I still have no idea how Lemmy really works, and I had to sign up for this instance - I don’t know, I don’t see a platform growing on that. But maybe that’s the point. I’m trying to engage though! The Voyager app’s “import sub” feature from Reddit is brilliant.

Never thought about communities following communities. It actually makes a lot of sense and would solve the fragmentation issue in an elegant and "democratic" way.

IDK, man. It's not that hard to just check a few of the communities and see which ones are active, and then post to those ones. And the benefit you get, for asking people to take literally a couple of minutes of effort to sort out how to get involved with some particular topic, is pretty significant.

I'm not trying to say not to make good solutions to it, but also, trying to make everything maximally easy carries a significant down side, in that it attracts people who want to put minimal effort into everything (including their posts and their interactions with others once they've arrived on the network.)

https://www.youtube.com/watch?v=QEJpZjg8GuA

Duplicates are a minor issue. That said, solution #2 (multi-comms) is considerably better than #3 (comms following comms).

The problems with #3 are:

• Topics are almost never as discrete as the author pretends them to be. Often they overlap, but only partially.
• Different comms have different rules, and in this situation rule enforcement becomes a mess.

There's no good solution for that. On the other hand, the problems the author associates with #2 are easy to solve, if users are allowed to share their multi-comms with each other as links:

• a new user might not know which comms to follow, but they can simply copy a multi-comm from someone who does
• good multi-comms are organically shared by users back and forth

Additionally, multi-comms address the root issue. The root issue is not that you got duplicate communities; it's that communities in general, even without duplicates, are hard to discover. Also note that the root issue is not exclusive to federated platforms, it pops up in Reddit too; it's a consequence of users being able to create comms by themselves.

About #1 (merging communities): to a certain extent users already do this. Nothing stops you from locking [!pancakes@a.com](/c/pancakes@a.com) with a pinned thread like "go to [!pancakes@b.com](/c/pancakes@b.com)".

This is a minor part of the text, but I feel in the mood to address it:

I post once to gauge interest then never post again because I got choice paralysis

The same users who get "choice paralysis" from deciding where to post are, typically, the ones who: can't be arsed to check rules before posting, can't be arsed to understand what someone else said before screeching, comment idiotic single-liners that add nothing but noise, whine "wah, TL;DR!" at anything with 100+ chars... because all those things backtrack to the same mindset: "thinking is too hard lol. I'm entitled to speak my empty mind, without thinking if I'm contributing or not lmao."

Is this really the sort of new user that we old users want to welcome here? Growth is important, but unrestricted growth regardless of cost is cancer.

Fully agree with solution three, federated communities is the way. Solution two is just dumb and is basically just the subbed feed

Multicommunities are/grouping communities is being discussed in this issue atm:

https://github.com/LemmyNet/lemmy/issues/818

This was already touched on earlier, but I wanted to add on a bit:

The idea comes from how Reddit handles it (MultiReddits) but from my experience it's a feature not many people made use of, and it sounds like a pain to have to constantly create and manage new multi-communities to group together duplicate communities. This shouldn't be a task that users have to manually do.

This is a pretty bad or maybe just naive take that IMO doesn't sum things in a productive way upon Multi-Reddits. That is-- 1) it arguably doesn't matter a bit how many people make use of it, as each person's MR is going to be a custom affair, and it works at the individual user level anyway, 2) on the contrary, it's no trouble at all to build your MR's either quickly or painstakingly, and you can spread that effort across weeks, months and even years. In the end, I find MR's fantastically useful as super-custom feeds that you can use to stay focused on a tight range of topics.

Unfortunately, these kinds of half-baked conclusions tend to suggest to me that OP doesn't have a whole lot of familiarity with either platform at this time. That said, there's a lot of interesting ideas in the article, it's just a little disappointing in various places.

I was actually thinking of something similar a few days ago. The conclusion I came to is "comms as users."

Communities being able to follow other communities is part of that. I think it'd be great.

I quite agree with the issue described and I 100% agree it's a critical one but, because none of the proposed solution seem to be ideal, I'm also wondering if this doesn't end up saying the right model, right in the sense that it will work with/feel much more simpler to most users, is a centralized system and not a federated one?