Should General Programming Questions Actually Be Posted To Biostar?

Question

[Pablo Pareja]

Hi all,

I'm noticing lately that more and more people are posting questions here that are not directly related to bioinformatics/computational biology but to general programming instead.

There are much more appropriate sites to post this kind of questions like http://stackoverflow.com/

I understand that a big number of bioinformaticians maybe don't know such sites, however I think that this shouldn't be a reason for accepting here in Biostar this kind of questions. Let me give you a couple of reasons justifying this:

• Most programming questions posted here are quite simple from the point of view of a computer scientist and are in the 99.9% of the cases already answered in stackoverflow (directly or indirectly).-
• The knowledge of the people registered there (regarding programming languages, apps architecture, etc...) is much broader.
• Questions strictly related to bioinformatics get somehow lost between all those regarding programming.

What do you think?

Cheers,
Pablo

[Pablo Marin-Garcia]

humm.. seems that meta questions are not available in Biostar (at least the links is missing in the top). I think that we should have a link to the google group in a more visible place then. The line at the bottom of the web page ' For general feedback, suggestions or to report a problem use the Biostar Central group' I think that get unnoticed by almost everyone.

[Michael Schubert]

This question should go to the biostar-central google group.

[Pablo Marin-Garcia]

@pablo_pareja, there is a lot of differences between a forum (like seqanswers) and a QA site (StackExchange). In order to be effective the site should only accept real specific questions and avoid open questions and dissertations. See the goals of creating SE site here: . Having say that Biostar has meta, comunity wiki and the the google group for this type of questions

[Pasta]

Don't get confused: Biostar is NOT a forum but a question-answer platform

[Pablo Pareja]

Hi pasta, what do you mean by "Biostar is NOT a fourm but a question-answer platform' ?

[Pablo Pareja]

@Michael Schubert didn't know such group existed...

[Pablo Marin-Garcia]

@pablo, there is a lot of differences between a forum (like seqanswers) and a QA site (StackExchange). In order to be effective the site should only accept real specific questions and avoid open questions and dissertations. See the goals of creating SE site here: http://www.youtube.com/watch?v=NWHfY_lvKIQ. Having say that there is meta, comunity wiki and the google group for this type of questions

[Pablo Pareja]

@pablo_marin_garcia I agree with you, I didn't realize there was such gooogle group till now.

[Neilfws]

Well, the way people learn about the Google Group is by asking a meta question here. Then we point them to the group and close the question :-) So, it doesn't really matter whether people know about it beforehand, since they do afterwards.

[Casey Bergman]

@pablo_pareja, I've modified the title slightly since your question is about whether questions should be posted. It is an entirely different matter whether answers are accepted ;).

[Pablo Pareja]

@Casey Bergman, You're right, I didn't realize it, thanks ;)

[Michael]

According to the rules, this question is off-topic and triggers subjective discussions so I should close it now, oh wait! I took part in the discussion myself... What do I do now?

Answer 1

[2184687-1231-83-]

I think they should be accepted and redirected to the best site in stackoverflow for an answer, if necessary.

Answer 2

[Istvan Albert]

For beginner it is often difficult to distinguish between a general programming question and a bioinformatics related programming question. Thus I would say that if the answer is simple then it is often easier to answer the question than redirect it to another site. If the question is complex then perhaps it is better to redirect it to a different forum.

In both cases we should mention that another forum may be more appropriate.

Answer 3

[Pablo Marin-Garcia]

I think this question should go to the Biostar Central group, so I have second thoughts answering this here but....as the Biostar Central group does not have many members, the Biostar users will not read about this discussion and as community wiki is getting discourage or misused in the StackExchange main stream, I am answering here:

I am slightly against having general programming questions here because they would dilute the biology (analysis) part of this site but, I have not set my mind about questions about biology frameworks API as Ensembl, bio(perl|python|java|ruby...), etc

[Pablo Marin-Garcia]

As I said I have mixed feelings. The thing is that StackExchange format is perfect for QA for frameworks, probably better than many of the mailing lists. Also has the nice tool of autocompletion and suggestion related questions. For example I use StackOverflow for ggplot2 (R graphics) questions instead the mailing list (that only use to see what is going on) because I found it more useful. So for questions like, how to retrieve all non-synonyms snps for gene x with ensembl API, is not too bad to ask it here. It has lower entry barrier (no mailing list signup etc)

[Pablo Pareja]

