Two questions were recently asked regarding research or studies about a specific topic:

Both were closed by the community in short order.

We've had a few other questions with the same type request in the past:



Closed and deleted

So there's a bit of an inconsistency in how they've been dealt with. Are these on-topic? If so, why? If not, why not?


2 Answers 2


I'm torn. Some of the questions seem good, most others don't. But I can't exactly figure out where that line is. I think it has to do with localization (time and space), topicality, and specificality of the question.

Some of the questions are vague or wide reaching. Like the ones about cowboy coding, chosing to study IT, and happiness in the IT industry. Either there's not a significant body of research, it's mostly polls/surveys, or they are so broad that there's a gratitutious amount of information out there. I don't see how these questions make the Internet a better place by being asked/answered.

Others are off-topic. The programming linguistics question is clearly asking about things from a linguist's perspective, not a programmer's perspective, for example.

A few are too localized in time and space. Things like geographic differences in work environments and market conditions fall into this category, especially in today's world where communication and cultural exchange is so easy and the technological landscape changes rapidly.

However, there are good ones. Deadlock resolution came up in not one, but two software engineering courses that I took as an undergraduate (concurrent systems and real-time/embedded systems). Coding productivity might also fall into this category - maximizing productivity of an employee is a concern of managers, but I'm not sold on it as it's very individualized and specific to a given programmer and team. There are also good questions about people making cases - I frequently need to present improvement suggestions in terms of software quality, productivity, and the bottom line (among other things), so questions like the debugging question that help identify reputable sources that can be cited in such a document might be useful.

As long as the question is about something that's on-topic, it's targeted to/about software development professionals, and it's specific enough to provide pointed answers, I don't see why it shouldn't stay (unless, of course, it has a better home somewhere else).

Like I said, I typically need to provide reputable, citable data as to why X is better/worse than Y (if such data exists), and "because Bob on Programmers told his success/failure story" isn't good enough, although it does help to make the point. The collective Internet (ie: Programmers) has read more reputable, published material than any single individual, and would be able to point me toward citable resources (including things that indicate the opposite).

I also wonder if there's a student angle here. I can't come up with a question that's not "do my research for me", but there might be something I'm not thinking of. I don't think Programmers should be a place for someone to ask the Internet for every resource they need to write a paper, but there might be some take on how questions like this can add value to students, especially those writing Masters or Ph.D. level work if they hit a block (especially in applied topics and case studies, rather than just academic research). But perhaps not.

Also, I don't know about other people, but I love when I get my IEEE Transactions on Software Engineering or ACM Transactions on Software Engineering and Methodology in the mail. Although take what I say with a grain of salt - I'm an empirical software engineer that specializes in process and methodology. So, to me, these numbers and research results drive what I do. I might be a bit biased.

  • 1
    The open ones seem to have a common theme. They are looking for specific data about a measurable condition. I see no reason why these would be off topic. The closed and closed/deleted questions seem to have a more abstract, hard to define, and/or relative conditions. Commented Dec 15, 2011 at 18:08

I don't think these are on-topic here: while they are about software development, the answers they are looking for require a specialty and expertise outside of that of a programmer.

That is, they're looking for statistics and data to support a claim, not for a software development solution. Programmers are not necessarily versed in the data that's been collected about their field: they're the subjects, not the researchers.

Coming up with evidence to support or dispute a claim is the domain of our sister site, Skeptics.SE. Of course, they have their own quality guidelines, and I don't think automatically migrating any question looking for data is going to work: it'd have to be done on a case-by-case basis in coordination with Skeptics. Particularly, claims have to be notable (i.e. X prominent figure believes Y, what evidence supports this), which most of these aren't.

  • However, it is remarkably difficult to ask a question on Skeptics.SE that the community considers a notable claim and I doubt many of these research type questions would be accepted there either.
    – Mark Booth
    Commented Dec 15, 2011 at 18:16
  • What is wrong with letting the community decide if they are off topic or not? Commented Dec 15, 2011 at 19:37
  • @Chad That's the purpose of this discussion, to generate community consensus on how to handle these types of questions. Most of the closed questions have been closed by regular members of the community determining they're not on-topic here: why are those questions closed and the others left open? Were they oversights or is there a type of data request question that's okay?
    – user8
    Commented Dec 15, 2011 at 19:43
  • I m thinking its the bad subjective part not the request for information that is the problem. Commented Dec 15, 2011 at 20:18
  • All though I do not see why the NonEnglish Language question was closed It seems to have been left open gotten an answer and closed 7 months afterword Commented Dec 15, 2011 at 20:20

You must log in to answer this question.