How useful must a question be to be allowed?

Question

I'll quote Jeff from this meta.Gaming question:

What value does this question have to anyone else on the planet, other than the one person who asked? In other words, what internet user is going to be searching for and finding this information?

It is my opinion that these [...] questions should be banned from all sites as a matter of policy.

They are the very definition of "too localized."

I was under the impression that "usefulness to the asker" was good enough for a Q&A site. It appears this isn't self-evident, however.

So, dear fellow unicorn watchers: how useful must a question be to the general public before an otherwise okay question can be allowed?

Answer 1

I think the important thing for a question is that it has to provide knowledge that others might find useful. There's two different levels to which a question can be primarily rooted as being valuable only to the author.

1. No one else is likely to have had the same issue due to choices the author made, or will ever want to have the same issue. This is specifically the kind of question where adapting the question to appeal to a broader audience changes the question to the point it would no longer help the original author. This kind of question is not only not useful, but it is also less likely to get any useful attention. No one is going to search for it, no one is going to answer it, and so it will just rot away. No matter how real the problem is for the author, if it is tied too close to the author then no one is going to be inspired to take their time and answer it. It's lose-lose.

2. No one else might actively think about this problem, but upon seeing the question they get their curiosity piqued. This has a strange effect in which few people will ever search for the question, but people who are already on the site and do run into the question. And they learn something. Sometimes the handy trick they might learn will be utterly worthless, but it's a cool piece of knowledge that makes someone satisfied in knowing "Hey, I can do this".

In both cases, the question will probably not get all that many votes (unless it is particularly creatively written, but that's a different thing altogether). But the latter is still useful to people. At bare minimum, I think that this is what is needed for questions that appeal primarily to the question author than the the world at large.

No one else might immediately find the question useful or search for it, but the question once read actually educates people with something new and useful that they feel good about learning. It should inspire a greater reaction than a shrug.

Answer 2

"Usefulness to the asker" is a given for any good-faith question that might be asked. A certain amount of filtering is necessary to avoid the signal-to-noise ratio becoming awful.

My litmus test for this is "Is it conceivable that someone else might have asked this?". That's a pretty generous threshold, and there are certainly cases where we need to be stricter. However, if a question is purely unique to the asker, then I think it is useless to the community.

Answer 3

My opinion is, as long as the information can be used by others (i.e no corner case), is on-topic, and that there is just a reasonable amount of good duplicate, the question belongs to the SE site it was posted on.

Answer 4

I was under the impression that "usefulness to the asker" was good enough for a Q&A site. It appears this isn't self-evident, however.

it is, if the title of your site happens to be Yahoo Answers.

TL;DR version -- questions about things you can't remember suck... hard.

I'll just ctrl+c, ctrl+v from my edited gaming meta question.

Mapping these ['help me identify this game' questions] to existing close reasons, there are two that fit.

1. Too Localized. What value do these questions have to anyone else on the planet, other than the one person who asked? In other words, what internet user is going to be searching for and finding information about a game one person vaguely remembers playing, as described in generic broad "um, I think it had.." terms?

2. Not a Real Question. Half-remembered descriptions of a game you vaguely remember playing, does not a real question make. This wouldn't be tolerated on any other site in our network. Unless the asker has demonstrated and documented that they've invested substantial effort to find this game, and given us something concrete that provides us with a reasonable chance of actually finding a game, it's simply not a real question -- it's a half-remembered question!

Conclusion: if the user can produce a screenshot or some other reasonably concrete identifying artifact from the game to work with, other than "I kinda remember some stuff about this game I once played.." I have less objections to these questions. I'm still not a fan of them, but I think that's a fair way to limit how many we have.