Post History

The experience we had on the physics site Somewhere Else may be instructive ([I've already written a little about it on Software Meta](https://software.codidact.com/posts/278783/278941#answer-278941)). In short, against my wishes they were allowed *but* because there was a fairly strong and well enforced notion of what a "good" answer would look like it worked out much better than I expected.

As a result of that experience my position now would be

> Not without a strong consensus about what rules will be used to establish quality control.

I don't believe we have enough steady participation for such a consensus to have emerged as yet.

I'd also like to note that PSE had rules mostly about the *answers* which has the advantage of directing the rules at people who are already familiar with the site, but it would not be unreasonable to have a few basic rules for questions as well.

In my opinion a good question in this realm is specific enough that proposed answers can be tested against the desired qualities on a point by point basis. For instance a user asks for a graduate level mechanics text in a terse style but featuring answers to exercises. If I wanted to suggests Goldstein's classic text, I could note that it (A) is a graduate level text, (B) is moderately but not extremely terse (and little wordier than L&L, for instance), (C.i) has fairly extensive exercises with (C.ii) final numeric answers for some, but no workings. Which would put a reader in a good position to judge if it met their needs.