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 |
|---|---|---|---|---|---|
14,960 | A while ago, I asked [In what war would one modern military vehicle make a difference?](https://worldbuilding.stackexchange.com/questions/12219/in-what-war-would-one-modern-military-vehicle-make-a-difference) I have now come up with a sort-of sequel.
A modern, Challenger II battle tank has been sent back in time to a past war. It was decided in my last question that this tank would indeed have some significant impact on the war. What was not shown is how long the tank would be effective for.
That's what I'm asking now. In this question, I'm looking for comparison of how long my Challenger could last in various historic wars, assuming:
* No restocking on fuel *or* parts from the present; the time machine has been destroyed
* Unlimited *knowledge* of the tank can be taken back by its crew and shared with anyone else who might need it
* In-period manufacturing techniques and technology
* The tank is rendered ineffective when it can no longer move or shoot.
For example, it might last longer in WWII because the factories would be more capable of producing armor for it and it might be able to run off fuel from the time. I doubt it would last so long in a war before vehicles were invented. | 2015/04/22 | [
"https://worldbuilding.stackexchange.com/questions/14960",
"https://worldbuilding.stackexchange.com",
"https://worldbuilding.stackexchange.com/users/2685/"
] | Other have pointed out what the weaknesses of this one tank are. But the big game changer here, especially when talking about the two world wars, is not the tank, but the "unlimited knowledge" about the tank.
Although it can be assumed to be impossible to re-create the challenger2 or a leopard2, or even an abrams, all that knowledge should still help the contemporary industry to build much better tanks.
So, while the individual tank will be rendered useless sooner or later, the side the crew chooses still has an enormous advantage. | Depending on availability of spare parts?
A tanks as good as it's parts, no parts, no service.
A modern tank is really not all that different than a past tank, subtle differences in armor, the biggest change is sensor technology to find targets.
So your hypothetical, if a Abrams M1 went back to the Battle of the Bulge, against Tiger's it'd rip it to shreds. An M829 Armor piericing discarding sabot can penetrate 540 mm (21 inches) of steel rolled armor. A tiger tank, had 100mm of armor. A HEAT M830 round will go thru a Tiger or Panzer tank. |
15,562,420 | Today I received code review comments from a senior developer, one thing he pointed out is that each function should have only one return statement. I wonder whether this is just code style issue or there is security issue, i.e. potential function stack problem or uninitialized variables. Can anyone provide an example(I prefer C\C++\C#) to show the cons of multiple return statement? Thanks a lot. | 2013/03/22 | [
"https://Stackoverflow.com/questions/15562420",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1437741/"
] | That flag was introduced in iOS 5.1, so you'll have to change your deployment target to 5.1 to be able to do that. | Use it:
* Double-click on your Target's name in Xcode to bring up the Target Info.
* Go to the Build/Summary tab and search for "iPhone OS Deployment Target".
* Pick whichever SDK you want to use as your minimum requirements.
One thing is that it won't tell you if it doesn't work on a certain version. You'll have to compare your methods with which ones are available in each version manually.
You can help curb this by changing your Base SDK too:
* Same place, search for "Base SDK".
* Change it to the lowest iPhone OS Device SDK you can/want.
(On Snow Leopard, this is 2.2.1; Leopard can go as low as 2.0 if you installed it with Xcode.) |
8,724,909 | I recently started to learn Clojure and my intention is to build a simple CMS using it for practice. Since Clojure is a purely functional language and it requires a new way of thinking it just came up for me that what if I use a NO-SQL database for my application like MongoDB?
Do you have any comparison in this topic? I searched on google but I failed to dig up any usable data regarding this matter. What are the pros and the cons?
**edit: I mentioned clojure and web development to put things in context. For example with clojure it is easier to write a domain-specific language than for example using java.** | 2012/01/04 | [
"https://Stackoverflow.com/questions/8724909",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/485337/"
] | Whatever the language, you will gain a lot from using a no-sql database if you observe some OOP principles that Domain Driven Design brings to the table. The idea of an Aggregate and the Aggregate Root allow you to form organization in a more document-centric way. Since no-sql data manipulation is tailored to key-value pairs, aggregates and aggregate roots are something to investigate.
Also if you feel like learning more on how to organize your architecture to involve both sql and no-sql, take a look at the CQRS "pattern". You can tailor the write side to use no-sql but keep traditional SQL or RDBMS on the read side.
CQRS: <http://martinfowler.com/bliki/CQRS.html>
DDD: <http://domaindrivendesign.org/>
Let me know if you have any questions. | I don't think this is really a Clojure question, or am I misinterpreting it? Are you looking for the pros and cons of using a NOSQL vs. SQL database for your web application?
Either way both approaches are easy to use in Clojure, there are libraries out there.
Check out: -
[congomongo](https://github.com/aboekhoff/congomongo) |
53,475 | I'm stuck. Can't figure out how I can extrude a closed curve onto a NURBS curve without it looking tapered.
I notice that when it's set to "2D" there is no tapering, but I want to do it on 3d.
Thanks for help.[](https://i.stack.imgur.com/1JFVa.png) | 2016/05/26 | [
"https://blender.stackexchange.com/questions/53475",
"https://blender.stackexchange.com",
"https://blender.stackexchange.com/users/24983/"
] | It's a known limitation of the current system, there is currently no elegant workaround for this as far as I know.
1. Either use only separate 2D curves for each segment (rotate them in object mode to their correct positions)
2. Add extra control vertex close to corners to minimize the tapering distance (clumsy and may not look good)
[](https://i.stack.imgur.com/EwsoO.gif)
3. Or use independent splines inside the same bezier curve for each segments (may cause intersection artifacts under certain circumstances)
[](https://i.stack.imgur.com/H4a05.gif)
Choose the lesser evil for your particular use case. | In the *Object Data* tab in the *Properties panel* (pictured in your screen shot), under *Shape > Resolution: > Preview U:*, change the number from 1 to 64 (the max). Leaving the *Render U:* at zero will cause it to use the *Preview U:* (just fine).
Also, make sure that you have added enough loop cuts to your cylinder so that it deforms properly. |
66,627 | I am a newbie in photoshop and learning it all by myself, recently I came across of this image and tried to recreate the effect (I used another image though). What I did:
1. Loaded the image on one layer
2. With the Marquee tool I selected bottom part of the image and layered via copy in a new layer
3. I converted it to a smart object
4. I applied lens blur to to the layer
here is the link: (Method 3: Blur the image)
<https://medium.com/@erikdkennedy/7-rules-for-creating-gorgeous-ui-part-2-430de537ba96#.39hutajug>
Is the method that I used correct? Or can anybody suggest a better one?
Thanks | 2016/02/07 | [
"https://graphicdesign.stackexchange.com/questions/66627",
"https://graphicdesign.stackexchange.com",
"https://graphicdesign.stackexchange.com/users/58931/"
] | First of all, it depends on the region but the most common of all copyright "laws" is [**the Berne Convention for the Protection of Literary and Artistic Works**](https://en.wikipedia.org/wiki/Berne_Convention).
Which basically states that you automatically have the copyright to any work you create even without notice on the work itself. There isn't an official register of copyright.
I started with "it depends on the region" therefore here is a **[List of parties to international copyright agreements](https://en.wikipedia.org/wiki/List_of_parties_to_international_copyright_agreements)**
That's about all there is to it from what I know, you can also apply for trademarks and patents if you have something that qualifies. *Also, if you are really worried about intellectual property, you might want to seek some legal advice from a lawyer that specializes on these kinds of things.* | This may be a long ago answered question but on reading the content I noticed there was no reference to "file info". This exists in AI (found under File> file info) and mirrors the way photoshop users add metadata to each file so that it follows the digital version at least where ever that goes. No protection from thieves as such but it labels your work as yours and enables you to tag the work as copyrighted with your details and much more. Apologies if you know this it can help as far as posting work on the internet though of course will not be displayed with printed work and then Alins input is correct. |
2,734,157 | Along with producing incorrect results, one of the worst fears in scientific programming is not being able to reproduce the results you've generated. What best practices help ensure your analysis is reproducible? | 2010/04/29 | [
"https://Stackoverflow.com/questions/2734157",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/38765/"
] | Post code, data, and results on the Internet. Write the URL in the paper.
Also, submit your code to "contests". For example, in music information retrieval, there is [MIREX](http://www.music-ir.org/mirex/2009/index.php/Main_Page). | Perhaps this is slightly off topic, but to follow @Jacques Carette lead regarding scientific computing specifics, it may be helpful to consult Verification & Validation ("V&V") literature for some specific questions, especially those that blur the line between reproducibility and correctness. Now that cloud computing is becoming more of an option for large simulation problems, reproducibility among random assortment of random CPUs will be more of a concern. Additionally, I don't know if it possible to fully separate "correctness" from "reproducibility" of your results because your results stemmed from your computational model. Even though your model seems to work on computational cluster A but doesn't on cluster B, you need to follow some guidelines to guarantee your work process for making this model is sound. Specific to reproducibility, there is some buzz in the V&V community to incorporate reproducibility error into overall model uncertainty (I will let the reader investigate this on their own).
For example, for computational fluid dynamics (CFD) work, the gold standard is [the ASME V&V guide](https://www.asme.org/products/codes-standards/v-v-20-2009-standard-verification-validation). For the applied multiphysics modeling and simulation people especially (with its general concepts applicable to the greater scientific computing community), this is an important standard to internalize. |
2,734,157 | Along with producing incorrect results, one of the worst fears in scientific programming is not being able to reproduce the results you've generated. What best practices help ensure your analysis is reproducible? | 2010/04/29 | [
"https://Stackoverflow.com/questions/2734157",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/38765/"
] | I'm a software engineer embedded in a team of research geophysicists and we're currently (as always) working to improve our ability to reproduce results upon demand. Here are a few pointers gleaned from our experience:
1. Put everything under version control: source code, input data sets, makefiles, etc
2. When building executables: we embed compiler directives in the executables themselves, we tag a build log with a UUID and tag the executable with the same UUID, compute checksums for executables, autobuild everything and auto-update a database (OK, it's just a flat file really) with build details, etc
3. We use Subversion's keywords to include revision numbers (etc) in every piece of source, and these are written into any output files generated.
4. We do lots of (semi-)automated regression testing to ensure that new versions of code, or new build variants, produce the same (or similar enough) results, and I'm working on a bunch of programs to quantify the changes which do occur.
5. My geophysicist colleagues do analyse the programs sensitivities to changes in inputs. I analyse their (the codes, not the geos) sensitivity to compiler settings, to platform and such like.
We're currently working on a workflow system which will record details of every job run: input datasets (including versions), output datasets, program (incl version and variant) used, parameters, etc -- what is commonly called provenance. Once this is up and running the only way to publish results will be by use of the workflow system. Any output datasets will contain details of their own provenance, though we haven't done the detailed design of this yet.
We're quite (perhaps too) relaxed about reproducing numerical results to the least-significant digit. The science underlying our work, and the errors inherent in the measurements of our fundamental datasets, do not support the validity of any of our numerical results beyond 2 or 3 s.f.
We certainly won't be publishing either code or data for peer-review, we're in the oil business. | Plenty of good suggestions already. I'll add (both from bitter experience---*before* publication, thankfully!),
1) Check your results for stability:
------------------------------------
* try several different subsets of the data
* rebin the input
* rebin the output
* tweak the grid spacing
* try several random seeds (if applicable)
***If it's not stable, you're not done.***
Publish the results of the above testing (or at least, keep the evidence and mention that you did it).
2) Spot check the intermediate results
--------------------------------------
Yes, you're probably going to develop the method on a small sample, then grind through the whole mess. Peak into the middle a few times while that grinding is going on. Better yet, where possible collect statistics on the intermediate steps and look for signs of anomalies therein.
***Again, any surprises and you've got to go back and do it again.***
And, again, retain and/or publish this.
---
Things already mentioned that I like include
* Source control---you need it for yourself anyway.
* Logging of build environment. Publication of the same is nice.
* Plan on making code and data available.
Another one no one has mentioned:
3) Document the code
--------------------
Yes, you're busy writing it, and probably busy designing it as you go along. But I don't mean a detailed document as much as a good explanation for all the surprises. You're going to need to write those up anyway, so think of it as getting a head start on the paper. And you can keep the documentation in source control so that you can freely throw away chunks that don't apply anymore---they'll be there if you need them back.
It wouldn't hurt to build a little README with build instructions and a "How to run" blurb. If you're going to make the code available, people are going to ask about this stuff... Plus, for me, checking back with it helps me stay on track. |
2,734,157 | Along with producing incorrect results, one of the worst fears in scientific programming is not being able to reproduce the results you've generated. What best practices help ensure your analysis is reproducible? | 2010/04/29 | [
"https://Stackoverflow.com/questions/2734157",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/38765/"
] | I'm a software engineer embedded in a team of research geophysicists and we're currently (as always) working to improve our ability to reproduce results upon demand. Here are a few pointers gleaned from our experience:
1. Put everything under version control: source code, input data sets, makefiles, etc
2. When building executables: we embed compiler directives in the executables themselves, we tag a build log with a UUID and tag the executable with the same UUID, compute checksums for executables, autobuild everything and auto-update a database (OK, it's just a flat file really) with build details, etc
3. We use Subversion's keywords to include revision numbers (etc) in every piece of source, and these are written into any output files generated.
4. We do lots of (semi-)automated regression testing to ensure that new versions of code, or new build variants, produce the same (or similar enough) results, and I'm working on a bunch of programs to quantify the changes which do occur.
5. My geophysicist colleagues do analyse the programs sensitivities to changes in inputs. I analyse their (the codes, not the geos) sensitivity to compiler settings, to platform and such like.
We're currently working on a workflow system which will record details of every job run: input datasets (including versions), output datasets, program (incl version and variant) used, parameters, etc -- what is commonly called provenance. Once this is up and running the only way to publish results will be by use of the workflow system. Any output datasets will contain details of their own provenance, though we haven't done the detailed design of this yet.
We're quite (perhaps too) relaxed about reproducing numerical results to the least-significant digit. The science underlying our work, and the errors inherent in the measurements of our fundamental datasets, do not support the validity of any of our numerical results beyond 2 or 3 s.f.
We certainly won't be publishing either code or data for peer-review, we're in the oil business. | Post code, data, and results on the Internet. Write the URL in the paper.
Also, submit your code to "contests". For example, in music information retrieval, there is [MIREX](http://www.music-ir.org/mirex/2009/index.php/Main_Page). |
2,734,157 | Along with producing incorrect results, one of the worst fears in scientific programming is not being able to reproduce the results you've generated. What best practices help ensure your analysis is reproducible? | 2010/04/29 | [
"https://Stackoverflow.com/questions/2734157",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/38765/"
] | publish the program code, make it available for review.
This is not directed at you by any means, but here is my rant:
If you do work sponsored by taxpayer money, if you publish the results in peer-reviewed journal, provide the source code, under open source license or in public domain.
I am tired of reading about this great algorithm somebody came up with, which they claim does x, but provide no way to verify/check source code. if I cannot see the code, I cannot verify you results, for algorithm implementations can be very drastic differences.
It is not moral in my opinion to keep work paid by taxpayers out of reach of fellow researchers. it's against science to push papers yet provide no tangible benefit to public in terms of usable work. | Record configuration parameters somehow (eg if you can set a certain variable to a certain value). This may be in the data output, or in version control.
If you're changing your program all the time (I am!), make sure you record what version of your program you're using. |
2,734,157 | Along with producing incorrect results, one of the worst fears in scientific programming is not being able to reproduce the results you've generated. What best practices help ensure your analysis is reproducible? | 2010/04/29 | [
"https://Stackoverflow.com/questions/2734157",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/38765/"
] | Plenty of good suggestions already. I'll add (both from bitter experience---*before* publication, thankfully!),
1) Check your results for stability:
------------------------------------
* try several different subsets of the data
* rebin the input
* rebin the output
* tweak the grid spacing
* try several random seeds (if applicable)
***If it's not stable, you're not done.***
Publish the results of the above testing (or at least, keep the evidence and mention that you did it).
2) Spot check the intermediate results
--------------------------------------
Yes, you're probably going to develop the method on a small sample, then grind through the whole mess. Peak into the middle a few times while that grinding is going on. Better yet, where possible collect statistics on the intermediate steps and look for signs of anomalies therein.
***Again, any surprises and you've got to go back and do it again.***
And, again, retain and/or publish this.
---
Things already mentioned that I like include
* Source control---you need it for yourself anyway.
* Logging of build environment. Publication of the same is nice.
* Plan on making code and data available.
Another one no one has mentioned:
3) Document the code
--------------------
Yes, you're busy writing it, and probably busy designing it as you go along. But I don't mean a detailed document as much as a good explanation for all the surprises. You're going to need to write those up anyway, so think of it as getting a head start on the paper. And you can keep the documentation in source control so that you can freely throw away chunks that don't apply anymore---they'll be there if you need them back.
It wouldn't hurt to build a little README with build instructions and a "How to run" blurb. If you're going to make the code available, people are going to ask about this stuff... Plus, for me, checking back with it helps me stay on track. | Post code, data, and results on the Internet. Write the URL in the paper.
Also, submit your code to "contests". For example, in music information retrieval, there is [MIREX](http://www.music-ir.org/mirex/2009/index.php/Main_Page). |
2,734,157 | Along with producing incorrect results, one of the worst fears in scientific programming is not being able to reproduce the results you've generated. What best practices help ensure your analysis is reproducible? | 2010/04/29 | [
"https://Stackoverflow.com/questions/2734157",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/38765/"
] | I'm a software engineer embedded in a team of research geophysicists and we're currently (as always) working to improve our ability to reproduce results upon demand. Here are a few pointers gleaned from our experience:
1. Put everything under version control: source code, input data sets, makefiles, etc
2. When building executables: we embed compiler directives in the executables themselves, we tag a build log with a UUID and tag the executable with the same UUID, compute checksums for executables, autobuild everything and auto-update a database (OK, it's just a flat file really) with build details, etc
3. We use Subversion's keywords to include revision numbers (etc) in every piece of source, and these are written into any output files generated.
4. We do lots of (semi-)automated regression testing to ensure that new versions of code, or new build variants, produce the same (or similar enough) results, and I'm working on a bunch of programs to quantify the changes which do occur.
5. My geophysicist colleagues do analyse the programs sensitivities to changes in inputs. I analyse their (the codes, not the geos) sensitivity to compiler settings, to platform and such like.
We're currently working on a workflow system which will record details of every job run: input datasets (including versions), output datasets, program (incl version and variant) used, parameters, etc -- what is commonly called provenance. Once this is up and running the only way to publish results will be by use of the workflow system. Any output datasets will contain details of their own provenance, though we haven't done the detailed design of this yet.
We're quite (perhaps too) relaxed about reproducing numerical results to the least-significant digit. The science underlying our work, and the errors inherent in the measurements of our fundamental datasets, do not support the validity of any of our numerical results beyond 2 or 3 s.f.
We certainly won't be publishing either code or data for peer-review, we're in the oil business. | I think a lot of the previous answers missed the "scientific computing" part of your question, and answered with very general stuff that applies to any science (make the data and method public, specialized to CS).
What they're missing is that you have to be even more specialized: you have to specific which version of the compiler you used, which switches were used when compiling, which version of the operating system you used, which versions of all the libraries you linked against, what hardware you are using, what else was going being run on your machine at the same time, and so forth. There are published papers out there where every one of these factors influenced the results in a non-trivial way.
For example (on Intel hardware) you could be using a library which uses the FPU's 80-bit floats, do an O/S upgrade, and now that library might now only use 64-bit doubles, and your results can drastically change if your problem was the least bit ill-conditioned.
A compiler upgrade might change the default rounding behaviour, or a single optimization might flip in which order 2 instructions get done, and again for ill-conditioned systems, boom, different results.
Heck, there are some funky stories of sub-optimal algorithms showing 'best' in practical tests because they were tested on a laptop which automatically slowed down the CPU when it overheated (which is what the optimal algorithm did).
None of these things are visible from the source code or the data. |
2,734,157 | Along with producing incorrect results, one of the worst fears in scientific programming is not being able to reproduce the results you've generated. What best practices help ensure your analysis is reproducible? | 2010/04/29 | [
"https://Stackoverflow.com/questions/2734157",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/38765/"
] | I'm a software engineer embedded in a team of research geophysicists and we're currently (as always) working to improve our ability to reproduce results upon demand. Here are a few pointers gleaned from our experience:
1. Put everything under version control: source code, input data sets, makefiles, etc
2. When building executables: we embed compiler directives in the executables themselves, we tag a build log with a UUID and tag the executable with the same UUID, compute checksums for executables, autobuild everything and auto-update a database (OK, it's just a flat file really) with build details, etc
3. We use Subversion's keywords to include revision numbers (etc) in every piece of source, and these are written into any output files generated.
4. We do lots of (semi-)automated regression testing to ensure that new versions of code, or new build variants, produce the same (or similar enough) results, and I'm working on a bunch of programs to quantify the changes which do occur.
5. My geophysicist colleagues do analyse the programs sensitivities to changes in inputs. I analyse their (the codes, not the geos) sensitivity to compiler settings, to platform and such like.
We're currently working on a workflow system which will record details of every job run: input datasets (including versions), output datasets, program (incl version and variant) used, parameters, etc -- what is commonly called provenance. Once this is up and running the only way to publish results will be by use of the workflow system. Any output datasets will contain details of their own provenance, though we haven't done the detailed design of this yet.
We're quite (perhaps too) relaxed about reproducing numerical results to the least-significant digit. The science underlying our work, and the errors inherent in the measurements of our fundamental datasets, do not support the validity of any of our numerical results beyond 2 or 3 s.f.
We certainly won't be publishing either code or data for peer-review, we're in the oil business. | Perhaps this is slightly off topic, but to follow @Jacques Carette lead regarding scientific computing specifics, it may be helpful to consult Verification & Validation ("V&V") literature for some specific questions, especially those that blur the line between reproducibility and correctness. Now that cloud computing is becoming more of an option for large simulation problems, reproducibility among random assortment of random CPUs will be more of a concern. Additionally, I don't know if it possible to fully separate "correctness" from "reproducibility" of your results because your results stemmed from your computational model. Even though your model seems to work on computational cluster A but doesn't on cluster B, you need to follow some guidelines to guarantee your work process for making this model is sound. Specific to reproducibility, there is some buzz in the V&V community to incorporate reproducibility error into overall model uncertainty (I will let the reader investigate this on their own).
For example, for computational fluid dynamics (CFD) work, the gold standard is [the ASME V&V guide](https://www.asme.org/products/codes-standards/v-v-20-2009-standard-verification-validation). For the applied multiphysics modeling and simulation people especially (with its general concepts applicable to the greater scientific computing community), this is an important standard to internalize. |
2,734,157 | Along with producing incorrect results, one of the worst fears in scientific programming is not being able to reproduce the results you've generated. What best practices help ensure your analysis is reproducible? | 2010/04/29 | [
"https://Stackoverflow.com/questions/2734157",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/38765/"
] | * Publish the original raw data online and make it freely available for download.
* Make the code base open source and available online for download.
* If randomization is used in optimization, then repeat the optimization several times, choosing the best value that results or use a fixed random seed, so that the same results are repeated.
* Before performing your analysis, you should split the data into a "training/analysis" dataset and a "testing/validation" dataset. Perform your analsysis on the "training" dataset, and make sure that the results that you get still hold on the "validation" dataset to ensure that your analysis is actually generalizable and isn't simply memorizing peculiarities of the dataset in question.
The first two points are incredibly important, because making the dataset available allows others to perform their own analyses on the same data, which increases the level of confidence in the validity of your own analyses. Additionally, making the dataset available online -- especially if you use linked data formats -- makes it possible for crawlers to aggregate your dataset with other datasets, thereby enabling analyses with larger data sets... in many types of research, the sample size is sometimes too small to be really confident about the results... but sharing your dataset makes it possible to construct very large datasets. Or, someone could use your dataset to validate the analysis that they performed on some other dataset.
Additionally, making your code open source makes it possible for the code and procedure to be reviewed by your peers. Often such reviews lead to the discovery of flaws or of the possibility for additional optimization and improvement. Most importantly, it allows other researchers to improve on your methods, without having to implement everything that you have already done from scratch. It very greatly accelerates the pace of research when researches can focus on just improvements and not on reinventing the wheel.
As for randomization... many algorithms rely on randomization to achieve their results. Stochastic and Monte Carlo methods are quite common, and while they have been proven to converge for certain cases, it is still possible to get different results. The way to ensure that you get the same results, is to have a loop in your code that invokes the computation some fixed number of times, and to choose the best result. If you use enough repititions, you can expect to find global or near-global optima instead of getting stuck in local optima. Another possibility is to use a predetermined seed, although that is not, IMHO, as good an approach since you could pick a seed that causes you to get stuck in local optima. In addition, there is no guarantee that random number generators on different platforms will generate the same results for that seed value. | Record configuration parameters somehow (eg if you can set a certain variable to a certain value). This may be in the data output, or in version control.
If you're changing your program all the time (I am!), make sure you record what version of your program you're using. |
2,734,157 | Along with producing incorrect results, one of the worst fears in scientific programming is not being able to reproduce the results you've generated. What best practices help ensure your analysis is reproducible? | 2010/04/29 | [
"https://Stackoverflow.com/questions/2734157",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/38765/"
] | Record configuration parameters somehow (eg if you can set a certain variable to a certain value). This may be in the data output, or in version control.
If you're changing your program all the time (I am!), make sure you record what version of your program you're using. | Perhaps this is slightly off topic, but to follow @Jacques Carette lead regarding scientific computing specifics, it may be helpful to consult Verification & Validation ("V&V") literature for some specific questions, especially those that blur the line between reproducibility and correctness. Now that cloud computing is becoming more of an option for large simulation problems, reproducibility among random assortment of random CPUs will be more of a concern. Additionally, I don't know if it possible to fully separate "correctness" from "reproducibility" of your results because your results stemmed from your computational model. Even though your model seems to work on computational cluster A but doesn't on cluster B, you need to follow some guidelines to guarantee your work process for making this model is sound. Specific to reproducibility, there is some buzz in the V&V community to incorporate reproducibility error into overall model uncertainty (I will let the reader investigate this on their own).
For example, for computational fluid dynamics (CFD) work, the gold standard is [the ASME V&V guide](https://www.asme.org/products/codes-standards/v-v-20-2009-standard-verification-validation). For the applied multiphysics modeling and simulation people especially (with its general concepts applicable to the greater scientific computing community), this is an important standard to internalize. |
2,734,157 | Along with producing incorrect results, one of the worst fears in scientific programming is not being able to reproduce the results you've generated. What best practices help ensure your analysis is reproducible? | 2010/04/29 | [
"https://Stackoverflow.com/questions/2734157",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/38765/"
] | I'm a software engineer embedded in a team of research geophysicists and we're currently (as always) working to improve our ability to reproduce results upon demand. Here are a few pointers gleaned from our experience:
1. Put everything under version control: source code, input data sets, makefiles, etc
2. When building executables: we embed compiler directives in the executables themselves, we tag a build log with a UUID and tag the executable with the same UUID, compute checksums for executables, autobuild everything and auto-update a database (OK, it's just a flat file really) with build details, etc
3. We use Subversion's keywords to include revision numbers (etc) in every piece of source, and these are written into any output files generated.
4. We do lots of (semi-)automated regression testing to ensure that new versions of code, or new build variants, produce the same (or similar enough) results, and I'm working on a bunch of programs to quantify the changes which do occur.
5. My geophysicist colleagues do analyse the programs sensitivities to changes in inputs. I analyse their (the codes, not the geos) sensitivity to compiler settings, to platform and such like.
We're currently working on a workflow system which will record details of every job run: input datasets (including versions), output datasets, program (incl version and variant) used, parameters, etc -- what is commonly called provenance. Once this is up and running the only way to publish results will be by use of the workflow system. Any output datasets will contain details of their own provenance, though we haven't done the detailed design of this yet.
We're quite (perhaps too) relaxed about reproducing numerical results to the least-significant digit. The science underlying our work, and the errors inherent in the measurements of our fundamental datasets, do not support the validity of any of our numerical results beyond 2 or 3 s.f.
