In theory, the advent of some sort of Community architecture on Steemit will help the development of communities out a lot. Of course, figuring out when that is actually likely to happen and how to make use of it is an entirely different issue.

The proposed mechanical changes to Steemit are – interesting, but not sufficient to the purpose, in my opinion.

Still, worrying about tag pollution before a tag actually exists in any number to be polluted is making a lot of assumption, which I'm not really down to do.

That argument is deployable against doing anything that someone hasn't done before. As such, I have to dismiss it out of hand as ridiculous.