1
$\begingroup$

I've noticed that the best Bioinformatics SE questions are where people ask about a specific problem they're having with their own data (or programs they're using). How would people feel about putting additional guidance in the "Ask Question" page, e.g. by adding guidance / tips to a right-hand sidebar?

Something like:

Hi, welcome to Bioinformatics StackExchange. Our community works best at answering specific questions about bioinformatics problems. Here are some tips to help get the best answers:

  1. Be specific about your problem; preferably ask a single question
  2. Add context - why are you trying to do this?
  3. Add detail; the more the better, as long as it's relevant
    • If there are input files, show the first few lines
    • If there are output files, show the first few lines
    • If there's an error, show the entire error message
    • If you're asking about issues with a program, version numbers can help
  4. Share your problem-solving research:
    • what did you try to overcome your problem?
    • why it did not work for you?

Thanks for asking. We hope you have a great time here!

Update: here is an example of a general question that lacks context and is not tied to a specific problem. Is it appropriate to ask these types of questions? Is it useful to the community to ask and/or answer questions like this?

$\endgroup$
2
$\begingroup$

I like the idea! A clear "Welcome to bioinformatics.SE" page/panel could really improve the quality of newcomers' questions.

I never had too much problems with a bit more general questions or hypothetical scenarios, the only one that gets to me is when people make no effort. So I would add to the list (ask for the research in a nice way)

  1. Share your problem-solving research:
    • what did you try to overcome your problem?
    • why it did not work for you?

We could also add a few link to good questions.

-- edit --

re generality: I belive there is a balance. "What is the current/recommended way to *build graph reference genome?" is within my boundaries, "How far we are from having a tricoder?" is not.

I guess I mean to say, is that majority of decisions in bioinformatics require a subjective opinion. Who has a capacity to go through all the new methods and do the benchmarking?! In reality, we use things that are good enough and we know that work and that's why I don't mind if answers require some opinions.

*replace by anything that has more than one soltuion.

$\endgroup$
4
  • $\begingroup$ If we're going to allow general questions, then this page also needs to be changed. $\endgroup$
    – gringer Mod
    May 7 at 20:22
  • $\begingroup$ I like the idea of adding links to examples, and the "share your research" bit. $\endgroup$
    – gringer Mod
    May 7 at 20:25
  • $\begingroup$ I tried to elaborate a bit more about the generality, not sure if I managed. That's perhaps the problem - it's hard to phrase how general questions are ok, so perhaps we can leave it as it is and just be forgiving? $\endgroup$
    – Kamil S Jaron Mod
    May 9 at 9:14
  • $\begingroup$ I think that with context relating to a specific issue both questions are okay (e.g. I'm doing a study that requires remote observation of blood biochemistry, and something like a tricorder would be really useful...). However, general questions without context lead to XY problems; it's much harder to find answers to satisfy the person asking the question, $\endgroup$
    – gringer Mod
    May 9 at 21:23
0
$\begingroup$

If you have a burning question that doesn't fit well within the StackExchange system (e.g. general discussion, advice, hypothetical situations, or opinions), we recommend posting on the bioinformatics subreddit.

Does bioinf.SE have a major problem with such questions? I'm not sure that there's any need for explicit text discouraging any particular kind of question. (Certainly it's good to encourage relevant details, evidence of research, etc, as the rest of your draft guidance does. But IMHO discouraging anything could remain unstated.)

In particular, I think official bioinf.SE pages (like the Ask a Question sidebar) should not be in the business of directing people to reddit. It might be appropriate to have a wiki page or similar listing all the other known bioinformatics Q&A-style resources; but we should not be recommending any one of them in particular.

$\endgroup$
3
  • $\begingroup$ It's a StackExchange thing (e.g. see this page); the format doesn't work well with questions where it's hard to evaluate useful and not useful answers. Such questions invite more comments than answers. $\endgroup$
    – gringer Mod
    May 10 at 21:49
  • 1
    $\begingroup$ I know what the reasoning is. I'm suggesting it's not very applicable to bioinf.SE, because (1) we don't have such a volume of questions that discouraging anything is particularly necessary; (2) the level of expertise here is such that the answers to such questions would likely be interesting nonetheless. Anyway, my point is not that such questions should be explicitly allowed — instead, that judging “fit within the StackExchange system” is a measure on a subjective scale, and we gain nothing from writing instructions trying to disallow such a nebulous category of questions. $\endgroup$ May 10 at 22:26
  • $\begingroup$ Okay, fair enough; I've got rid of the general question paragraph. We will need to get that page changed, though. $\endgroup$
    – gringer Mod
    May 11 at 4:08

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .