Luke's feedback on the permissioned data proposal is pretty interesting. The current proposal has a kind of locational element to permissions, where a record's URI reflects the access control, and I can understand why that would be jarring. I'm talking with the team about whether we think that's something that could change, and what a change like that would cost. We're still in the collecting feedback phase so it's good to hear how it hits people.
Regarding his other concern, I would've loved to make atproto a localfirst protocol -- that's the field the atproto team came from -- but we had to pick & choose where to spend our complexity budget. I wouldn't say it's a nonstarter for the ecosystem; you can use something like iroh in conjuction with atproto to share identities -- dial people at their dids or domain names -- and then commit state to the atproto repo once it's ready to leave your local sync mesh. Maybe that sort of thing makes its way into future versions of the protocol. But, engineering often involves trades, and for the v1 that was one of the trades we had to make.
a meta-note that i think deserves attention: we never ever seen such a level of engagement around a social media protocol as this. count the number of mastodon protocol discussions that have happened, or dispora, or activitypub. they exist as background. matrix maybe comes close if you look at raw count anywhere, but i'd say has no-where near the visibility, doesn't surface/emerge into the public. atproto is succeeding at being a discussed protocol. at being a socialized social media protocol.
and that's just amount of discussion. but what about engagement by the people working on it? nothing like this, not by high level people, discussing in open, talking about protocols. you don't see permissioned data blog serieses coming out about the direction. you don't see the same blogsphere posts. you don't see the CTO coming to chat about it.
a huge amount of this is just that Bluesky people really work hard to do the right thing, are humble and trying hard to build good protocols, good network design, trying to build the most distributable most easy to get the content network out there that works the best. i think also though, the ability of the systems / protocols here to be expressed lends itself to being discussable, has good primitives, that make it fruitful and interesting for things like Permissioned Data, or things like the sync 1.1 protocol, to be generally understandable, good primitives, that are discussable.
luke's points about local first are really good. this is hard challenge for "Authenticated Transfer Protocol", at protocol, which prioritized a sense of identity and data coupling, that is harder to do in local context, that is harder to distribute without the online verification. great concern, great area to focus on. i look forward to more people experimenting with interesting PDS architectures & seeing what if anything might possibly relax to accommodate.