted accurately. If you're concerned that people might click through on this link in ignorance, adding a warning that it's clearly anti-whatever is easy enough, but I don't see any value in removing the link entirely.
In the second case, however, I see absolutely no value to the link: It's basically just an unsolicited advertisement and an attempt to drive people to a third-party site. When such links are tangential to the actual post and can be removed entirely without affecting anything, they should simply be removed.
The third case is where things get tricky, and these should probably be taken on a case-by-case basis. If such a site is being referenced, it should be clear exactly what value (if any) this reference has to the actual post. Anti-whatever sites are typically so full of bias, fallacies and bad rhetoric that they are just terrible sources period (from any academic perspective): If the entire answer is built around references to such a site, whether it's directly linked or not the answer is probably just a terrible answer and should be downvoted into oblivion. And when such posts are simply a better-disguised version of the second case (i.e. active attempt to drive traffic to the third-party site, rather than an honest attempt to answer the question) they should probably be deleted entirely. | If the site is marked by a reputable organization as a hate group, then I think we should not link to them. Instead, leave the link in footnotes without the http:// like I have done in [this question](https://islam.stackexchange.com/questions/33364/answering-the-protagonists-of-the-misinterperted-islamic-jihad).
linking to them helps their site by driving more traffic to them. We are here to promote rational prespectives of Islam not drive traffic to irrational sites. |
1,004 | I came here to seek truth and I assume others also look for the same, which means I should expect beliefs against mine and rejection of my basics in religion. All good, but I've seen recently where some posts link to sites in which explicitly hate-speech and insulting to other beliefs is common.
In the interest of providing particulars, the incident which prompted this discussion was seeing an answer reference material from <http://shia.bs> (which is clearly anti-Shi'a). I don't want to link to that answer since I'm not interested in calling it (or the author who posted it) out in particular, I just want a general discussion on how and when such sites should be used, if at all.
For me this is very disappointing since this goes against rational reasoning and Q&A norms. And we are helping such sites spread hate and hurt the unity and peace between Muslims, which is for sure tricks of our enemies.
Should we forbid such sites? | 2014/04/15 | [
"https://islam.meta.stackexchange.com/questions/1004",
"https://islam.meta.stackexchange.com",
"https://islam.meta.stackexchange.com/users/-1/"
] | First, we should recognize this is in the gray area between censorship and not hosting offensive material, and people will reasonably disagree with one another. I therefore expect that no policy will be a perfect solution. Thus...
**Summary**: I recommend:
1. respecting the original author's decision to link or not link,
2. optionally adding an unobtrusive and impartial-sounding adjective such as "...the Shia-critical site [link]..." if relevant, and
3. adding a pertinent snippet from the site (to reduce the need to access the site).
[Monica Cellio](https://meta.stackexchange.com/a/291082/351283), diamond moderator at Mi Yodeya (i.e., Judaism.SE), proposed removing links to actively hostile sites, which seems reasonable to me.
---
**Should we forbid linking to anti-any-religious groups sites?**
Forbidding these links goes against the sectarian nature of the site (this is a question and answer site; it's neither pro-XYZ nor anti-XYZ). While it's against StackExchange policy to host offensive material (see: [Stack Exchange Network Terms of Service](http://stackexchange.com/legal)), this is counterbalanced by StackExchange's general attitude towards censorship. In this situation, Islam.SE doesn't host the content of those links.
*These links provide context*. It may be that visiting the site is needed to understand a post, and not providing the link is an obstacle. Providing a pertinent snippet from the site, thereby giving a self-contained answer/question, reduces the need for someone to click on the link. (And many anti-[anything] sites are prone to link rot.)
It's going to be a judgement call (i.e., a matter of opinion). In these cases, we should respect the original author's decision; they best understand whether or not a link is needed to understand their post. Otherwise, we risk arguments, edit wars, and antipathy towards the site. (There's going to be exceptions, e.g. if the link encourages violence.)
We can optionally add some adjective e.g. "...the Shia-critical site [link]...", something that's both unobtrusive and is not itself offensive. Then reader can make an informed decision as to whether or not they want to click the link.
---
A similar question was asked on meta.Judaism.SE [Censoring comments](https://judaism.meta.stackexchange.com/questions/1632/), and it was argued against for the following reasons: (a) an implementation nightmare, (b) a bureaucratic nightmare, (c) unnecessary, (d) bad practice for this site.
I asked the question at meta.SE: [Should we edit out useful links to likely offensive websites?](https://meta.stackexchange.com/questions/291046/should-we-edit-out-useful-links-to-likely-offensive-websites) Hopefully, we can get a better perspective from more experienced sites and people more familiarity with StackExchange policy. The question got this response from [Monica Cellio](https://meta.stackexchange.com/a/291082/351283):
>
> We almost always do this in the case of a site that is actively hostile, abusive, or fraudulent (as opposed to just wrong). We feel no obligation to provide visibility and promotion for hate sites, for example.
>
>
> ...
>
>
> For sites that aren't like that but are more mildly problematic, often we keep links but make sure the nature of the link is clear without clicking.
>
>
>
This is somewhat consistent with goldPseudo's [answer](https://islam.meta.stackexchange.com/a/1863/17163). | I think the real question here is, **why is the link there in the first place**?
