Bridgy doesn’t scrape anything. It works the same as any other ActivityPub instance, the only difference is that it converts some JSON from one format to another.
It also converts edit and deletion events, so in your scenario it would relay that you want your poem or photo deleted.
This isn’t a web scrapper that reposts content like all the bots reposting Reddit threads to Lemmy. This is a protocol translator between federated networks that speak different languages.
That’s exactly the point. Isolationists instances can always defederate bridges if they don’t like them. This outrage is them imposing their will on the rest of the fediverse.
This insane isolationism from the vocal minority will kill ActivityPub. The fact that the author is now backing down and switching to an opt-in system is infuriating. Makes want to fork the project and host a copy of the bridge that’s opt-out.
The good news is that none of the large instances are going for these insane policies. Small instances and solo instances can defederate themselves into irrelevance all they want, just like beehaw did.
This reply from July shows they’re still working on it, it’s just challenging: https://nitter.net/photomatt/status/1676415784548184065
Is Tumblr still working on this? It’s a little depressing to think Facebook of all companies might beat you to it.
Yeah 😕 They are orders of magnitude bigger and better capitalized than us. Not an excuse, just reality.
It makes sense to pause registrations if an instance is actively overloaded, but there’s no reason it can’t keep growing if more hardware is added and the performance issues are resolved.
Not entirely sure what you mean by that. It deletes posts in the same way that any other ActivityPub server does, by federating the deletion request.
It’s up to the receiving servers to handle that request and delete the post. You can easily have an ActivityPub native server that doesn’t honor those requests.