https://daringfireball.net/linked/2023/06/19/not-that-kind-of-open
@daringfireball Openness isn’t merely a protocol thing. It’s a delicate power balance.
When an entity larger than all other combined joins, it may become "too big to fail" and start dictating terms.
Gmail unilaterally dictates who can use SMTP. GitHub became the center of decentralised git. Systems tend to centralise.
@kornel @daringfireball Yes, and the way to fight with GMail isn’t to “block” gmail from other email services. That hurts the other email services, not Gmail.
Meta’s threads doesn’t *need* ActivityPub. It has Instagram’s far larger network. Blocking them from your mastodon instance only hurts you, not them. It resigns Mastodon to forever being that weird niche rather than part of the mainstream.
Integrate and use it to make yourself mainstream!
@oblomov @kornel @daringfireball Let’s say that’s true. It’s plausible that Meta wants to do the whole “embrace, extend, extinguish” thing.
Defederating them early won’t fix anything. Meta has a huge user base. Threads doesn’t need AP to “bootstrap” content.
It will either steal users from Mastodon, or it won’t. De-federating will incentive more users to switch away from Mastodon. Integrating will let them stay because of the integration.
@modulusshift @oblomov @kornel @daringfireball I understand how it works. You missed the rest of point. Defederating jumps you directly to the “you only get any experience on their instance” step. So yeah, it’s counter-productive.
Many apps are being built on AP that are “extending” in all sorts of different ways already. These other apps will just have to compete with Meta.
@nmn inasmuch as this instance is part of the fediverse, sure. I doubt it will be in practice, so much as a parasitic blob that incidentally connects to the fediverse just enough to suck it dry.
in short: I don't think it's going to make using the broader fediverse any better, and I suspect it's going to eventually attempt to kill it like any centralized-brain company would.