In my experience, this sort of issue typically shows up in one of three use cases:
1. A question seeking the context and/or veracity of a claim made on the site:
>
> According to **badsite.foo**, Muslims eat babies for lunch. Does this actually happen?
>
>
>
2. A poster is actively trying to drive traffic to the site:
>
> To find out what Muslims *really* believe about babies, check out **badsite.foo**.
>
>
>
3. A poster is actually using it as a reference:
>
> According to one hadith (e.g. as posted on **badsite.foo**), babies are particularly delicious. While Muslim scholars have unanimously declared this particular narration as *mawdu*, it is still commonly accepted by certain fringe groups.
>
>
>
In the first case, I don't see anything particularly wrong with that. Such a link provides valuable context which is useful for (a) making sure that the OP is accurately understanding and representing the claim and (b) determining exactly what arguments were presented to *make* this claim so they can be verified/refuted accurately. If you're concerned that people might click through on this link in ignorance, adding a warning that it's clearly anti-whatever is easy enough, but I don't see any value in removing the link entirely.
In the second case, however, I see absolutely no value to the link: It's basically just an unsolicited advertisement and an attempt to drive people to a third-party site. When such links are tangential to the actual post and can be removed entirely without affecting anything, they should simply be removed.
The third case is where things get tricky, and these should probably be taken on a case-by-case basis. If such a site is being referenced, it should be clear exactly what value (if any) this reference has to the actual post. Anti-whatever sites are typically so full of bias, fallacies and bad rhetoric that they are just terrible sources period (from any academic perspective): If the entire answer is built around references to such a site, whether it's directly linked or not the answer is probably just a terrible answer and should be downvoted into oblivion. And when such posts are simply a better-disguised version of the second case (i.e. active attempt to drive traffic to the third-party site, rather than an honest attempt to answer the question) they should probably be deleted entirely. |
463 | There are a lot of retro-computers with BASIC as an integral part of the operating system. The IBM 5100 was one of the first computers with a separate memory space for BASIC, followed by the Commodore PET, Apple II and TRS-80.
But why was BASIC built into their operating systems, using memory? From [this memory map](http://sta.c64.org/cbm64mem.html), the Commodore 64 had memory completely reserved for BASIC. There were computers, like the Altair 8800, that provided BASIC as software that could be loaded into memory when necessary, and these computers both worked and sold well.
So why was BASIC built into so many computers' operating systems? | 2016/05/11 | [
"https://retrocomputing.stackexchange.com/questions/463",
"https://retrocomputing.stackexchange.com",
"https://retrocomputing.stackexchange.com/users/278/"
] | Thraka is right that BASIC served the role of operating system on many 8-bit micros. *Some* OS was needed, that was beyond doubt. But at the time dedicated operating systems were either very limited (say, Atari DOS), overly complex, big and expensive for the tiny computers (Unix), or - for the middle ground, that was "just right" - in their infancy (CP/M).
BASIC, on the other hand, was touted THE beginner's programming language.
At the time - 8-bit micros were primarily aimed at the kids and teens market. They were "home computers", where the dad could write and print letters, mom could design crosstitch images, but 90% of the time kids would be playing games. But if the computer was to be more than a glorified gaming console with hardly-used extra functions, it needed a marketing point that would set it apart from the consoles: **"It Teaches Your Kids Programming"** was the critical sales point. And so it needed a kid-friendly programming language.
BASIC had only one serious competitor for that position at the time: [Logo](https://en.wikipedia.org/wiki/Logo_%28programming_language%29), and while Logo was generally considered a much better learning tool\* but it hardly yielded towards "general use" - you could do some quite fancy stuff easily (drawing a [Koch Snowflake](https://en.wikipedia.org/wiki/Koch_snowflake) was less than a screen long program; not a packed screen too) but didn't yield well towards "common use", and was considerably slower. Other languages than the two either employed concepts too advanced for kids, or were too complex to implement on 8-bit micros (leaving enough memory for actual software too!).
...also - Microsoft Basic was an important language for "professional environment". Learning its simpler variants was considered (usually by clueless parents) an important building block towards building a career of using it professionally. Nobody ever wrote serious business software in Logo. Therefore the marketing appeal of Basic - despite its factual shortcomings - overshadowed Logo as THE language for learning.
On top of that, implementing OS procedures as BASIC commands was simple, natural and took little effort, little memory and allowed to integrate or script them as BASIC programs easily. Logo didn't yield itself to that purpose nearly as well.
And so, BASIC being simultaneously a comfortable OS environment, a scripting tool, a teaching tool, and - above all - an important marketing point - became the language of choice.
\* (e.g. providing recursion, which was between very difficult and impossible in Basic, lacking local variables, parametric function calls and rarely implementing easy to access arrays which could serve as a crutch for these). Also, the Turtle Graphics, which was excellent for teaching the concept of "giving commands", and building a library of procedures, but made utilizing the gfx as "pixel canvas" difficult. | As it says in Wikipedia [BASIC](https://en.wikipedia.org/wiki/BASIC):
>
> The introduction of the first microcomputers in the mid-1970s was the start of explosive growth for BASIC. It had the advantage that it was fairly well known to the young designers and computer hobbyists who took an interest in microcomputers.
>
> ... BASIC was one of the few languages that was both high-level enough to be usable by those without training and small enough to fit into the microcomputers of the day, making it the de facto standard programming language on early microcomputers.