We certainly won't be publishing either code or data for peer-review, we're in the oil business. | Record configuration parameters somehow (eg if you can set a certain variable to a certain value). This may be in the data output, or in version control.
If you're changing your program all the time (I am!), make sure you record what version of your program you're using. |
497,807 | The sentence, "A younger me sat outside," though colloquially acceptable appears to be contrary to the rules of English pronomial declension. However, the allegedly more correct sentence "A younger I sat outside" seems jarringly wrong.
Is there citation for the former in formal prose, or the latter at all? Do any style guides comment on the issue? Perhaps the pronoun takes on a more abstract, conceptual definition of the self that is conveyed only by "me" and not by "I." | 2019/05/09 | [
"https://english.stackexchange.com/questions/497807",
"https://english.stackexchange.com",
"https://english.stackexchange.com/users/347712/"
] | This phrasing appears a lot. In a subject position, phrases like "a younger me" function as ordinary singular noun phrases rather than as pronouns.
First, from Wil McCarthy. "The Policeman's Daughter." *Analog*, vol. 125, iss. 6, June 2005, p. 8-34, as found on the [Corpus of Contemporary American English](https://www.english-corpora.org/coca/).
>
> Even **a younger me**, **a green me** fresh out of school, **is better qualified** than most attorneys to wrestle this particular alligator.
>
>
>
In this case, *me* (the first person pronoun in the objective case) is being treated as a noun referring to the speaker. The verb *is* corresponds to any singular third person noun phrase, rather than to a first person noun phrase, further signalling that this isn't a usual pronominal usage. Finally, it takes an indefinite article, specifying that this is one of several possible **me**s who fits the criterion of each adjective (younger, green). Such objectification works in a speculative fiction story that involves the technology of copying oneself.
Second, from Tom Ough. "[Sir Richard Branson: 'If anything goes wrong, I put it out of my mind and move on](https://www.telegraph.co.uk/men/thinking-man/sir-richard-branson-anything-goes-wrong-put-mind-move/).' " *The Telegraph*, 26 Nov. 2017.
>
> **A younger me might have been disappointed to know** I’ve remained a dreadful driver.
>
>
>
Because the adjective modifies something about the speaker (younger, fitter, stronger), the usage commonly precedes modal verb phrases like "might have ..." and "would have ..." ([News on the Web](https://www.english-corpora.org/now/) features several further results like this from credible publications.) Talking about oneself in this way allows a speaker to speculate about their performance without using a more verbose turn of phrase:
>
> When I was younger, I might have been disappointed to know ...
>
>
> A younger version of me might have been disappointed to know ...
>
>
>
---
I can't find the usage "a younger me" earlier than the 20th century ([Google Books search](https://www.google.com/search?q=%22a%20younger%20me%22&tbm=bks&source=lnt&tbs=cdr:1,cd_min:1900,cd_max:1999&sa=X&ved=0ahUKEwj40_Wn0-HiAhWnTt8KHe1mCUgQpwUIIg&biw=1252&bih=600&dpr=1.09)). While adjective-me constructions are much older than that ("And make a conquest of [unhappy me](http://www.opensourceshakespeare.org/views/plays/play_view.php?WorkID=pericles&Act=1&Scene=4&Scope=scene)" is from Shakespeare), the me-as-reference-to-myself crops up in the 19th century (Thomas Carlyle wrote in 1828, "[Haunted and blinded by some shadow of his own little Me](https://books.google.com/books?id=QGYJAAAAQAAJ&pg=PA115&lpg=PA115&dq=Haunted%20and%20blinded%20by%20some%20shadow%20of%20his%20own%20little%20Me.&source=bl&ots=294TKx9F_K&sig=ACfU3U2xSu4hF_eiA_uMRWrp2nscGHOwJw&hl=en&sa=X&ved=2ahUKEwjimOiH1eHiAhUlnOAKHZmWCbwQ6AEwA3oECAgQAQ#v=onepage&q=Haunted%20and%20blinded%20by%20some%20shadow%20of%20his%20own%20little%20Me.&f=false)"). Here's another example from Rudyard Kipling, "Divided Destinies," orig. 1886 ([here from 1900](https://books.google.com/books?id=BNkeAAAAMAAJ&pg=PA192&lpg=PA192&dq=Kipling%20%20An%20inscrutable%20Decree%20Makes%20thee%20a%20gleesome%20fleasome%20Thou,%20and%20me%20a%20wretched%20Me.&source=bl&ots=3KH_C-0T1l&sig=ACfU3U0PkPlEnWCwYOapT6t2hGV2uOAG-g&hl=en&sa=X&ved=2ahUKEwjc3vP91eHiAhUDJt8KHZv4AoQQ6AEwBHoECAkQAQ#v=onepage&q=Kipling%20%20An%20inscrutable%20Decree%20Makes%20thee%20a%20gleesome%20fleasome%20Thou%2C%20and%20me%20a%20wretched%20Me.&f=false)):
>
> An inscrutable Decree Makes thee a gleesome fleasome Thou, and me **a wretched Me**.
>
>
>
Note the switch to the nominative from "thee" to "thou" but the use of "me" both times. This specific usage is peculiar to "me."
Why "me" and not "I?" Dictionaries mainly indicate that I is unlikely to be encountered in this way. The [Oxford English Dictionary](https://www-oed-com.libproxy.ung.edu/view/Entry/115378?rskey=kO2cOt&result=6#eid), "me, pron.1, n., and adj." documents several uses of "me" with adjectives and various articles under "B. n." (the Shakespeare, Carlyle, and Kipling quotes are all from here). However, there is no specific entry for uses like "a younger me," perhaps because it is recent enough to not be documented by lexicographers.
Meanwhile, there appears to be even less on noun uses of "[I](https://www.oed.com/view/Entry/90671)" with adjectives - a few entries, a few examples, and none in the spirit of "a younger I." In the example sentences, the "a/an" article never appears, and the adjectives are primarily "other," showing a more restrictive pattern than examples of "me." So the OED suggests - in an absence of precedence - that "a younger I" wouldn't work.
Corpus searches provide further evidence of absence. "A younger I" doesn't turn up in COCA or NOW as a distinct noun phrase.
Why are "a younger I" and similar uses rare or nonexistent? It could be arbitrary. It could also be something about the case structure of the first person pronoun that noun-me comes from: speakers may use the objective case to create distance between I-as-speaker and me-as-addressee. Since in these examples "I" is the position of the speaker or writer, *me* creates a discontinuity between the speaker and the younger/green/better qualified/unhappy/wretched self. In other words, **me** provides further disambiguation for readers who would expect the speaker's own thoughts and actions to come in first-person "I" in the nominative case. | *The Cambridge Grammar of the English Language* by Huddleston and Pullum has a subsection "Pre-head internal dependents normally excluded too" (Page 430):
>
> Pronouns do not normally allow internal pre-head dependents: \**Extravagant he bought a new car*; \**I met interesting them all*. The qualification ‘normally’ caters for one minor exception, the use of a few adjectives such as *lucky*, *poor*, *silly* with the core personal pronouns:
>
>
> [12] i *Lucky you! No one noticed you had gone home early.*
>
>
> ii *They decided it would have to be done by poor old me.*
>
>
> The adjective is semantically non-restrictive, and the NP characteristically stands alone as an exclamation, as in [i]. **It can be integrated into clause structure, as in [ii], but not as subject** (\**Poor you have got the night shift again*). **The pronoun must be in accusative or plain case** (compare *Silly me!* and \**Silly I!*).
>
>
>
(Emphasis mine.) |
620,835 | As an layman and outsider who has read some of Dirac, I want an understanding of how important absolute size is to quantum mechanics - like wondering if it is a necessary or sufficient condition (along these lines).
As far as I understand things, like all good theories, quantum mechanics is a mix of empirical data (many of which can't be explained classically), and seasoned induction/intuition/logic. This is where I want to see how absolute size fits in. Is the notion of there being absolute smallness (a scale where there is no way to cause a non-neglibile disturbance upon interaction with it by any means) doing most of the legwork for quantum mechanics?
Would I come up with something resembling quantum mechanics if I ran similar experiments with massive apparatuses like scattering bowling balls, so that I were disturbing *everything* non-negligibly?
Then, why can I not quantify the disturbance (i.e. knowing the momentum, time, etc of my bowling balls) and retrieve a *realistic* picture (as in realism). Sure I may disturb any system I measure like measuring a bird's velocity, but I know the weight of the bird roughly (all birds' momentums are within a few orders of magnitude), I know the details of my bowling ball, so can't I retrieve a realistic, deterministic picture of the world? Here it seems like if you can quantify the disturbance then the non-neglibility or not of it seems no longer important.
Why can I not quantify the disturbances then in quantum mechanics and regain determinism during the measurement process?
I would never dream of anything other than realism in the bowling ball world. Where do I need to begin to ponder something other than realism, as I think is required in the Copenhagen-like interpretations? The fact that there is no sub-photon scale? There is a sub-bowling-ball scale. Is this difference where and why quantum mechanics gets "weird"? That there are limits to all empirical investigations. But if I can reproduce so to speak, a lot of quantum mechanics with a bowling ball world, why not believe quantum mechanics can be made likewise deterministic. There must be some other weirdness than absolute size right? So is absolute size a red herring, neither sufficient nor necessary? And the legwork is really superposition of states and entanglement, which must be understood agnostic to absolute size? And thus those are what force us to question realism? | 2021/03/13 | [
"https://physics.stackexchange.com/questions/620835",
"https://physics.stackexchange.com",
"https://physics.stackexchange.com/users/255220/"
] | There is a fundamental constant of nature that establishes the scale over which quantum effects become dominant and readily measured with special tools. It is called *Planck's constant* and it is a very, very tiny number, which means that quantum effects like uncertainty only kick in at very, very tiny length scales.
It is of course possible to apply those same quantum uncertainty rules to macroscopic objects like bowling balls, but the tininess of Planck's constant guarantees that at those large scales, the quantum effects are so very, very, *very* tiny that there is no possible way to measure them- as pointed out by PM 2Ring in his comment.
Note here that if Planck's constant were zero, all quantum effects would vanish, and if it were big, then we would experience quantum weirdness in our everyday lives. The physicist George Gamow wrote a series of books for non-physicists in which the protagonist, Mister Thompkins, gets to experience worlds in which for example Planck's constant is instead large, and explores the (bizarre) consequences. | When you go bowling do you invariably score a strike every time? I would guess not.
All the inaccuracies in your aiming and throwing of the ball are quantum effects becoming visible in the chaotic system that your body is. By practicing you can improve the signal to noise ratio of your performance, but you can never reach 100%. |
8,475,265 | I need to copy files to and from some windows network share, like \\compname\admin$ for example.
Right now we use JNI and open a connection using WNetAddConnection2 and then WinAPIs CopyFile to operate.
Source machine has Windows OS.
Isn't there another simpler Java way? | 2011/12/12 | [
"https://Stackoverflow.com/questions/8475265",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/579828/"
] | [JCIFS](http://jcifs.samba.org/) is an Open Source client library that implements the CIFS/SMB networking protocol in 100% Java. | A solution we have involves the j-interop library and WMI. May be worth a look. However, we are coming from Linux copying to Windows. I don't know what platform your source is from the question. |
456,690 | I have a KY-005 infrared transmitter connected to my Arduino Uno. But currently I can't get it to work. On [this](https://arduinomodules.info/ky-005-infrared-transmitter-sensor-module/) website it says that you must connect the led directly to a digital pin on the Arduino, which I have not done (I connected it in series with a 220 ohm resistor. Because i don’t Thrust the led can handle it)
Is it correct that you can/must connect it directly?
Update: i hooked up the led directly to the arduino pin 3 (not +5V) and that did. it now works | 2019/09/08 | [
"https://electronics.stackexchange.com/questions/456690",
"https://electronics.stackexchange.com",
"https://electronics.stackexchange.com/users/230525/"
] | According to your link "The KY-005 Infrared Transmitter Module consists of **just** a 5mm IR LED.". This is confirmed in [this YouTube video](https://www.youtube.com/watch?v=hYw4CTPsUNg) (image below is from the video).
[](https://i.stack.imgur.com/JbqBK.jpg)
Infrared is not visible to the human eye, but you can check to see if it is emitting
light using the camera in a cell phone or laptop computer.
Connecting it directly to the Arduino I/O pin produces the most light, but relies on the internal resistance of the port to limit current. This is safe provided no other pins are driving heavy loads (ATmega328 absolute maximum current for all MCU pins combined is 200mA). It should still work with a resistor, but with shorter range. With 220Ω it should draw about 15mA, which is around half the current it draws with a direct connection.
Note: do **not** connect the 'signal' input directly to +5V, as it will burn out the LED! | The operating voltage of the KY-005 is 5V, according to the web site, and it can be driven directly from an Arduino output. You should be able to detect it with the series resistor, perhaps you have a software problem. |
107,827 | First of all, I am aware that its impossible for a planet with less gravity than the Earth, to naturally maintain an atmosphere with the same terrestrial gases, in an equal or greater proportion. Therefore, this proportion of gases is maintained artificially.
Second, although I have notions of what the ballistic coefficient implies, I have no idea how to calculate it or in what proportion its affected by the atmospheric density and gases of an atmosphere.
Following with this line of thought, we have the force of gravity of the planet X that is 8.04 m / s².
Let us assume then that the atmospheric pressure of this planet, whose atmosphere is being altered artificially, is approximately 2 atm.
Now with all the elements previously presented, I would like to know two things:
* 1) Would this atmosphere sustain intelligent life in a realistic way?
It is necessary to know if a human civilization could develop there.
* 2) Would the effect of this atmosphere be enough to significantly
affect the ballistic coefficient, to the point where its
preferable to a civilization that has discovered firearms, have to
continue using crossbows for more effective shots at a distance?
Note:Take into account side effects that may present apparently inert gases at high levels of partial pressure, such as nitrogen narcosis. | 2018/03/25 | [
"https://worldbuilding.stackexchange.com/questions/107827",
"https://worldbuilding.stackexchange.com",
"https://worldbuilding.stackexchange.com/users/48880/"
] | >
> an atmosphere with the same terrestrial gases, in an equal or greater proportion.
>
>
>
78% Nitrogen
21% Oxygen
01% Other stuff (mostly Argon)
>
> the atmospheric pressure of this planet, whose atmosphere is being altered artificially, is approximately **2 atm**.
>
>
>
Note that since the *partial pressure* of oxygen would effectively twice that of Earth, fires would burn faster.
>
> Would this atmosphere sustain intelligent life in a realistic way?
>
>
>
Native life would/could have adapted to it, so there's no reason to say "no" to this hypothetical question. So... **yes.**
>
> It is necessary to know if a human civilization could develop there.
>
>
>
If these are "humans which evolved on Earth, and then transported to this planet", then the answer is **maybe.** That's because nitrogen narcosis can begin even at relatively low depths when scuba diving, and 2 ATM of pressure is the same as diving to 10m.
Thus, enough people just might be loopy enough to die off within a few generations.
>
> Would the effect of this atmosphere be enough to significantly affect the ballistic coefficient, to the point where its preferable to a civilization that has discovered firearms, have to continue using crossbows for more effective shots at a distance?
>
>
>
The question is flawed, because we didn't develop firearms because they shot *better* and *farther*. We developed them because they were **easier**.
~~Note that since there's so much more oxygen at 2 ATM, the gunpowder would burn a *lot* faster, generating a *lot* more pressure in the gun breaches. It's very possible **that** would have delayed the development of firearms (and cannons) until the development of better metallurgy.~~ | Firearms were adopted because they were easier to train large numbers of people to use quickly than long or recurve bows. They replaced crossbows because a crossbow bolt, fired from a steel crossbow and drawn with a spanning mechanism could provide about 200J of energy at the target, while an arquebus imparted enough energy to the projectile to deliver 1000J of energy, an *order of magnitude* difference.
[](https://i.stack.imgur.com/UfJFX.jpg)
*Spanning (or drawing) a steel crossbow. This gets you to 200J of energy*
[](https://i.stack.imgur.com/7K2is.jpg)
*A Medieval soldier with a firearm. A 1000J of energy can deal with even heavy plate armour*
Any crossbow which can deliver that much energy would be either far too large to carry or use in any practical manner, or require such a massive and powerful spanning mechanism that would also be impractical for use in battle outside of special situations.
So regardless of if you are fighting on Earth, the hypothetical planet, or even the vacuum of space (both a crossbow and a firearm will work in a vacuum), a firearm will always outperform a bow. |
107,827 | First of all, I am aware that its impossible for a planet with less gravity than the Earth, to naturally maintain an atmosphere with the same terrestrial gases, in an equal or greater proportion. Therefore, this proportion of gases is maintained artificially.
Second, although I have notions of what the ballistic coefficient implies, I have no idea how to calculate it or in what proportion its affected by the atmospheric density and gases of an atmosphere.
Following with this line of thought, we have the force of gravity of the planet X that is 8.04 m / s².
Let us assume then that the atmospheric pressure of this planet, whose atmosphere is being altered artificially, is approximately 2 atm.
Now with all the elements previously presented, I would like to know two things:
* 1) Would this atmosphere sustain intelligent life in a realistic way?
It is necessary to know if a human civilization could develop there.
* 2) Would the effect of this atmosphere be enough to significantly
affect the ballistic coefficient, to the point where its
preferable to a civilization that has discovered firearms, have to
continue using crossbows for more effective shots at a distance?
Note:Take into account side effects that may present apparently inert gases at high levels of partial pressure, such as nitrogen narcosis. | 2018/03/25 | [
"https://worldbuilding.stackexchange.com/questions/107827",
"https://worldbuilding.stackexchange.com",
"https://worldbuilding.stackexchange.com/users/48880/"
] | >
> Would the effect of this atmosphere be enough to significantly affect the ballistic coefficient
>
>
>
No, and at a first glance neither would the higher oxygen pressure.
But you might imagine have large quantities of dry pollen floating in the denser atmosphere. This would require a great care in lighting fires (fireplaces would need something like a [flame arrester](https://en.wikipedia.org/wiki/Flame_arrester) mesh).
But in the open, firing a gun *might* be a deadly mistake - unless it had just stopped raining, or the weather conditions or season or geographical features otherwise allowed it. The combination of high oxygen pressure and suspended combustible particles would turn the volume around the shooter into a [**dust bomb**](https://en.wikipedia.org/wiki/Coal_dust#Explosions), a weak form of fuel-air bomb. With the shooter at the center.
In Earth atmosphere, it takes some doing for a dust explosion to take place - but if memory serves, a bunch of probationary cooks engaging in a flour battle near cooking fires in a kitchen in Paris was enough to kill one and maim two others, and wreck the kitchen.
Combine some local plant like [lycopodium](https://en.wikipedia.org/wiki/Lycopodium_powder), but *more so*, and a higher oxygen availability, and wild gun shooting loses a lot of its appeal. | Firearms were adopted because they were easier to train large numbers of people to use quickly than long or recurve bows. They replaced crossbows because a crossbow bolt, fired from a steel crossbow and drawn with a spanning mechanism could provide about 200J of energy at the target, while an arquebus imparted enough energy to the projectile to deliver 1000J of energy, an *order of magnitude* difference.
[](https://i.stack.imgur.com/UfJFX.jpg)
*Spanning (or drawing) a steel crossbow. This gets you to 200J of energy*
[](https://i.stack.imgur.com/7K2is.jpg)
*A Medieval soldier with a firearm. A 1000J of energy can deal with even heavy plate armour*
Any crossbow which can deliver that much energy would be either far too large to carry or use in any practical manner, or require such a massive and powerful spanning mechanism that would also be impractical for use in battle outside of special situations.
So regardless of if you are fighting on Earth, the hypothetical planet, or even the vacuum of space (both a crossbow and a firearm will work in a vacuum), a firearm will always outperform a bow. |
23,439 | I need to run 4 instances of Mac OS X desktop (10.4 to 10.7) for our continuous integration setup (so they need to be on all the time). I've used PC hypervisors in the past (XenServer, ESXi, etc) but never for a Mac.
Is it possible to run those guest operating systems on a Mac Mini hypervisor?
***Edit:*** Ideally, it should be managable remotely (like XenServer and ESXi), desktop virtualization software (like VirtualBox) isn't really what I had in mind. | 2011/08/28 | [
"https://apple.stackexchange.com/questions/23439",
"https://apple.stackexchange.com",
"https://apple.stackexchange.com/users/10732/"
] | You might take a look at VirtualBox which includes a GNU GPL version as well. It offers limited/experimental capabilities to run virtualized OS X VMs, but I think Apple generally forbids virtualizing OS X via their licensing/usage terms. | There are licensing issues with the scheme that you propose. See this [previous answer](https://apple.stackexchange.com/questions/19939/where-can-i-read-the-full-lion-eula) regarding the Mac OS X EULA, which will lead you to additional information. Summary: you can't do what you want as virtualization, as versions prior to 10.7 were not available for virtualization at all per their license.
Neither VMWare or Parallels will talk publicly about circumventing the Apple EULA for pre-10.7 versions. VMWare Fusion actively prevents virtualizing the OS X client versions (there are one or more workarounds that I've seen; Google is your friend here).
I know of no hypervisor software that will allow you to accomplish your goal, either. Sorry. |
23,439 | I need to run 4 instances of Mac OS X desktop (10.4 to 10.7) for our continuous integration setup (so they need to be on all the time). I've used PC hypervisors in the past (XenServer, ESXi, etc) but never for a Mac.
Is it possible to run those guest operating systems on a Mac Mini hypervisor?
***Edit:*** Ideally, it should be managable remotely (like XenServer and ESXi), desktop virtualization software (like VirtualBox) isn't really what I had in mind. | 2011/08/28 | [
"https://apple.stackexchange.com/questions/23439",
"https://apple.stackexchange.com",
"https://apple.stackexchange.com/users/10732/"
] | You might take a look at VirtualBox which includes a GNU GPL version as well. It offers limited/experimental capabilities to run virtualized OS X VMs, but I think Apple generally forbids virtualizing OS X via their licensing/usage terms. | You have confusion between PC and Mac, they are both x86/x64, they are the same as John Hodgman and Justin Long have demonstrated before. The compatibility concern should only arise with older PowerPC based Macs, and the EFI boot process instead of BIOS.
Therefore XenServer would run fine, ESX 4.0 does not support EFI as discussed here:
<http://communities.intel.com/thread/3909>
Recently discussion about vSphere apparently supporting EFI on XServe:
<http://lists.apple.com/archives/macos-x-server/2011/Jul/msg00080.html>
>
> What's New in VMware vSphere 5?
> \* Support for Apple products -- vSphere 5 supports Apple Xserve
> servers running OS X Server 10.6 (Snow Leopard) as a guest operating
> system.
>
>
> To read for yourself, the What's New PDF can be downloaded from this
> web page "*VMware vSphere for Small and Midsized Busienss*":
>
>
> <http://www.vmware.com/products/vsphere/small-business/overview.html>
>
>
> |
23,439 | I need to run 4 instances of Mac OS X desktop (10.4 to 10.7) for our continuous integration setup (so they need to be on all the time). I've used PC hypervisors in the past (XenServer, ESXi, etc) but never for a Mac.
Is it possible to run those guest operating systems on a Mac Mini hypervisor?
***Edit:*** Ideally, it should be managable remotely (like XenServer and ESXi), desktop virtualization software (like VirtualBox) isn't really what I had in mind. | 2011/08/28 | [
"https://apple.stackexchange.com/questions/23439",
"https://apple.stackexchange.com",
"https://apple.stackexchange.com/users/10732/"
] | You might take a look at VirtualBox which includes a GNU GPL version as well. It offers limited/experimental capabilities to run virtualized OS X VMs, but I think Apple generally forbids virtualizing OS X via their licensing/usage terms. | I currently run 10.7 (was originally 10.6 then upgraded it) in ESXi 5 on my Mac Mini (Upgraded from 4GB RAM to 16GB) without any issues.
The performance isn't great when using the actual desktop, but for your requirement (I'm doing the same, using the VMs as build bots) it works absolutely fine. |
23,439 | I need to run 4 instances of Mac OS X desktop (10.4 to 10.7) for our continuous integration setup (so they need to be on all the time). I've used PC hypervisors in the past (XenServer, ESXi, etc) but never for a Mac.
Is it possible to run those guest operating systems on a Mac Mini hypervisor?
***Edit:*** Ideally, it should be managable remotely (like XenServer and ESXi), desktop virtualization software (like VirtualBox) isn't really what I had in mind. | 2011/08/28 | [
"https://apple.stackexchange.com/questions/23439",
"https://apple.stackexchange.com",
"https://apple.stackexchange.com/users/10732/"
] | There are licensing issues with the scheme that you propose. See this [previous answer](https://apple.stackexchange.com/questions/19939/where-can-i-read-the-full-lion-eula) regarding the Mac OS X EULA, which will lead you to additional information. Summary: you can't do what you want as virtualization, as versions prior to 10.7 were not available for virtualization at all per their license.
Neither VMWare or Parallels will talk publicly about circumventing the Apple EULA for pre-10.7 versions. VMWare Fusion actively prevents virtualizing the OS X client versions (there are one or more workarounds that I've seen; Google is your friend here).
I know of no hypervisor software that will allow you to accomplish your goal, either. Sorry. | You have confusion between PC and Mac, they are both x86/x64, they are the same as John Hodgman and Justin Long have demonstrated before. The compatibility concern should only arise with older PowerPC based Macs, and the EFI boot process instead of BIOS.
Therefore XenServer would run fine, ESX 4.0 does not support EFI as discussed here:
<http://communities.intel.com/thread/3909>
Recently discussion about vSphere apparently supporting EFI on XServe:
<http://lists.apple.com/archives/macos-x-server/2011/Jul/msg00080.html>
>
> What's New in VMware vSphere 5?
> \* Support for Apple products -- vSphere 5 supports Apple Xserve
> servers running OS X Server 10.6 (Snow Leopard) as a guest operating
> system.
>
>
> To read for yourself, the What's New PDF can be downloaded from this
> web page "*VMware vSphere for Small and Midsized Busienss*":
>
>
> <http://www.vmware.com/products/vsphere/small-business/overview.html>
>
>
> |
23,439 | I need to run 4 instances of Mac OS X desktop (10.4 to 10.7) for our continuous integration setup (so they need to be on all the time). I've used PC hypervisors in the past (XenServer, ESXi, etc) but never for a Mac.
Is it possible to run those guest operating systems on a Mac Mini hypervisor?
***Edit:*** Ideally, it should be managable remotely (like XenServer and ESXi), desktop virtualization software (like VirtualBox) isn't really what I had in mind. | 2011/08/28 | [
"https://apple.stackexchange.com/questions/23439",
"https://apple.stackexchange.com",
"https://apple.stackexchange.com/users/10732/"
] | There are licensing issues with the scheme that you propose. See this [previous answer](https://apple.stackexchange.com/questions/19939/where-can-i-read-the-full-lion-eula) regarding the Mac OS X EULA, which will lead you to additional information. Summary: you can't do what you want as virtualization, as versions prior to 10.7 were not available for virtualization at all per their license.
Neither VMWare or Parallels will talk publicly about circumventing the Apple EULA for pre-10.7 versions. VMWare Fusion actively prevents virtualizing the OS X client versions (there are one or more workarounds that I've seen; Google is your friend here).
I know of no hypervisor software that will allow you to accomplish your goal, either. Sorry. | I currently run 10.7 (was originally 10.6 then upgraded it) in ESXi 5 on my Mac Mini (Upgraded from 4GB RAM to 16GB) without any issues.
The performance isn't great when using the actual desktop, but for your requirement (I'm doing the same, using the VMs as build bots) it works absolutely fine. |
23,439 | I need to run 4 instances of Mac OS X desktop (10.4 to 10.7) for our continuous integration setup (so they need to be on all the time). I've used PC hypervisors in the past (XenServer, ESXi, etc) but never for a Mac.
Is it possible to run those guest operating systems on a Mac Mini hypervisor?
***Edit:*** Ideally, it should be managable remotely (like XenServer and ESXi), desktop virtualization software (like VirtualBox) isn't really what I had in mind. | 2011/08/28 | [
"https://apple.stackexchange.com/questions/23439",
"https://apple.stackexchange.com",
"https://apple.stackexchange.com/users/10732/"
] | You have confusion between PC and Mac, they are both x86/x64, they are the same as John Hodgman and Justin Long have demonstrated before. The compatibility concern should only arise with older PowerPC based Macs, and the EFI boot process instead of BIOS.
