The design of the activity pub protocol (of which lemmy, kbin, beehaw, mastodon, etc. are built) is such that there could never be a central authority that controls the entire fediverse.

Consider the world wide web, which is the biggest federated service in history. There is no central administrative body that can exist for all websites because anyone can set up their own web server and host whatever they want on it. The same is true for lemmy instances.

The beauty of this system is that people who want a more curated experience are free to use instances that ban the kinds of behaviors or views that they find problematic. Compare that to Reddit where hate subs get removed only when they become a financial liability.

Essentially. none.

Admins can block users on individual servers. If individual servers are causing problems, Admins can defederate them, which causes more issues then it resolve usually.

There, isn't much you can do to ensure an individual problem-user stays banned.

Non-profits are not an antidote to corporate control. Non-profits rely on grants which are typically coming from corporations or governments, with strings attached. While there are a handful of legit non-profits, the majority of them are just ways for people with money to burn to expand their influence on various issues. Especially when they are focused on setting standards like this, as opposed to turning donations directly into meals / supplies / shelters etc.

There have been discussions of doing this on Mastodon as well, but they have been met with resistance. There are several instances which have incorporated themselves as one form of non-profit or another (depending on jurisdiction), but the centralization of block-lists or introduction of federation-wide governance bodies has not taken off for various reasons.

In general, it is good practice to look at the block-lists from a few instances you vibe with and use that as a starting point, but realize when it comes to politics, everyone is a crank and that simply copy/pasting their lists isn't the greatest approach. A lot of instances are shit and belong on those lists (the bigots), but several end up there because people are allergic to nuance, while several are omitted because people think a subject is nuanced when it really isn't.

Also, I wouldn't worry about an instance getting too big at this stage. Some communities are big. Some communities are small. Some have irreconcilable differences. As we stumble upon these differences, there will be splits. New instances get started up every day. People will have disagreements about the way admins / mods handle things. This is how it goes. Having 100,000 homogeneous instances of equal size does not indicate a healthy network. The way these things split and group up is driven by culture and politics. The ideal place to land has nothing to do with the size of the instance, but your trust that the administration won't cleave your community in half for arbitrary reasons.

Who gets to decide what is acceptable speech? Who watches the watchers?

Why is the world no longer ok with the fact that people hold differing opinions on things, and feel the need to segregate themselves away from people that they disagree with?

This was the fundamental problem with reddit IMO - the echo chamber was very real. The community guidelines and upvote/downvote mechanic were such that you basically ended up with a hivemind consensus on any given issue simply because any view not inline with that consensus was hidden, obliterated or banned. That's not healthy discourse.

The same is true for any echo chamber, and the same will be true for people who elect to hide in a walled garden over at beehaw and pretend that there simply IS no differing opinion on their chosen issues.

The problem is, you only serve to delude yourself. You go about your life believing that the world is on your side, when in reality more often than not you are part of an extremely small minority.

I think the current solution of multiple instances will work fine and is far preferable to what you are suggesting, which is a simple duplication of the global censorship policies that kill any genuine discussion on reddit around hot-button issues.

With respect to how it works in the microblogging corner of Fedi, the tendency is to be actively collaborative, and aggregate some moderation resources, sometimes through backchannels, other times through a tag like #FediBlock - all of which have political implications that have been years-long meta discussions. The emphasis, at least among instances that want to moderate heavily, is on allowing users to feel undisturbed in their own space and not be challenged on literally everything they say, but to still expand that space where it makes sense.

I'm not sure the exact same dynamic will take place over here. The existence of many distinct spaces on the same instance mitigates a major initial problem Mastodon faced in its early waves: when you literally put everyone leaving Twitter on the same public timeline, old grudges spark up and they start campaigns to harass each other off the platform. That's how it came to pass that Mastodon ended up with a ton of user privacy features, and over years, instances warring over ideology and trying to colonize each other, which of course ends in mutual blocking.

In our case I think there's a good chance for small aggregator instances that just "do one thing well" to thrive and see a lot of external traffic, while not having to moderate their entire comments section, since you can opt to not federate that - not your site, not your concern.

I had an idea that servers could issue out "equity" or voting shares. Decisions like moderation teams, federation, or other choices could be voted on based on users stake.

These shares could have other uses as well.