>
>
>
In other words, it was easy to learn even for those with no programming experience, and the interpreter memory footprint was small, which was important due to the cost of memory.
(I don't exactly recall, but back in the 80s an 8k bit RAM chip in C64 cost several dollars.)
The use of an interpreter was also important from an ease of use point. While other existing languages were similar and not so difficult to learn (for example FORTRAN), they had to be compiled.
This means that instead of getting instant results (or errors), you had to write the code, compile it (hopefully no errors otherwise you had to change the code again and compile again), and then run it. If any errors, back to square one you go. |
463 | There are a lot of retro-computers with BASIC as an integral part of the operating system. The IBM 5100 was one of the first computers with a separate memory space for BASIC, followed by the Commodore PET, Apple II and TRS-80.
But why was BASIC built into their operating systems, using memory? From [this memory map](http://sta.c64.org/cbm64mem.html), the Commodore 64 had memory completely reserved for BASIC. There were computers, like the Altair 8800, that provided BASIC as software that could be loaded into memory when necessary, and these computers both worked and sold well.
So why was BASIC built into so many computers' operating systems? | 2016/05/11 | [
"https://retrocomputing.stackexchange.com/questions/463",
"https://retrocomputing.stackexchange.com",
"https://retrocomputing.stackexchange.com/users/278/"
] | Having BASIC available for the machine was a selling point so early adopters wouldn't have to wait for software to become available--they could write what they need themselves, and they wouldn't need to learn machine language to do it.
>
> [Bill] Gates [believed](https://en.wikipedia.org/wiki/Altair_BASIC#Origin_and_development) that, by providing a BASIC interpreter for
> the new [MITS Altair] computer, they could make it more attractive to
> hobbyists.
>
>
> | During the late 1970s and early 1980s, ROM was cheaper than RAM. The amount of useful work that can be accomplished with a BASIC program that occupies a certain amount of RAM will be often greater than the amount that could be accomplished with the same quantity of machine code, so a computer with 16K of ROM and 8K of RAM could be more useful than one with 2K of ROM and 16K of RAM, even though the latter computer would cost more.
Further, editing a BASIC program "in place" is much more efficient than trying to do likewise with a machine-language program. In-place editing of machine code is possible, but for the most part generating efficient machine code requires creating a text file and passing it through an assembler. This in turn generally entails a long slow process every time one wants to modify the code and test it again. By contrast, in-place editing of a BASIC program is easy, and for many kinds of programs cutting the edit-build-test cycle from five minutes to five seconds is a huge benefit which vastly outweighs the difference in execution speed.
The need for BASIC largely disappeared when RAM got cheap enough that platforms like Turbo Pascal could allow the editor, compiler, source code, executable, and data to all be kept in memory simultaneously. Basic's primary benefit was the ability to keep everything in RAM, and once such ability became available in a compiled language there was no longer a reason to tolerate the slowness or semantic limitations of interpreted BASIC. |
463 | There are a lot of retro-computers with BASIC as an integral part of the operating system. The IBM 5100 was one of the first computers with a separate memory space for BASIC, followed by the Commodore PET, Apple II and TRS-80.
But why was BASIC built into their operating systems, using memory? From [this memory map](http://sta.c64.org/cbm64mem.html), the Commodore 64 had memory completely reserved for BASIC. There were computers, like the Altair 8800, that provided BASIC as software that could be loaded into memory when necessary, and these computers both worked and sold well.
So why was BASIC built into so many computers' operating systems? | 2016/05/11 | [
"https://retrocomputing.stackexchange.com/questions/463",
"https://retrocomputing.stackexchange.com",
"https://retrocomputing.stackexchange.com/users/278/"
] | As far as I know it was included because it was essentially the operating system interface. When you installed MS-DOS on a PC it provided you with commands and allowed you to run programs that executed machine code on the machine.
BASIC back then was similar. It provided you with a command prompt, operating system-level commands like reading/writing data to disk, a printer, tape, do simple things like print to the screen, and read in user input.
This way you had a general set of commands with an interpreter that could be ported to other CPUs and memory configurations with minimal changes. You just needed to setup which commands were available and what machine specific code was run for those command.
Essentially a portable, programmable, operating system. | In addition to the answers above, something to keep in mind is that there wasn't a ton of commercial software available at the time compared to today and computers were not yet things that everyone needed or even found useful for their jobs. It's not like today where you could buy a computer, load it with Office and other software, use it to do your job, maybe even run your business and turn it into a multimillion dollar operation, and never touch a line of code.
Meanwhile when I went to college in 1995 I had fellow freshmen who could barely get their PC turned on but they knew how to program in Pascal because the American school system at the time taught programming because they didn't know what else to do with these computer things everyone is buying, but they knew that it was going to be a "thing" in the future.
So one of the reasons a lot of these old toy computers came with BASIC built in was because, compared to today, they didn't do much else. And they also tended to not run the software from the previous generation or any other computers (if you think about it, Apple's never completely shaken this mindset - they have few qualms about breaking backwards compatibility for machines more than a few years old and they sell a computer that can't run the software from the competition's platforms). Notice how it wasn't until the DOS/Windows cycle started happening that backwards compatibility became a big thing and then sales of computer took off. There were other factors too, like price, but it inspired a lot of consumer confidence when you could be reasonably certain your stuff could come along for the ride. |
463 | There are a lot of retro-computers with BASIC as an integral part of the operating system. The IBM 5100 was one of the first computers with a separate memory space for BASIC, followed by the Commodore PET, Apple II and TRS-80.