Maybe I'm a bit radical but I wouldn't include questions about biology frameworks either since most of them have their own mailing-lists (I just checked biojava site and it seems there are two different lists). I think posting this kind of questions is duplicating information already available or even preventing some people to access it (if you have a question about biojava for example, you'd go to biojava mailing list, not here)

Answer 4

[Michael]

You are making up a problem where I see none. Neither is the number of pure programming questions overwhelming, nor is it diluting the content. So what is the problem? If the question turns out to be about pure programming, either because the poster had a biological question in mind or even mistook biostar for a programming related site (might also happen for any R related questions, or it might turn out the OP had a biological problem but this turns out to be irrelevant for the solution), you can either rectify the misunderstanding or simply answer the question if you can, to help someone out and maybe collect some points.

It's important imho to relax about this and decide this on a case by case basis, not trying to set up overly bureaucratic rules to carve a decision in stone. I wouldn't encourage folks to ask pure programming questions if they know for sure, but if in doubt I would encourage to ask and then we will figure out a solution. Remember, we are trying to be nice, our aim is not first and foremost to turn down other peoples questions.

Edit (nice edit Casey, though the semantics of the question has now changed): Now the answer is simply No! Still, if you have a biological problem behind your question and you are not sure, try find out, and if still unsure but have good reasons to ask, just ask, worst thing that can happen is that the question is closed.

[Farhat]

I agree with this. While there are occasional pure programming questions, they are not changing the overall milieu of Biostar. As long as it is related to bioinformatics, even if tangentially, I say let it be in for now.

[Pablo Pareja]

I disagree with this, maybe you don't see a problem here but I somehow do. What I expect from Biostar is a site where questions provide some actual biological/bioninformatics knowledge/contribution, not a place where all kind of questions which are indirectly related to bionformatics are posted. Maybe taking the latter approach would result into more platform users but the quality of the content would be lower. I just don't think that a 'more quantity than quality' approach would be the right one.

[Michael]

Pablo, how do you distinguish between questions that are 'worthy' because they provide new contributions or not? You see, this evaluation is a rather subjective one. In order to make a decision you have to see the question first. I would love to see that closing questions would be a more democratic process like in a voting to close process where a certain number of votes is required. That is not possible atm. I value your appreciation of biostar and your expectations of it a a high-quality resource is surely justified, however your perception what is appropriate is solely your subjectiv view.

[Pablo Pareja]

Michael, my appreciations about what I expect of Biostar are obviously subjective (they're just that, opinions + expectations); there's no need to clarify that. However I think there's a pretty simple rule for separating questions that provide contributions from those which don't: whenever there's a question whose relationship with biology/bioinformatics is so indirect that it can be applied to any field like here: How To Count The Number Of Elements In A Column Other Than 'X' In R

I think it's pretty obvious saying there's no contribution

[Michael]

Yes, that question is a good example, but, it is also the only one at the moment, at least in what I saw. If you want to contribute, you can by adding a comment suggesting to divert to stackoverflow or suggesting to close. After you become moderator, you could do even more. for that I think you need simply to gather about 1000-2000 reputation or so, actually not quite sure how this 'promotion' works.

Answer 5

[Neilfws]

One might argue that a lot of bioinformatics problems are, essentially, "general programming" questions. It's just that the program is being applied to biological data. For example if I have a problem involving an array of gene identifiers, is that a problem in array handling or a problem in gene identifiers?

So long as the question mentions what the bioinformatic context is, I have no problem with programming questions.

[Pablo Pareja]

well, it's not that I would have a problem answering that kind of question either, I'd just say that this would probably not be the right place to post it. Maybe the thing is just that I have a different idea of what Biostar should tend to be/evolve to. I'd like to mention http://mathoverflow.net/ as an example of what I mean. There you cannot find such 'trivial' programming questions (in or out of maths context) but rather actual Mathematics related questions.

Answer 6

[Tiffani]

I am a new coder, and I thought Biostar was an Informatics place only, If I have actual programming questions, I normally just google it or go to another place. I could see people maybe asking questions like "What are commonly used languages with Bioinformatics" and stuff like that but otherwise I think other code questions should be for google.

Answer 7

[Matt Chambers]

I think general programming questions should be strongly discouraged if only because users are FAR FAR more likely to get a rapid, high quality answer from StackOverflow (often already asked and answered). This site only exists because of the success of SO at answering general programming questions!