Almost everything is available. You can run your own account host, feed generators, moderation services, app servers (appview, relay) and most code is open. The only thing not open is a bunch of custom scaling optimizations (like database configurations) and configuration for the official recommendation algorithm & spam filtering mod tools, and stuff like that. All the rest is available, and the things that's missing aren't necessary unless you want to match their user count (but then you can probably build it yourself)
Natanael
joined 1 month ago
He's not a shareholder, and also it's a public benefit corporation so shareholders have less power over the board
He never had ownership. The investment was in the form of a contract to build the protocol, not buying shares.
Jack Dorsey never had ownership (just directed an investment) and left the board (didn't agree with moderation, lol)
Jay also isn't majority owner.
It's a public benefit corporation too so they don't have a profit requirement.
The harder parts with decentralizing content-addressed systems like it is scaling open spaces (like how a microblog is technically one big shared space). You need big caches and big indexes. They're working actively on making it easier for others to run those app servers. There's already a few independent projects building them. Federating account hosting and feed generation and moderation services are all live already
There's a new option available now for reply controls, you can limit it to just people who follow you. While it's a very low bar, it's enough of a threshold for most randoms to not bother following just to reply to you
Bluesky is open source though
As long as these types pay attention to what the scientists tell them and explain it to others accurately, they're helpful
Now you have to reread this too
Some people only browse global feeds and downvote stuff as if they're trying to train the Netflix recommendation algorithm, completely ignoring the rules of the community it originates from
Hashing alone if it's just usernames isn't enough. Need something like keyed hashes, but then malicious servers can lie about numbers of votes.
Otherwise you need something ridiculously overengineered like public but encrypted logs of user actions and Zero-knowledge proofs of correctness mapping everything to a distinct existing user without revealing who it is.
As I mentioned in another post: for consistency is better to have each server count total votes from their own users, send a signed & timestamped message with the count to the host of the post being voted on. Then the host can display a consistent vote count to everybody that shows where votes are coming from without manipulation of external votes.
Each individual server can lie about its count, but not by too much or else it will be detected and the server can get defederated (or have its votes ignored).
Especially in federated networks where the data isn't under access control, doubly so if the privacy extension is optional
Have you heard of bridgy?