But why was BASIC built into their operating systems, using memory? From [this memory map](http://sta.c64.org/cbm64mem.html), the Commodore 64 had memory completely reserved for BASIC. There were computers, like the Altair 8800, that provided BASIC as software that could be loaded into memory when necessary, and these computers both worked and sold well.
So why was BASIC built into so many computers' operating systems? | 2016/05/11 | [
"https://retrocomputing.stackexchange.com/questions/463",
"https://retrocomputing.stackexchange.com",
"https://retrocomputing.stackexchange.com/users/278/"
] | Having BASIC available for the machine was a selling point so early adopters wouldn't have to wait for software to become available--they could write what they need themselves, and they wouldn't need to learn machine language to do it.
>
> [Bill] Gates [believed](https://en.wikipedia.org/wiki/Altair_BASIC#Origin_and_development) that, by providing a BASIC interpreter for
> the new [MITS Altair] computer, they could make it more attractive to
> hobbyists.
>
>
> | A lot of great information has been given.
One more thing to consider is what the machines were like. Most had no decent non-volatile storage - no hard drives, many didn't have diskette drives either.
Even the original IBM-PC came with a cassette tape interface - they used a standard audio cassette recorder/player. Even when they had a diskette drive, the drive was often painfully slow. (Looking at you, Commodore 64.)
Having a programming language built in meant that you only needed to save/load the very small program, and it meant that every instance of a given computer (ie, C64, TRS-80 Model III, TI/99, etc) had the exact same language available on it.
This had value because computer magazines actually had programs printed in them, and the readers would laboriously type them in, debug the typos, and run them. |
463 | There are a lot of retro-computers with BASIC as an integral part of the operating system. The IBM 5100 was one of the first computers with a separate memory space for BASIC, followed by the Commodore PET, Apple II and TRS-80.
But why was BASIC built into their operating systems, using memory? From [this memory map](http://sta.c64.org/cbm64mem.html), the Commodore 64 had memory completely reserved for BASIC. There were computers, like the Altair 8800, that provided BASIC as software that could be loaded into memory when necessary, and these computers both worked and sold well.
So why was BASIC built into so many computers' operating systems? | 2016/05/11 | [
"https://retrocomputing.stackexchange.com/questions/463",
"https://retrocomputing.stackexchange.com",
"https://retrocomputing.stackexchange.com/users/278/"
] | Having a programming language built-in gave you a multi-purpose tool into your hands at the flick of a switch (power on).
As to the choice of BASIC vs. other programming languages, microcomputer BASIC dialects are - despite some of their shortcomings - quite accessible to novice programmers. They are a bit like using English in an imperative style ("PRINT this", "DO that"), so not hard to learn. There was a lot of BASIC literature around at that time and sufficiently general example programs could be ported to different systems without too much work. Many other languages are more cryptic to someone not used to it.
As they were usually included on the ROM chips, they didn't take away any RAM if you chose not to use BASIC. | Having BASIC available for the machine was a selling point so early adopters wouldn't have to wait for software to become available--they could write what they need themselves, and they wouldn't need to learn machine language to do it.
>
> [Bill] Gates [believed](https://en.wikipedia.org/wiki/Altair_BASIC#Origin_and_development) that, by providing a BASIC interpreter for
> the new [MITS Altair] computer, they could make it more attractive to
> hobbyists.
>
>
> |
463 | There are a lot of retro-computers with BASIC as an integral part of the operating system. The IBM 5100 was one of the first computers with a separate memory space for BASIC, followed by the Commodore PET, Apple II and TRS-80.
But why was BASIC built into their operating systems, using memory? From [this memory map](http://sta.c64.org/cbm64mem.html), the Commodore 64 had memory completely reserved for BASIC. There were computers, like the Altair 8800, that provided BASIC as software that could be loaded into memory when necessary, and these computers both worked and sold well.
So why was BASIC built into so many computers' operating systems? | 2016/05/11 | [
"https://retrocomputing.stackexchange.com/questions/463",
"https://retrocomputing.stackexchange.com",
"https://retrocomputing.stackexchange.com/users/278/"
] | In addition to the answers above, something to keep in mind is that there wasn't a ton of commercial software available at the time compared to today and computers were not yet things that everyone needed or even found useful for their jobs. It's not like today where you could buy a computer, load it with Office and other software, use it to do your job, maybe even run your business and turn it into a multimillion dollar operation, and never touch a line of code.
Meanwhile when I went to college in 1995 I had fellow freshmen who could barely get their PC turned on but they knew how to program in Pascal because the American school system at the time taught programming because they didn't know what else to do with these computer things everyone is buying, but they knew that it was going to be a "thing" in the future.
So one of the reasons a lot of these old toy computers came with BASIC built in was because, compared to today, they didn't do much else. And they also tended to not run the software from the previous generation or any other computers (if you think about it, Apple's never completely shaken this mindset - they have few qualms about breaking backwards compatibility for machines more than a few years old and they sell a computer that can't run the software from the competition's platforms). Notice how it wasn't until the DOS/Windows cycle started happening that backwards compatibility became a big thing and then sales of computer took off. There were other factors too, like price, but it inspired a lot of consumer confidence when you could be reasonably certain your stuff could come along for the ride. | Once upon a time, 16 kilobytes of DRAM was expensive. The original 8080 Altair came with up to 4 kbytes of memory on a single memory board.
