I recently asked, are there any premade outdoor worlds suitable for testing 3D graphics? The question was put on hold:

Questions asking for resources are considered too broad here, unfortunately (because they are list-of-link generators); a good place to ask if the suggestions above aren't sufficient would be GDNet.

I agree that I'm basically asking for a list of links. I don't agree that that's a fundamentally non-worthwhile thing for this site, though it might be prone to difficulties.

In other words, there are bad (unproductive) list-of-link questions and good (productive, helpful, well suited to the community's interest and expertise) list-of-link questions.

The only prior discussion I can find on this is a Meta question from 2010: Should we allow "list of X"-questions? The consensus back then was that such questions were allowed but needed to be Community Wikis, which sounds reasonable to me.

  • 1
    \$\begingroup\$ Can you post your proposal for what we should/should not allow as an answer? That way it's clearer to tell if votes mean "this is a good question to ask" versus "this is a good solution to adopt." \$\endgroup\$ – user1430 Oct 16 '18 at 21:14

There are two primary reasons we don't support these kinds of questions.

  • Our "don't ask" page (and indeed all SE sites, as this one is not under our direct control) says to "avoid asking subjective questions where every answer is equally valid...". While the existence of some resource is an objective fact, that's not where the issue is. The subjectivity comes from the fact that any resource that exists answers the question, and there's nothing else for the site's machinery (voting) to chew on except popularity, which can be very subjective.

  • The answers are link-only and devoid of almost any other content, and we destroy such answers per network policy.

Any guidelines that propose exemptions to the "lists of resources are off-topic" policy would need to thread those needles, as well as be something that the community itself can uphold (that is, not require constant moderator attention). I'm not sure I can think of any such guidelines that don't radically transform the question into something that is already on-topic, but perhaps somebody else could.

As for the community wiki angle; we don't use CW as a crutch for otherwise-bad questions. Community wiki is really, at this point in the network's lifecycle, something of a vestigial element. We don't really have a defined use for it, and all the attempts I can recall to try to institute a use haven't really gained any traction.

  • \$\begingroup\$ The spirit of the rule against link-only answers is to prevent people from simply deferring to an external resource, e.g. "here, read this book for your answer." It doesn't make sense to apply it where the external resource itself is the answer. ("What paper was this algorithm originally published in?") \$\endgroup\$ – Maxpm Oct 17 '18 at 2:30
  • \$\begingroup\$ That’s not all though, it’s also to ensure that the answers remain valid and useful on this site. If the link goes away (which definitely happens), the answer can become varying degrees of useless. \$\endgroup\$ – user1430 Oct 17 '18 at 13:56
  • \$\begingroup\$ Which is exactly the problem that CWs are intended to fix, isn’t it? It seems like the tools are available but for some reason we’re very reluctant to use them. No wonder they’re “vestigial,” right? :) I don’t know what you mean by “crutch” – a question needing CW doesn’t make it any less of a question and it doesn’t make it any less worthwhile to the community. \$\endgroup\$ – Maxpm Oct 17 '18 at 19:14
  • \$\begingroup\$ @Maxpm No, community wikis don't fix or address link rot at all. Link rot can be corrected on any answer, CW or no, by all users with the appropriate reputation, just like CW wiki questions. The rot has still happened, and may not be correctable, thus rendering the answer and its votes useless. \$\endgroup\$ – user1430 Oct 17 '18 at 19:18
  • \$\begingroup\$ Very few questions "need" to be CW (which is why users cannot mark them as such any more), and similarly so for answers. SE's intent for CW answers is that they are building block for a community to share in the development of an answer. But there's no need for that if the only sharing we do is updating links. Existing edits work fine; CW solves none of the other concerns with link-only answers. \$\endgroup\$ – user1430 Oct 17 '18 at 19:22

There are two nuances to my question that I think make the difference:

  1. The information isn't readily available elsewhere. It's not, for example, like asking "what are some free 3D game engines?" which can be answered by the first page of Google.
  2. It's not subjective, because it doesn't ask for the "best" resource, but rather any resource that reasonably fits the goal. The hope was for that broad frame to inspire answers catering to different tradeoffs. For example, "this scene's complexity is representative of contemporary games," or "this scene looks dramatically different under HDR," or whatever.

Given these criteria, or criteria like them, I think exceptions could be carved out of the broad rule against asking for resources.

Stack Overflow has questions like The Definitive C++ Book Guide and List and they don't cause the world to come crashing down. (And evidently there is no binding network-wide policy forbidding them.)

  • 1
    \$\begingroup\$ That SO book question was locked (like some other early SO questions) as a recognition that it has historical value even though the site policy has moved beyond permitting questions like it, however. \$\endgroup\$ – user1430 Oct 17 '18 at 13:58
  • 1
    \$\begingroup\$ I appreciate that you've gone to the trouble searching for precedent. In this particular case, the precursors you've found are not necessarily in sync with the community standards. Admittedly, it would be good if there was a way for the system to reflect that. \$\endgroup\$ – Pikalek Oct 17 '18 at 14:55

You must log in to answer this question.

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