@ccorn Well, it is a judgement call and often not easy to make where to post - whether the local meta or the main meta.

It probably depends on where you feel more comfortable. For example, if people on Mathematics Meta know you, they are more likely to take the suggestion seriously. Similarly, if you are regular on Meta Stack Exchange and are above 10k reputation there, it's natural that users understand from this that you are already an experienced user and they take the suggestion more seriously than from a newbie.

For example, not too long ago I have made a feature request first on Meta Mathematics (Limit for number of questions where tag can be added by tag-creator soon after the creation of tag) and only then I have made a post on Meta Stack Exchange (Do not allow massive retagging by the tag-creator shortly after creation of a new tag).

I you look at the last footnote in my post on the local meta - I have tried to explain there why I chose to do it this way.

But very briefly, some possible advantages I saw in this:

1. This was related to some problems that we have experienced shortly before that on this site. So I was expecting that users from Mathematics SE would be more interested in the topic.

2. I am much less familiar with Meta Stack Exchange than with Mathematics Meta. I simply felt more comfortable here.

3. My plan was to post it later to the main meta if the feedback on the local meta is positive. (And in fact I did this a bit later.) Which meant that when posting on Meta Stack Exchange I had another advantage - I was able to link to a post on a local meta as a supporting evidence that there is a site which has problems with this and that this specific suggestion gaind some support there.

I agree that in general it is probably better to post on Meta Stack Exchange if the changes would be network wide. But sometimes discussing it locally first might be good - if nothing else, it might help you fine tune your proposal a bit, at least if you get some feedback on local meta. And sometimes you might even get a response from users who are Meta.SE regulars pointing out that the same topic has already been discussed there, which is also useful.

I should explicitly point out that this comes from a user who has posted a few feature requests (local meta, main meta), but none of them was implemented. So I cannot claim too much experience in these matters.

Maybe something like this (whether to ask on local meta) might be a reasonable question on meta. (But which one?)

But it might be useful to check whether there were some related discussions in the past.

I was able to find this on Meta Stack Exchange: Can bug reports and feature requests requiring SE Dev attention be reported on per-site metas, or do they need to all go here?

I am currently experimenting with SEDE. Seems like median time to close answered questions is around 7h, only 4% are below 20min. I'd like to narrow this down to questions with "missing context" close reason, but I cannot find the joins associating a post with its close reason.

I have mentioned this also in the SEDE room, but the room is almost inactive, so I do not see getting a response there as very likely.

@MartinSleziak Thanks!

> Off-topic questions receiving answers is a common problem but it is not a severe one; many questions, once closed, will be cleaned up by the system automatically. Those that aren't can still be deleted manually by moderators and users with delete-vote privileges (of which there are currently 718). That should be plenty of people to handle closing and deleting even a third of the ~500 questions a day that Math.SE receives.

This seems to me as overly optimistic estimate of number of users who both have time and are willing to devote it to clean-up tasks. (Not to mention that at least some of the users counted there are probably are no longer active; or have some periods of activity and inacitivity.)