The only programming languages (higher than ASM) that could fit in that small a memory were Tiny BASIC (Palo Alto Tiny BASIC, Altair BASIC (Gates and Allen) and Woz's Integer BASIC, et.al.) and Forth. The only alternatives were hand coding assembly language, or cross-compiling on a larger minicomputer. Basic was more well known and far more commercially viable than entering programs in binary via front panel switches. Later personal computers continued with what the market was familiar with (from the books and magazine articles at that time), and came with BASIC loaded into ROM.
So the answer is:
BASIC (with a very simple OS for I/O) came in ROM on many early personal computers because mass-produced masked ROM was cheaper than RAM, on a per byte basis. 64k address space was considered huge at the time, so dedicating a portion of it to ROM was not considered wasteful. And BASIC in ROM allowed the computer to be usable on power-up, instead of after loading software via paper-tape, cassette tape, or some other media. |
463 | There are a lot of retro-computers with BASIC as an integral part of the operating system. The IBM 5100 was one of the first computers with a separate memory space for BASIC, followed by the Commodore PET, Apple II and TRS-80.
But why was BASIC built into their operating systems, using memory? From [this memory map](http://sta.c64.org/cbm64mem.html), the Commodore 64 had memory completely reserved for BASIC. There were computers, like the Altair 8800, that provided BASIC as software that could be loaded into memory when necessary, and these computers both worked and sold well.
So why was BASIC built into so many computers' operating systems? | 2016/05/11 | [
"https://retrocomputing.stackexchange.com/questions/463",
"https://retrocomputing.stackexchange.com",
"https://retrocomputing.stackexchange.com/users/278/"
] | Thraka is right that BASIC served the role of operating system on many 8-bit micros. *Some* OS was needed, that was beyond doubt. But at the time dedicated operating systems were either very limited (say, Atari DOS), overly complex, big and expensive for the tiny computers (Unix), or - for the middle ground, that was "just right" - in their infancy (CP/M).
BASIC, on the other hand, was touted THE beginner's programming language.
At the time - 8-bit micros were primarily aimed at the kids and teens market. They were "home computers", where the dad could write and print letters, mom could design crosstitch images, but 90% of the time kids would be playing games. But if the computer was to be more than a glorified gaming console with hardly-used extra functions, it needed a marketing point that would set it apart from the consoles: **"It Teaches Your Kids Programming"** was the critical sales point. And so it needed a kid-friendly programming language.
BASIC had only one serious competitor for that position at the time: [Logo](https://en.wikipedia.org/wiki/Logo_%28programming_language%29), and while Logo was generally considered a much better learning tool\* but it hardly yielded towards "general use" - you could do some quite fancy stuff easily (drawing a [Koch Snowflake](https://en.wikipedia.org/wiki/Koch_snowflake) was less than a screen long program; not a packed screen too) but didn't yield well towards "common use", and was considerably slower. Other languages than the two either employed concepts too advanced for kids, or were too complex to implement on 8-bit micros (leaving enough memory for actual software too!).
...also - Microsoft Basic was an important language for "professional environment". Learning its simpler variants was considered (usually by clueless parents) an important building block towards building a career of using it professionally. Nobody ever wrote serious business software in Logo. Therefore the marketing appeal of Basic - despite its factual shortcomings - overshadowed Logo as THE language for learning.
On top of that, implementing OS procedures as BASIC commands was simple, natural and took little effort, little memory and allowed to integrate or script them as BASIC programs easily. Logo didn't yield itself to that purpose nearly as well.
And so, BASIC being simultaneously a comfortable OS environment, a scripting tool, a teaching tool, and - above all - an important marketing point - became the language of choice.
\* (e.g. providing recursion, which was between very difficult and impossible in Basic, lacking local variables, parametric function calls and rarely implementing easy to access arrays which could serve as a crutch for these). Also, the Turtle Graphics, which was excellent for teaching the concept of "giving commands", and building a library of procedures, but made utilizing the gfx as "pixel canvas" difficult. | Because at the time a personal computer without BASIC was a total non-starter. A personal computer without BASIC would seem as silly then as a computer without a web browser would seem today.
We considered BASIC interpreters essential in the 1970's and early 1980's because the modern concept of the personal computer as a machine on which one runs ready-made application programs was only beginning to emerge. BASIC was invented in 1964. Microsoft BASIC was released in 1975. Wordstar was released in 1978, Visicalc in 1979, Wordperfect in 1979, Microsoft Word in 1983, Lotus 123 in 1983. If you were a user then, you might have heard of some of these programs, you might be able to get a copy of one of them, and it might run on your computer (but probably would not).
Making practical use of a computer without programming was still very much a new idea. BASIC was seen as a language which end users would use to solve problems. I learned it from the 1975 book "Making BASIC Work For You". It is all about how to solve business accounting and record-keeping problems. The examples are things which today would be done using spreadsheets and other office applications.
