I used to periodically retag questions with deprecated tags. This led to this critical comment.

As I said many times, moderators could have at least prevent deprecated tags from being added to new questions: The existing deprecated tags should be blacklisted.

This discussion did not get much response - but the only comments left there seem to be more supportive than negative: What is the correct way to handle questions with deprecated tags?

Perhaps also the fact that the information that the tags are deprecated was added by a moderator into the tag-info and at least one of the moderators contributed to removals of deprecated tags can be taken as a hint that this is somewhat official stance. (At the moment I do not have access to SEDE - I do not know why - so I won't be able to get actual counts.)

But I don't remember whether I've seen some of the moderators saying something about this explicitly. (MO mods are in general rather silent on meta, compared to Mathematics.)

Of course, I can imagine that some users might object if there's too much bumping. I am not sure to which extent the numbers mentioned here should still be considered as the recommendation: Do we have an unofficial quota on how many old questions one should bump for minor edits in a single day? (Maybe if the site has grown, it is to be expected that there might be more bumping.)

BTW the question which cause the above comments still has no top-level tag.

About bumping, I also understand that frequent bumping (i.e. several days in a row) is somewhat inconvenient and I did so in the last few weeks. I especially did so in creating tags. To summarize:

(1) I created semigroups-of-operators. This one had been suggested earlier here. Before merging between 'semigroups' and 'monoids', the tag 'semigroup' was quite clearly split into two parts, say 35% semigroups of operators and 65% semigroups (I haven't checked count). After merging it drown the operators part as a smaller proportion, incoherent with the remainder of the tag.

The split was trivial to do (slightly less trivial was improving tags at each case). Basically, "semigroup" in the analytic/operator side really means "1-parameter", and for instance the discrete (monogenous) semigroup generated by a single operator is not called "semigroup" by this community. The related tags were mostly functional analysis and pde (although [fa.functional-analysis] appears in a few questions really pertaining to groups-and-monoids).

In this way I bumped 100 questions in few days (8?) which is a lot.

Next I created ordered-groups which is much closer to my own field, and tagged in this way about 50 questions. I think it's useful because there was no keyword gathering such questions. It also shows that "ordered group" is used in different meanings by different communities (for instance for people working in valued fields it means abelian, and this is a significant proportion of these questions). Also this part suffers from a somewhat incoherent use of tags pertaining to orderings.

I also created the tags relational-structures and modular-lie-algebras which are subjects I'm interested in; both bumped about 10-15 questions (although certainly more questions could have the first tag!)

About deprecated tag chasing (discrete-mathematics, abstract-algebra, geometry), I retagged in priority those with no other tag. And more recently some highly upvoted. (The tag abstract-algebra tends, even before being deprecated, to attract low-quality questions, but there were significant exceptions.)

specific-question meta The tag bumping seems suitable for this one: Do we have an unofficial quota on how many old questions one should bump for minor edits in a single day?

Maybe adding the tag (and bumping questions) would increase the likelihood that some users might add answer/comment there. (But it is more likely that nothing will be added - after all, meta is not that active.)

@MartinSleziak Yes it sounds like a reasonable tag to add to this question on meta.

linked-questions The question "Infinite tensor products" links to: cardinality of product modulo direct sum.

linked-questions And "Limit for divergent sequences" links to Infinite tensor products.

linked-question The question "Infinite tensor products" is also linked in this answer: Why are finiteness conditions important (and how to recognize them)?

linked-questions The same questions is linked in "(Z/n)^(I) is a direct summand of (Z/n)^I" and "$\infty$-ary tensor product on a category".

linked-questions And it is also linked in these two answers: Monoids with infinite products and Vector spaces without natural bases.

@YCor It's possible that I am stating the obvious - but I'll mention one disadvantage of the link you have given is that it will always show the most recent messages in the room - so after some time, what a user following that see will be related to something completely els.

Possible alternatives would be to link to a specific day or a specific message in the chat transcript.

It is possible also to created a conversation consisting of several consecutive messages - that is especially useful if the messages you want to link span more than one day.

I will also mention that even if the conversation is created, it is possible to make it later into the longer conversation (without changing the link to the conversation) - simply by removing it and then creating the conversation with the same name.

@MartinSleziak @MartinSleziak thanks: I erased my comment with the inconvenient link, would you post a comment with a better link?

@YCor Here is a link to the conversation: chat.stackexchange.com/rooms/10243/conversation/… (As I have mentioned before, if the discussion here in chat continues, it is possible to make the conversation longer while the same link still works.)

Should I post the comment with this link, or do you prefer to post it youself?

(I see that you suggested above that I should post the comment - but since you are already part of that conversation, it seems a bit more natural to me if you post it. That's the reason why I asked again, just to be sure.)