I’ve been thinking in the back of my mind for a while now, that the Yarn.social / twtxt + ActivityPub integration was a mistake and a. bad idea. I’m starting to consider it a complete failure.
@movq@www.uninformativ.de For several reasons; I think the pull based model is honestly better in every regard, the deliberate slow nature of Twtxt/Yarn is actually beneficial and discovery/connections made through “people you know” is more genuine (rather then pushing crap to randos you don’t really know or have never known)
Not to mention, actually implementing Activity Pub is just so hard to get right 🤦♂️
I understand your thoughts on this, but I would not call it a failure - because you learned a lot from it, and lots of things worked as well.
And there are alternatives for those who needs\wants activitypub, so I think also yarn\twtxt benefits from you focusing on that instead of dealing with the frustrations of activitypub integration. And maybe it’ll feel a bit better to put that on the backburner? :)
@stigatle@yarn.stigatle.no I agree 👌