Therefore XenServer would run fine, ESX 4.0 does not support EFI as discussed here:
<http://communities.intel.com/thread/3909>
Recently discussion about vSphere apparently supporting EFI on XServe:
<http://lists.apple.com/archives/macos-x-server/2011/Jul/msg00080.html>
>
> What's New in VMware vSphere 5?
> \* Support for Apple products -- vSphere 5 supports Apple Xserve
> servers running OS X Server 10.6 (Snow Leopard) as a guest operating
> system.
>
>
> To read for yourself, the What's New PDF can be downloaded from this
> web page "*VMware vSphere for Small and Midsized Busienss*":
>
>
> <http://www.vmware.com/products/vsphere/small-business/overview.html>
>
>
> | I currently run 10.7 (was originally 10.6 then upgraded it) in ESXi 5 on my Mac Mini (Upgraded from 4GB RAM to 16GB) without any issues.
The performance isn't great when using the actual desktop, but for your requirement (I'm doing the same, using the VMs as build bots) it works absolutely fine. |
401,807 | The new survey is chock full of the usual questions trying to determine which demographic groups (race, age, sex, etc.) I belong to.
Apparently there haven't been any lessons learned from the last demographic debacle.
Stack Overflow was, is, and should forever be free of these kinds of demographic distinctions. By focusing on any demographic group or issue exclusively, you bring discord to a site that never existed in the first place.
Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
Stack Overflow is a programming website. The only interest its veteran participants has is helping other people with their software development questions. We don't care whether those people are black, white, female or Martian, because those personal characteristics are irrelevant. | 2020/10/05 | [
"https://meta.stackoverflow.com/questions/401807",
"https://meta.stackoverflow.com",
"https://meta.stackoverflow.com/users/102937/"
] | The goal of collecting these statistics is to judge people by the color of their skin, the particular genitals they possess, or the gender they state rather than by the content of their character. We know this to be the case because any disparity found in the data is immediately attributed to prejudice, rather than further investigated to see if it correlates to behaviors instead. Any grievance someone attributes to their characteristics is immediately validated by SO staff, whether it holds water or not. "Disadvantaged" or "marginalized" (or in more traditional terminology, "oppressed") groups are instantly judged positively, whether an individual who is a member of that group has produced quality content or not, and those who disagree with this outlook are ostracized irrespective of the character they have demonstrated (e.g. Monica). In short, nothing has changed within SO: it has, as a matter of policy, embraced an ideology that insists people are judged by class and characteristics rather than actions. 2019 demonstrated we can no longer assume good faith on the part of SO. | The demographic data harvested by the survey is patently unfit for any purpose. Even assuming everyone who completed the survey filled in the demographic section absolutely honestly, there's no guarantee that those people are proportionally representative of Stack Overflow's userbase - i.e. the survey suffers from an implicit selection bias.
Which really begs the question why Stack Exchange Inc. continues to insist on collecting this data, especially when they employ a data scientist who (one hopes) would know said data's uselessness. At this point I can't really conceive of a valid and/or benign reason, and since SE Inc. continues to refuse to explain what they are doing or intend to do with this data, it all begins to appear rather ominous.
I would strongly advise everyone to refrain from participating in future surveys until SE Inc. clarifies what they're using this data for. If you do decide to participate, under **absolutely no circumstances** should you provide truthful demographic data - either omit that section entirely, or take creative liberties. |
401,807 | The new survey is chock full of the usual questions trying to determine which demographic groups (race, age, sex, etc.) I belong to.
Apparently there haven't been any lessons learned from the last demographic debacle.
Stack Overflow was, is, and should forever be free of these kinds of demographic distinctions. By focusing on any demographic group or issue exclusively, you bring discord to a site that never existed in the first place.
Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
Stack Overflow is a programming website. The only interest its veteran participants has is helping other people with their software development questions. We don't care whether those people are black, white, female or Martian, because those personal characteristics are irrelevant. | 2020/10/05 | [
"https://meta.stackoverflow.com/questions/401807",
"https://meta.stackoverflow.com",
"https://meta.stackoverflow.com/users/102937/"
] | The goal of collecting these statistics is to judge people by the color of their skin, the particular genitals they possess, or the gender they state rather than by the content of their character. We know this to be the case because any disparity found in the data is immediately attributed to prejudice, rather than further investigated to see if it correlates to behaviors instead. Any grievance someone attributes to their characteristics is immediately validated by SO staff, whether it holds water or not. "Disadvantaged" or "marginalized" (or in more traditional terminology, "oppressed") groups are instantly judged positively, whether an individual who is a member of that group has produced quality content or not, and those who disagree with this outlook are ostracized irrespective of the character they have demonstrated (e.g. Monica). In short, nothing has changed within SO: it has, as a matter of policy, embraced an ideology that insists people are judged by class and characteristics rather than actions. 2019 demonstrated we can no longer assume good faith on the part of SO. | **TL;DR** Did the Code of Conduct changes work, or are the problems they were supposed to address still there?
I'm going to take a contrary position to what the question is advocating and say that the value of this data really depends on what Stack Exchange does with it. (I don't have much confidence that they will, in fact, do the right thing with it, but I suppose that it's possible).
I would like to point out that some of the groups that they ask about have [indicated](https://meta.stackexchange.com/a/334577/343810) that they do, in fact, feel that there's a problem with their demographic being welcomed. For all the uproar caused by the Code of Conduct changes last fall, there's been remarkably little follow-up on whether the changes that were implemented were actually effective at helping these groups. At a minimum, I'd like to know whether the CoC changes helped the situation or whether they still feel that there's a problem. This data could, if interpreted properly, help answer that. |
401,807 | The new survey is chock full of the usual questions trying to determine which demographic groups (race, age, sex, etc.) I belong to.
Apparently there haven't been any lessons learned from the last demographic debacle.
Stack Overflow was, is, and should forever be free of these kinds of demographic distinctions. By focusing on any demographic group or issue exclusively, you bring discord to a site that never existed in the first place.
Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
Stack Overflow is a programming website. The only interest its veteran participants has is helping other people with their software development questions. We don't care whether those people are black, white, female or Martian, because those personal characteristics are irrelevant. | 2020/10/05 | [
"https://meta.stackoverflow.com/questions/401807",
"https://meta.stackoverflow.com",
"https://meta.stackoverflow.com/users/102937/"
] | The goal of collecting these statistics is to judge people by the color of their skin, the particular genitals they possess, or the gender they state rather than by the content of their character. We know this to be the case because any disparity found in the data is immediately attributed to prejudice, rather than further investigated to see if it correlates to behaviors instead. Any grievance someone attributes to their characteristics is immediately validated by SO staff, whether it holds water or not. "Disadvantaged" or "marginalized" (or in more traditional terminology, "oppressed") groups are instantly judged positively, whether an individual who is a member of that group has produced quality content or not, and those who disagree with this outlook are ostracized irrespective of the character they have demonstrated (e.g. Monica). In short, nothing has changed within SO: it has, as a matter of policy, embraced an ideology that insists people are judged by class and characteristics rather than actions. 2019 demonstrated we can no longer assume good faith on the part of SO. | I'm not going to defend SE/SO and say that they doing this purely for the best of intentions, but evidently there's a lot of people that aren't being served well by the [Code of Conduct](https://stackoverflow.com/conduct).
The company recently released a response to a letter they received about the Code of Conduct.
[Responding to the Lavender Letter and commitments moving forward](https://meta.stackexchange.com/questions/354978/responding-to-the-lavender-letter-and-commitments-moving-forward)
The "Lavender Letter" was written and approved by a fairly large amount of people who had gotten bigoted and otherwise hate filled reactions to things they posted. I've had to flag more than a few comments that were less than kind towards others and myself, so I know it happens.
[Dear Stack Exchange: a statement and a letter from your moderators](https://meta.stackexchange.com/questions/334575/dear-stack-exchange-a-statement-and-a-letter-from-your-moderators/334577#334577)
Completely ignoring people's/user's demographics ignores how they may be either underserved, discriminated against, or otherwise shown to be unwelcome (or worse) by this group of sites.
>
> “If you are neutral in situations of injustice, you have chosen the side of the oppressor”
>
> – Desmond Tutu
>
> [...]
>
> “We must take sides. Neutrality helps the oppressor, never the victim. Silence encourages the tormentor, never the tormented”
>
> – Elie Wiesel
>
> [...]
>
> “Washing one’s hands of the conflict between the powerful and the powerless means to side with the powerful, not to be neutral ”
>
> – Paulo Freire
>
> [...]
>
> “The hottest place in Hell is reserved for those who remain neutral in times of great moral conflict…[an individual] who accepts evil without protesting against it is really cooperating with it”
>
> – Martin Luther King Jr.
>
>
>
<https://organizingchange.org/here-is-how-moral-leaders-approach-neutrality/>
>
> Silence is the Voice of Complicity
>
>
>
<https://medium.com/equality-includes-you/silence-is-the-voice-of-complicity-a21525aa2d01>
Having SE/SO be transparent how they use the data they collect would definitely be a major benefit, so we know exactly how that data is used. I'm simply using [Hanlon's Razor](https://en.wikipedia.org/wiki/Hanlon%27s_razor) here, rather than any insider knowledge of how the data is actually used, so the other Answers believing the worst could be correct. But I prefer to assume the best, instead of the worst, in people and businesses until I am proven wrong.
And yes, I'm aware of what went down with [Monica Cellio](https://judaism.meta.stackexchange.com/questions/5193/stack-overflow-inc-sinat-chinam-and-the-goat-for-azazel). I'm also aware that making mistakes happens and doesn't necessarily show a systemic or habitual pattern of abuse, even when a single mistake is doubled down on and not corrected.
Now, before you comment, I'm going to ask you to remember the [Code of Conduct](https://stackoverflow.com/conduct). Also, is your comment based on an knee jerk or emotional reaction, or is it based on something more substantial? I say that because I've seen some rather heated comments on other Answers and the Question itself in this thread, which is part of why I'm posting my own Answer, even though I know it's likely the unpopular opinion and likely to get negative reactions.
Just remember that certain types of negative reaction to this Answer may just prove me right. |
401,807 | The new survey is chock full of the usual questions trying to determine which demographic groups (race, age, sex, etc.) I belong to.
Apparently there haven't been any lessons learned from the last demographic debacle.
Stack Overflow was, is, and should forever be free of these kinds of demographic distinctions. By focusing on any demographic group or issue exclusively, you bring discord to a site that never existed in the first place.
Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
Stack Overflow is a programming website. The only interest its veteran participants has is helping other people with their software development questions. We don't care whether those people are black, white, female or Martian, because those personal characteristics are irrelevant. | 2020/10/05 | [
"https://meta.stackoverflow.com/questions/401807",
"https://meta.stackoverflow.com",
"https://meta.stackoverflow.com/users/102937/"
] | Fundamentally, the nature of a field such as software development, or any kind of software/hardware interaction, is that neither the software nor the hardware *care* about the race, gender, age, or background that their operator is taking part of programming them in.
[Not gonna deny that it'd be *smart* every now and again](https://www.iflscience.com/technology/this-racist-soap-dispenser-reveals-why-diversity-in-tech-is-muchneeded/) to factor those differences in, but the blunt reality is that these cold machines don't care who you are or what you are. They only function.
Stack Overflow, as a consequence of Q&A, is an attempt to isolate and cut out all of the extra fluff about empathy and ethics, since both of those things are subjective and cannot be **concretely** answered. We'll have opinions, but opinions aren't **facts**, and Q&A is about **facts**.
Stack Overflow is entirely predicated on the notion of *code* or something that can be explained without knowing anything about the person that's asking it, or where they're from. In essence, we don't care *by design*; the fact that a person has written a poor question has nothing to do with their background in the slightest. (Well, maybe a bit on the English comprehension side, but there are legion non-native English speakers who are able to write questions effectively here.)
Surveys and questionnaires are an attempt by Stack Exchange to identify who the people are that are interacting with the machine. This means that you *do* get questions about your age, gender, and race. **That makes complete sense.**
The problem that I see with a survey that asks these questions is,
what *question* are you trying to answer?
-----------------------------------------
This may represent the chief disconnect between staff and the larger Meta Stack Overflow community.
Staff see the users are **people** with diverse backgrounds and walks of life, and value this.
Power users see users as another person with **another coding problem**, with no opportunity or chance for the chit-chat.
Again, that is by design. I'm not here to play therapist or hear about how oppressive the field is for people of color. I'm here to answer your technical question, and that's about it.
The problem is that the company [clearly has a different question in mind](https://meta.stackexchange.com/a/354804/175248) when engaging groups of people that use the site, as opposed to the people who are, for lack of a better phrase, left ["holding the bag"](https://www.collinsdictionary.com/us/dictionary/english/be-left-holding-the-bag) on content moderation and curation.
Until we agree with the question they're trying to answer, we're ***never*** going to like these surveys. | **TL;DR** Did the Code of Conduct changes work, or are the problems they were supposed to address still there?
I'm going to take a contrary position to what the question is advocating and say that the value of this data really depends on what Stack Exchange does with it. (I don't have much confidence that they will, in fact, do the right thing with it, but I suppose that it's possible).
I would like to point out that some of the groups that they ask about have [indicated](https://meta.stackexchange.com/a/334577/343810) that they do, in fact, feel that there's a problem with their demographic being welcomed. For all the uproar caused by the Code of Conduct changes last fall, there's been remarkably little follow-up on whether the changes that were implemented were actually effective at helping these groups. At a minimum, I'd like to know whether the CoC changes helped the situation or whether they still feel that there's a problem. This data could, if interpreted properly, help answer that. |
401,807 | The new survey is chock full of the usual questions trying to determine which demographic groups (race, age, sex, etc.) I belong to.
Apparently there haven't been any lessons learned from the last demographic debacle.
Stack Overflow was, is, and should forever be free of these kinds of demographic distinctions. By focusing on any demographic group or issue exclusively, you bring discord to a site that never existed in the first place.
Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
Stack Overflow is a programming website. The only interest its veteran participants has is helping other people with their software development questions. We don't care whether those people are black, white, female or Martian, because those personal characteristics are irrelevant. | 2020/10/05 | [
"https://meta.stackoverflow.com/questions/401807",
"https://meta.stackoverflow.com",
"https://meta.stackoverflow.com/users/102937/"
] | Fundamentally, the nature of a field such as software development, or any kind of software/hardware interaction, is that neither the software nor the hardware *care* about the race, gender, age, or background that their operator is taking part of programming them in.
[Not gonna deny that it'd be *smart* every now and again](https://www.iflscience.com/technology/this-racist-soap-dispenser-reveals-why-diversity-in-tech-is-muchneeded/) to factor those differences in, but the blunt reality is that these cold machines don't care who you are or what you are. They only function.
Stack Overflow, as a consequence of Q&A, is an attempt to isolate and cut out all of the extra fluff about empathy and ethics, since both of those things are subjective and cannot be **concretely** answered. We'll have opinions, but opinions aren't **facts**, and Q&A is about **facts**.
Stack Overflow is entirely predicated on the notion of *code* or something that can be explained without knowing anything about the person that's asking it, or where they're from. In essence, we don't care *by design*; the fact that a person has written a poor question has nothing to do with their background in the slightest. (Well, maybe a bit on the English comprehension side, but there are legion non-native English speakers who are able to write questions effectively here.)
Surveys and questionnaires are an attempt by Stack Exchange to identify who the people are that are interacting with the machine. This means that you *do* get questions about your age, gender, and race. **That makes complete sense.**
The problem that I see with a survey that asks these questions is,
what *question* are you trying to answer?
-----------------------------------------
This may represent the chief disconnect between staff and the larger Meta Stack Overflow community.
Staff see the users are **people** with diverse backgrounds and walks of life, and value this.
Power users see users as another person with **another coding problem**, with no opportunity or chance for the chit-chat.
Again, that is by design. I'm not here to play therapist or hear about how oppressive the field is for people of color. I'm here to answer your technical question, and that's about it.
The problem is that the company [clearly has a different question in mind](https://meta.stackexchange.com/a/354804/175248) when engaging groups of people that use the site, as opposed to the people who are, for lack of a better phrase, left ["holding the bag"](https://www.collinsdictionary.com/us/dictionary/english/be-left-holding-the-bag) on content moderation and curation.
Until we agree with the question they're trying to answer, we're ***never*** going to like these surveys. | >
> Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
>
>
>
In theory, this information can be used to do both of those things, or at least to do them *better*.
I say "in theory" because after over a year of these surveys we still have very little idea of how the data is being used.
Positive potential
------------------
Here's what you'd want to see: demographic information used to identify biases in some part of the system, features or instructions or checks that disproportionately affect members of some sub-group.
A somewhat-obvious example might be a situation where the automated quality checks trigger on certain phrasings that are a strong predictor of quality issues *for the majority of writers* but a weak predictor for a sub-group with different default writing styles (we ran into this exact problem face-first when we started rolling out International sites, so I would be surprised if it *didn't* exist on the English site as well!).
A more subtle example might involve more indirection: imagine a bias that kept certain groups out of /review, resulting in an emergent bias against a programming language that was disproportionately used by those groups. That sort of thing is *really* hard to tease out, but invaluable if you can identify it!
Negative potential
------------------
OTOH... Do we really need a reminder of [how disastrous poor analysis can be when put to use](https://meta.stackexchange.com/questions/343802/what-data-about-meta-has-eluded-stack-exchange-until-recently), especially if built on poor data?
Even with the best of intentions, work done based on incorrect analysis has the potential to actually *hurt* the subgroups being analyzed. Or hurt everyone. Or hurt everyone resulting in blame being laid at the feet of a subgroup that never asked for special treatment.
We've seen *all of this* in the past, and it sure ain't pretty.
Conclusion: the need for oversight
----------------------------------
The potential for good is enormous, but data is never benign once it is being put to use. This is why it is critical to ask questions about how such data is being used: we cannot afford to extend trust or assume good intentions here. I brought up [the failure to report on results](https://meta.stackexchange.com/questions/351024/stack-overflow-2020-q12-company-commitments-report-card) repeatedly in the past, but it continues - therefore, I have come to regard these surveys with intense suspicion, as one should always do when an organization is taking pains to obscure how they handle sensitive data.
Let's not treat this as a distraction or a benign effort; this is a serious undertaking with the potential for both great good and great harm.
Addendum: demographics as a starting point
------------------------------------------
Came across this talk today by C++ veteran and former SO user ThePhD:
<https://www.youtube.com/embed/vaLKm9FE8oo>
The first half covers your concerns and the relevance of demographic information. Crucially, it highlights the importance of accurate information, of having a clear hypothesis up front, and of further research dedicated to proving or disproving that hypothesis if demographics give it credence. I can't really do it justice, worth a listen if you have the time. |
401,807 | The new survey is chock full of the usual questions trying to determine which demographic groups (race, age, sex, etc.) I belong to.
Apparently there haven't been any lessons learned from the last demographic debacle.
Stack Overflow was, is, and should forever be free of these kinds of demographic distinctions. By focusing on any demographic group or issue exclusively, you bring discord to a site that never existed in the first place.
Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
Stack Overflow is a programming website. The only interest its veteran participants has is helping other people with their software development questions. We don't care whether those people are black, white, female or Martian, because those personal characteristics are irrelevant. | 2020/10/05 | [
"https://meta.stackoverflow.com/questions/401807",
"https://meta.stackoverflow.com",
"https://meta.stackoverflow.com/users/102937/"
] | The demographic data harvested by the survey is patently unfit for any purpose. Even assuming everyone who completed the survey filled in the demographic section absolutely honestly, there's no guarantee that those people are proportionally representative of Stack Overflow's userbase - i.e. the survey suffers from an implicit selection bias.
Which really begs the question why Stack Exchange Inc. continues to insist on collecting this data, especially when they employ a data scientist who (one hopes) would know said data's uselessness. At this point I can't really conceive of a valid and/or benign reason, and since SE Inc. continues to refuse to explain what they are doing or intend to do with this data, it all begins to appear rather ominous.
I would strongly advise everyone to refrain from participating in future surveys until SE Inc. clarifies what they're using this data for. If you do decide to participate, under **absolutely no circumstances** should you provide truthful demographic data - either omit that section entirely, or take creative liberties. | I'm not going to defend SE/SO and say that they doing this purely for the best of intentions, but evidently there's a lot of people that aren't being served well by the [Code of Conduct](https://stackoverflow.com/conduct).
The company recently released a response to a letter they received about the Code of Conduct.
[Responding to the Lavender Letter and commitments moving forward](https://meta.stackexchange.com/questions/354978/responding-to-the-lavender-letter-and-commitments-moving-forward)
The "Lavender Letter" was written and approved by a fairly large amount of people who had gotten bigoted and otherwise hate filled reactions to things they posted. I've had to flag more than a few comments that were less than kind towards others and myself, so I know it happens.
[Dear Stack Exchange: a statement and a letter from your moderators](https://meta.stackexchange.com/questions/334575/dear-stack-exchange-a-statement-and-a-letter-from-your-moderators/334577#334577)
Completely ignoring people's/user's demographics ignores how they may be either underserved, discriminated against, or otherwise shown to be unwelcome (or worse) by this group of sites.
>
> “If you are neutral in situations of injustice, you have chosen the side of the oppressor”
>
> – Desmond Tutu
>
> [...]
>
> “We must take sides. Neutrality helps the oppressor, never the victim. Silence encourages the tormentor, never the tormented”
>
> – Elie Wiesel
>
> [...]
>
> “Washing one’s hands of the conflict between the powerful and the powerless means to side with the powerful, not to be neutral ”
>
> – Paulo Freire
>
> [...]
>
> “The hottest place in Hell is reserved for those who remain neutral in times of great moral conflict…[an individual] who accepts evil without protesting against it is really cooperating with it”
>
> – Martin Luther King Jr.
>
>
>
<https://organizingchange.org/here-is-how-moral-leaders-approach-neutrality/>
>
> Silence is the Voice of Complicity
>
>
>
<https://medium.com/equality-includes-you/silence-is-the-voice-of-complicity-a21525aa2d01>
Having SE/SO be transparent how they use the data they collect would definitely be a major benefit, so we know exactly how that data is used. I'm simply using [Hanlon's Razor](https://en.wikipedia.org/wiki/Hanlon%27s_razor) here, rather than any insider knowledge of how the data is actually used, so the other Answers believing the worst could be correct. But I prefer to assume the best, instead of the worst, in people and businesses until I am proven wrong.
And yes, I'm aware of what went down with [Monica Cellio](https://judaism.meta.stackexchange.com/questions/5193/stack-overflow-inc-sinat-chinam-and-the-goat-for-azazel). I'm also aware that making mistakes happens and doesn't necessarily show a systemic or habitual pattern of abuse, even when a single mistake is doubled down on and not corrected.
Now, before you comment, I'm going to ask you to remember the [Code of Conduct](https://stackoverflow.com/conduct). Also, is your comment based on an knee jerk or emotional reaction, or is it based on something more substantial? I say that because I've seen some rather heated comments on other Answers and the Question itself in this thread, which is part of why I'm posting my own Answer, even though I know it's likely the unpopular opinion and likely to get negative reactions.
Just remember that certain types of negative reaction to this Answer may just prove me right. |
401,807 | The new survey is chock full of the usual questions trying to determine which demographic groups (race, age, sex, etc.) I belong to.
Apparently there haven't been any lessons learned from the last demographic debacle.
Stack Overflow was, is, and should forever be free of these kinds of demographic distinctions. By focusing on any demographic group or issue exclusively, you bring discord to a site that never existed in the first place.
Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
Stack Overflow is a programming website. The only interest its veteran participants has is helping other people with their software development questions. We don't care whether those people are black, white, female or Martian, because those personal characteristics are irrelevant. | 2020/10/05 | [
"https://meta.stackoverflow.com/questions/401807",
"https://meta.stackoverflow.com",
"https://meta.stackoverflow.com/users/102937/"
] | >
> Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
>
>
>
In theory, this information can be used to do both of those things, or at least to do them *better*.
I say "in theory" because after over a year of these surveys we still have very little idea of how the data is being used.
Positive potential
------------------
Here's what you'd want to see: demographic information used to identify biases in some part of the system, features or instructions or checks that disproportionately affect members of some sub-group.
A somewhat-obvious example might be a situation where the automated quality checks trigger on certain phrasings that are a strong predictor of quality issues *for the majority of writers* but a weak predictor for a sub-group with different default writing styles (we ran into this exact problem face-first when we started rolling out International sites, so I would be surprised if it *didn't* exist on the English site as well!).
A more subtle example might involve more indirection: imagine a bias that kept certain groups out of /review, resulting in an emergent bias against a programming language that was disproportionately used by those groups. That sort of thing is *really* hard to tease out, but invaluable if you can identify it!
Negative potential
------------------
OTOH... Do we really need a reminder of [how disastrous poor analysis can be when put to use](https://meta.stackexchange.com/questions/343802/what-data-about-meta-has-eluded-stack-exchange-until-recently), especially if built on poor data?
Even with the best of intentions, work done based on incorrect analysis has the potential to actually *hurt* the subgroups being analyzed. Or hurt everyone. Or hurt everyone resulting in blame being laid at the feet of a subgroup that never asked for special treatment.
We've seen *all of this* in the past, and it sure ain't pretty.
Conclusion: the need for oversight
----------------------------------
The potential for good is enormous, but data is never benign once it is being put to use. This is why it is critical to ask questions about how such data is being used: we cannot afford to extend trust or assume good intentions here. I brought up [the failure to report on results](https://meta.stackexchange.com/questions/351024/stack-overflow-2020-q12-company-commitments-report-card) repeatedly in the past, but it continues - therefore, I have come to regard these surveys with intense suspicion, as one should always do when an organization is taking pains to obscure how they handle sensitive data.
Let's not treat this as a distraction or a benign effort; this is a serious undertaking with the potential for both great good and great harm.
Addendum: demographics as a starting point
------------------------------------------
Came across this talk today by C++ veteran and former SO user ThePhD:
<https://www.youtube.com/embed/vaLKm9FE8oo>
The first half covers your concerns and the relevance of demographic information. Crucially, it highlights the importance of accurate information, of having a clear hypothesis up front, and of further research dedicated to proving or disproving that hypothesis if demographics give it credence. I can't really do it justice, worth a listen if you have the time. | **TL;DR** Did the Code of Conduct changes work, or are the problems they were supposed to address still there?
I'm going to take a contrary position to what the question is advocating and say that the value of this data really depends on what Stack Exchange does with it. (I don't have much confidence that they will, in fact, do the right thing with it, but I suppose that it's possible).
I would like to point out that some of the groups that they ask about have [indicated](https://meta.stackexchange.com/a/334577/343810) that they do, in fact, feel that there's a problem with their demographic being welcomed. For all the uproar caused by the Code of Conduct changes last fall, there's been remarkably little follow-up on whether the changes that were implemented were actually effective at helping these groups. At a minimum, I'd like to know whether the CoC changes helped the situation or whether they still feel that there's a problem. This data could, if interpreted properly, help answer that. |
401,807 | The new survey is chock full of the usual questions trying to determine which demographic groups (race, age, sex, etc.) I belong to.
Apparently there haven't been any lessons learned from the last demographic debacle.
Stack Overflow was, is, and should forever be free of these kinds of demographic distinctions. By focusing on any demographic group or issue exclusively, you bring discord to a site that never existed in the first place.
Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
Stack Overflow is a programming website. The only interest its veteran participants has is helping other people with their software development questions. We don't care whether those people are black, white, female or Martian, because those personal characteristics are irrelevant. | 2020/10/05 | [
"https://meta.stackoverflow.com/questions/401807",
"https://meta.stackoverflow.com",
"https://meta.stackoverflow.com/users/102937/"
] | **TL;DR** Did the Code of Conduct changes work, or are the problems they were supposed to address still there?
I'm going to take a contrary position to what the question is advocating and say that the value of this data really depends on what Stack Exchange does with it. (I don't have much confidence that they will, in fact, do the right thing with it, but I suppose that it's possible).
