Stricter guidance on homebrew answers with no demonstrated evidence of use

Question

I am aware that there are guidelines regarding how to handle answers that suggest homebrew solutions where no evidence of their use has been shown. My understanding is these are generally strongly discouraged, but that there is no clear instruction as to exactly what to do with them.

There is a previous meta question on this subject - Is homebrew an acceptable answer to a question? - but my question is whether our stance should be stricter.

I ask this because of the existence of our very strict rules on game-rec answers and the delete policy for anything that doesn't meet them. How is a homebrew answer that shows no evidence of actual use any different to a game-rec answer that shows no evidence? Is there any inherent value in such homebrew answers that is not present in the latter, or should we delete them in the same way? Is there a middle ground?

Answer 1

Game-rec questions are a special category. They're perpetually only tentatively allowed, and the heavy moderation applied to them is the minimum necessary to keep it that way. Since this is special handling, it doesn't set automatic precedents - so our policies there don't set precedent for other parts of the site.

We don't need special policy for homebrew answers. We can handle them as any other answer that needs to follow GSBS requirements and the Back It Up! Principle. That means downvoting if they're poor quality (e.g. not backing it up), and moderation comes in when it's needed, but answers are not automatically under threat of deletion if they don't provide evidence. Unless a moderator feels like it's necessary.

Answer 2

We only need special, more draconian rules if there's a big problem. Game-recs were a big problem - many of them, lots of answers, terribly low quality.

It has not come to my attention that we have rampant homebrew answering causing a similar problem, but if you wanted to collect evidence thereof it could cause us to reevaluate that.