There’s also the small matter of all the signs that tell you what lane you want being jammed up in the girders, so you can’t even see them except for 0.5 seconds when you’re almost directly under them.

…of course there’s not enough room in the picture to add that bit, which seems like a bit of a metaphor.

I would definitely consider that a serious potential issue, if for no other reason than so many communities will likely find a use for tags based on the nature of the community structure.

For example, I could see a ton of communities having tags for things like modposts, new member intros, meta topics, memes, questions, reviews, how-to’s/tutorials, guides, etc. and that’s just for broad post types that would apply to thousands of communities.

I think letting users manually make their own multi-lems, perhaps with the ability for communities to sort of team up to make uber-lems of closely related communities to help users discover more of them…but sub, unsub, multi, and un-multi as they see fit…is likely the best approach.

Maybe don’t take disagreement so personally?

I too would like to do this myself and not have AI or anyone else decide for me what content gets lumped together.

I predict that this is also an issue that will slowly resolve itself over time, as critical masses of users gradually coalesce around one community, or more…but only if the extras are distinct in some way…which would very specifically be made more difficult by the sort of programming you’re proposing.

I’m not saying there’s no merit in your suggestion, only that it may not be the one-size-fits-all solution that you seem to think it is

Exactly.

It’s analogous to the way that Reddit knowingly allowing some subs to exist repelled some users.

Most were able to get past it and simply not subscribe to subs they found objectionable, but I’m sure many people just stayed away once they learned that certain subs existed and were very much known about by Reddit admins.

One key difference here is the way that your instance is able to enforce rules and to some extent influence and filter your user experience, and that’s worth consideration too.

I’m also curious if and how an instance like lemmy.ml can, for example, delete comments, ban users, take down content in cases of cross-instance interaction. Could the admins of lemmy.ml, for example, ban a user from another instance from Lemmy completely? From their local communities? Could they remove that person’s comments? Can they prevent their own users from seeing content they don’t like on other instances? Can they moderate content from their users that is posted to communities on other instances?