I would like to point out that some of the groups that they ask about have [indicated](https://meta.stackexchange.com/a/334577/343810) that they do, in fact, feel that there's a problem with their demographic being welcomed. For all the uproar caused by the Code of Conduct changes last fall, there's been remarkably little follow-up on whether the changes that were implemented were actually effective at helping these groups. At a minimum, I'd like to know whether the CoC changes helped the situation or whether they still feel that there's a problem. This data could, if interpreted properly, help answer that. | I'm not going to defend SE/SO and say that they doing this purely for the best of intentions, but evidently there's a lot of people that aren't being served well by the [Code of Conduct](https://stackoverflow.com/conduct).
The company recently released a response to a letter they received about the Code of Conduct.
[Responding to the Lavender Letter and commitments moving forward](https://meta.stackexchange.com/questions/354978/responding-to-the-lavender-letter-and-commitments-moving-forward)
The "Lavender Letter" was written and approved by a fairly large amount of people who had gotten bigoted and otherwise hate filled reactions to things they posted. I've had to flag more than a few comments that were less than kind towards others and myself, so I know it happens.
[Dear Stack Exchange: a statement and a letter from your moderators](https://meta.stackexchange.com/questions/334575/dear-stack-exchange-a-statement-and-a-letter-from-your-moderators/334577#334577)
Completely ignoring people's/user's demographics ignores how they may be either underserved, discriminated against, or otherwise shown to be unwelcome (or worse) by this group of sites.
>
> “If you are neutral in situations of injustice, you have chosen the side of the oppressor”
>
> – Desmond Tutu
>
> [...]
>
> “We must take sides. Neutrality helps the oppressor, never the victim. Silence encourages the tormentor, never the tormented”
>
> – Elie Wiesel
>
> [...]
>
> “Washing one’s hands of the conflict between the powerful and the powerless means to side with the powerful, not to be neutral ”
>
> – Paulo Freire
>
> [...]
>
> “The hottest place in Hell is reserved for those who remain neutral in times of great moral conflict…[an individual] who accepts evil without protesting against it is really cooperating with it”
>
> – Martin Luther King Jr.
>
>
>
<https://organizingchange.org/here-is-how-moral-leaders-approach-neutrality/>
>
> Silence is the Voice of Complicity
>
>
>
<https://medium.com/equality-includes-you/silence-is-the-voice-of-complicity-a21525aa2d01>
Having SE/SO be transparent how they use the data they collect would definitely be a major benefit, so we know exactly how that data is used. I'm simply using [Hanlon's Razor](https://en.wikipedia.org/wiki/Hanlon%27s_razor) here, rather than any insider knowledge of how the data is actually used, so the other Answers believing the worst could be correct. But I prefer to assume the best, instead of the worst, in people and businesses until I am proven wrong.
And yes, I'm aware of what went down with [Monica Cellio](https://judaism.meta.stackexchange.com/questions/5193/stack-overflow-inc-sinat-chinam-and-the-goat-for-azazel). I'm also aware that making mistakes happens and doesn't necessarily show a systemic or habitual pattern of abuse, even when a single mistake is doubled down on and not corrected.
Now, before you comment, I'm going to ask you to remember the [Code of Conduct](https://stackoverflow.com/conduct). Also, is your comment based on an knee jerk or emotional reaction, or is it based on something more substantial? I say that because I've seen some rather heated comments on other Answers and the Question itself in this thread, which is part of why I'm posting my own Answer, even though I know it's likely the unpopular opinion and likely to get negative reactions.
Just remember that certain types of negative reaction to this Answer may just prove me right. |
401,807 | The new survey is chock full of the usual questions trying to determine which demographic groups (race, age, sex, etc.) I belong to.
Apparently there haven't been any lessons learned from the last demographic debacle.
Stack Overflow was, is, and should forever be free of these kinds of demographic distinctions. By focusing on any demographic group or issue exclusively, you bring discord to a site that never existed in the first place.
Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
Stack Overflow is a programming website. The only interest its veteran participants has is helping other people with their software development questions. We don't care whether those people are black, white, female or Martian, because those personal characteristics are irrelevant. | 2020/10/05 | [
"https://meta.stackoverflow.com/questions/401807",
"https://meta.stackoverflow.com",
"https://meta.stackoverflow.com/users/102937/"
] | Fundamentally, the nature of a field such as software development, or any kind of software/hardware interaction, is that neither the software nor the hardware *care* about the race, gender, age, or background that their operator is taking part of programming them in.
[Not gonna deny that it'd be *smart* every now and again](https://www.iflscience.com/technology/this-racist-soap-dispenser-reveals-why-diversity-in-tech-is-muchneeded/) to factor those differences in, but the blunt reality is that these cold machines don't care who you are or what you are. They only function.
Stack Overflow, as a consequence of Q&A, is an attempt to isolate and cut out all of the extra fluff about empathy and ethics, since both of those things are subjective and cannot be **concretely** answered. We'll have opinions, but opinions aren't **facts**, and Q&A is about **facts**.
Stack Overflow is entirely predicated on the notion of *code* or something that can be explained without knowing anything about the person that's asking it, or where they're from. In essence, we don't care *by design*; the fact that a person has written a poor question has nothing to do with their background in the slightest. (Well, maybe a bit on the English comprehension side, but there are legion non-native English speakers who are able to write questions effectively here.)
Surveys and questionnaires are an attempt by Stack Exchange to identify who the people are that are interacting with the machine. This means that you *do* get questions about your age, gender, and race. **That makes complete sense.**
The problem that I see with a survey that asks these questions is,
what *question* are you trying to answer?
-----------------------------------------
This may represent the chief disconnect between staff and the larger Meta Stack Overflow community.
Staff see the users are **people** with diverse backgrounds and walks of life, and value this.
Power users see users as another person with **another coding problem**, with no opportunity or chance for the chit-chat.
Again, that is by design. I'm not here to play therapist or hear about how oppressive the field is for people of color. I'm here to answer your technical question, and that's about it.
The problem is that the company [clearly has a different question in mind](https://meta.stackexchange.com/a/354804/175248) when engaging groups of people that use the site, as opposed to the people who are, for lack of a better phrase, left ["holding the bag"](https://www.collinsdictionary.com/us/dictionary/english/be-left-holding-the-bag) on content moderation and curation.
Until we agree with the question they're trying to answer, we're ***never*** going to like these surveys. | I'm not going to defend SE/SO and say that they doing this purely for the best of intentions, but evidently there's a lot of people that aren't being served well by the [Code of Conduct](https://stackoverflow.com/conduct).
The company recently released a response to a letter they received about the Code of Conduct.
[Responding to the Lavender Letter and commitments moving forward](https://meta.stackexchange.com/questions/354978/responding-to-the-lavender-letter-and-commitments-moving-forward)
The "Lavender Letter" was written and approved by a fairly large amount of people who had gotten bigoted and otherwise hate filled reactions to things they posted. I've had to flag more than a few comments that were less than kind towards others and myself, so I know it happens.
[Dear Stack Exchange: a statement and a letter from your moderators](https://meta.stackexchange.com/questions/334575/dear-stack-exchange-a-statement-and-a-letter-from-your-moderators/334577#334577)
Completely ignoring people's/user's demographics ignores how they may be either underserved, discriminated against, or otherwise shown to be unwelcome (or worse) by this group of sites.
>
> “If you are neutral in situations of injustice, you have chosen the side of the oppressor”
>
> – Desmond Tutu
>
> [...]
>
> “We must take sides. Neutrality helps the oppressor, never the victim. Silence encourages the tormentor, never the tormented”
>
> – Elie Wiesel
>
> [...]
>
> “Washing one’s hands of the conflict between the powerful and the powerless means to side with the powerful, not to be neutral ”
>
> – Paulo Freire
>
> [...]
>
> “The hottest place in Hell is reserved for those who remain neutral in times of great moral conflict…[an individual] who accepts evil without protesting against it is really cooperating with it”
>
> – Martin Luther King Jr.
>
>
>
<https://organizingchange.org/here-is-how-moral-leaders-approach-neutrality/>
>
> Silence is the Voice of Complicity
>
>
>
<https://medium.com/equality-includes-you/silence-is-the-voice-of-complicity-a21525aa2d01>
Having SE/SO be transparent how they use the data they collect would definitely be a major benefit, so we know exactly how that data is used. I'm simply using [Hanlon's Razor](https://en.wikipedia.org/wiki/Hanlon%27s_razor) here, rather than any insider knowledge of how the data is actually used, so the other Answers believing the worst could be correct. But I prefer to assume the best, instead of the worst, in people and businesses until I am proven wrong.
And yes, I'm aware of what went down with [Monica Cellio](https://judaism.meta.stackexchange.com/questions/5193/stack-overflow-inc-sinat-chinam-and-the-goat-for-azazel). I'm also aware that making mistakes happens and doesn't necessarily show a systemic or habitual pattern of abuse, even when a single mistake is doubled down on and not corrected.
Now, before you comment, I'm going to ask you to remember the [Code of Conduct](https://stackoverflow.com/conduct). Also, is your comment based on an knee jerk or emotional reaction, or is it based on something more substantial? I say that because I've seen some rather heated comments on other Answers and the Question itself in this thread, which is part of why I'm posting my own Answer, even though I know it's likely the unpopular opinion and likely to get negative reactions.
Just remember that certain types of negative reaction to this Answer may just prove me right. |
401,807 | The new survey is chock full of the usual questions trying to determine which demographic groups (race, age, sex, etc.) I belong to.
Apparently there haven't been any lessons learned from the last demographic debacle.
Stack Overflow was, is, and should forever be free of these kinds of demographic distinctions. By focusing on any demographic group or issue exclusively, you bring discord to a site that never existed in the first place.
Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
Stack Overflow is a programming website. The only interest its veteran participants has is helping other people with their software development questions. We don't care whether those people are black, white, female or Martian, because those personal characteristics are irrelevant. | 2020/10/05 | [
"https://meta.stackoverflow.com/questions/401807",
"https://meta.stackoverflow.com",
"https://meta.stackoverflow.com/users/102937/"
] | The demographic data harvested by the survey is patently unfit for any purpose. Even assuming everyone who completed the survey filled in the demographic section absolutely honestly, there's no guarantee that those people are proportionally representative of Stack Overflow's userbase - i.e. the survey suffers from an implicit selection bias.
Which really begs the question why Stack Exchange Inc. continues to insist on collecting this data, especially when they employ a data scientist who (one hopes) would know said data's uselessness. At this point I can't really conceive of a valid and/or benign reason, and since SE Inc. continues to refuse to explain what they are doing or intend to do with this data, it all begins to appear rather ominous.
I would strongly advise everyone to refrain from participating in future surveys until SE Inc. clarifies what they're using this data for. If you do decide to participate, under **absolutely no circumstances** should you provide truthful demographic data - either omit that section entirely, or take creative liberties. | >
> We don't care whether those people are black, white, female or Martian, because those personal characteristics are irrelevant.
>
>
>
If we as a community are truly judging content based solely on the content itself, where are all the black, white, female or Martian contributors?
The demographics of those participating in tech (including Stack Overflow) are highly skewed and in no way representative of the population as a whole.
That naturally leads to two questions. Which groups are not participating and why not?
I don't know Stack Overflow's internal policy, but it doesn't seem like too big of a leap to think that the demographics questions are at least in part influenced by the above statements.
>
> Instead of separating people into demographic groups and attempting to achieve some utopian state of absolute equality, how about improving the onboarding process for new users? Or getting to know your communities better?
>
>
>
I would argue that attempting to find answers to the two questions above addresses both of the problems you mention here in a more significant and lasting way than any UI changes ever will. |
36,478 | I accidentally left a delivery of NADPH Tetrasodium Salt (Santa Cruz Biotechnology) at room temperature for a little over half an hour.
In addition, when I opened up the package, I briefly held the bottle by the body instead of the cap (which would have prevented body heat from contacting it).
It was packed with airbags, but no ice packs, which is why I am asking this: is the powder still stable and usable? | 2015/07/30 | [
"https://biology.stackexchange.com/questions/36478",
"https://biology.stackexchange.com",
"https://biology.stackexchange.com/users/15955/"
] | In my experience, half an hour at room temperature will make absolutely no difference. Boehringer Mannheim (now [part of Roche](http://www.nytimes.com/1997/05/27/business/roche-plans-an-11-billion-acquisition.html)), who at one time supplied the best NADPH, used to recommend storage at 4o C.
By A340 callibration, NAD(P)H is typically about 85% 'pure' based on dry weight measurements, and Sigma will try to tell you that the remainder is mainly water. That is, if you calculate a concentration based on the dry weight, the A340 will be about 85% what you expect.
Even if some water is absorbed from the atmosphere, this will make little difference as you are probably going to calculate the concentration based on
the A340.
As [User 243](https://biology.stackexchange.com/users/15878/243) has [pointed out](https://biology.stackexchange.com/a/36483/1136), if you are concerned that the NAD(P)H is degraded or oxidized, then an A340
measurement will probably set things right.
If you are concerned that all the A340 absorbing material is not NAD(P)H, or that a significant proportion is in the α-form (Oppenheimer & Kaplan, 1975) or is otherwise enzymically inactive, the best thing is an enzymic calibration.
This is done by adding limiting amounts of NAD(P)H to an assay system containing excess substrate and calibrating enzyme, and measuring the total decrease in absorbance (Racker, 1957).
A popular calibrating enzyme for NADH is aldehyde dehydrogenase ([EC 1.2.1.3](http://www.chem.qmul.ac.uk/iubmb/enzyme/EC1/2/1/3.html)), as the reaction is practically irreversible, and it makes it easy to 'drive' the calibrating reaction to completing (and you don't have to worry about equilibria).
As I said in a previous post, NAD(P)H is UNSTABLE IN ACID (Oppenheimer and Kaplan, 1974). To put it crudely, 'the head (nicotinamide moiety) falls off', causing a bleaching of the absorbance at 340nm. Making 100mM acetate, pH 5, 100 micromolar in NAD(P)H produces a very significant decrease in A340 absorbance with time (which may be very easily monitored in a spectrophotometer).
Just because all the A340 material is due to NAD(P)H, this does not necessarily mean that the commercial preparation is pure. In fact, far from it. Dalziel (1963) showed that commercial preparations of NADH then available contained a competitive inhibitor of many dehydrogenases, most probably adenosine diphosphate ribose.
Accepting Dalziel's value of 17 600 M-1cm-1 for the extinction coefficient for NADH at 260nm, and assuming that NADPH has an identical value, then highly purified samples of NAD(P)H should have an A260/A340 ratio of about 2.83, and values **higher** than this may indicate the presence of impurities that absorb at 260nm, but not at 340nm. ADP-ribose is a prime example, but NAD(P)+ is another possibility.
I notice that in one of the [links](https://www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/Sigma/Product_Information_Sheet/2/n7505pis.pdf) given by [AliceD](https://biology.stackexchange.com/users/9943/aliced), Sigma quote a A260/A340 ratio of 2.32. So is the higher value due to the presence of residual NAD(P)+, or something else (a breakdown product?), or both?
You may need to use chromatography, typically on an anion-exchange resin, to obtain NAD(P)H free of impurities.
**References**
Dalziel, K. (1963). The purification of nicotinamide adenine dinucleotide and the kinetic effects of nucleotide impurities. J. Biol. Chem. 238, 1538 - 1543. [[pdf](http://www.jbc.org/content/238/4/1538.full.pdf)]
Oppenheimer, N. J. & Kaplan, N. O. (1974). Structure of the primary acid rearrangement product of reduced nicotinamide adenine dinucleotide (NADH). Biochemistry 13, 4675 - 4685.[[pubmed](http://www.ncbi.nlm.nih.gov/pubmed/?term=Structure+of+the+primary+acid+rearrangement+product+of+reduced+nicotinamide+adenine+dinucleotide+%28NADH%29)]
Oppenheimer, N. J. & Kaplan, N. O. (1975). The alpha, beta-epimerization of reduced nicotinamide adenine dinucleotide. Arch. Biochem. Biophys.166, 526 - 535.
Racker, E. (1949). Aldehyde dehydrogenase, a diphosphopyridine nucleotide-linked enzyme. J. Biol. Chem. 177, 883 - 892. [[pdf](http://www.jbc.org/content/177/2/883.full.pdf)]
Racker, E. (1957). Spectrophotometric enzymatic methods for the determination of aldehydes and ketoaldehydes. Methods Enzymol. III, 293 - 296.
**Other References**
Ciotti, M. M. & Kaplan, N. O. (1957). Procedures for determination of pyridine nucleotides. Methods Enzymol. III, 890 - 899
Horecker, B. L. & Kornberg, A. (1948). The extinction coefficients of the reduced band of pyridine nucleotides. J. Biol. Chem. 175, 385 - 390. [The definitive determination of the extinction coefficient at 340nm] [[pdf](http://www.jbc.org/content/175/1/385.full.pdf)]
Kornberg, A. & Pricer, W. E. (1950). On the structure of triphosphopyridine nucleotide. J. Biol. Chem. 186, 557 - 567.
[This paper shows that the 'extra' phosphate in NADP(H) is attached to the 2' position of a ribose. In Coenzyme A, of course, the phosphate attached to the ribose is on the 3' position. [[pdf](http://www.jbc.org/content/186/2/557.full.pdf+html)]
Pullman, M. E., San Pietro, A. & Colowick, S. P. (1954). On the structure of reduced diphosphopyridine nucleotide. J. Biol. Chem. 206, 129 - 141 [[pdf](http://www.jbc.org/content/206/1/129.long)] | I think when they sent it to you without dry ice, it is probably OK to store it at room temperature.
Sigma seems to advise on their [NADPH](https://www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/Sigma/Product_Information_Sheet/2/n7505pis.pdf) to store [dehydrated NADPH](http://www.sigmaaldrich.com/catalog/product/sigma/n9910?lang=en®ion=AU) at room temperature, while they advise to store the [hydrated](http://www.sigmaaldrich.com/catalog/product/sigma/n1630?lang=en®ion=AU) [forms](http://www.sigmaaldrich.com/catalog/product/sigma/n7505?lang=en®ion=AU) at -20oC. Likely the presence or absence of molecular water in the material is crucial for its storage.
disclaimer: I am not an organic chemist and when you are doing crucial experiments with your material and you doubt its quality, I'd recommend to chuck this batch away and order new. Likely, a single day of futile experimentation in the lab due to bad reagents is more expensive than the purchase of another jar of NADPH. |
36,478 | I accidentally left a delivery of NADPH Tetrasodium Salt (Santa Cruz Biotechnology) at room temperature for a little over half an hour.
In addition, when I opened up the package, I briefly held the bottle by the body instead of the cap (which would have prevented body heat from contacting it).
It was packed with airbags, but no ice packs, which is why I am asking this: is the powder still stable and usable? | 2015/07/30 | [
"https://biology.stackexchange.com/questions/36478",
"https://biology.stackexchange.com",
"https://biology.stackexchange.com/users/15955/"
] | In my experience, half an hour at room temperature will make absolutely no difference. Boehringer Mannheim (now [part of Roche](http://www.nytimes.com/1997/05/27/business/roche-plans-an-11-billion-acquisition.html)), who at one time supplied the best NADPH, used to recommend storage at 4o C.
By A340 callibration, NAD(P)H is typically about 85% 'pure' based on dry weight measurements, and Sigma will try to tell you that the remainder is mainly water. That is, if you calculate a concentration based on the dry weight, the A340 will be about 85% what you expect.
Even if some water is absorbed from the atmosphere, this will make little difference as you are probably going to calculate the concentration based on
the A340.
As [User 243](https://biology.stackexchange.com/users/15878/243) has [pointed out](https://biology.stackexchange.com/a/36483/1136), if you are concerned that the NAD(P)H is degraded or oxidized, then an A340
measurement will probably set things right.
If you are concerned that all the A340 absorbing material is not NAD(P)H, or that a significant proportion is in the α-form (Oppenheimer & Kaplan, 1975) or is otherwise enzymically inactive, the best thing is an enzymic calibration.
This is done by adding limiting amounts of NAD(P)H to an assay system containing excess substrate and calibrating enzyme, and measuring the total decrease in absorbance (Racker, 1957).
A popular calibrating enzyme for NADH is aldehyde dehydrogenase ([EC 1.2.1.3](http://www.chem.qmul.ac.uk/iubmb/enzyme/EC1/2/1/3.html)), as the reaction is practically irreversible, and it makes it easy to 'drive' the calibrating reaction to completing (and you don't have to worry about equilibria).
As I said in a previous post, NAD(P)H is UNSTABLE IN ACID (Oppenheimer and Kaplan, 1974). To put it crudely, 'the head (nicotinamide moiety) falls off', causing a bleaching of the absorbance at 340nm. Making 100mM acetate, pH 5, 100 micromolar in NAD(P)H produces a very significant decrease in A340 absorbance with time (which may be very easily monitored in a spectrophotometer).
Just because all the A340 material is due to NAD(P)H, this does not necessarily mean that the commercial preparation is pure. In fact, far from it. Dalziel (1963) showed that commercial preparations of NADH then available contained a competitive inhibitor of many dehydrogenases, most probably adenosine diphosphate ribose.
Accepting Dalziel's value of 17 600 M-1cm-1 for the extinction coefficient for NADH at 260nm, and assuming that NADPH has an identical value, then highly purified samples of NAD(P)H should have an A260/A340 ratio of about 2.83, and values **higher** than this may indicate the presence of impurities that absorb at 260nm, but not at 340nm. ADP-ribose is a prime example, but NAD(P)+ is another possibility.
I notice that in one of the [links](https://www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/Sigma/Product_Information_Sheet/2/n7505pis.pdf) given by [AliceD](https://biology.stackexchange.com/users/9943/aliced), Sigma quote a A260/A340 ratio of 2.32. So is the higher value due to the presence of residual NAD(P)+, or something else (a breakdown product?), or both?
You may need to use chromatography, typically on an anion-exchange resin, to obtain NAD(P)H free of impurities.
**References**
Dalziel, K. (1963). The purification of nicotinamide adenine dinucleotide and the kinetic effects of nucleotide impurities. J. Biol. Chem. 238, 1538 - 1543. [[pdf](http://www.jbc.org/content/238/4/1538.full.pdf)]
Oppenheimer, N. J. & Kaplan, N. O. (1974). Structure of the primary acid rearrangement product of reduced nicotinamide adenine dinucleotide (NADH). Biochemistry 13, 4675 - 4685.[[pubmed](http://www.ncbi.nlm.nih.gov/pubmed/?term=Structure+of+the+primary+acid+rearrangement+product+of+reduced+nicotinamide+adenine+dinucleotide+%28NADH%29)]
Oppenheimer, N. J. & Kaplan, N. O. (1975). The alpha, beta-epimerization of reduced nicotinamide adenine dinucleotide. Arch. Biochem. Biophys.166, 526 - 535.
Racker, E. (1949). Aldehyde dehydrogenase, a diphosphopyridine nucleotide-linked enzyme. J. Biol. Chem. 177, 883 - 892. [[pdf](http://www.jbc.org/content/177/2/883.full.pdf)]
Racker, E. (1957). Spectrophotometric enzymatic methods for the determination of aldehydes and ketoaldehydes. Methods Enzymol. III, 293 - 296.
**Other References**
Ciotti, M. M. & Kaplan, N. O. (1957). Procedures for determination of pyridine nucleotides. Methods Enzymol. III, 890 - 899
Horecker, B. L. & Kornberg, A. (1948). The extinction coefficients of the reduced band of pyridine nucleotides. J. Biol. Chem. 175, 385 - 390. [The definitive determination of the extinction coefficient at 340nm] [[pdf](http://www.jbc.org/content/175/1/385.full.pdf)]
Kornberg, A. & Pricer, W. E. (1950). On the structure of triphosphopyridine nucleotide. J. Biol. Chem. 186, 557 - 567.
[This paper shows that the 'extra' phosphate in NADP(H) is attached to the 2' position of a ribose. In Coenzyme A, of course, the phosphate attached to the ribose is on the 3' position. [[pdf](http://www.jbc.org/content/186/2/557.full.pdf+html)]
Pullman, M. E., San Pietro, A. & Colowick, S. P. (1954). On the structure of reduced diphosphopyridine nucleotide. J. Biol. Chem. 206, 129 - 141 [[pdf](http://www.jbc.org/content/206/1/129.long)] | You could measure OD at 340nm. If OD340 is much lower than expected, NADPH is oxidized and does not have much biochemical activity.
<http://www.bmglabtech.com/media/35216/1043734.pdf> |
226,626 | I created a live Kali USB for my Chromebook which single boots ubuntu mate. I was wondering, if I made a live kali(or other linux distribution) for booting on a Mac, so using .img instead of .iso would the .img still work on my Chromebook?
In simpler terms will a live USB for OSX still work on a Chromebook or windows. | 2016/02/08 | [
"https://apple.stackexchange.com/questions/226626",
"https://apple.stackexchange.com",
"https://apple.stackexchange.com/users/169373/"
] | No, the disk should still be able to boot.
There isn't actually a difference between the operating system in the .img and .iso. They're just different ways of storing the same information.
Kali(or any other \*nix) distro is an operating system, just as Windows and OS X are. Whether or not an operating system will run on a computer is solely dependent on the hardware, not the software already on the computer. | If your USB "drive" is large enough you should be able to partition it into two volumes. Then install a linux distribution that will run on Mac hardware on one and your ubuntu for your Chromebook on the other. I'm not familiar enough with the boot process of the Chromebook to know if there would be any other snags. |
15,298 | Most of the commenters at my WordPress blog do not have a Gravatar account. Hence the comment section is almost always filled with the same old mystery man avatars.
I am looking for custom gravatar generators for commenters without their own avatar. So far I have only found Identicons, MonsterID’s, and Wavatars. I find some of the avatars generated by these engines to be these ugly, scary and OTT. Also these do not blend with the blog theme and stick out like a sore thumb.
Can some one point me to a avatar generator that can generate nice looking avatars? | 2011/04/21 | [
"https://wordpress.stackexchange.com/questions/15298",
"https://wordpress.stackexchange.com",
"https://wordpress.stackexchange.com/users/2587/"
] | It depends on what you call "nice" of course - it's hard to create something meaningful with an algorithm.
Someone once created [a unicorn-generator](https://unicornify.appspot.com/), which was [used on April 1, 2010 all over Stack Overflow](https://meta.stackexchange.com/questions/37328/my-godits-full-of-unicorns). Your unicorn looks like this:
 | The only other one supported by Gravatar is the retro generator. This is built into newer versions of WP, just select it in Settings->Discussion. |
270,813 | For liable, a sentence in which it is used correctly;
>
> Such a figure is liable to be attacked as a blasphemer.
>
>
>
For likely:
>
> What he told me is likely to be true. (Not liable. It sounds weird)
>
>
>
Ok I know I can't be 100% sure of things just because of how it sounds. I want a reason from a grammatical view point.
Here is a sentence where liable is **less** suitable than likely. (According to my exam practice book. It doesn't explain why. But its words definitely got some weight to it.)
>
> That book is liable to become a best seller because it is well written, full of suspense, and very entertaining,
>
>
> | 2015/09/02 | [
"https://english.stackexchange.com/questions/270813",
"https://english.stackexchange.com",
"https://english.stackexchange.com/users/128704/"
] | 'Liable' means (in one sense), 'open to', 'capable of', with no necessary connotation of the likelihood or probability of the event. So, "Such a figure is open to/capable of being [= liable to be] attacked ....".
'Likely', on the other hand, so far as I can imagine right now, always suggests a connotation of probability (likelihood). So, "What he told me is probably [= likely to be] true."
In your third example, the context makes it clear that the 'likely' sense of liable is being used, rather than the 'capable of' sense: *any* book (much as I hate to admit it) is 'capable of' becoming a best seller, so a reader of the sentence understands the sense of 'liable' used is not merely 'open to/capable of' (that being understood) but 'likely to'.
>
> liable
>
>
> adj.
>
>
> ....
>
>
> 3. probable, likely, *or capable*
>
>
>
(from [The Free Dictionary](http://www.thefreedictionary.com/liable), italics mine)
Then, 'capable' (adj.):
>
> 1. Having capacity or ability ....
>
>
>
(op. cit.)
Observe that stating the capacity exists is materially different than stating the capacity **will** be realized.
