Great article from @pluralistic explaining why he backs the #FreeOurFeeds campaign.
https://pluralistic.net/2025/01/23/defense-in-depth/
It makes more sense now, but can someone please explain to me why is costs SO much money to stand up another #Bluesky relay?
Not sure but I think ... scale.
" In July 2024, running a Relay on ATProto already required 1 terabyte of storage. But more alarmingly, just a four months later in November 2024, running a relay now requires approximately 5 terabytes of storage. That is a nearly 5x increase in just four months."
This is from a great article: https://dustycloud.org/blog/how-decentralized-is-bluesky/ from @cwebber
@Tallish_Tom @patrickleavy @pluralistic @cwebber Yup. That's the central problem of centralized vs. decentralized systems. Centralized systems offer better discoverability of content at the cost of massive scaling costs. Decentralized scales better, but inherently makes discoverability harder. Except that in practice decentralized works well enough for your own network of friends that it's indistinguishable from centralized.
@tknarr @Tallish_Tom @pluralistic @cwebber interesting...
I would rather keep it "small, local, relevant" so #Mastodon is fine for me.
If that's the money it takes for #bluesky to go global, on another instance, then fine. Someone should further develop the bridge, though, and we can all stay on mastodon and interact with those on bluesky. Happy days.
@patrickleavy @tknarr @Tallish_Tom @pluralistic @cwebber
If there are lots of servers on AT, no single server has to have massive scalability or $ because it all will be spread out. Eventually.
@CassandraVert @patrickleavy @Tallish_Tom @pluralistic @cwebber That's the thing, Bluesky intends for their Relay to remain the central one indefinitely with other Relays relegated to serving isolated special-purpose groups. That gives them control over which PDSes can federate. There's also plc.directory, which is also centralized and controls who and what can exist in the network.