The computer buyer was also different. The average person did not think that he ever would or ever could operate a computer and so had no desire to own a home computer. They were bought by hobbyists who valiantly tried to show their friends that the machines had some practical value (which was hard since most tasks took two or three times as long on a computer) and by parents who realized that microcomputers, even if useless now, were the wave of the future and wanted their children to "learn programming". In both cases it was essential that the computer be programmable. |
463 | There are a lot of retro-computers with BASIC as an integral part of the operating system. The IBM 5100 was one of the first computers with a separate memory space for BASIC, followed by the Commodore PET, Apple II and TRS-80.
But why was BASIC built into their operating systems, using memory? From [this memory map](http://sta.c64.org/cbm64mem.html), the Commodore 64 had memory completely reserved for BASIC. There were computers, like the Altair 8800, that provided BASIC as software that could be loaded into memory when necessary, and these computers both worked and sold well.
So why was BASIC built into so many computers' operating systems? | 2016/05/11 | [
"https://retrocomputing.stackexchange.com/questions/463",
"https://retrocomputing.stackexchange.com",
"https://retrocomputing.stackexchange.com/users/278/"
] | As far as I know it was included because it was essentially the operating system interface. When you installed MS-DOS on a PC it provided you with commands and allowed you to run programs that executed machine code on the machine.
BASIC back then was similar. It provided you with a command prompt, operating system-level commands like reading/writing data to disk, a printer, tape, do simple things like print to the screen, and read in user input.
This way you had a general set of commands with an interpreter that could be ported to other CPUs and memory configurations with minimal changes. You just needed to setup which commands were available and what machine specific code was run for those command.
Essentially a portable, programmable, operating system. | Once upon a time, 16 kilobytes of DRAM was expensive. The original 8080 Altair came with up to 4 kbytes of memory on a single memory board.
The only programming languages (higher than ASM) that could fit in that small a memory were Tiny BASIC (Palo Alto Tiny BASIC, Altair BASIC (Gates and Allen) and Woz's Integer BASIC, et.al.) and Forth. The only alternatives were hand coding assembly language, or cross-compiling on a larger minicomputer. Basic was more well known and far more commercially viable than entering programs in binary via front panel switches. Later personal computers continued with what the market was familiar with (from the books and magazine articles at that time), and came with BASIC loaded into ROM.
So the answer is:
BASIC (with a very simple OS for I/O) came in ROM on many early personal computers because mass-produced masked ROM was cheaper than RAM, on a per byte basis. 64k address space was considered huge at the time, so dedicating a portion of it to ROM was not considered wasteful. And BASIC in ROM allowed the computer to be usable on power-up, instead of after loading software via paper-tape, cassette tape, or some other media. |
463 | There are a lot of retro-computers with BASIC as an integral part of the operating system. The IBM 5100 was one of the first computers with a separate memory space for BASIC, followed by the Commodore PET, Apple II and TRS-80.
But why was BASIC built into their operating systems, using memory? From [this memory map](http://sta.c64.org/cbm64mem.html), the Commodore 64 had memory completely reserved for BASIC. There were computers, like the Altair 8800, that provided BASIC as software that could be loaded into memory when necessary, and these computers both worked and sold well.
So why was BASIC built into so many computers' operating systems? | 2016/05/11 | [
"https://retrocomputing.stackexchange.com/questions/463",
"https://retrocomputing.stackexchange.com",
"https://retrocomputing.stackexchange.com/users/278/"
] | Nobody so far has said the magic words, which is **[Microsoft BASIC](https://en.wikipedia.org/wiki/Microsoft_BASIC)**. First developed for the Altair 8800 (the first commercially successful personal computer!), Microsoft spent a lot of energy making sure their BASIC would work on every personal computer in the 1970s and licensing it widely. It ran on [CP/M](https://en.wikipedia.org/wiki/CP/M), it ran on the [TRS-80](https://en.wikipedia.org/wiki/TRS-80#BASIC), and -- of course -- the [IBM PC](https://en.wikipedia.org/wiki/IBM_Personal_Computer). Apple's [AppleSoft BASIC](https://en.wikipedia.org/wiki/Applesoft_BASIC#Background), [Commodore BASIC](https://en.wikipedia.org/wiki/Commodore_BASIC) and [Amiga BASIC](https://en.wikipedia.org/wiki/AmigaBASIC) were all originally licensed from Microsoft. This meant that if you were manufacturing a personal computer in the 1970s, the first question any customer would ask was whether their existing Microsoft BASIC programs would work on it -- essentially, the first standardized programming language/execution platform! | A lot of great information has been given.
One more thing to consider is what the machines were like. Most had no decent non-volatile storage - no hard drives, many didn't have diskette drives either.
Even the original IBM-PC came with a cassette tape interface - they used a standard audio cassette recorder/player. Even when they had a diskette drive, the drive was often painfully slow. (Looking at you, Commodore 64.)
Having a programming language built in meant that you only needed to save/load the very small program, and it meant that every instance of a given computer (ie, C64, TRS-80 Model III, TI/99, etc) had the exact same language available on it.
This had value because computer magazines actually had programs printed in them, and the readers would laboriously type them in, debug the typos, and run them. |
463 | There are a lot of retro-computers with BASIC as an integral part of the operating system. The IBM 5100 was one of the first computers with a separate memory space for BASIC, followed by the Commodore PET, Apple II and TRS-80.
But why was BASIC built into their operating systems, using memory? From [this memory map](http://sta.c64.org/cbm64mem.html), the Commodore 64 had memory completely reserved for BASIC. There were computers, like the Altair 8800, that provided BASIC as software that could be loaded into memory when necessary, and these computers both worked and sold well.
