Questions about looking for a given data structure (or a library to do a very specific task) are usually left open on Stack Overflow, even though they have no definitive answer, because it is such a common task applicable to many programmers. I can't even count the number of times an SO question has saved me hours of research looking for the right library.

Yet, this 4-1/2 year old question, with 68 upvotes, has been closed. This upsets me because, when I was looking for a C# priority queue, this page was a better resource than the other first top 20 results on google combined!

And now that it's closed, links to newer and better priority queue implementations can no longer be posted. In essence, by closing that question we are forcibly causing it to be stale and outdated. That is not good for anyone, especially when that link is the top result on google.

Can we please have this question reopened? More importantly, can we get a specific ruling stating that these sorts of question should be allowed since, despite having no definitive answer, they are some of the most useful on-topic questions on the site?

Disclosure: I recently wrote a priority queue implementation which I believe could help a lot of people

  • 3
    It's a shopping question, asking for recommendations. That is not a good fit for the SE network. Popularity does not equate a good SO question. – Martijn Pieters Jul 1 '13 at 7:57
  • 2
    Better question, why isn't it deleted? – casperOne Jul 1 '13 at 12:10
  • @casperOne: with only 32k views and 69 upvotes (over 4 and a half years (!)), the question was clearly not popular enough to be closed until recently. It has not yet been deleted for pretty much the same reasons. – Martijn Pieters Jul 1 '13 at 14:02
  • @MartijnPieters So in other words, it should have been deleted, we just didn't find it yet. – casperOne Jul 1 '13 at 14:04
  • ...it is days like this that I'm ashamed to be a part of this community. – BlueRaja Jul 1 '13 at 15:59

It was closed because it is a shopping question, a subjective request for a recommendation. As such it is not constructive and not a good fit for the Stack Exchange network.

Today, it'd be closed as 'Primarily opinion-based' instead:

  • primarily opinion-based — Many good questions generate some degree of opinion based on expert experience, but answers to this question will tend to be almost entirely based on opinions, rather than facts, references, or specific expertise.

It should stay closed. Especially since it'll otherwise attract promotion for every new Queue implementation out there.

Don't confuse the overal Stack Overflow Google ranking with a thumbs-up for a specific question. Neither should you conflate usefulness to you with fit; just because the answers help you doesn't mean the question won't attract bad answers and a lot of work for the community to keep the question cleared up.

  • Fortunately, most others do not share that strict dogmatism, or some of the most useful questions on the site would be closed! I agree that shopping recommendations are off-topic, but I disagree that this is a shopping recommendation; it's more akin to "what are my options for <specific task>", which are on-topic (see, for example, half the questions tagged skyrim on gaming.SE!). If you don't agree that these questions are useful, try searching for "<common programming task> <language>" in google - the most relevant/useful hit is almost always SO! – BlueRaja Jul 1 '13 at 8:05
  • 2
    That's because SO has a sky-high google ranking, most stuff on Stack Overflow is in the top results if the keywords match well enough. That doesn't make such questions a good fit for the format. – Martijn Pieters Jul 1 '13 at 8:07
  • But that also doesn't make them not a good fit for the format. Given the high quality of the answers and the extremely high usefulness of both the questions and its answers, this specific example clearly is a good fit for the format. – BlueRaja Jul 1 '13 at 8:08
  • The question should be narrowed in focus and be problem oriented. "I am trying to implement a frobnar widget processing algorithm, but the standard queue implementation doesn't let me fruzzle the ambams. I tried to work around that by doing x and y, but that resulted in the following exception. What to do?" leads to much more detailed answers. – Martijn Pieters Jul 1 '13 at 8:09
  • If those answers happen to recommend other queue implementations that solve the problem, so be it. – Martijn Pieters Jul 1 '13 at 8:09
  • 6
    @BlueRaja: And don't conflate usefulness to you with the question being a good fit for our format. They are not the same thing. – Martijn Pieters Jul 1 '13 at 8:10

Neither popularity nor usefulness to you (or even a lot of other people) is a solid indicator of usefulness to the SE network. In fact, sometimes popular questions can degrade the overall experience for everybody because they derail the focus. SE thrives on high signal to noise ratio closely scoped QnA. If you start allowing other things in, the value on those things starts to decrease.

A few meme or humor posts might get a lot of attention, but then your user base is distracted and your experts go elsewhere. We want the people seriously interested in a specific subject matter, not just volumes of people.

In the case of shopping questions, these need constantly updating and revising to be useful. As you noted in your reasons for re-opening it it, this question stops being useful if it stops being updated. That isn't a good kind of question to have on the network when everything else can be archived and remain useful.

Somebody else needs to come up with a software platform to handle shopping questions. The SE model of QnA is not optimized for it and allowing them would detract from what it is optimized for.

  • Who is talking about meme or humor posts? This is a programming question highly relevant and useful to programmers; how could that not be useful for the SE network!? "this question stops being useful if it stops being updated" - That is also true (much more true, in fact) of literally every question which involves any API (see: any old question on jquery, which will often recommend a deprecated function). – BlueRaja Jul 1 '13 at 8:13
  • @BlueRaja I didn't say this was a case of a humor question, I gave that as an example of the general reasoning behind popular questions not always being a good thing. – Caleb Jul 1 '13 at 9:44
  • ...But it's not popular because it's funny or a meme, it's popular because it's so useful to so many people. Surely that must be a separate case? – BlueRaja Jul 1 '13 at 15:44
  • @BlueRaja Not really. Raw popularity of a single question for any reason can be disruptive to other questions no matter what the reason for its particular popularity is. Just because it's useful in a different way doesn't make it not a miss-fit for the venue. – Caleb Jul 1 '13 at 16:00

You must log in to answer this question.

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