See also the usage note concerning 'liable':
>
> Usage: The use of liable to to mean likely to was formerly considered incorrect, but is now acceptable.
>
>
>
(op. cit.)
Edit
----
Your edit of the question puts a *slightly* different complexion on the third example. The list of reasons in the example make it more than abundantly clear that 'likely to' is the sense used. For this reason, the preferable choice of terms is 'likely', rather than 'capable', because 'likely' is the more precise term for communicating that meaning. | The reason for this is that liable has negative implications. From The American Dictionary of the English Language by Noah Webster we have [this usage note](http://edl.byu.edu/webster/term/2261351) for this sense of the word:
>
> Liable, in this sense, is always applied to evils. We never say, a man is liable to happiness or prosperity, but he is liable to disease, calamities, censure; he is liable to err, to sin, to fall.
>
>
>
However, the definition I'll give is rom the Century Dictionary and Cyclopedia 1890-1914 since it's a little more direct:
>
> Having an aptitude or tendency; subject; exposed, as to the doing or occurring of something evil, injurious, or erroneous: as, we are constantly liable to accidents; your plans are liable to defeat.
>
>
>
This doesn't really seem to indicate that liable is very likely but perhaps more probable than not or at least more probable than desired. [The American Heritage Dictionary, 5th Edition](http://www.thefreedictionary.com/Liable) does liken the word to likely though and notes that liable in the sense of likeliness should be reserved for when the subject is likely to experience an unfavorable outcome. The only time you'd say liable like this is '[Springtime for Hitler](https://en.wikipedia.org/wiki/The_Producers_%281968_film%29#Plot) is liable to become a huge success.' Otherwise, being a bestseller would be a good thing. That's the main reason why it's considered wrong: Liable is reserved for emphasizing potential problems and bad consequences.
I have further speculations through synonymy with the idea that it must happen, referencing the sense of being "bound" as A.D.E.L. and C.D.C. state in their primary sense of legal obligation. If it is bound to happen, it must happen as if it is destined since it is limited to that outcome alone. See some of the [2010 Random House Kernerman Webster's College Dictionary's definitions of bound](http://www.thefreedictionary.com/bound) definitions for that sense. "This book is bound to become a bestseller." is certainly a statement I've seen in marketing sometimes and "This book is liable to become a bestseller" is occasionally used in its place. However this paragraph is mostly just doubtful guesswork.. |
270,813 | For liable, a sentence in which it is used correctly;
>
> Such a figure is liable to be attacked as a blasphemer.
>
>
>
For likely:
>
> What he told me is likely to be true. (Not liable. It sounds weird)
>
>
>
Ok I know I can't be 100% sure of things just because of how it sounds. I want a reason from a grammatical view point.
Here is a sentence where liable is **less** suitable than likely. (According to my exam practice book. It doesn't explain why. But its words definitely got some weight to it.)
>
> That book is liable to become a best seller because it is well written, full of suspense, and very entertaining,
>
>
> | 2015/09/02 | [
"https://english.stackexchange.com/questions/270813",
"https://english.stackexchange.com",
"https://english.stackexchange.com/users/128704/"
] | 'Liable' means (in one sense), 'open to', 'capable of', with no necessary connotation of the likelihood or probability of the event. So, "Such a figure is open to/capable of being [= liable to be] attacked ....".
'Likely', on the other hand, so far as I can imagine right now, always suggests a connotation of probability (likelihood). So, "What he told me is probably [= likely to be] true."
In your third example, the context makes it clear that the 'likely' sense of liable is being used, rather than the 'capable of' sense: *any* book (much as I hate to admit it) is 'capable of' becoming a best seller, so a reader of the sentence understands the sense of 'liable' used is not merely 'open to/capable of' (that being understood) but 'likely to'.
>
> liable
>
>
> adj.
>
>
> ....
>
>
> 3. probable, likely, *or capable*
>
>
>
(from [The Free Dictionary](http://www.thefreedictionary.com/liable), italics mine)
Then, 'capable' (adj.):
>
> 1. Having capacity or ability ....
>
>
>
(op. cit.)
Observe that stating the capacity exists is materially different than stating the capacity **will** be realized.
See also the usage note concerning 'liable':
>
> Usage: The use of liable to to mean likely to was formerly considered incorrect, but is now acceptable.
>
>
>
(op. cit.)
Edit
----
Your edit of the question puts a *slightly* different complexion on the third example. The list of reasons in the example make it more than abundantly clear that 'likely to' is the sense used. For this reason, the preferable choice of terms is 'likely', rather than 'capable', because 'likely' is the more precise term for communicating that meaning. | I have always interpreted that "liable" formally implied "legal responsibility" for one's actions when they cause damage to people, or people's property, so that compensation or even punishment would be required if the case were to be judged by an authorised "party".
In my case of learning &understanding "liable" I never included any aspect of "probability". The information presented in the texts above seems to indicate an entirely different meaning now-a days that has nothing to do with legal responsibility and accountability! |
270,813 | For liable, a sentence in which it is used correctly;
>
> Such a figure is liable to be attacked as a blasphemer.
>
>
>
For likely:
>
> What he told me is likely to be true. (Not liable. It sounds weird)
>
>
>
Ok I know I can't be 100% sure of things just because of how it sounds. I want a reason from a grammatical view point.
Here is a sentence where liable is **less** suitable than likely. (According to my exam practice book. It doesn't explain why. But its words definitely got some weight to it.)
>
> That book is liable to become a best seller because it is well written, full of suspense, and very entertaining,
>
>
> | 2015/09/02 | [
"https://english.stackexchange.com/questions/270813",
"https://english.stackexchange.com",
"https://english.stackexchange.com/users/128704/"
] | The reason for this is that liable has negative implications. From The American Dictionary of the English Language by Noah Webster we have [this usage note](http://edl.byu.edu/webster/term/2261351) for this sense of the word:
>
> Liable, in this sense, is always applied to evils. We never say, a man is liable to happiness or prosperity, but he is liable to disease, calamities, censure; he is liable to err, to sin, to fall.
>
>
>
However, the definition I'll give is rom the Century Dictionary and Cyclopedia 1890-1914 since it's a little more direct:
>
> Having an aptitude or tendency; subject; exposed, as to the doing or occurring of something evil, injurious, or erroneous: as, we are constantly liable to accidents; your plans are liable to defeat.
>
>
>
This doesn't really seem to indicate that liable is very likely but perhaps more probable than not or at least more probable than desired. [The American Heritage Dictionary, 5th Edition](http://www.thefreedictionary.com/Liable) does liken the word to likely though and notes that liable in the sense of likeliness should be reserved for when the subject is likely to experience an unfavorable outcome. The only time you'd say liable like this is '[Springtime for Hitler](https://en.wikipedia.org/wiki/The_Producers_%281968_film%29#Plot) is liable to become a huge success.' Otherwise, being a bestseller would be a good thing. That's the main reason why it's considered wrong: Liable is reserved for emphasizing potential problems and bad consequences.
I have further speculations through synonymy with the idea that it must happen, referencing the sense of being "bound" as A.D.E.L. and C.D.C. state in their primary sense of legal obligation. If it is bound to happen, it must happen as if it is destined since it is limited to that outcome alone. See some of the [2010 Random House Kernerman Webster's College Dictionary's definitions of bound](http://www.thefreedictionary.com/bound) definitions for that sense. "This book is bound to become a bestseller." is certainly a statement I've seen in marketing sometimes and "This book is liable to become a bestseller" is occasionally used in its place. However this paragraph is mostly just doubtful guesswork.. | I have always interpreted that "liable" formally implied "legal responsibility" for one's actions when they cause damage to people, or people's property, so that compensation or even punishment would be required if the case were to be judged by an authorised "party".
In my case of learning &understanding "liable" I never included any aspect of "probability". The information presented in the texts above seems to indicate an entirely different meaning now-a days that has nothing to do with legal responsibility and accountability! |
4,040,652 | Looking around my understanding, Sencha is purely JS based.
I am wondering if there is any way for me to use content from PHP + MySql driven website? | 2010/10/28 | [
"https://Stackoverflow.com/questions/4040652",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/344631/"
] | In short: yes, through [XMLHttpRequests](https://developer.mozilla.org/en/xmlhttprequest) to your server. Sencha Touch is a Javascript framework that runs on your *frontend*, since it's mobile probably on the phone your'e targeting. PHP (& MySQL) runs on a *backend*, your server. Your phone app would then access the website where the PHP-app runs and fetch/store data, probably through a [RESTful API](http://en.wikipedia.org/wiki/Representational_State_Transfer). | You could try using Zend\_Rest\_Server on PHP side and Ext.RestProxy in order to integrate SenchaTouch with PHP. |
72,113 | There's a team lead who chooses to address the entire team whenever there is a problem with one employee. It's usually trivial to tell who is being referred to because the issue will be with something like, "two hour lunches", "chatting too much with staff who are not on break", "forgetting to clock out", etc. Situations where one can think to oneself, "Ah yeah, that'd be Steve" or "Yep, that's Alice."
I'm not really concerned with how to handle this situation at my own workplace but as a relatively young (aka inexperienced) professional looking to one day get into management, I'm curious how effective of a tool this is. **Is it better to have an individual meeting with the employee who has behavior that needs to be addressed or is it a wise tactic to simply address the whole team?**
Clearly the simple answer is, "It depends" but I think we can agree that we're intelligent enough to discern the obvious examples of that statement. I'd like to know, by-and-large, which is the more effective tactic? | 2016/07/26 | [
"https://workplace.stackexchange.com/questions/72113",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/48146/"
] | Negative feedback and correction should almost *always* be done in private when it concerns a single, identifiable individual.
The "it depends" is pretty minimal in situations like what you are describing. If it's a *single* employee, you really need to deal with it directly to that individual.
No one likes being reprimanded. Particularly in front of their entire team.
Group reprimands should be an absolute last resort (even for a single person). Ideally, only after having dealt with the individual in question and appropriate consequences given. | If someone doesn't want to be "The person who takes a two hour lunch break." then they need to stop doing it. The boss isn't mentioning names and is just taking the most efficient route of addressing this in front of everyone so they all know the rules. It can help a manager to have group buy-in for the rules. Also, it benefits the group if they know the culprit is not getting away with it. Many people won't say anything to management, but will criticize them behind their back for letting other people slack.
For things that are either not common/specific to a particular job or task that no one else does, very serious offenses and/or are of a private nature (sexual-harassment, which I realize is also very serious), they should be handled in private.
At some point, it needs to be determined why you have a team and discuss things as a group. Otherwise, deal with everyone individually and send out a memo for things pertaining to everyone. No sense discussing them as a group if the interpretation or administration rules are going to be handled one-on-one. |
72,113 | There's a team lead who chooses to address the entire team whenever there is a problem with one employee. It's usually trivial to tell who is being referred to because the issue will be with something like, "two hour lunches", "chatting too much with staff who are not on break", "forgetting to clock out", etc. Situations where one can think to oneself, "Ah yeah, that'd be Steve" or "Yep, that's Alice."
I'm not really concerned with how to handle this situation at my own workplace but as a relatively young (aka inexperienced) professional looking to one day get into management, I'm curious how effective of a tool this is. **Is it better to have an individual meeting with the employee who has behavior that needs to be addressed or is it a wise tactic to simply address the whole team?**
Clearly the simple answer is, "It depends" but I think we can agree that we're intelligent enough to discern the obvious examples of that statement. I'd like to know, by-and-large, which is the more effective tactic? | 2016/07/26 | [
"https://workplace.stackexchange.com/questions/72113",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/48146/"
] | Is it really a problem where the group is letting standards slip and needs to be reigned in even if one or two people are pushing the envelope further or is it really an issue with a single person?
With your two hour lunch example, I'd be hard-pressed to believe that would really be something that only one person was guilty of. Most likely, people on the team gradually started taking the "lunch hour" as a rough guideline rather than a hard limit and an hour turned into 70 minutes with an occasional 90 minute lunch on Friday. Sure, perhaps one person is particularly egregious about it but frequently you'd have other people on the team that have gotten the implicit message that you don't care too much about exactly how long their lunch is. If you need to counteract that implicit message, you have to address the team. Maybe you address the worst offender privately as well for emphasis particularly if it continues.
On the other hand, if there literally is only one person that needs the correction, talking to the team is likely to be counterproductive. If only one person is forgetting to clock out, addressing the team may well make that person believe that everyone on the team is being forgetful and that they're not an outlier. And you'll probably have someone else that is generally very conscientious but forgot one day last month who thinks that the message is aimed at him and will panic. Taking the one person that has the issue aside will be far more productive. | Speaking, now, *"strictly as an individual ..."*
>
> "If you mean to talk to ME, then, *by gawd,* talk to *ME!*"
>
>
>
**Don't you dare(!)** put me into a "group-meeting situation" in which I, *growing ever more red-in-the-face by the second,* am subjected to the **public(!) humiliation** of "being referred-to *obliquely"* by someone who **hadn't the *guts(!)*** ... *(ahem)* ... to speak to me (in gracious privacy) by *name.*
>
> "You got a beef with ME? Then, *by gawd,* talk to ME about it. *In ... private!*"
>
>
>
"Yeah, it's not too much to ask."
If you have an issue with an individual employee, then that is a *person-to-person* encounter, and you should absolutely treat it as such. (Perhaps, with a representative of HR in attendance. By all means, discuss the matter first with HR, before you involve the employee.)
You should only present "team matters" to "the team." |
72,113 | There's a team lead who chooses to address the entire team whenever there is a problem with one employee. It's usually trivial to tell who is being referred to because the issue will be with something like, "two hour lunches", "chatting too much with staff who are not on break", "forgetting to clock out", etc. Situations where one can think to oneself, "Ah yeah, that'd be Steve" or "Yep, that's Alice."
I'm not really concerned with how to handle this situation at my own workplace but as a relatively young (aka inexperienced) professional looking to one day get into management, I'm curious how effective of a tool this is. **Is it better to have an individual meeting with the employee who has behavior that needs to be addressed or is it a wise tactic to simply address the whole team?**
Clearly the simple answer is, "It depends" but I think we can agree that we're intelligent enough to discern the obvious examples of that statement. I'd like to know, by-and-large, which is the more effective tactic? | 2016/07/26 | [
"https://workplace.stackexchange.com/questions/72113",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/48146/"
] | Is it really a problem where the group is letting standards slip and needs to be reigned in even if one or two people are pushing the envelope further or is it really an issue with a single person?
With your two hour lunch example, I'd be hard-pressed to believe that would really be something that only one person was guilty of. Most likely, people on the team gradually started taking the "lunch hour" as a rough guideline rather than a hard limit and an hour turned into 70 minutes with an occasional 90 minute lunch on Friday. Sure, perhaps one person is particularly egregious about it but frequently you'd have other people on the team that have gotten the implicit message that you don't care too much about exactly how long their lunch is. If you need to counteract that implicit message, you have to address the team. Maybe you address the worst offender privately as well for emphasis particularly if it continues.
On the other hand, if there literally is only one person that needs the correction, talking to the team is likely to be counterproductive. If only one person is forgetting to clock out, addressing the team may well make that person believe that everyone on the team is being forgetful and that they're not an outlier. And you'll probably have someone else that is generally very conscientious but forgot one day last month who thinks that the message is aimed at him and will panic. Taking the one person that has the issue aside will be far more productive. | There is a simple rule of thumb that I used when addressing my team. This is, of course, assuming that the issue is not major, in which case you immediately do a one on one, or depending what happened with HR involved.
Otherwise:
* For the first offence or two, then I would address the group. Make it general and moderately light. If you *are* speaking to the entire team, ensure that you are not singling out any one person.
* If I have a repeat offender, then I'd take them to the side and speak to them individually.
Of course, discretion sometimes *is* required, and that's a normal part of becoming an experienced manager. |
72,113 | There's a team lead who chooses to address the entire team whenever there is a problem with one employee. It's usually trivial to tell who is being referred to because the issue will be with something like, "two hour lunches", "chatting too much with staff who are not on break", "forgetting to clock out", etc. Situations where one can think to oneself, "Ah yeah, that'd be Steve" or "Yep, that's Alice."
I'm not really concerned with how to handle this situation at my own workplace but as a relatively young (aka inexperienced) professional looking to one day get into management, I'm curious how effective of a tool this is. **Is it better to have an individual meeting with the employee who has behavior that needs to be addressed or is it a wise tactic to simply address the whole team?**
Clearly the simple answer is, "It depends" but I think we can agree that we're intelligent enough to discern the obvious examples of that statement. I'd like to know, by-and-large, which is the more effective tactic? | 2016/07/26 | [
"https://workplace.stackexchange.com/questions/72113",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/48146/"
] | Speaking, now, *"strictly as an individual ..."*
>
> "If you mean to talk to ME, then, *by gawd,* talk to *ME!*"
>
>
>
**Don't you dare(!)** put me into a "group-meeting situation" in which I, *growing ever more red-in-the-face by the second,* am subjected to the **public(!) humiliation** of "being referred-to *obliquely"* by someone who **hadn't the *guts(!)*** ... *(ahem)* ... to speak to me (in gracious privacy) by *name.*
>
> "You got a beef with ME? Then, *by gawd,* talk to ME about it. *In ... private!*"
>
>
>
"Yeah, it's not too much to ask."
If you have an issue with an individual employee, then that is a *person-to-person* encounter, and you should absolutely treat it as such. (Perhaps, with a representative of HR in attendance. By all means, discuss the matter first with HR, before you involve the employee.)
You should only present "team matters" to "the team." | Praise in public, criticize in private. Simple. |
72,113 | There's a team lead who chooses to address the entire team whenever there is a problem with one employee. It's usually trivial to tell who is being referred to because the issue will be with something like, "two hour lunches", "chatting too much with staff who are not on break", "forgetting to clock out", etc. Situations where one can think to oneself, "Ah yeah, that'd be Steve" or "Yep, that's Alice."
I'm not really concerned with how to handle this situation at my own workplace but as a relatively young (aka inexperienced) professional looking to one day get into management, I'm curious how effective of a tool this is. **Is it better to have an individual meeting with the employee who has behavior that needs to be addressed or is it a wise tactic to simply address the whole team?**
Clearly the simple answer is, "It depends" but I think we can agree that we're intelligent enough to discern the obvious examples of that statement. I'd like to know, by-and-large, which is the more effective tactic? | 2016/07/26 | [
"https://workplace.stackexchange.com/questions/72113",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/48146/"
] | There is a simple rule of thumb that I used when addressing my team. This is, of course, assuming that the issue is not major, in which case you immediately do a one on one, or depending what happened with HR involved.
Otherwise:
* For the first offence or two, then I would address the group. Make it general and moderately light. If you *are* speaking to the entire team, ensure that you are not singling out any one person.
* If I have a repeat offender, then I'd take them to the side and speak to them individually.
Of course, discretion sometimes *is* required, and that's a normal part of becoming an experienced manager. | Praise in public, criticize in private. Simple. |
72,113 | There's a team lead who chooses to address the entire team whenever there is a problem with one employee. It's usually trivial to tell who is being referred to because the issue will be with something like, "two hour lunches", "chatting too much with staff who are not on break", "forgetting to clock out", etc. Situations where one can think to oneself, "Ah yeah, that'd be Steve" or "Yep, that's Alice."
I'm not really concerned with how to handle this situation at my own workplace but as a relatively young (aka inexperienced) professional looking to one day get into management, I'm curious how effective of a tool this is. **Is it better to have an individual meeting with the employee who has behavior that needs to be addressed or is it a wise tactic to simply address the whole team?**
Clearly the simple answer is, "It depends" but I think we can agree that we're intelligent enough to discern the obvious examples of that statement. I'd like to know, by-and-large, which is the more effective tactic? | 2016/07/26 | [
"https://workplace.stackexchange.com/questions/72113",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/48146/"
] | Is it really a problem where the group is letting standards slip and needs to be reigned in even if one or two people are pushing the envelope further or is it really an issue with a single person?
With your two hour lunch example, I'd be hard-pressed to believe that would really be something that only one person was guilty of. Most likely, people on the team gradually started taking the "lunch hour" as a rough guideline rather than a hard limit and an hour turned into 70 minutes with an occasional 90 minute lunch on Friday. Sure, perhaps one person is particularly egregious about it but frequently you'd have other people on the team that have gotten the implicit message that you don't care too much about exactly how long their lunch is. If you need to counteract that implicit message, you have to address the team. Maybe you address the worst offender privately as well for emphasis particularly if it continues.
On the other hand, if there literally is only one person that needs the correction, talking to the team is likely to be counterproductive. If only one person is forgetting to clock out, addressing the team may well make that person believe that everyone on the team is being forgetful and that they're not an outlier. And you'll probably have someone else that is generally very conscientious but forgot one day last month who thinks that the message is aimed at him and will panic. Taking the one person that has the issue aside will be far more productive. | If someone doesn't want to be "The person who takes a two hour lunch break." then they need to stop doing it. The boss isn't mentioning names and is just taking the most efficient route of addressing this in front of everyone so they all know the rules. It can help a manager to have group buy-in for the rules. Also, it benefits the group if they know the culprit is not getting away with it. Many people won't say anything to management, but will criticize them behind their back for letting other people slack.
For things that are either not common/specific to a particular job or task that no one else does, very serious offenses and/or are of a private nature (sexual-harassment, which I realize is also very serious), they should be handled in private.
At some point, it needs to be determined why you have a team and discuss things as a group. Otherwise, deal with everyone individually and send out a memo for things pertaining to everyone. No sense discussing them as a group if the interpretation or administration rules are going to be handled one-on-one. |
72,113 | There's a team lead who chooses to address the entire team whenever there is a problem with one employee. It's usually trivial to tell who is being referred to because the issue will be with something like, "two hour lunches", "chatting too much with staff who are not on break", "forgetting to clock out", etc. Situations where one can think to oneself, "Ah yeah, that'd be Steve" or "Yep, that's Alice."
I'm not really concerned with how to handle this situation at my own workplace but as a relatively young (aka inexperienced) professional looking to one day get into management, I'm curious how effective of a tool this is. **Is it better to have an individual meeting with the employee who has behavior that needs to be addressed or is it a wise tactic to simply address the whole team?**
Clearly the simple answer is, "It depends" but I think we can agree that we're intelligent enough to discern the obvious examples of that statement. I'd like to know, by-and-large, which is the more effective tactic? | 2016/07/26 | [
"https://workplace.stackexchange.com/questions/72113",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/48146/"
] | Negative feedback and correction should almost *always* be done in private when it concerns a single, identifiable individual.
The "it depends" is pretty minimal in situations like what you are describing. If it's a *single* employee, you really need to deal with it directly to that individual.
No one likes being reprimanded. Particularly in front of their entire team.
Group reprimands should be an absolute last resort (even for a single person). Ideally, only after having dealt with the individual in question and appropriate consequences given. | Speaking, now, *"strictly as an individual ..."*
>
> "If you mean to talk to ME, then, *by gawd,* talk to *ME!*"
>
>
>
**Don't you dare(!)** put me into a "group-meeting situation" in which I, *growing ever more red-in-the-face by the second,* am subjected to the **public(!) humiliation** of "being referred-to *obliquely"* by someone who **hadn't the *guts(!)*** ... *(ahem)* ... to speak to me (in gracious privacy) by *name.*
>
> "You got a beef with ME? Then, *by gawd,* talk to ME about it. *In ... private!*"
>
>
>
"Yeah, it's not too much to ask."
If you have an issue with an individual employee, then that is a *person-to-person* encounter, and you should absolutely treat it as such. (Perhaps, with a representative of HR in attendance. By all means, discuss the matter first with HR, before you involve the employee.)
You should only present "team matters" to "the team." |
72,113 | There's a team lead who chooses to address the entire team whenever there is a problem with one employee. It's usually trivial to tell who is being referred to because the issue will be with something like, "two hour lunches", "chatting too much with staff who are not on break", "forgetting to clock out", etc. Situations where one can think to oneself, "Ah yeah, that'd be Steve" or "Yep, that's Alice."
I'm not really concerned with how to handle this situation at my own workplace but as a relatively young (aka inexperienced) professional looking to one day get into management, I'm curious how effective of a tool this is. **Is it better to have an individual meeting with the employee who has behavior that needs to be addressed or is it a wise tactic to simply address the whole team?**
Clearly the simple answer is, "It depends" but I think we can agree that we're intelligent enough to discern the obvious examples of that statement. I'd like to know, by-and-large, which is the more effective tactic? | 2016/07/26 | [
"https://workplace.stackexchange.com/questions/72113",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/48146/"
] | Negative feedback and correction should almost *always* be done in private when it concerns a single, identifiable individual.
The "it depends" is pretty minimal in situations like what you are describing. If it's a *single* employee, you really need to deal with it directly to that individual.
No one likes being reprimanded. Particularly in front of their entire team.
Group reprimands should be an absolute last resort (even for a single person). Ideally, only after having dealt with the individual in question and appropriate consequences given. | Praise in public, criticize in private. Simple. |
72,113 | There's a team lead who chooses to address the entire team whenever there is a problem with one employee. It's usually trivial to tell who is being referred to because the issue will be with something like, "two hour lunches", "chatting too much with staff who are not on break", "forgetting to clock out", etc. Situations where one can think to oneself, "Ah yeah, that'd be Steve" or "Yep, that's Alice."
I'm not really concerned with how to handle this situation at my own workplace but as a relatively young (aka inexperienced) professional looking to one day get into management, I'm curious how effective of a tool this is. **Is it better to have an individual meeting with the employee who has behavior that needs to be addressed or is it a wise tactic to simply address the whole team?**
Clearly the simple answer is, "It depends" but I think we can agree that we're intelligent enough to discern the obvious examples of that statement. I'd like to know, by-and-large, which is the more effective tactic? | 2016/07/26 | [
"https://workplace.stackexchange.com/questions/72113",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/48146/"
] | Is it really a problem where the group is letting standards slip and needs to be reigned in even if one or two people are pushing the envelope further or is it really an issue with a single person?
With your two hour lunch example, I'd be hard-pressed to believe that would really be something that only one person was guilty of. Most likely, people on the team gradually started taking the "lunch hour" as a rough guideline rather than a hard limit and an hour turned into 70 minutes with an occasional 90 minute lunch on Friday. Sure, perhaps one person is particularly egregious about it but frequently you'd have other people on the team that have gotten the implicit message that you don't care too much about exactly how long their lunch is. If you need to counteract that implicit message, you have to address the team. Maybe you address the worst offender privately as well for emphasis particularly if it continues.
On the other hand, if there literally is only one person that needs the correction, talking to the team is likely to be counterproductive. If only one person is forgetting to clock out, addressing the team may well make that person believe that everyone on the team is being forgetful and that they're not an outlier. And you'll probably have someone else that is generally very conscientious but forgot one day last month who thinks that the message is aimed at him and will panic. Taking the one person that has the issue aside will be far more productive. | Praise in public, criticize in private. Simple. |
72,113 | There's a team lead who chooses to address the entire team whenever there is a problem with one employee. It's usually trivial to tell who is being referred to because the issue will be with something like, "two hour lunches", "chatting too much with staff who are not on break", "forgetting to clock out", etc. Situations where one can think to oneself, "Ah yeah, that'd be Steve" or "Yep, that's Alice."
I'm not really concerned with how to handle this situation at my own workplace but as a relatively young (aka inexperienced) professional looking to one day get into management, I'm curious how effective of a tool this is. **Is it better to have an individual meeting with the employee who has behavior that needs to be addressed or is it a wise tactic to simply address the whole team?**
Clearly the simple answer is, "It depends" but I think we can agree that we're intelligent enough to discern the obvious examples of that statement. I'd like to know, by-and-large, which is the more effective tactic? | 2016/07/26 | [
"https://workplace.stackexchange.com/questions/72113",
"https://workplace.stackexchange.com",
"https://workplace.stackexchange.com/users/48146/"
] | Is it really a problem where the group is letting standards slip and needs to be reigned in even if one or two people are pushing the envelope further or is it really an issue with a single person?
With your two hour lunch example, I'd be hard-pressed to believe that would really be something that only one person was guilty of. Most likely, people on the team gradually started taking the "lunch hour" as a rough guideline rather than a hard limit and an hour turned into 70 minutes with an occasional 90 minute lunch on Friday. Sure, perhaps one person is particularly egregious about it but frequently you'd have other people on the team that have gotten the implicit message that you don't care too much about exactly how long their lunch is. If you need to counteract that implicit message, you have to address the team. Maybe you address the worst offender privately as well for emphasis particularly if it continues.
