The current wording of the "recommend a tool" close reason is like so (formatting omitted):

Questions asking us to recommend or find a tool, library or favorite off-site resource are off-topic for Stack Overflow as they tend to attract opinionated answers and spam. Instead, describe the problem and what has been done so far to solve it.

This close reason nicely covers the cases where a user wants us to provide a recommendation. But I feel it could become even more versatile if we amend it to also forbid providing comparisons of libraries/tools, another fairly common type of off-topic question.

Something like:

Questions asking us to recommend or find a tool, library or favorite off-site resource, or to provide comparisons between them, are off-topic for Stack Overflow as they tend to attract opinionated answers and spam. Instead, describe the problem and what has been done so far to solve it.

The wording would have to be less awkward, but you get the general idea.

Comparison questions also attract opinionated responses, and possibly spam from companies promoting their products. Even worse, they tend to become outdated very quickly.


The most appropriate close reason for a "Which is better, Foo or Bar?" question would be "primarily opinion-based".

The most appropriate close reason for a "What are all of the differences between Foo and Bar?" question would be "too broad".

  • 1
    As the close reasons currently stand, yes this is true. But isn't "recommend a tool" just a more specific type of "primarily opinion-based"? We have the latter as a standalone close reason, and I think it might be worthwhile to strengthen it by adding the "comparison" clause. – Chris Laplante Feb 19 '14 at 17:13
  • @ChrisLaplante Then you're just be generalizing it to the point where it's no longer serving it's purpose. You're right; it's a subset of opinion based because it's both such a common issue and because it lets the close reason provide a more targeted response to that issue. Generalizing it to cover two fairly different problems means it can no longer provide that easy targeted response. It defeats the purpose of having it as as a separate reason from "opinion based" in the first place. – Servy Feb 19 '14 at 17:16

A variation of this I find annoying, and which isn't exclusive to tools, is "Which is faster, x or y?"

Not because it's opinion-based. For the most part, with enough fixed criteria, it's actually very easy to prove which one is faster. But without those fixed criteria, it can very much be It Depends™.

In most cases, it took the user longer to type up the question (never mind wait for answers) than it would have taken for them to just test it themselves. Especially since testing it themselves will make sure that many of the criteria above are not different between the tester and the person interested in the results. (For example, my desktop has more memory than most, so my local testing may yield different results than others' if one method favors/requires large amounts of memory)

I never know what to do with those types of questions, but I do feel that they should be closed. So I would happily welcome a "which is better?" close reason, even if it is combined with another, existing, similar close reason.

  • "This question appears to be off-topic because it lacks sufficient information to diagnose the problem. Describe your problem in more detail or include a minimal example in the question itself." likely applies to many of those types of questions. That or "unclear" if the criteria aren't specific enough to evaluate them. – Servy Feb 19 '14 at 18:15
  • @Servy I don't agree that that quite fits - sometimes those questions have a lot of detail and aren't really requiring anyone to diagnose the problem. The problem is pretty clear: the user doesn't know which is faster: <code sample a, which may be very complete> or <code sample b, which may be very complete>. – Aaron Bertrand Feb 19 '14 at 18:16
  • In the general case that's correct. I'm simply saying that from my personal experiences, most of those types of questions happen to not provide enough information to answer them, hence the "It depends" response. It's certainly possible for them to have enough information, just rare. – Servy Feb 19 '14 at 18:17
  • 1
    @Servy All I'm saying is that if we define a reason that explicitly states that comparison questions are not on-topic because they're hard to answer, they won't wonder how they should add more detail to the question to get it re-opened, if we use your suggested close reason instead. It's about clarity. – Aaron Bertrand Feb 19 '14 at 18:18
  • I'm not sure if I'd want to say that every single type of "how do these two code snippets compare" question is offtopic. I'd think that there can be some cases where enough information is provided, and the relevant comparison is narrow enough to fit in an SO answer, to make it possible to adequately compare two possible solutions to a problem. It's certainly hard to do well, and if it's too hard I can understand rejecting those rare gems to avoid having to deal with the crap, but it does make me a bit wary. – Servy Feb 19 '14 at 18:21
  • 1
    @Servy I'm not suggesting that every comparison question should be closed (with that reason or with a less clear reason). I'm saying that when such a question should be closed, there should be a clear reason that justifies its closure. – Aaron Bertrand Feb 19 '14 at 18:21
  • @Servy and I am talking about your comment on my answer, not your answer. If I were talking about your answer, I would probably comment on your answer, not respond to your comment on my own. – Aaron Bertrand Feb 19 '14 at 18:23

You must log in to answer this question.

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