Mastodon Feed
jonny@neuromatch.social ("jonny (nonvenomous)") wrote:
there are lots of cool tech things happening there and i am pro-atproto, but after watching it happen a few times, it is sort of inescapable how one of the 'dev draws' that people who move from working on AP to atproto express as "the atproto userbase is not so stifling" is equally well expressed as "atproto users have no prior expectation of privacy and respect for consent." all-public-everything is already baked in at a protocol level except for bespoke implementations, so the floor is very low.
the AP space is sort of stagnant because it's stuck in a single program (also, as always, respect to the masto devs holding it down) which prohibits people from being able to self-sort into programs that can actually respect their privacy needs, so a lot has to be done by social convention and climate. as a result, people working on AP stuff end up stepping on a bunch of rakes - often, charitably, with every intention of not doing so - because we have the expectations and desire for privacy without being able to take the steps we need to make it actually enforceable.
I see a lot of the cooler atproto work (which is cool and i am a fan of, i genuinely think Blacksky is the coolest group working in the whole extended space) re-arriving at stuff we have had on the fedi for a long time. on the AP fedi we can't do some of the more public things like they are doing in no small part because the people who don't want that kind of thing can't protect themselves from it, and with an even finer grain, choose when and in what context to opt in.
so, the "AP is a bunch of stodgy privacy zealots" argument (which i am fine with, i am a stodgy privacy zealot), in my opinion is largely just a reflection of "AP is stymied by protocol and implementation where it's difficult for people to not use mastodon forever." If it was much easier for people to move around, experiment, temporarily duck into a different, more privacy-respecting software stack, exist across multiple, and so on, then a lot of the "yet another opt-out scraper" problems disappear by giving people the power to prevent that in the first place.
I think LD needs tooling, but is mostly just "unused" in the AP fedi, and it should be a very interesting protocol space to work in, but that one absence of migration and mutability at an account level is a protocol killer.