On the other hand, if there literally is only one person that needs the correction, talking to the team is likely to be counterproductive. If only one person is forgetting to clock out, addressing the team may well make that person believe that everyone on the team is being forgetful and that they're not an outlier. And you'll probably have someone else that is generally very conscientious but forgot one day last month who thinks that the message is aimed at him and will panic. Taking the one person that has the issue aside will be far more productive. | Negative feedback and correction should almost *always* be done in private when it concerns a single, identifiable individual.
The "it depends" is pretty minimal in situations like what you are describing. If it's a *single* employee, you really need to deal with it directly to that individual.
No one likes being reprimanded. Particularly in front of their entire team.
Group reprimands should be an absolute last resort (even for a single person). Ideally, only after having dealt with the individual in question and appropriate consequences given. |
566 | In principle, by placing a GNSS-receiver on all extremities of a spacecraft (or aircraft, for that matter), one can determine the orientation of the satellite. Google [lists some studies](https://www.google.com/search?q=gnss+based+attitude+determination), but have such systems been used in real spacecraft? On the one hand, I'd think cubesats might go for such a low-cost solution, on the other hand, the smaller the spacecraft, the larger the error in attitude determination by using GNSS. The technique is not listed on [the Wikipedia page on attitude control](http://en.wikipedia.org/wiki/Attitude_control). Has this been used in practice at all? | 2013/07/21 | [
"https://space.stackexchange.com/questions/566",
"https://space.stackexchange.com",
"https://space.stackexchange.com/users/33/"
] | The ISS does it. Source: <http://spacestationlive.jsc.nasa.gov/handbooks/adcoHandbook.pdf>
>
> Attitude Determination
>
>
> How am I currently oriented?
>
>
> The ISS also uses GPS to determine how the ISS is
> oriented, or facing, as it orbits the Earth. **This
> orientation, or attitude, can be determined by
> measuring the difference in time that the GPS signals
> are received by four antennas**. These antennas
> receive the same GPS signal at slightly different times,
> with the signal traveling at a constant speed (the
> speed of light).
>
>
> To calculate attitude, at least four of the nearly 24 GPS
> satellites in orbit must be in view of these antennas.
> **The GPS receiver calculates the ISS attitude about
> once a minute**, providing information on where the ISS
> is pointed at that point in time.
>
>
>
Now, in more detail on that:
<http://sunnyday.mit.edu/16.355/Gomez-GPS-Lessons.pdf>
>
> The ISS GPS antenna array consists of four antennas in a
> 3 meter by 1.5 meter rectangle on the S0 element of the
> ISS main truss. The array center is slightly to the port
> side (4 meters) of the vehicle centerline, as shown in
> Figure 10.
>
>
>
>
> **GPS alone can not meet the attitude
> accuracy requirements. The multipath environment on
> ISS is such that the unfiltered GPS attitude solutions can
> not meet the 0.5 degree requirement. The unfiltered GPS
> attitude solutions are used as an input in the ISS’s on
> board software attitude filter.**
>
>
> | The new Soyuz MS does too.
See <https://en.wikipedia.org/wiki/Soyuz_MS>
>
> Instead of relying on ground stations for orbital determination and
> correction, the now included Satellite Navigation System ASN-K
> (Russian: (АСН-К, Аппаратура Спутниковой Навигации) relying on GLONASS
> and GPS signals for navigation.[2][11] It uses four fixed antennas to
> achieve a positioning accuracy of 5 m (16 ft), with the objective to
> reduce that number to as little as 3 cm (1.2 in) and an attitude
> accuracy of 0.5°.[12]
>
>
> |
2,533,981 | I am a long time agile advocated but one of the things that bothers me about Agile is that a lot of agile practitioners, especially the younger ones, have thrown out or are missing a whole lot of good (non Scrum, non XP) practices. Alistair Cockburn's style of writing Use Cases springs to mind; orthogonal arrays (pairwise testing) is another.
I read mostly Agile related books and articles and work with mostly Agile folk ... is there anything I'm missing? | 2010/03/28 | [
"https://Stackoverflow.com/questions/2533981",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/228721/"
] | It might be interesting in 5-10 years time to see how maintainable these systems are when nobody wrote down why a particular decision was made and all the people involved have left. | >
> (...) have thrown out or are missing a whole lot of good (non Scrum, non XP) practices.
>
>
>
Scrum is not prescriptive, it's up to you to choose how to do things. In other words, nothing forces you to use User Stories for example (even if User Stories work for lots of teams, there is no consensus) so feel free to use (light) use-cases if you think they are more appropriate in your context. To illustrate this, Jeff Sutherland reported he would never use User Stories again for PDA device projects (they use some kind of "light specifications" in his current company). And the same applies for testing, use whatever works for you. To summarize, if you find XP not flexible enough, use something else... and inspect and adapt. |
2,533,981 | I am a long time agile advocated but one of the things that bothers me about Agile is that a lot of agile practitioners, especially the younger ones, have thrown out or are missing a whole lot of good (non Scrum, non XP) practices. Alistair Cockburn's style of writing Use Cases springs to mind; orthogonal arrays (pairwise testing) is another.
I read mostly Agile related books and articles and work with mostly Agile folk ... is there anything I'm missing? | 2010/03/28 | [
"https://Stackoverflow.com/questions/2533981",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/228721/"
] | It might be interesting in 5-10 years time to see how maintainable these systems are when nobody wrote down why a particular decision was made and all the people involved have left. | Iterative development.
In practice, agile teams may do iterations (or anything for that matter, agile is a kind of "true scotsman"), but agile processes don't require or define iterative development sufficiently.
Take RUP, for example - clumsy and bloated, it does compile a few good methods for long-term development that agile misses.
---
On a general note, agile is a way to steer clear of problems: how to avoid long term planning, how to keep teams small, tasks short, customers involved, etc. It works more often than not, but sometimes you have to face and solve problems: how to reach strict deadline, make big team work, achieve distant and complex goals, make customer refine requirements. That's when one needs to look beyond agile. |
2,533,981 | I am a long time agile advocated but one of the things that bothers me about Agile is that a lot of agile practitioners, especially the younger ones, have thrown out or are missing a whole lot of good (non Scrum, non XP) practices. Alistair Cockburn's style of writing Use Cases springs to mind; orthogonal arrays (pairwise testing) is another.
I read mostly Agile related books and articles and work with mostly Agile folk ... is there anything I'm missing? | 2010/03/28 | [
"https://Stackoverflow.com/questions/2533981",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/228721/"
] | It might be interesting in 5-10 years time to see how maintainable these systems are when nobody wrote down why a particular decision was made and all the people involved have left. | >
> is there anything I'm missing?
>
>
>
Yes, I think a lot, but only if you are interested in Softawre Development Processes.
I like this paraphrase:
>
> Each project should be as agile as possible but not more agile.
>
>
>
Not every project can be agile... but I think 80%+ can.
I see Agile as "[car of the year](http://www.caroftheyear.org/)". It is very well suited for most of the people, but if you need/want something special, for example car able to speed 300KM/H or car able to carry 20 tons of goods you need something else.
There is also so many cases when one may want something else than "car of the year" that requires a book to write them down :-) I recommend you [Agility and Discipline Made Easy: Practices from OpenUP and RUP](https://rads.stackoverflow.com/amzn/click/com/0321321308). In this book you'll find many "missing parts" very well illustrated. The key to understanding is that Agility is only a (requested) property of software development process which sometimes cannot be achieved. The book describes several Key Development Principles (which are basis for RUP) and explains which level of "ceremony" and "iterativeness" follows from using them on different levels of adoption.
**An example**
Practice: Automate change management and change propagation
In your project you may require very advanced and strict change management and decide to "Automate change management and change propagation" by implementing custom or re-configuring existing tools and by using Change and Control Board.
Effect: This most probably increase level of "ceremony" in your project. |
2,533,981 | I am a long time agile advocated but one of the things that bothers me about Agile is that a lot of agile practitioners, especially the younger ones, have thrown out or are missing a whole lot of good (non Scrum, non XP) practices. Alistair Cockburn's style of writing Use Cases springs to mind; orthogonal arrays (pairwise testing) is another.
I read mostly Agile related books and articles and work with mostly Agile folk ... is there anything I'm missing? | 2010/03/28 | [
"https://Stackoverflow.com/questions/2533981",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/228721/"
] | >
> is there anything I'm missing?
>
>
>
Yes, I think a lot, but only if you are interested in Softawre Development Processes.
I like this paraphrase:
>
> Each project should be as agile as possible but not more agile.
>
>
>
Not every project can be agile... but I think 80%+ can.
I see Agile as "[car of the year](http://www.caroftheyear.org/)". It is very well suited for most of the people, but if you need/want something special, for example car able to speed 300KM/H or car able to carry 20 tons of goods you need something else.
There is also so many cases when one may want something else than "car of the year" that requires a book to write them down :-) I recommend you [Agility and Discipline Made Easy: Practices from OpenUP and RUP](https://rads.stackoverflow.com/amzn/click/com/0321321308). In this book you'll find many "missing parts" very well illustrated. The key to understanding is that Agility is only a (requested) property of software development process which sometimes cannot be achieved. The book describes several Key Development Principles (which are basis for RUP) and explains which level of "ceremony" and "iterativeness" follows from using them on different levels of adoption.
**An example**
Practice: Automate change management and change propagation
In your project you may require very advanced and strict change management and decide to "Automate change management and change propagation" by implementing custom or re-configuring existing tools and by using Change and Control Board.
Effect: This most probably increase level of "ceremony" in your project. | >
> (...) have thrown out or are missing a whole lot of good (non Scrum, non XP) practices.
>
>
>
Scrum is not prescriptive, it's up to you to choose how to do things. In other words, nothing forces you to use User Stories for example (even if User Stories work for lots of teams, there is no consensus) so feel free to use (light) use-cases if you think they are more appropriate in your context. To illustrate this, Jeff Sutherland reported he would never use User Stories again for PDA device projects (they use some kind of "light specifications" in his current company). And the same applies for testing, use whatever works for you. To summarize, if you find XP not flexible enough, use something else... and inspect and adapt. |
2,533,981 | I am a long time agile advocated but one of the things that bothers me about Agile is that a lot of agile practitioners, especially the younger ones, have thrown out or are missing a whole lot of good (non Scrum, non XP) practices. Alistair Cockburn's style of writing Use Cases springs to mind; orthogonal arrays (pairwise testing) is another.
I read mostly Agile related books and articles and work with mostly Agile folk ... is there anything I'm missing? | 2010/03/28 | [
"https://Stackoverflow.com/questions/2533981",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/228721/"
] | >
> is there anything I'm missing?
>
>
>
Yes, I think a lot, but only if you are interested in Softawre Development Processes.
I like this paraphrase:
>
> Each project should be as agile as possible but not more agile.
>
>
>
Not every project can be agile... but I think 80%+ can.
I see Agile as "[car of the year](http://www.caroftheyear.org/)". It is very well suited for most of the people, but if you need/want something special, for example car able to speed 300KM/H or car able to carry 20 tons of goods you need something else.
There is also so many cases when one may want something else than "car of the year" that requires a book to write them down :-) I recommend you [Agility and Discipline Made Easy: Practices from OpenUP and RUP](https://rads.stackoverflow.com/amzn/click/com/0321321308). In this book you'll find many "missing parts" very well illustrated. The key to understanding is that Agility is only a (requested) property of software development process which sometimes cannot be achieved. The book describes several Key Development Principles (which are basis for RUP) and explains which level of "ceremony" and "iterativeness" follows from using them on different levels of adoption.
**An example**
Practice: Automate change management and change propagation
In your project you may require very advanced and strict change management and decide to "Automate change management and change propagation" by implementing custom or re-configuring existing tools and by using Change and Control Board.
Effect: This most probably increase level of "ceremony" in your project. | Iterative development.
In practice, agile teams may do iterations (or anything for that matter, agile is a kind of "true scotsman"), but agile processes don't require or define iterative development sufficiently.
Take RUP, for example - clumsy and bloated, it does compile a few good methods for long-term development that agile misses.
---
On a general note, agile is a way to steer clear of problems: how to avoid long term planning, how to keep teams small, tasks short, customers involved, etc. It works more often than not, but sometimes you have to face and solve problems: how to reach strict deadline, make big team work, achieve distant and complex goals, make customer refine requirements. That's when one needs to look beyond agile. |
9,739,114 | I've got a problem with my old app and would like to know how to fix it. I found some advices, but I don't understand them. I need step-by-step instructions.
When I open my app: <http://apps.facebook.com/ako-dobre-poz-jeadci/> it says: "This application does not yet support secure browsing (HTTPS)."
And when I open my app with turned off secure browsing, it says: "The app "Ako dobre poznáš Miša P.?" is temporarily unavailable due to an issue with its third-party developer. We are investigating the situation and apologize for any inconvenience."
Thanks for any ideas. | 2012/03/16 | [
"https://Stackoverflow.com/questions/9739114",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1274220/"
] | Your app has to support https. They sent out notifications for this about 6 months ago.
As mentioned in the comments, you probably also need to [migrate to OAuth 2.0](http://developers.facebook.com/docs/oauth2-https-migration/).
You really have to keep on top of your facebook apps, as they are notorious for quickly deprecating and removing API components. To put it bluntly, you got Zucker punched. | Your app (and all of its resources) needs to be hosted on a secured domain (e.g.: <https://www.example.com>) with a valid SSL certificate |
9,739,114 | I've got a problem with my old app and would like to know how to fix it. I found some advices, but I don't understand them. I need step-by-step instructions.
When I open my app: <http://apps.facebook.com/ako-dobre-poz-jeadci/> it says: "This application does not yet support secure browsing (HTTPS)."
And when I open my app with turned off secure browsing, it says: "The app "Ako dobre poznáš Miša P.?" is temporarily unavailable due to an issue with its third-party developer. We are investigating the situation and apologize for any inconvenience."
Thanks for any ideas. | 2012/03/16 | [
"https://Stackoverflow.com/questions/9739114",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1274220/"
] | Your app has to support https. They sent out notifications for this about 6 months ago.
As mentioned in the comments, you probably also need to [migrate to OAuth 2.0](http://developers.facebook.com/docs/oauth2-https-migration/).
You really have to keep on top of your facebook apps, as they are notorious for quickly deprecating and removing API components. To put it bluntly, you got Zucker punched. | In order to fix it, you have to buy and setup an SSL certificate. (it is now required since 1 October 2011 : [News](http://www.wpcode.net/fb-ssl.html/))
After, you have to add your https url in your app settings.
Check Facebook authentification guide for dev for more information about it:
[Authentification Guide - Facebook](https://developers.facebook.com/docs/authentication/) |
9,739,114 | I've got a problem with my old app and would like to know how to fix it. I found some advices, but I don't understand them. I need step-by-step instructions.
When I open my app: <http://apps.facebook.com/ako-dobre-poz-jeadci/> it says: "This application does not yet support secure browsing (HTTPS)."
And when I open my app with turned off secure browsing, it says: "The app "Ako dobre poznáš Miša P.?" is temporarily unavailable due to an issue with its third-party developer. We are investigating the situation and apologize for any inconvenience."
Thanks for any ideas. | 2012/03/16 | [
"https://Stackoverflow.com/questions/9739114",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1274220/"
] | Your app has to support https. They sent out notifications for this about 6 months ago.
As mentioned in the comments, you probably also need to [migrate to OAuth 2.0](http://developers.facebook.com/docs/oauth2-https-migration/).
You really have to keep on top of your facebook apps, as they are notorious for quickly deprecating and removing API components. To put it bluntly, you got Zucker punched. | You need to purchase and setup an SSL certificate and then plug your https url into your app's settings page. See their [migration guide](http://developers.facebook.com/docs/oauth2-https-migration/) where it says this is now required.
Source: [Secure browsing is not supported in my app](https://stackoverflow.com/questions/8603127/secure-browsing-is-not-supported-in-my-app). |
9,739,114 | I've got a problem with my old app and would like to know how to fix it. I found some advices, but I don't understand them. I need step-by-step instructions.
When I open my app: <http://apps.facebook.com/ako-dobre-poz-jeadci/> it says: "This application does not yet support secure browsing (HTTPS)."
And when I open my app with turned off secure browsing, it says: "The app "Ako dobre poznáš Miša P.?" is temporarily unavailable due to an issue with its third-party developer. We are investigating the situation and apologize for any inconvenience."
Thanks for any ideas. | 2012/03/16 | [
"https://Stackoverflow.com/questions/9739114",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1274220/"
] | Your app has to support https. They sent out notifications for this about 6 months ago.
As mentioned in the comments, you probably also need to [migrate to OAuth 2.0](http://developers.facebook.com/docs/oauth2-https-migration/).
You really have to keep on top of your facebook apps, as they are notorious for quickly deprecating and removing API components. To put it bluntly, you got Zucker punched. | You need to edit the settings in your app to supply an https URL. I created a very nice step by step guide that takes you through this and many other setup steps, setting up a Java WebService, Persistence, and registering for FREE web hosting (that you can use for your https address), resulting in a completed demo App with some sample FB calls. Goes from absolute 0 to fully functioning FB app with this 3 part blog series (part 1 starts here):
<http://ocpsoft.org/opensource/creating-a-facebook-app-setup-and-tool-installation/> |
9,739,114 | I've got a problem with my old app and would like to know how to fix it. I found some advices, but I don't understand them. I need step-by-step instructions.
When I open my app: <http://apps.facebook.com/ako-dobre-poz-jeadci/> it says: "This application does not yet support secure browsing (HTTPS)."
And when I open my app with turned off secure browsing, it says: "The app "Ako dobre poznáš Miša P.?" is temporarily unavailable due to an issue with its third-party developer. We are investigating the situation and apologize for any inconvenience."
Thanks for any ideas. | 2012/03/16 | [
"https://Stackoverflow.com/questions/9739114",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1274220/"
] | You need to edit the settings in your app to supply an https URL. I created a very nice step by step guide that takes you through this and many other setup steps, setting up a Java WebService, Persistence, and registering for FREE web hosting (that you can use for your https address), resulting in a completed demo App with some sample FB calls. Goes from absolute 0 to fully functioning FB app with this 3 part blog series (part 1 starts here):
<http://ocpsoft.org/opensource/creating-a-facebook-app-setup-and-tool-installation/> | Your app (and all of its resources) needs to be hosted on a secured domain (e.g.: <https://www.example.com>) with a valid SSL certificate |
9,739,114 | I've got a problem with my old app and would like to know how to fix it. I found some advices, but I don't understand them. I need step-by-step instructions.
When I open my app: <http://apps.facebook.com/ako-dobre-poz-jeadci/> it says: "This application does not yet support secure browsing (HTTPS)."
And when I open my app with turned off secure browsing, it says: "The app "Ako dobre poznáš Miša P.?" is temporarily unavailable due to an issue with its third-party developer. We are investigating the situation and apologize for any inconvenience."
Thanks for any ideas. | 2012/03/16 | [
"https://Stackoverflow.com/questions/9739114",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1274220/"
] | You need to edit the settings in your app to supply an https URL. I created a very nice step by step guide that takes you through this and many other setup steps, setting up a Java WebService, Persistence, and registering for FREE web hosting (that you can use for your https address), resulting in a completed demo App with some sample FB calls. Goes from absolute 0 to fully functioning FB app with this 3 part blog series (part 1 starts here):
<http://ocpsoft.org/opensource/creating-a-facebook-app-setup-and-tool-installation/> | In order to fix it, you have to buy and setup an SSL certificate. (it is now required since 1 October 2011 : [News](http://www.wpcode.net/fb-ssl.html/))
After, you have to add your https url in your app settings.
Check Facebook authentification guide for dev for more information about it:
[Authentification Guide - Facebook](https://developers.facebook.com/docs/authentication/) |
9,739,114 | I've got a problem with my old app and would like to know how to fix it. I found some advices, but I don't understand them. I need step-by-step instructions.
When I open my app: <http://apps.facebook.com/ako-dobre-poz-jeadci/> it says: "This application does not yet support secure browsing (HTTPS)."
And when I open my app with turned off secure browsing, it says: "The app "Ako dobre poznáš Miša P.?" is temporarily unavailable due to an issue with its third-party developer. We are investigating the situation and apologize for any inconvenience."
Thanks for any ideas. | 2012/03/16 | [
"https://Stackoverflow.com/questions/9739114",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/1274220/"
] | You need to edit the settings in your app to supply an https URL. I created a very nice step by step guide that takes you through this and many other setup steps, setting up a Java WebService, Persistence, and registering for FREE web hosting (that you can use for your https address), resulting in a completed demo App with some sample FB calls. Goes from absolute 0 to fully functioning FB app with this 3 part blog series (part 1 starts here):
<http://ocpsoft.org/opensource/creating-a-facebook-app-setup-and-tool-installation/> | You need to purchase and setup an SSL certificate and then plug your https url into your app's settings page. See their [migration guide](http://developers.facebook.com/docs/oauth2-https-migration/) where it says this is now required.
Source: [Secure browsing is not supported in my app](https://stackoverflow.com/questions/8603127/secure-browsing-is-not-supported-in-my-app). |
16,053 | How do Electronic & Mechanical Voting machines work? And how do they provide accurate votes? Is it overly complicated machine, or is it surprisingly simple? | 2017/03/02 | [
"https://politics.stackexchange.com/questions/16053",
"https://politics.stackexchange.com",
"https://politics.stackexchange.com/users/12707/"
] | This is a very broad question, so any answers are going to be either very high-level or really long. I'm going to go for the former, but I encourage you to read the links I'm including, and then ask followup questions either here or on [Information Security](https://security.stackexchange.com/questions/tagged/electronic-voting) about specifics.
---
First of all, there are many voting machines in use across the country. Each state sets its own rules, purchases its own machines, and conducts its own elections (even if the *day* is Federally mandated). [This page](https://www.verifiedvoting.org/resources/voting-equipment/) has a great deal of high-level information about the various voting machines, and you can drill down through the site to read more about each kind. One thing to note is that purely mechanical voting machines have been retired across the country (and I miss them!) - every machine is at least partially digital now.
Fundamentally, though, a voting machine consists of [two parts](https://en.wikipedia.org/wiki/Voting_machine): Voter input, and tallying. Voter input can be any way of inputting data into a computer (keyboard, touch screen, buttons, scantron, etc.). Tallying just means that the computer will keep a record of all the votes it sees. Depending on the state's laws and machine in question, this could be the official record, or simply a preliminary count used for quick election-night reporting.
Given how simple the basic functionality is, anyone with relevant knowledge can theoretically build a voting machine. However, the hard part is getting a state to use it, which usually means getting it certified in whatever manner the state requires. [This page](http://www.ncsl.org/research/elections-and-campaigns/voting-system-standards-testing-and-certification.aspx) goes through a lot of the desired features in a voting machine, the groups that set the (voluntary!) standards that machines should meet, and which states require which levels of testing. | I can think of at least three kinds of voting machines.
1. [Mechanical, with levers](http://www.fec.gov/pages/lever.htm). You pull down the lever for the candidate or option that you want. There usually are restrictions keeping you from selecting multiple choices in the same race (unless of course you are allowed to do that). When you finish making your selections, you press a button that records your choices and updates the internal tallies.
2. [Electronic scanner](https://en.wikipedia.org/wiki/Optical_scan_voting_system). Feed a paper ballot into a machine which scans the ballot to determine the choices. Usually keeps tallies as it goes.
3. Touch screen. The modern electronic version of the mechanical voting machine. Often adds a confirmation screen showing the voter's choices. Confirmation clears the screen and updates the internal tallies. May or [may not](http://www.reuters.com/article/us-usa-election-machines-idUSKCN11Q0EU) leave a paper trail.
And that's just the machines. Provisional and absentee ballots are often paper based and counted by hand.
Voting machines work in many different ways. How should voting machines work may well be a better question but is more discussion-based than objective. Some of the controversies involve
1. Should there be a paper trail?
2. If there is a paper trail, should it be voter verified? I.e. should we show the paper result to the voter and let them confirm or reject it.
3. If there is a paper trail, how important is it that we not be able to determine who cast the vote?
4. Should we be able to cancel votes cast by invalid voters?
5. How important are quick results?
6. How complicated are the counting systems?
7. How important are unambiguous results? Florida in 2000 suggested that this was pretty important.
8. How important is it that voting be intuitive?
These things contradict each other. A paper trail is easier to recount, but it is also easier to determine who cast which vote. Paper ballots are both anonymous and verifiable, but they are harder to count and less intuitive as the counting system gets more complicated. Anonymous votes prevent post-election cancellation if the voter is invalid. Voter drawn paper ballots are not unambiguous. Computer printed paper ballots require voter examination to be sure of casting the votes as intended, but that makes the system more complicated.
This is the general background. If you want something more specific, you'd need to ask a narrower question. |
12,854 | If the FBI reopens the investigation into Hillary Clinton's use of a private email server or they open an investigation into Hillary Clinton's mishandling of classified material, is it possible to postpone the election? Either to allow the democrats to come up with another candidate or for Hillary to get cleared of the allegations?
If it is, who would be triggering it? | 2016/10/30 | [
"https://politics.stackexchange.com/questions/12854",
"https://politics.stackexchange.com",
"https://politics.stackexchange.com/users/9930/"
] | The Current President's Term Must End
-------------------------------------
No matter what else happens, [Article II](https://www.law.cornell.edu/constitution/articleii), Section I of the Constitution sets the President and Vice President's terms at 4 years. Whether there is a new election or not, they must vacate their offices January 20.
Congress sets the date of the election
--------------------------------------
The same article gives Congress the ability to set the specific date of the election. They have met this obligation through a [law](https://www.law.cornell.edu/uscode/text/3/1) (US Code Title 3, Chapter 1, Section 1) which says:
>
> The electors of President and Vice President shall be appointed, in each State, on the Tuesday next after the first Monday in November, in every fourth year succeeding every election of a President and Vice President.
>
>
>
Would it be reasonable to pass a new law?
-----------------------------------------
The election is currently scheduled to take place in 9 days. Should Congress pass a law unusually quickly, the President would need to sign it before it had any legitimacy (or the President would have to veto it and Congress would need to override that veto). Unless the President and Congress are both completely on board, this is highly unlikely to happen.
For discussion, let's suppose they passed this act and it became effective on Monday, October 31st. At this point, states and parties have already heavily advertised the date of the election, election sites have been chosen and election workers hired (or recruited) for those dates. Changing the date of the election would require a massive amount of coordination and likely be perceived as unreasonable by the states and parties.
State Laws
----------
Many states also have laws which set the election date. For example, Kansas has [KSA 25-101(a)](http://kslegislature.org/li/b2015_16/statute/025_000_0000_chapter/025_001_0000_article/025_001_0001_section/025_001_0001_k/), which requires an election every-other year on the same date as the federal election.
State legislatures may not be in session, which means it would be impractical for them to change their laws this quickly. We can only speculate what would happen, but it would likely mean a sleepless night for Attorney Generals across the nation. It could mean that state elections happen on the required date, but federal elections happen on a new day.
Why would they?
---------------
Accusation of having committed a crime does not prevent someone from running for the presidency or being elected. Nothing in someone's criminal history prevents them from being president. There would be no (legal) reason to change the election date.
If voters decide this issue is important, they can choose not to vote or vote for someone else. | It appears that federal elections could be postponed via an act of congress:
<http://classroom.synonym.com/can-federal-election-postponed-20816.html>
I'll leave it to you to decide how plausible that would be with our current congress. |
52,735 | The tags [xsl](/questions/tagged/xsl "show questions tagged 'xsl'") and [xslt](/questions/tagged/xslt "show questions tagged 'xslt'") seem to be used pretty interchangeably on StackOverflow. In most, if not all cases, I feel that [xslt](/questions/tagged/xslt "show questions tagged 'xslt'") is how the question should be tagged.
In [many cases](https://stackoverflow.com/questions/tagged/xslt+xsl), questions are tagged as both.
