Could you please explain why it would not be a problem if that would happen with Mastadon, I don't exactly understand how that works differently?
Viewing post in How do you feel about BlueSky app?
Theoretically both Bluesky and Mastodon work like emails: everyone can set their own, provided they have the resources for that, and they should be able to communicate with compatible email services.
But while Mastodon delivers those “emails” directly to the destination address, Bluesky sends to a middleman which then sends to the destination.
Now, if a Mastodon site is compromised, people in other sites can avoid it. But if Bluesky’s middleman is compromised, it’s much harder to avoid it as many interconnected sites depend on it.
Also just asked and apparently, according to Grok, multiple such “middlemen” can be set for a single Bluesky instance. If that is correct*, I still can think of some fail points, though then I’d need to figure out how Bluesky handles identifying connection throttling, shadow bans and delivered data’s integrity.
*I don’t fully trust AIs’ quality and it’s not an area I know a bunch to sprinkle questions for quality control