4

For personal reasons I am no longer interested in contributing new questions and answers to Stack Overflow; but I think this idea would have much broader applicability than my specific use case.

Suppose I notice that a particular question is repeatedly, poorly asked. Ideally, there would be a canonical so that duplicates can be properly closed, but I can't find one. I have a basic idea of what the question should say and what would be in scope and what wouldn't be; but actually drafting a proper example for the question and clearly explaining the underlying problem (in a way that people who actually have the problem could understand!) is more work than I have time for - not to mention writing an answer.

I'd like to have the ability to outsource and reward that work, by placing a bounty that solicits a high quality question as well as an answer. Simply waiting for it to appear isn't going to work, because the people qualified to write the question (and, presumably, the answer as well) will have their own ideas.

4
  • 2
    I like the idea. Although I don't see it currently fitting the model. However, perhaps it should be made to fit. It's essentially placing a bounty for a Q as well as the A. This could theoretically be fulfilled if somebody finds a good existing Q&A pair (or maybe an existing question?) that was probably missed. Details to be determined, maybe it doesn't even need to be exactly "bounty" (wager some of your rep) but I feel there should be a system for rewarding organizing the library we have. As opposed to the current system: adding more and more and more and more and more and more to it.
    – VLAZ
    Commented Dec 19, 2023 at 11:51
  • I second Nij. Users capable of doing it properly would not be motivated by rep. Commented Dec 19, 2023 at 11:57
  • Ask Different has a meta list pointing to canonical QAs. It was kicked off by a mod & then added to by other users, often by the one posting the answer, if not also the question. It has no reward structure. apple.meta.stackexchange.com/questions/2418/… You could start to build a list on the meta of your choice - leaving undefined 'future' questions to be linked later.
    – Tetsujin
    Commented Dec 19, 2023 at 16:09
  • 2
    @Nij would you mind writing it up as an answer? Commented Dec 20, 2023 at 3:16

2 Answers 2

1

There's really no need for a new feature. Just go to the site's Meta and ask for someone to write whatever type of post you want to see. You don't have to announce you're planning to give a bounty to the answer, but you could. (Note: unfortunately you can't give a bounty to the person who asked a question, but if it's self-answered well, the problem is moot.)

On some meta sites, there are already threads dedicated to this type of request. See for example:

-2

What do you get when someone notices too many duplicates that should redirect to each other, and starts a mission to create the master duplicate of them all? Yet another duplicate that should redirect.

Canonical questions are not typically the product of a single genius user who only needed a little extrinsic motivation to finally make the effort. They are the result of collaboration and discussion by a range of users over an extended period.

A bounty can't (currently) be awarded to multiple people like a Nobel prize, but rhis isn't the kind of thing a bounty is appropriate for; this is overarching curating that the community as a whole requires and is responsible for. Rewarding individuals for the collective result is not entirely justified without other conditions to quality and contribution.

Perhaps the company could start a programme of identifying significant clusters of duplicates and providing support for the mini-projects of developing a key canonical Q&A, perhaps with merging of the answers selected as best representatives, and closing them all to it for good.

Perhaps they'll just let the proliferation continue and the community will have to sort it themselves on meta as usual. I know which is likely to work faster and better at this point.

2
  • 1
    I mean, most of the canonicals i've used the most were in fact made for the purpose of being a canonical rather than just organically popping into existence. That's not to say it's more common for them to be fabricated, but it's certainly not a fringe idea or one that hasn't been successful.
    – Kevin B
    Commented Dec 20, 2023 at 23:09
  • I do though like the idea of using "technology" to aggregate commonly asked topics that don't yet commonly get dupe closed toward a common target in a way that we can then use to identify question areas that could use a dupe target cleanup. That's something the community can do with SEDE if the company wasn't interested in that.
    – Kevin B
    Commented Dec 20, 2023 at 23:10

You must log in to answer this question.

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