Could the two tags be merged into [xslt](/questions/tagged/xslt "show questions tagged 'xslt'")? | 2010/06/06 | [
"https://meta.stackexchange.com/questions/52735",
"https://meta.stackexchange.com",
"https://meta.stackexchange.com/users/147400/"
] | The semantic distinction between "XSL" and "XSLT" may be useful in some contexts, but SO's tag taxonomy isn't one of them.
I browsed through the top 30 or so questions tagged with "xsl" to see if I could find one that was about XSL and not XSLT, and couldn't. In all cases, the questioner was using "XSL" and "XSLT" interchangeably. This shouldn't be terribly surprising; people to whom the distinction between the two is significant don't ask a lot of questions about either of them.
There are times when I'm adamant that the proper terminology be used (do not say "nodes" when you mean "elements" in my hearing), but I think that cleaving to this particular distinction, in this particular context, probably does no good and may do a little harm. | XSL and XSLT are two separate things. XSL is "Extensible Stylesheet Language", while XSLT is specifically "XSL Transforms". People can (and often do, on SO) use XSL alone to refer to XSLT, but it can also be used for XSL-FO (XSL Formatting Objects).
Think of it like how we have a separate C# tag and a C#4.0 tag. That kind of relationship isn't so much ambiguity, as it is a superclass.
Keeping them separate, I think, would be preferable. |
52,735 | The tags [xsl](/questions/tagged/xsl "show questions tagged 'xsl'") and [xslt](/questions/tagged/xslt "show questions tagged 'xslt'") seem to be used pretty interchangeably on StackOverflow. In most, if not all cases, I feel that [xslt](/questions/tagged/xslt "show questions tagged 'xslt'") is how the question should be tagged.
In [many cases](https://stackoverflow.com/questions/tagged/xslt+xsl), questions are tagged as both.
Could the two tags be merged into [xslt](/questions/tagged/xslt "show questions tagged 'xslt'")? | 2010/06/06 | [
"https://meta.stackexchange.com/questions/52735",
"https://meta.stackexchange.com",
"https://meta.stackexchange.com/users/147400/"
] | From <http://www.w3.org/Style/XSL/>
>
> XSL is a family of recommendations for
> defining XML document transformation
> and presentation. It consists of three
> parts:
>
>
> * XSL Transformations (XSLT) a language for transforming XML
> * the XML Path Language (XPath) an expression language used by XSLT to access or refer to parts of an XML document. (XPath is also used by the XML Linking specification)
> * XSL Formatting Objects (XSL-FO) an XML vocabulary for specifying formatting semantics
>
>
>
Then, unless people is asking about this family relationship, history, etc., the question is erroneously tagged.
Since there is no feature that automatically add a label to another, where it can be corrected if appropriate, I think it should stay this synonym. | Since that all the tag leaders agreed that it makes sense to have this synonym in place. I went ahead and created the synonym AND merged the tags.
In the vast majority of cases users mean xslt when they tag stuff xsl, hence I placed this synonym and merged. |
52,735 | The tags [xsl](/questions/tagged/xsl "show questions tagged 'xsl'") and [xslt](/questions/tagged/xslt "show questions tagged 'xslt'") seem to be used pretty interchangeably on StackOverflow. In most, if not all cases, I feel that [xslt](/questions/tagged/xslt "show questions tagged 'xslt'") is how the question should be tagged.
In [many cases](https://stackoverflow.com/questions/tagged/xslt+xsl), questions are tagged as both.
Could the two tags be merged into [xslt](/questions/tagged/xslt "show questions tagged 'xslt'")? | 2010/06/06 | [
"https://meta.stackexchange.com/questions/52735",
"https://meta.stackexchange.com",
"https://meta.stackexchange.com/users/147400/"
] | XSL and XSLT are two separate things. XSL is "Extensible Stylesheet Language", while XSLT is specifically "XSL Transforms". People can (and often do, on SO) use XSL alone to refer to XSLT, but it can also be used for XSL-FO (XSL Formatting Objects).
Think of it like how we have a separate C# tag and a C#4.0 tag. That kind of relationship isn't so much ambiguity, as it is a superclass.
Keeping them separate, I think, would be preferable. | Since that all the tag leaders agreed that it makes sense to have this synonym in place. I went ahead and created the synonym AND merged the tags.
In the vast majority of cases users mean xslt when they tag stuff xsl, hence I placed this synonym and merged. |
52,735 | The tags [xsl](/questions/tagged/xsl "show questions tagged 'xsl'") and [xslt](/questions/tagged/xslt "show questions tagged 'xslt'") seem to be used pretty interchangeably on StackOverflow. In most, if not all cases, I feel that [xslt](/questions/tagged/xslt "show questions tagged 'xslt'") is how the question should be tagged.
In [many cases](https://stackoverflow.com/questions/tagged/xslt+xsl), questions are tagged as both.
Could the two tags be merged into [xslt](/questions/tagged/xslt "show questions tagged 'xslt'")? | 2010/06/06 | [
"https://meta.stackexchange.com/questions/52735",
"https://meta.stackexchange.com",
"https://meta.stackexchange.com/users/147400/"
] | The semantic distinction between "XSL" and "XSLT" may be useful in some contexts, but SO's tag taxonomy isn't one of them.
I browsed through the top 30 or so questions tagged with "xsl" to see if I could find one that was about XSL and not XSLT, and couldn't. In all cases, the questioner was using "XSL" and "XSLT" interchangeably. This shouldn't be terribly surprising; people to whom the distinction between the two is significant don't ask a lot of questions about either of them.
There are times when I'm adamant that the proper terminology be used (do not say "nodes" when you mean "elements" in my hearing), but I think that cleaving to this particular distinction, in this particular context, probably does no good and may do a little harm. | **We must be practical**. We are not academic writers preparing a scientific tome on the origin and meanings of words in a particular topic-world.
**The vast majority** (much more than 90%) **of questions tagged "xsl" are xslt questions**. People, who ask xsl-fo questions typically tag them "xsl-fo" and such questions happen to be quite rare.
**So, we have a practical need here** -- in order to better serve all our users, who are not aware of the subtle problem that xsl isn't the same as xslt, and whose questions may stay unnoticed for a prolonged time because xslt experts sometimes have a lot of questions tagged "xslt", it is really useful if at SO "xsl" be permanently established as a synonym of "xslt".
**Let's avoid all the etimological, philosophical and historical aspects and just be practical**. |
52,735 | The tags [xsl](/questions/tagged/xsl "show questions tagged 'xsl'") and [xslt](/questions/tagged/xslt "show questions tagged 'xslt'") seem to be used pretty interchangeably on StackOverflow. In most, if not all cases, I feel that [xslt](/questions/tagged/xslt "show questions tagged 'xslt'") is how the question should be tagged.
In [many cases](https://stackoverflow.com/questions/tagged/xslt+xsl), questions are tagged as both.
Could the two tags be merged into [xslt](/questions/tagged/xslt "show questions tagged 'xslt'")? | 2010/06/06 | [
"https://meta.stackexchange.com/questions/52735",
"https://meta.stackexchange.com",
"https://meta.stackexchange.com/users/147400/"
] | The semantic distinction between "XSL" and "XSLT" may be useful in some contexts, but SO's tag taxonomy isn't one of them.
I browsed through the top 30 or so questions tagged with "xsl" to see if I could find one that was about XSL and not XSLT, and couldn't. In all cases, the questioner was using "XSL" and "XSLT" interchangeably. This shouldn't be terribly surprising; people to whom the distinction between the two is significant don't ask a lot of questions about either of them.
There are times when I'm adamant that the proper terminology be used (do not say "nodes" when you mean "elements" in my hearing), but I think that cleaving to this particular distinction, in this particular context, probably does no good and may do a little harm. | From <http://www.w3.org/Style/XSL/>
>
> XSL is a family of recommendations for
> defining XML document transformation
> and presentation. It consists of three
> parts:
>
>
> * XSL Transformations (XSLT) a language for transforming XML
> * the XML Path Language (XPath) an expression language used by XSLT to access or refer to parts of an XML document. (XPath is also used by the XML Linking specification)
> * XSL Formatting Objects (XSL-FO) an XML vocabulary for specifying formatting semantics
>
>
>
Then, unless people is asking about this family relationship, history, etc., the question is erroneously tagged.
Since there is no feature that automatically add a label to another, where it can be corrected if appropriate, I think it should stay this synonym. |
31,654 | From the beginning of the film, it looks that Yubaba is searching for Chihiro with the clear intention to turn her into animal.
What reasons prevent Yubaba to do so when Chihiro suddenly comes to her office? While being rude, Yubaba still talks to Chihiro and at the end offers to sign the job contract. Looks like Yubaba is restricted by some rules she must follow. Why, and what are these rules? | 2016/04/24 | [
"https://anime.stackexchange.com/questions/31654",
"https://anime.stackexchange.com",
"https://anime.stackexchange.com/users/23662/"
] | She cannot refuse her as long as chihiro asks. Yubaba also has a rule not to turn down anyone asking for a job and therefore has to give her the job and can't be turned into a piglet as she is under her service. | It was two reasons actually: Yubaba's oath and...... Big Baby's tantrum!
In the English dub, Yubaba said: "*I cant believe I took that oath. To give a job to anyone who asks. Ridiculous! I hate being so nice all the time.*"
So yeah she can't turn down a job request but she can intimidate you to change your mind.
Yubaba was having fun intimidating Chihiro. Until Big Baby woke up. She doesn't want to deal with Big Baby's tantrum, so she gave in to Chihiro's incessant request.
[](https://i.stack.imgur.com/cZtOS.png)
One thing I find funny during that scene is, she told Chihiro that she doesn't want her because she's a spoiled cry-baby, with no manners. Hey lady, you just described your son! |
6,982,607 | I am planning to build a website for courses online, teachers can create their courses and put their exams, students can enroll, view courses and apply for exams ..
I am a bit confused, to build it with DDN or Drupal ? which is easier and more powerful ?
I have no problem with .NET C# or PHP, although I see C# more easy to code with,
what I need to know, which CMS is better for my case ?
Which CMS is more robust & complete ?
Thanks in advance. | 2011/08/08 | [
"https://Stackoverflow.com/questions/6982607",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/396681/"
] | Based on what you are trying to build DotNetNuke will only really give you authentication and basic page creation out of the box. All those other elements will have to be custom coded in .NET.
If you use Drupal 7 you can build majority of that using CCK, Views and workflow without really having to do a ton of programming if you don't want to.
I have been doing DotNetNuke since 2003 and have developed tons of modules for it, including some publicly available at [www.dnnspot.com](http://www.dnnspot.com)
If I was in your shoes I would do Drupal, but I know both pretty well.
ACTUALLY - If 'I' was in your shoes I would do Ruby on Rails now ;) but ... Drupal would work pretty well and you could get alot if not all of this done without slinging much code. Drupal is crazy. | DNN is now built on C# but has always been on .NET. I'm in a similar position as I code in PHP and .NET and have used Drupal and have been working on DNN for the last 4 years. I find that Module Development in DNN is wonderful.
i hope that helps. |
6,982,607 | I am planning to build a website for courses online, teachers can create their courses and put their exams, students can enroll, view courses and apply for exams ..
I am a bit confused, to build it with DDN or Drupal ? which is easier and more powerful ?
I have no problem with .NET C# or PHP, although I see C# more easy to code with,
what I need to know, which CMS is better for my case ?
Which CMS is more robust & complete ?
Thanks in advance. | 2011/08/08 | [
"https://Stackoverflow.com/questions/6982607",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/396681/"
] | Of course, either platform has its pros and cons. What you're looking at building is essentially a Learning Management System, or LMS. There are a few existing LMS solutions built specifically for DotNetNuke. They include [Engage: Campus](http://www.engagesoftware.com/Products/Modules/Engage_Campus.aspx), [NetLearn](http://www.snowcovered.com/Snowcovered2/Default.aspx?tabid=242&PackageID=18056), and [Accord](http://www.snowcovered.com/Snowcovered2/Default.aspx?tabid=242&PackageID=22737).
I would tell you to go with which ever platform offers you the most features for managing this kind of content for the best return on your investment, and offers you the best comfort level. If you're familiar with .Net and are comfortable in that arena, then stick with DotNetNuke. Otherwise go with another platform. However, having nearly 8 years of experience in developing solutions with DotNetNuke, and 13+ overall in developing software solutions (including PHP-based ones), if I have a choice, I always go with DotNetNuke. (And not just because I am an employee for a year now.)
I find that DotNetNuke has the most stable builds and releases overall. It's security team and features are second to none. It's community and ecosystem is unrivaled. If you're looking for a pre-built solution for any kind of business problem, chances are that the 10,000+ extensions in the [Store](http://www.snowcovered.com/) or [Forge](http://forge.dotnetnuke.com) will get you up and running with less development time and less expense. The forge has free (open source) extensions, and the median price of a store extension is $89 (LMS's are quite more expensive though).
The eco-system also boasts over 800 known ISV's out there. So chances are, if you're looking for support, not only is there an official company to provide it and back you up when something goes wrong, there are numerous other options out there in terms of companies that specialize in DotNetNuke.
One last thing on the releases... DotNetNuke has a known roadmap, a frequent release cycle, and an outstanding reputation for backwards compatibility. Your extensions will continue to work on the platform as you upgrade to take advantage of the latest and greatest features added to the platform, usability updates, bug fixes, and security updates.
As a Sales Engineer at DNN Corp, I hear all kinds of use cases, and I see DotNetNuke used a lot in educational institutions. They all use it to varying degrees as not only a CMS, but also a LMS, social platform (even in schools), and more. Most schools are also standardized in some way on the Microsoft stack too, so integration is much more straight-forward using DotNetNuke.
I could go on, but this is as much time as I have to give you an informed response. | DNN is now built on C# but has always been on .NET. I'm in a similar position as I code in PHP and .NET and have used Drupal and have been working on DNN for the last 4 years. I find that Module Development in DNN is wonderful.
i hope that helps. |
6,982,607 | I am planning to build a website for courses online, teachers can create their courses and put their exams, students can enroll, view courses and apply for exams ..
I am a bit confused, to build it with DDN or Drupal ? which is easier and more powerful ?
I have no problem with .NET C# or PHP, although I see C# more easy to code with,
what I need to know, which CMS is better for my case ?
Which CMS is more robust & complete ?
Thanks in advance. | 2011/08/08 | [
"https://Stackoverflow.com/questions/6982607",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/396681/"
] | Of course, either platform has its pros and cons. What you're looking at building is essentially a Learning Management System, or LMS. There are a few existing LMS solutions built specifically for DotNetNuke. They include [Engage: Campus](http://www.engagesoftware.com/Products/Modules/Engage_Campus.aspx), [NetLearn](http://www.snowcovered.com/Snowcovered2/Default.aspx?tabid=242&PackageID=18056), and [Accord](http://www.snowcovered.com/Snowcovered2/Default.aspx?tabid=242&PackageID=22737).
I would tell you to go with which ever platform offers you the most features for managing this kind of content for the best return on your investment, and offers you the best comfort level. If you're familiar with .Net and are comfortable in that arena, then stick with DotNetNuke. Otherwise go with another platform. However, having nearly 8 years of experience in developing solutions with DotNetNuke, and 13+ overall in developing software solutions (including PHP-based ones), if I have a choice, I always go with DotNetNuke. (And not just because I am an employee for a year now.)
I find that DotNetNuke has the most stable builds and releases overall. It's security team and features are second to none. It's community and ecosystem is unrivaled. If you're looking for a pre-built solution for any kind of business problem, chances are that the 10,000+ extensions in the [Store](http://www.snowcovered.com/) or [Forge](http://forge.dotnetnuke.com) will get you up and running with less development time and less expense. The forge has free (open source) extensions, and the median price of a store extension is $89 (LMS's are quite more expensive though).
The eco-system also boasts over 800 known ISV's out there. So chances are, if you're looking for support, not only is there an official company to provide it and back you up when something goes wrong, there are numerous other options out there in terms of companies that specialize in DotNetNuke.
One last thing on the releases... DotNetNuke has a known roadmap, a frequent release cycle, and an outstanding reputation for backwards compatibility. Your extensions will continue to work on the platform as you upgrade to take advantage of the latest and greatest features added to the platform, usability updates, bug fixes, and security updates.
As a Sales Engineer at DNN Corp, I hear all kinds of use cases, and I see DotNetNuke used a lot in educational institutions. They all use it to varying degrees as not only a CMS, but also a LMS, social platform (even in schools), and more. Most schools are also standardized in some way on the Microsoft stack too, so integration is much more straight-forward using DotNetNuke.
I could go on, but this is as much time as I have to give you an informed response. | Based on what you are trying to build DotNetNuke will only really give you authentication and basic page creation out of the box. All those other elements will have to be custom coded in .NET.
If you use Drupal 7 you can build majority of that using CCK, Views and workflow without really having to do a ton of programming if you don't want to.
I have been doing DotNetNuke since 2003 and have developed tons of modules for it, including some publicly available at [www.dnnspot.com](http://www.dnnspot.com)
If I was in your shoes I would do Drupal, but I know both pretty well.
ACTUALLY - If 'I' was in your shoes I would do Ruby on Rails now ;) but ... Drupal would work pretty well and you could get alot if not all of this done without slinging much code. Drupal is crazy. |
6,982,607 | I am planning to build a website for courses online, teachers can create their courses and put their exams, students can enroll, view courses and apply for exams ..
I am a bit confused, to build it with DDN or Drupal ? which is easier and more powerful ?
I have no problem with .NET C# or PHP, although I see C# more easy to code with,
what I need to know, which CMS is better for my case ?
Which CMS is more robust & complete ?
Thanks in advance. | 2011/08/08 | [
"https://Stackoverflow.com/questions/6982607",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/396681/"
] | Based on what you are trying to build DotNetNuke will only really give you authentication and basic page creation out of the box. All those other elements will have to be custom coded in .NET.
If you use Drupal 7 you can build majority of that using CCK, Views and workflow without really having to do a ton of programming if you don't want to.
I have been doing DotNetNuke since 2003 and have developed tons of modules for it, including some publicly available at [www.dnnspot.com](http://www.dnnspot.com)
If I was in your shoes I would do Drupal, but I know both pretty well.
ACTUALLY - If 'I' was in your shoes I would do Ruby on Rails now ;) but ... Drupal would work pretty well and you could get alot if not all of this done without slinging much code. Drupal is crazy. | Full featured Learning Management Systems (LMS) require substantial development (man decades), especially if you want to support SCORM eLearning content. Both DNN and Drupal will offer you most of the add on functionality that you will need in addition to the LMS features - the ability to build web pages, rich text editors, etc.
I would evaluate the different the LMS available for each platform and make your choice based on how robust the LMS are. In addition, evaluate how easily the LMS integrates with add on modules such as forums, wikis, etc. Most of our clients select our LMS without any knowledge of the portal framework beneath it.
Based on these evaluation metrics I am confident that you will find the [Interzoic Accord LMS](http://www.interzoic.com) running on DNN to be a superior choice.
Chris |
6,982,607 | I am planning to build a website for courses online, teachers can create their courses and put their exams, students can enroll, view courses and apply for exams ..
I am a bit confused, to build it with DDN or Drupal ? which is easier and more powerful ?
I have no problem with .NET C# or PHP, although I see C# more easy to code with,
what I need to know, which CMS is better for my case ?
Which CMS is more robust & complete ?
Thanks in advance. | 2011/08/08 | [
"https://Stackoverflow.com/questions/6982607",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/396681/"
] | Of course, either platform has its pros and cons. What you're looking at building is essentially a Learning Management System, or LMS. There are a few existing LMS solutions built specifically for DotNetNuke. They include [Engage: Campus](http://www.engagesoftware.com/Products/Modules/Engage_Campus.aspx), [NetLearn](http://www.snowcovered.com/Snowcovered2/Default.aspx?tabid=242&PackageID=18056), and [Accord](http://www.snowcovered.com/Snowcovered2/Default.aspx?tabid=242&PackageID=22737).
I would tell you to go with which ever platform offers you the most features for managing this kind of content for the best return on your investment, and offers you the best comfort level. If you're familiar with .Net and are comfortable in that arena, then stick with DotNetNuke. Otherwise go with another platform. However, having nearly 8 years of experience in developing solutions with DotNetNuke, and 13+ overall in developing software solutions (including PHP-based ones), if I have a choice, I always go with DotNetNuke. (And not just because I am an employee for a year now.)
I find that DotNetNuke has the most stable builds and releases overall. It's security team and features are second to none. It's community and ecosystem is unrivaled. If you're looking for a pre-built solution for any kind of business problem, chances are that the 10,000+ extensions in the [Store](http://www.snowcovered.com/) or [Forge](http://forge.dotnetnuke.com) will get you up and running with less development time and less expense. The forge has free (open source) extensions, and the median price of a store extension is $89 (LMS's are quite more expensive though).
The eco-system also boasts over 800 known ISV's out there. So chances are, if you're looking for support, not only is there an official company to provide it and back you up when something goes wrong, there are numerous other options out there in terms of companies that specialize in DotNetNuke.
One last thing on the releases... DotNetNuke has a known roadmap, a frequent release cycle, and an outstanding reputation for backwards compatibility. Your extensions will continue to work on the platform as you upgrade to take advantage of the latest and greatest features added to the platform, usability updates, bug fixes, and security updates.
As a Sales Engineer at DNN Corp, I hear all kinds of use cases, and I see DotNetNuke used a lot in educational institutions. They all use it to varying degrees as not only a CMS, but also a LMS, social platform (even in schools), and more. Most schools are also standardized in some way on the Microsoft stack too, so integration is much more straight-forward using DotNetNuke.
I could go on, but this is as much time as I have to give you an informed response. | Full featured Learning Management Systems (LMS) require substantial development (man decades), especially if you want to support SCORM eLearning content. Both DNN and Drupal will offer you most of the add on functionality that you will need in addition to the LMS features - the ability to build web pages, rich text editors, etc.
I would evaluate the different the LMS available for each platform and make your choice based on how robust the LMS are. In addition, evaluate how easily the LMS integrates with add on modules such as forums, wikis, etc. Most of our clients select our LMS without any knowledge of the portal framework beneath it.
Based on these evaluation metrics I am confident that you will find the [Interzoic Accord LMS](http://www.interzoic.com) running on DNN to be a superior choice.
Chris |
41,425 | A spaceship of several kilometers length is accelerating to a very high fraction of $c$ (basically as close as they can possibly get).
***Which problems can the machinery and the crew encounter?***
And as follow-up question resulting from this:
***Is a multiple kilometer long ship traveling at near light-speed feasible?***
To clarify, I'm interested on how travelling at near-light-speed is different to travelling at "normal" speed. Dodging obstacles is not in the scope of this question and [has already been addressed](https://worldbuilding.stackexchange.com/q/6335/108).
---
I expect the following things to be problematic, but don't know if they are actually. It would be nice if you could address them, as well as adding other problems you see.
* Moving fuel to the engine. It has to be decelerated to move from the tank to the drive at the back of the ship. What happens with that energy?
* Moving inside the ship. If it is possible for a human to actually accelerate enough to move, I expect moving down the same corridor in two directions can already be quite an adventure.
* Communicating. If a sensor picks up a problem at the drive, the signal has to travel to a mechanics/computer console, be evaluated (maybe in slow human brains) and then be reacted on.
* Turning the ship to brake. During the turning, different parts of the ship move at slightly different speeds. | 2016/05/09 | [
"https://worldbuilding.stackexchange.com/questions/41425",
"https://worldbuilding.stackexchange.com",
"https://worldbuilding.stackexchange.com/users/108/"
] | Travelling at near-light speed is **indistinguishable** from being at rest.
---------------------------------------------------------------------------
This is the [principle of Galilean relativity](https://en.wikipedia.org/wiki/Galilean_invariance), and is preserved under special relativity: there is no feasible experiment to determine whether you are stationary or moving at a constant velocity. So there are *no* practical problems that would occur simply from moving at near-light speed that wouldn't occur at rest, and vice-versa.
Since everything inside the ship is moving *with* the ship at the same speed (this is why you aren't flung backwards when you jump while on a moving train), neither the crew nor the fuel nor the communications will be affected: for them, the world will be exactly as it would be it were not moving at all. It's the world *outside* the ship that would appear to experience the odd effects of Lorentz contraction, time dilation, redshifting, and what-not: everything inside the ship is its own protected bubble. (Of course, someone *outside* the ship would claim it was really the ship itself experiencing these odd effects, and that's perfectly fine).
The only time engineering is required is to handle the following cases:
* **Acceleration**: This is the only time when frame invariance is relaxed. Deceleration will cause massive stresses on your ship building materials, depending on how quickly you want to come to an abrupt halt. You should anticipate long deceleration times, something your crew will have to take into account. Conversely, accelerating to light speed will also take a long time if you want to avoid enormous compression stress. **This includes rotation**. A full treatment of what the world looks like for such rotating frames can be explored by examining [the Kerr metric](https://en.wikipedia.org/wiki/Kerr_metric).
* **Navigation**: Your starboard computers will register large amounts of redshifting from nearby stars, which can skew your metrics of where you are or what you are dealing with if you're using [Hubble's law](https://en.wikipedia.org/wiki/Hubble%27s_law). A correction factor must be applied to all of your sensors to account for this - this can be easily calculated by knowledge of the ship's velocity.
* **Synchronisation with off-ship clocks**: For the crew inside the ship, it will appear as if life outside the ship is moving at a much slower rate. As a consequence, significant discrepancies between clocks at the ship's port of arrival and the ship's internal clocks will occur. This affects messages sent to the ship and any attempt to synchronise with the world outside. Again, a correction factor can be calculated and employed but *must* be known in advance.
* **Knowledge of ship velocity**: While this is on the surface not hard to solve - just measure the velocity of things moving relative to the ship - the devil lies in the details. Objects moving relative to the ship have their own velocity *prior* to the ship's acceleration, which skews measurements. One way out is to use [standard candles](http://hyperphysics.phy-astr.gsu.edu/hbase/astro/stdcand.html), which we already currently use to estimate distances - simply compare the luminous intensity from a standard candle at one second and the luminous intensity at another, and use it to work out the distance that has been travelled. From this you can work out velocity.
You may also want to consider general relativistic effects: a ship moving at that speed has an incredible amount of kinetic energy, and will thus exert a significant gravitational influence in its wake (assuming its mass is already sizable to begin with). You'd have to figure out a decent flight plan that doesn't leave *too* much damage behind. | AndreiROM is right, thanks to our friend inertia the only time you'd notice anything is while the ship is actively thrusting during acceleration and deceleration, or if you turned really sharply.
This actually has something of a beneficial nature because you could build the ship to where the floor is "down" during thrusting, and so you get something like gravity.
Imagine the ship is sitting on it's drive in Earth gravity. Build the decks so that the floors are where gravity is pulling, and have lifts/ladders to get between decks.
While the drive is operational it will be pushing the floor up toward your feet with the net effect that it will feel like there is gravity. How much gravity depends on how hard the ship is thrusting.
It would be more like a several kilometer high skyscraper instead of a several kilometer long ship. If the decks are laid out parallel to the direction of travel then you will be pushed back toward the back of the ship any time the drive is going, and this would severely limit the amount of thrusting you could do to get up to your desired speed, unless artificial gravity is a thing.
This question has some interesting numbers when it comes to travel times, in case it helps you any:
<https://space.stackexchange.com/questions/840/how-fast-will-1g-get-you-there> |
41,425 | A spaceship of several kilometers length is accelerating to a very high fraction of $c$ (basically as close as they can possibly get).
***Which problems can the machinery and the crew encounter?***
And as follow-up question resulting from this:
***Is a multiple kilometer long ship traveling at near light-speed feasible?***
To clarify, I'm interested on how travelling at near-light-speed is different to travelling at "normal" speed. Dodging obstacles is not in the scope of this question and [has already been addressed](https://worldbuilding.stackexchange.com/q/6335/108).
---
I expect the following things to be problematic, but don't know if they are actually. It would be nice if you could address them, as well as adding other problems you see.
* Moving fuel to the engine. It has to be decelerated to move from the tank to the drive at the back of the ship. What happens with that energy?
* Moving inside the ship. If it is possible for a human to actually accelerate enough to move, I expect moving down the same corridor in two directions can already be quite an adventure.