So why was BASIC built into so many computers' operating systems? | 2016/05/11 | [
"https://retrocomputing.stackexchange.com/questions/463",
"https://retrocomputing.stackexchange.com",
"https://retrocomputing.stackexchange.com/users/278/"
] | A lot of great information has been given.
One more thing to consider is what the machines were like. Most had no decent non-volatile storage - no hard drives, many didn't have diskette drives either.
Even the original IBM-PC came with a cassette tape interface - they used a standard audio cassette recorder/player. Even when they had a diskette drive, the drive was often painfully slow. (Looking at you, Commodore 64.)
Having a programming language built in meant that you only needed to save/load the very small program, and it meant that every instance of a given computer (ie, C64, TRS-80 Model III, TI/99, etc) had the exact same language available on it.
This had value because computer magazines actually had programs printed in them, and the readers would laboriously type them in, debug the typos, and run them. | BASIC was first used at Dartmouth College in 1964. It was the first readily available programming language on time-share systems (the predecessor of cloud computing). In 1975 Microsoft released a version of Basic for the first consumer level microcomputer, the MITS Altair 8800. Apple, Commodore, Tandy and then IBM PC (in 1981) followed suit with their microcomputers. By the time IBM released the PC, Apple had already acquired considerable marketshare in schools and thus acquired lots of young budding programmers who mostly only knew BASIC.
In its day, BASIC was as indispensable as a web browser is today. |
66,560 | >
> Placed in front of you with a feast meat for a king,
>
> Offered by the clown that comes with a new, neat thing.
>
>
> Tastes from the sea when the red girl comes this way,
>
> Always good with whatever John catches that day.
>
>
> To tempt you with what the cows try to murder,
>
> Or at the place where you won't wait further.
>
>
>
What am I? | 2018/05/31 | [
"https://puzzling.stackexchange.com/questions/66560",
"https://puzzling.stackexchange.com",
"https://puzzling.stackexchange.com/users/48593/"
] | Based on Francesco Roggia's observation that
>
> the initial letters spell **potato**,
>
>
>
you are
>
> **french fries**.
>
>
>
Placed in front of you with a feast meat for a king,
>
> Burger King
>
>
>
Offered by the clown that comes with a new, neat thing.
>
> McDonald's
>
>
>
Tastes from the sea when the red girl comes this way,
>
> Wendy's
>
>
>
Always good with whatever John catches that day.
>
> Long John Silver's
>
>
>
To tempt you with what the cows try to murder,
>
> Chick-fil-A
>
>
>
Or at the place where you won't wait further.
>
> In-N-Out Burger [last clue added with hint from author]
>
>
> | I'm just guessing here:
>
> Is it something about a Potato? Just because of the first letter of the rhyme maybe that's a hint to the right answer. I don't really know what is a potato about with each line of the riddle, i let anyone else guess it
>
>
> |
35,299 | I'm a developer and I'm just so used to using Spanish (Argentina) as my keyboard layout on my PC both at home and at work. Now I want to develop on my Macbook Pro too, but the only Spanish layouts available are "regular" and ISO, both of which are basically the same as the one I use, except I have to press the alt key to input characters that are very common for me, like {}[].
How can I set my keyboard layout to Spanish (Argentina)? Thanks a lot! | 2012/01/02 | [
"https://apple.stackexchange.com/questions/35299",
"https://apple.stackexchange.com",
"https://apple.stackexchange.com/users/1949/"
] | Would the Windows Latin American layout mentioned in this article meet your needs?
<http://m10lmac.blogspot.com/2007/02/more-ways-to-type-spanish.html> | You might find [this question](https://apple.stackexchange.com/questions/19490/change-keyboard-shortcut-for-and-in-german-keyboard-layout) to be helpful for your problem. The two tools referred to in the answers are [Ukulele](http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=ukelele) and [Spark](http://www.shadowlab.org/softwares/spark.php).
Ukulele allows you to edit keyboard layouts (thus creating the layout you want), while Spark creates shortcut key combinations, which you could use to remap a few keys to create the layout you want. |
35,299 | I'm a developer and I'm just so used to using Spanish (Argentina) as my keyboard layout on my PC both at home and at work. Now I want to develop on my Macbook Pro too, but the only Spanish layouts available are "regular" and ISO, both of which are basically the same as the one I use, except I have to press the alt key to input characters that are very common for me, like {}[].
How can I set my keyboard layout to Spanish (Argentina)? Thanks a lot! | 2012/01/02 | [
"https://apple.stackexchange.com/questions/35299",
"https://apple.stackexchange.com",
"https://apple.stackexchange.com/users/1949/"
] | You might find [this question](https://apple.stackexchange.com/questions/19490/change-keyboard-shortcut-for-and-in-german-keyboard-layout) to be helpful for your problem. The two tools referred to in the answers are [Ukulele](http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&item_id=ukelele) and [Spark](http://www.shadowlab.org/softwares/spark.php).
