@AlexanderGruber Thanks for the response. The way I see it there are two aspects of this.

One is immediate situation - it would be good to somehow stop repeated bumping and editing/retagging wars. That's up to mods and their communication with the involved users.

The other thing is how to prevent the problems with the same sets of tags to come up again in the future. What would be a good way to try to get some definitive answer on which of these tags are actually useful and should stay on the site and which ones should be removed. (Either by consensus of the community or - if that's not a realistic goal - decision by the mods.)

Without mentioning specific tags, I'll at least say that I would consider some of those tags useful. And about some of them I do not actually know enough about that specific topic/method to be able to judge whether they are useful (in fact, I would not be able even to judge on which questions to used them).

@MartinSleziak Even for tags for which some consensus was achieved (like Adding contest-math tag to questions retroactively): Michael Rozenberg will always add (or re-add) tags if he finds it appropriate. Recent examples are math.stackexchange.com/q/2168689, math.stackexchange.com/q/906972, and math.stackexchange.com/q/2095843, which are not related to any mathematical contest.

There is also this related discussion: Is it incorrect if I add [tag:contest-math] to my questions, even if it isn't explicitly from a contest?

It is not surprising that people may have different views on what belong under that tag. Many users who are experienced with mathematical contests can consider it as: "I know it when I see it".

@MartinR I suppose you saw Michael Rozenberg's message in this room: chat.stackexchange.com/transcript/19138/2018/10/25 (He tried to ping you there - not noticing that you are not pingable. That's the reason why I've mentioned that message in crude.)

@MartinSleziak Yes, I saw that, but I have given up discussing with Michael directly. We had several discussions about tagging/editing/closing questions, and none of them ended in a way that both sides were convinced.

I think I recall that he has already been suspended. (Even more than once, I think) So from that viewpoint things do not seem great. For people interested in (inequalities) he is probably a very valuable contributor.