* Communicating. If a sensor picks up a problem at the drive, the signal has to travel to a mechanics/computer console, be evaluated (maybe in slow human brains) and then be reacted on.
* Turning the ship to brake. During the turning, different parts of the ship move at slightly different speeds. | 2016/05/09 | [
"https://worldbuilding.stackexchange.com/questions/41425",
"https://worldbuilding.stackexchange.com",
"https://worldbuilding.stackexchange.com/users/108/"
] | Travelling at near-light speed is **indistinguishable** from being at rest.
---------------------------------------------------------------------------
This is the [principle of Galilean relativity](https://en.wikipedia.org/wiki/Galilean_invariance), and is preserved under special relativity: there is no feasible experiment to determine whether you are stationary or moving at a constant velocity. So there are *no* practical problems that would occur simply from moving at near-light speed that wouldn't occur at rest, and vice-versa.
Since everything inside the ship is moving *with* the ship at the same speed (this is why you aren't flung backwards when you jump while on a moving train), neither the crew nor the fuel nor the communications will be affected: for them, the world will be exactly as it would be it were not moving at all. It's the world *outside* the ship that would appear to experience the odd effects of Lorentz contraction, time dilation, redshifting, and what-not: everything inside the ship is its own protected bubble. (Of course, someone *outside* the ship would claim it was really the ship itself experiencing these odd effects, and that's perfectly fine).
The only time engineering is required is to handle the following cases:
* **Acceleration**: This is the only time when frame invariance is relaxed. Deceleration will cause massive stresses on your ship building materials, depending on how quickly you want to come to an abrupt halt. You should anticipate long deceleration times, something your crew will have to take into account. Conversely, accelerating to light speed will also take a long time if you want to avoid enormous compression stress. **This includes rotation**. A full treatment of what the world looks like for such rotating frames can be explored by examining [the Kerr metric](https://en.wikipedia.org/wiki/Kerr_metric).
* **Navigation**: Your starboard computers will register large amounts of redshifting from nearby stars, which can skew your metrics of where you are or what you are dealing with if you're using [Hubble's law](https://en.wikipedia.org/wiki/Hubble%27s_law). A correction factor must be applied to all of your sensors to account for this - this can be easily calculated by knowledge of the ship's velocity.
* **Synchronisation with off-ship clocks**: For the crew inside the ship, it will appear as if life outside the ship is moving at a much slower rate. As a consequence, significant discrepancies between clocks at the ship's port of arrival and the ship's internal clocks will occur. This affects messages sent to the ship and any attempt to synchronise with the world outside. Again, a correction factor can be calculated and employed but *must* be known in advance.
* **Knowledge of ship velocity**: While this is on the surface not hard to solve - just measure the velocity of things moving relative to the ship - the devil lies in the details. Objects moving relative to the ship have their own velocity *prior* to the ship's acceleration, which skews measurements. One way out is to use [standard candles](http://hyperphysics.phy-astr.gsu.edu/hbase/astro/stdcand.html), which we already currently use to estimate distances - simply compare the luminous intensity from a standard candle at one second and the luminous intensity at another, and use it to work out the distance that has been travelled. From this you can work out velocity.
You may also want to consider general relativistic effects: a ship moving at that speed has an incredible amount of kinetic energy, and will thus exert a significant gravitational influence in its wake (assuming its mass is already sizable to begin with). You'd have to figure out a decent flight plan that doesn't leave *too* much damage behind. | One rather big problem that you haven't mentioned is the impact of material in the interstellar medium. While gas and dust in space is extremely diffuse, at near-light speeds even a few hydrogen atoms, let alone dust particles or micrometeors, will carry significant energy that will wear away the hull of the ship over time. You're going to want to put some sort of shielding up at the front of the ship to absorb the damage. Also, unlike slower ships which can be shaped like anything, it actually benefits near-light-speed ships to have a streamlined shape to reduce friction and damage. |
41,425 | A spaceship of several kilometers length is accelerating to a very high fraction of $c$ (basically as close as they can possibly get).
***Which problems can the machinery and the crew encounter?***
And as follow-up question resulting from this:
***Is a multiple kilometer long ship traveling at near light-speed feasible?***
To clarify, I'm interested on how travelling at near-light-speed is different to travelling at "normal" speed. Dodging obstacles is not in the scope of this question and [has already been addressed](https://worldbuilding.stackexchange.com/q/6335/108).
---
I expect the following things to be problematic, but don't know if they are actually. It would be nice if you could address them, as well as adding other problems you see.
* Moving fuel to the engine. It has to be decelerated to move from the tank to the drive at the back of the ship. What happens with that energy?
* Moving inside the ship. If it is possible for a human to actually accelerate enough to move, I expect moving down the same corridor in two directions can already be quite an adventure.
* Communicating. If a sensor picks up a problem at the drive, the signal has to travel to a mechanics/computer console, be evaluated (maybe in slow human brains) and then be reacted on.
* Turning the ship to brake. During the turning, different parts of the ship move at slightly different speeds. | 2016/05/09 | [
"https://worldbuilding.stackexchange.com/questions/41425",
"https://worldbuilding.stackexchange.com",
"https://worldbuilding.stackexchange.com/users/108/"
] | AndreiROM is right, thanks to our friend inertia the only time you'd notice anything is while the ship is actively thrusting during acceleration and deceleration, or if you turned really sharply.
This actually has something of a beneficial nature because you could build the ship to where the floor is "down" during thrusting, and so you get something like gravity.
Imagine the ship is sitting on it's drive in Earth gravity. Build the decks so that the floors are where gravity is pulling, and have lifts/ladders to get between decks.
While the drive is operational it will be pushing the floor up toward your feet with the net effect that it will feel like there is gravity. How much gravity depends on how hard the ship is thrusting.
It would be more like a several kilometer high skyscraper instead of a several kilometer long ship. If the decks are laid out parallel to the direction of travel then you will be pushed back toward the back of the ship any time the drive is going, and this would severely limit the amount of thrusting you could do to get up to your desired speed, unless artificial gravity is a thing.
This question has some interesting numbers when it comes to travel times, in case it helps you any:
<https://space.stackexchange.com/questions/840/how-fast-will-1g-get-you-there> | One rather big problem that you haven't mentioned is the impact of material in the interstellar medium. While gas and dust in space is extremely diffuse, at near-light speeds even a few hydrogen atoms, let alone dust particles or micrometeors, will carry significant energy that will wear away the hull of the ship over time. You're going to want to put some sort of shielding up at the front of the ship to absorb the damage. Also, unlike slower ships which can be shaped like anything, it actually benefits near-light-speed ships to have a streamlined shape to reduce friction and damage. |
5,742 | I know that keeping /home in different partitions will preserve my data.
But what about the user-specific configuration files? Are always forward-compatible?
Are there other issues as well? | 2009/05/07 | [
"https://serverfault.com/questions/5742",
"https://serverfault.com",
"https://serverfault.com/users/2165/"
] | I'd like to add a me-too to Mike Arthur's answer. I've used a Linux desktop since 1997 and have migrated my home directory along the way. Most apps just work.
The only problems I've had was when I switched distributions. Upgrading from RedHat to newer RedHat or RedHat to Fedora or Fedora to newer Fedora is usually really easy. But when I switched from RedHat to Mandrake (now Mandriva) and Mandrake to Fedora it was not a pretty sight. Mandrake and Fedora both used a customized KDE directory structure that was incompatible with each other. As a consequence at that time I deleted most of my .kde directory to recreate the settings. But since then I've not really had problems.
In any case, you can always just erase whatever's in your home directory that you don't want after a system update. I find it extremely convenient that my data is separate from my programs. Even if you have to blow away your whole home directory every two years, as a Fedora user that would save me about 2-3 backup-restore cycles (Fedora updates every 6 months).
You can also set up a directory for use with programs. I typically have /usr/local as a separate partition. I install anything I compile from scratch there. This doesn't work as flawlessly as in the data case because many times the distros are not binary compatible with each other. But at least your programs are all there and you can assess them as needed. | It's mostly up to the individual apps to upgrade their per-user config files. Upgrading a server is mostly dependent upon the distro you're using, and can range from having to pick out individual apps to rebuild and reinstall (a al Slackware) to simply telling the package manager to upgrade everything to the latest rev (a la Ubuntu, debian, and redhat variants) |
5,742 | I know that keeping /home in different partitions will preserve my data.
But what about the user-specific configuration files? Are always forward-compatible?
Are there other issues as well? | 2009/05/07 | [
"https://serverfault.com/questions/5742",
"https://serverfault.com",
"https://serverfault.com/users/2165/"
] | It's mostly up to the individual apps to upgrade their per-user config files. Upgrading a server is mostly dependent upon the distro you're using, and can range from having to pick out individual apps to rebuild and reinstall (a al Slackware) to simply telling the package manager to upgrade everything to the latest rev (a la Ubuntu, debian, and redhat variants) | For config files:
Some GNU/Linux distributions have a system in place that will show you diff's as it updates the configuration files. (eg: gentoo has etc-update which will show you a diff and ask you to keep the old config, use the new one, or interactively merge the two.)
If you don't have such luxery with the distro you are using, make backups of /etc and the relevant config files, and diff/merge them manually.
Always read the docs and manpages of whatever you're editing the config files for. |
5,742 | I know that keeping /home in different partitions will preserve my data.
But what about the user-specific configuration files? Are always forward-compatible?
Are there other issues as well? | 2009/05/07 | [
"https://serverfault.com/questions/5742",
"https://serverfault.com",
"https://serverfault.com/users/2165/"
] | I'd like to add a me-too to Mike Arthur's answer. I've used a Linux desktop since 1997 and have migrated my home directory along the way. Most apps just work.
The only problems I've had was when I switched distributions. Upgrading from RedHat to newer RedHat or RedHat to Fedora or Fedora to newer Fedora is usually really easy. But when I switched from RedHat to Mandrake (now Mandriva) and Mandrake to Fedora it was not a pretty sight. Mandrake and Fedora both used a customized KDE directory structure that was incompatible with each other. As a consequence at that time I deleted most of my .kde directory to recreate the settings. But since then I've not really had problems.
In any case, you can always just erase whatever's in your home directory that you don't want after a system update. I find it extremely convenient that my data is separate from my programs. Even if you have to blow away your whole home directory every two years, as a Fedora user that would save me about 2-3 backup-restore cycles (Fedora updates every 6 months).
You can also set up a directory for use with programs. I typically have /usr/local as a separate partition. I install anything I compile from scratch there. This doesn't work as flawlessly as in the data case because many times the distros are not binary compatible with each other. But at least your programs are all there and you can assess them as needed. | I've been using Linux on desktop and server for about 5 years and had the same home directory for pretty much that whole time and had pretty much flawless upgrades of the dotfiles in my home directory. |
5,742 | I know that keeping /home in different partitions will preserve my data.
But what about the user-specific configuration files? Are always forward-compatible?
Are there other issues as well? | 2009/05/07 | [
"https://serverfault.com/questions/5742",
"https://serverfault.com",
"https://serverfault.com/users/2165/"
] | I've been using Linux on desktop and server for about 5 years and had the same home directory for pretty much that whole time and had pretty much flawless upgrades of the dotfiles in my home directory. | For config files:
Some GNU/Linux distributions have a system in place that will show you diff's as it updates the configuration files. (eg: gentoo has etc-update which will show you a diff and ask you to keep the old config, use the new one, or interactively merge the two.)
If you don't have such luxery with the distro you are using, make backups of /etc and the relevant config files, and diff/merge them manually.
Always read the docs and manpages of whatever you're editing the config files for. |
5,742 | I know that keeping /home in different partitions will preserve my data.
But what about the user-specific configuration files? Are always forward-compatible?
Are there other issues as well? | 2009/05/07 | [
"https://serverfault.com/questions/5742",
"https://serverfault.com",
"https://serverfault.com/users/2165/"
] | I'd like to add a me-too to Mike Arthur's answer. I've used a Linux desktop since 1997 and have migrated my home directory along the way. Most apps just work.
The only problems I've had was when I switched distributions. Upgrading from RedHat to newer RedHat or RedHat to Fedora or Fedora to newer Fedora is usually really easy. But when I switched from RedHat to Mandrake (now Mandriva) and Mandrake to Fedora it was not a pretty sight. Mandrake and Fedora both used a customized KDE directory structure that was incompatible with each other. As a consequence at that time I deleted most of my .kde directory to recreate the settings. But since then I've not really had problems.
In any case, you can always just erase whatever's in your home directory that you don't want after a system update. I find it extremely convenient that my data is separate from my programs. Even if you have to blow away your whole home directory every two years, as a Fedora user that would save me about 2-3 backup-restore cycles (Fedora updates every 6 months).
You can also set up a directory for use with programs. I typically have /usr/local as a separate partition. I install anything I compile from scratch there. This doesn't work as flawlessly as in the data case because many times the distros are not binary compatible with each other. But at least your programs are all there and you can assess them as needed. | For config files:
Some GNU/Linux distributions have a system in place that will show you diff's as it updates the configuration files. (eg: gentoo has etc-update which will show you a diff and ask you to keep the old config, use the new one, or interactively merge the two.)
If you don't have such luxery with the distro you are using, make backups of /etc and the relevant config files, and diff/merge them manually.
Always read the docs and manpages of whatever you're editing the config files for. |
1,994,174 | Am very to new to web development.
I have the web pages; web pages are developed in html, CSS Style sheet.
For Example
I have the ftp domain or crystal.com for hosting my web page
For hosting my web pages, I have to create setup file for hosting my web page or simply post my html files.
Can any one tell the procedure, what are the things I have to do for the web hosting?
Note: I am not asking about Web hosting domain. Already I have the domain, but how to post my html files in that domain.
Need Help? | 2010/01/03 | [
"https://Stackoverflow.com/questions/1994174",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/128071/"
] | Ftp is the best way to publish your files. Use something like [filezilla](http://filezilla-project.org) and read the help docs. You will upload the files to a folder (usually something like public\_html) and then the files correspond to your domain. Example:
you upload a file:
/public\_html/blog/index.htm
then go to your site:
<http://www.yourdomain.com/blog/index.htm>
Here is [a good tutorial on FTP](http://www.siteground.com/tutorials/ftp/). It's pretty easy once you get the hang of it. | You need to set up your domain provider's DNS server to point to the server where your files are located. For example, I have my domain (brasee.com) through GoDaddy.com, and I host my site locally. So I go to the GoDaddy website and update their DNS server to forward brasee.com requests to my server's IP address. |
5,711,407 | I have a screen that I need to remove some items on smaller screens because they will not fit.
I search for items in the layout to populate with live data in code, such as populating names etc. There are 5 items that need to be populated in my large layout but only 2 in my small layout as three I removed because they won't fit. How in my Activity can I tell that I am in "Small mode" and not search the layout for the three id's that I want to skip and not populate?
Do I need to just search for them and catch the errors and ignore or is there a better way?
Thanks | 2011/04/19 | [
"https://Stackoverflow.com/questions/5711407",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/467569/"
] | It looks like in this circumstance the widgets will return null when I attempt to find them if the layout chosen doesn't support that widget. So I can just detect if its null and if so skip populating the widget. | You will want to utilize a combo of the resource directories (high, medium, low) and follow the guidelines on how to support multiple screen sizes from the android documentation. Also you can get the screen dimensions if you need to anything specific in the code. See these links.
[Get screen dimensions in pixels](https://stackoverflow.com/questions/1016896/android-how-to-get-screen-dimensions)
<http://developer.android.com/guide/practices/screens_support.html> |
2,548,841 | I want to create a high performance server in C# which could take about ~10k clients. Now i started writing a TcpServer with C# and for each client-connection i open a new thread. I also use one thread to accept the connections. So far so good, works fine.
The server has to deserialize AMF incoming objects do some logic ( like saving the position of a player ) and send some object back ( serializing objects ). I am not worried about the serializing/deserializing part atm.
My main concern is that I will have a lot of threads with 10k clients and i've read somewhere that an OS can only hold like a few hunderd threads.
Are there any sources/articles available on writing a decent async threaded server ? Are there other possibilties or will 10k threads work fine ? I've looked on google, but i couldn't find much info about design patterns or ways which explain it clearly | 2010/03/30 | [
"https://Stackoverflow.com/questions/2548841",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/284401/"
] | You're going to run into a number of problems.
1. You can't spin up 10,000 threads for a couple of reasons. It'll trash the kernel scheduler. If you're running a 32-bit, then the default stack address space of 1MB means that 10k threads will reserve about 10GB of address space. That'll fail.
2. You can't use a simple select system either. At it's heart, select is O(N) for the number of sockets. With 10k sockets, that's bad.
3. You can use IO Completion Ports. This is the scenario they're designed for. To my knowledge there is no stable, managed IO Completion port library. You'll have to write your own using P/Invoke or Managed C++. Have fun. | You definitely don't want a thread per request. Even if you have fewer clients, the overhead of creating and destroying threads will cripple the server, and there's no way you'll get to 10,000 threads; the OS scheduler will die a horrible death long before then.
There are numerous articles online about asynchronous server programming in C# (e.g., [here](http://www.codeguru.com/Csharp/Csharp/cs_network/sockets/article.php/c8781/)). Just google around a bit. |
2,548,841 | I want to create a high performance server in C# which could take about ~10k clients. Now i started writing a TcpServer with C# and for each client-connection i open a new thread. I also use one thread to accept the connections. So far so good, works fine.
The server has to deserialize AMF incoming objects do some logic ( like saving the position of a player ) and send some object back ( serializing objects ). I am not worried about the serializing/deserializing part atm.
My main concern is that I will have a lot of threads with 10k clients and i've read somewhere that an OS can only hold like a few hunderd threads.
Are there any sources/articles available on writing a decent async threaded server ? Are there other possibilties or will 10k threads work fine ? I've looked on google, but i couldn't find much info about design patterns or ways which explain it clearly | 2010/03/30 | [
"https://Stackoverflow.com/questions/2548841",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/284401/"
] | The way to write an efficient multithreaded server is to use I/O completion ports (using a thread per request is quite inefficient, as @Marcelo mentions).
If you use the asynchronous version of the .NET socket class, you get this for free. See [this question](https://stackoverflow.com/questions/539940/how-does-net-make-use-of-io-threads-or-io-completion-ports) which has pointers to documentation. | You definitely don't want a thread per request. Even if you have fewer clients, the overhead of creating and destroying threads will cripple the server, and there's no way you'll get to 10,000 threads; the OS scheduler will die a horrible death long before then.
There are numerous articles online about asynchronous server programming in C# (e.g., [here](http://www.codeguru.com/Csharp/Csharp/cs_network/sockets/article.php/c8781/)). Just google around a bit. |
2,548,841 | I want to create a high performance server in C# which could take about ~10k clients. Now i started writing a TcpServer with C# and for each client-connection i open a new thread. I also use one thread to accept the connections. So far so good, works fine.
The server has to deserialize AMF incoming objects do some logic ( like saving the position of a player ) and send some object back ( serializing objects ). I am not worried about the serializing/deserializing part atm.
My main concern is that I will have a lot of threads with 10k clients and i've read somewhere that an OS can only hold like a few hunderd threads.
Are there any sources/articles available on writing a decent async threaded server ? Are there other possibilties or will 10k threads work fine ? I've looked on google, but i couldn't find much info about design patterns or ways which explain it clearly | 2010/03/30 | [
"https://Stackoverflow.com/questions/2548841",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/284401/"
] | You're going to run into a number of problems.
1. You can't spin up 10,000 threads for a couple of reasons. It'll trash the kernel scheduler. If you're running a 32-bit, then the default stack address space of 1MB means that 10k threads will reserve about 10GB of address space. That'll fail.
2. You can't use a simple select system either. At it's heart, select is O(N) for the number of sockets. With 10k sockets, that's bad.
3. You can use IO Completion Ports. This is the scenario they're designed for. To my knowledge there is no stable, managed IO Completion port library. You'll have to write your own using P/Invoke or Managed C++. Have fun. | You want to look into using [IO completion ports](http://msdn.microsoft.com/en-us/library/aa365198(VS.85).aspx). You basically have a threadpool and a queue of IO operations.
>
> I/O completion ports provide an
> efficient threading model for
> processing multiple asynchronous I/O
> requests on a multiprocessor system.
> When a process creates an I/O
> completion port, the system creates an
> associated queue object for requests
> whose sole purpose is to service these
> requests. Processes that handle many
> concurrent asynchronous I/O requests
> can do so more quickly and efficiently
> by using I/O completion ports in
> conjunction with a pre-allocated
> thread pool than by creating threads
> at the time they receive an I/O
> request.
>
>
> |
2,548,841 | I want to create a high performance server in C# which could take about ~10k clients. Now i started writing a TcpServer with C# and for each client-connection i open a new thread. I also use one thread to accept the connections. So far so good, works fine.
The server has to deserialize AMF incoming objects do some logic ( like saving the position of a player ) and send some object back ( serializing objects ). I am not worried about the serializing/deserializing part atm.
My main concern is that I will have a lot of threads with 10k clients and i've read somewhere that an OS can only hold like a few hunderd threads.
Are there any sources/articles available on writing a decent async threaded server ? Are there other possibilties or will 10k threads work fine ? I've looked on google, but i couldn't find much info about design patterns or ways which explain it clearly | 2010/03/30 | [
"https://Stackoverflow.com/questions/2548841",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/284401/"
] | You're going to run into a number of problems.
1. You can't spin up 10,000 threads for a couple of reasons. It'll trash the kernel scheduler. If you're running a 32-bit, then the default stack address space of 1MB means that 10k threads will reserve about 10GB of address space. That'll fail.
2. You can't use a simple select system either. At it's heart, select is O(N) for the number of sockets. With 10k sockets, that's bad.
3. You can use IO Completion Ports. This is the scenario they're designed for. To my knowledge there is no stable, managed IO Completion port library. You'll have to write your own using P/Invoke or Managed C++. Have fun. | The way to write an efficient multithreaded server is to use I/O completion ports (using a thread per request is quite inefficient, as @Marcelo mentions).
If you use the asynchronous version of the .NET socket class, you get this for free. See [this question](https://stackoverflow.com/questions/539940/how-does-net-make-use-of-io-threads-or-io-completion-ports) which has pointers to documentation. |
2,548,841 | I want to create a high performance server in C# which could take about ~10k clients. Now i started writing a TcpServer with C# and for each client-connection i open a new thread. I also use one thread to accept the connections. So far so good, works fine.
The server has to deserialize AMF incoming objects do some logic ( like saving the position of a player ) and send some object back ( serializing objects ). I am not worried about the serializing/deserializing part atm.
My main concern is that I will have a lot of threads with 10k clients and i've read somewhere that an OS can only hold like a few hunderd threads.
Are there any sources/articles available on writing a decent async threaded server ? Are there other possibilties or will 10k threads work fine ? I've looked on google, but i couldn't find much info about design patterns or ways which explain it clearly | 2010/03/30 | [
"https://Stackoverflow.com/questions/2548841",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/284401/"
] | The way to write an efficient multithreaded server is to use I/O completion ports (using a thread per request is quite inefficient, as @Marcelo mentions).
If you use the asynchronous version of the .NET socket class, you get this for free. See [this question](https://stackoverflow.com/questions/539940/how-does-net-make-use-of-io-threads-or-io-completion-ports) which has pointers to documentation. | You want to look into using [IO completion ports](http://msdn.microsoft.com/en-us/library/aa365198(VS.85).aspx). You basically have a threadpool and a queue of IO operations.
>
> I/O completion ports provide an
> efficient threading model for
> processing multiple asynchronous I/O
> requests on a multiprocessor system.
> When a process creates an I/O
> completion port, the system creates an
> associated queue object for requests
> whose sole purpose is to service these
> requests. Processes that handle many
> concurrent asynchronous I/O requests
> can do so more quickly and efficiently
> by using I/O completion ports in
> conjunction with a pre-allocated
> thread pool than by creating threads
> at the time they receive an I/O
> request.
>
>
> |
48,630 | So I am tired of having to do 2 clicks in Chrome to go forward but not backwards.
Are there any web browsers that look and work as close to Apple's safari?
Ideally I would like the navigation bar(back,home,recent apps buttons) to be hidden at the bottom similar to Android Games. | 2018/02/13 | [
"https://softwarerecs.stackexchange.com/questions/48630",
"https://softwarerecs.stackexchange.com",
"https://softwarerecs.stackexchange.com/users/6490/"
] | [Opera mini](https://play.google.com/store/apps/details?id=com.opera.mini.native) has it. It also looks a little like Safari.
(You can see the forward button on the bottom left of the screenshot)
[](https://i.stack.imgur.com/Lx8qOm.jpg) | [CM Browser](https://play.google.com/store/apps/details?id=com.ksmobile.cb&hl=en) is a very good browser that has forward and backward buttons in plain sight. I'm using it for at least 4 years and I'm really happy with it and also its security features. |
48,630 | So I am tired of having to do 2 clicks in Chrome to go forward but not backwards.
Are there any web browsers that look and work as close to Apple's safari?
Ideally I would like the navigation bar(back,home,recent apps buttons) to be hidden at the bottom similar to Android Games. | 2018/02/13 | [
"https://softwarerecs.stackexchange.com/questions/48630",
"https://softwarerecs.stackexchange.com",
"https://softwarerecs.stackexchange.com/users/6490/"
] | [CM Browser](https://play.google.com/store/apps/details?id=com.ksmobile.cb&hl=en) is a very good browser that has forward and backward buttons in plain sight. I'm using it for at least 4 years and I'm really happy with it and also its security features. | My app [Subhash Browser & Feed Reader](https://play.google.com/store/apps/details?id=com.vsubhash.droid.subhashbrowser&hl=en) has a toolbar with all kinds of functions including Back and Forward history navigation. The home button is used to quit the browser but there is a Favorites button that you can use. The toolbar is at the top. You can hide it by shaking the device or using an option from the menu. I am not sure how Apple Safari is but this app is for Android.
[](https://i.stack.imgur.com/CSL8E.png) |
48,630 | So I am tired of having to do 2 clicks in Chrome to go forward but not backwards.
Are there any web browsers that look and work as close to Apple's safari?
Ideally I would like the navigation bar(back,home,recent apps buttons) to be hidden at the bottom similar to Android Games. | 2018/02/13 | [
"https://softwarerecs.stackexchange.com/questions/48630",
"https://softwarerecs.stackexchange.com",
"https://softwarerecs.stackexchange.com/users/6490/"
] | [Opera mini](https://play.google.com/store/apps/details?id=com.opera.mini.native) has it. It also looks a little like Safari.
(You can see the forward button on the bottom left of the screenshot)
[](https://i.stack.imgur.com/Lx8qOm.jpg) | My app [Subhash Browser & Feed Reader](https://play.google.com/store/apps/details?id=com.vsubhash.droid.subhashbrowser&hl=en) has a toolbar with all kinds of functions including Back and Forward history navigation. The home button is used to quit the browser but there is a Favorites button that you can use. The toolbar is at the top. You can hide it by shaking the device or using an option from the menu. I am not sure how Apple Safari is but this app is for Android.
[](https://i.stack.imgur.com/CSL8E.png) |
59,625,210 | I want to automatically Re-trigger a failed pipeline using the If Condition Activity (dynamic content).
Process :
* Pipeline 1 running at a schedule time with trigger 1 - works
* If pipeline 1 fails, scheduled trigger 2 will run pipeline 2 - works
* **Pipeline 2 should contain if condition to check if pipeline 1 failed** - This is the Issue
* If pipeline 1 failed, then rerun else ignore - need to fix this
How can this be done?
All help is appreciated.
Thank you. | 2020/01/07 | [
"https://Stackoverflow.com/questions/59625210",
"https://Stackoverflow.com",
"https://Stackoverflow.com/users/8865285/"
] | I can give you an idea,
For example. Your pipeline1 failed by some reasons. At this time, you can create a file to Azure Storage Blob.
[](https://i.stack.imgur.com/5dVeT.png)
(Here is an example, you can use the activities what you want to use.)
Then create the trigger2 triggered by a blob been created. | can't you do it with "Execute Pipeline" activity?
For example:
You create a simple pipeline named "Rerun main pipeline" and use "Execute Pipeline" inside that and link it to "main pipeline". Then in main pipeline, you add failure output and link it to "Rerun main pipeline". |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.