Ukulele allows you to edit keyboard layouts (thus creating the layout you want), while Spark creates shortcut key combinations, which you could use to remap a few keys to create the layout you want. | I created this project in github for Latin American keyboard in macOS, maybe it helps you. <https://github.com/neosergio/Latam-Keyboard> |
35,299 | I'm a developer and I'm just so used to using Spanish (Argentina) as my keyboard layout on my PC both at home and at work. Now I want to develop on my Macbook Pro too, but the only Spanish layouts available are "regular" and ISO, both of which are basically the same as the one I use, except I have to press the alt key to input characters that are very common for me, like {}[].
How can I set my keyboard layout to Spanish (Argentina)? Thanks a lot! | 2012/01/02 | [
"https://apple.stackexchange.com/questions/35299",
"https://apple.stackexchange.com",
"https://apple.stackexchange.com/users/1949/"
] | Would the Windows Latin American layout mentioned in this article meet your needs?
<http://m10lmac.blogspot.com/2007/02/more-ways-to-type-spanish.html> | I created this project in github for Latin American keyboard in macOS, maybe it helps you. <https://github.com/neosergio/Latam-Keyboard> |
165,915 | I applied for a job and during the various stages of the job interview the hiring person told me that they will get back to me in a certain amount of time. However, after every round of the interview the hiring person overstepped that limit. One week turned into two, two weeks turned into five. **And this happened after every single interview round.**
Even though the job sounds interesting and is with a reputable company, I wonder how they will treat me on the actual job if if they overstep their maximum allotted time to respond to me after every single interview round.
The position I'm applying for is relatively new in the company, so I think they're taking their time with the job applicants, but still, overstepping the time limit to get back to me on three separate occasions is non-professional to say the least. They also didn't acknowledge the delay.
**Should I consider this kind of behavior a red flag?** | 2020/10/13 | [
"https://workplace.stackexchange.com/questions/165915",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/120871/"
] | I would take it as a potential red flag regarding deadlines and accountability as they relate to corporate culture. I would definitely be frustrated being part of an organization with that in its culture.
That said, there are legit reasons for delays. A realistic possibility that comes to mind is you didn't make the cut between rounds but a higher ranked person dropped out.
The best thing you can do is ask to the reasoning at the time of offering/next interview round. I would just phrase it, "Is this a normal part of the company culture or were you experiencing exceptional circumstances?" | I would say that unless you apply for a job, the quality of their HR is not that important. Look for red flags with the people you are working with, they are more important. You can get an excellent job in a company with rubbish HR. |
165,915 | I applied for a job and during the various stages of the job interview the hiring person told me that they will get back to me in a certain amount of time. However, after every round of the interview the hiring person overstepped that limit. One week turned into two, two weeks turned into five. **And this happened after every single interview round.**
Even though the job sounds interesting and is with a reputable company, I wonder how they will treat me on the actual job if if they overstep their maximum allotted time to respond to me after every single interview round.
The position I'm applying for is relatively new in the company, so I think they're taking their time with the job applicants, but still, overstepping the time limit to get back to me on three separate occasions is non-professional to say the least. They also didn't acknowledge the delay.
**Should I consider this kind of behavior a red flag?** | 2020/10/13 | [
"https://workplace.stackexchange.com/questions/165915",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/120871/"
] | I would take it as a potential red flag regarding deadlines and accountability as they relate to corporate culture. I would definitely be frustrated being part of an organization with that in its culture.
That said, there are legit reasons for delays. A realistic possibility that comes to mind is you didn't make the cut between rounds but a higher ranked person dropped out.
The best thing you can do is ask to the reasoning at the time of offering/next interview round. I would just phrase it, "Is this a normal part of the company culture or were you experiencing exceptional circumstances?" | Yes and no.
Yes since in normal times it would be obvious that you should get a proper estimate. No since hiring in Corona times may be more complicated and subject ton constantly changing rules (e.g. is there anybody on furlough who could do the work, prove to your boss that you checked, prove to working council that you believe it....) |
165,915 | I applied for a job and during the various stages of the job interview the hiring person told me that they will get back to me in a certain amount of time. However, after every round of the interview the hiring person overstepped that limit. One week turned into two, two weeks turned into five. **And this happened after every single interview round.**
Even though the job sounds interesting and is with a reputable company, I wonder how they will treat me on the actual job if if they overstep their maximum allotted time to respond to me after every single interview round.
The position I'm applying for is relatively new in the company, so I think they're taking their time with the job applicants, but still, overstepping the time limit to get back to me on three separate occasions is non-professional to say the least. They also didn't acknowledge the delay.
**Should I consider this kind of behavior a red flag?** | 2020/10/13 | [
"https://workplace.stackexchange.com/questions/165915",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/120871/"
] | >
> Would you consider an offer from a company which always took more time
> to get back to you after every single round of the interview?
>
>
>
I would consider an offer from a company no matter how much time it took to get back to me. In the big scheme of things, that time is a sunk cost and isn't relevant to the merit of the job itself. Once hired, I wouldn't be going through the interview process (with any delays) again.
That said, if the delay bothers you enough, you can easily dismiss the offer without even considering it. Since you used the term "non-professional to say the least" you have to decide if what you saw during the interview process extends to all the other processes in the company, or was perhaps an outlier. Think back over the interview - what you heard and what you saw - to help you decide.
For me, I might weigh that bother against all the other aspects of the company/job/offer and make a decision based on the whole. | I would say that unless you apply for a job, the quality of their HR is not that important. Look for red flags with the people you are working with, they are more important. You can get an excellent job in a company with rubbish HR. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.