After reading a bit about Usenet, it seems to me as if the whole Fediverse seems to be just a reinvention of Usenet.
What’s the big difference?
After reading a bit about Usenet, it seems to me as if the whole Fediverse seems to be just a reinvention of Usenet.
What’s the big difference?
Actual attempt at an answer!
ActivityPub has actors and activities. These are very broadly defined - yes, a user is an actor, but so is a magazine in kbin. A like, a thread, and a microblog are all activities. These come from an actor, and they are sent to and cc’d to other actors in the fediverse.
NNTP, however, is not actor to actor, it’s server to server, to my understanding.
In practice, the way this is implemented here, it’s not that much of a practical difference, but it’s interesting to know.
The other difference is that NNTP servers would forward messages to their other known NNTP servers, essentially creating a distributed network of information. Per the ActivityPub protocol however, no instance is obligated to do that on ActivityPub. The only obligation for forwarding is if a) The values of to, cc, and/or audience contain a Collection owned by the server (e.g. followers is a Collection) AND The values of inReplyTo, object, target and/or tag are objects owned by the server. So basically if I receive something from lemmy.world user actor, to lemmy.world community actor… Even if kbin.social hasn’t received it and errored out, I have no obligation as the.coolest.zone to send it out to them.
From my experience Usenet required you to post in a specific NewsGroup. They had threads but if you posted something in Class1Railroads about your modeling it wouldn’t necessarily be seen by ModelRailroads. The threads got really messy with some quoting the whole thread before adding their comment at the bottom.
Fediverse gives the posts a chance to break out and be picked up by others. Post in Kbin/Lemmy in a specific group/magazine and someone in Mastadon might see it and reply without being subscribed to the specific group. They don’t have to wade through the whole conversation to read your post.
Thanks, at least one who got what I was on aboutˆˆ
Ok, so we got a push vs pull model and a bit more differentiation in the protocol. So there is at least some improvement on the concept. When reading about it, it felt like yet another reinvention, but looks like there is at least some improvement on the idea. Thanks for the summary!