@lnxw37a2 @allison @mewmew @luna @cadence
IIRC at the very start they had a FAQ why do we need a new protocol and can't just use an existing one, and they argued that XMPP's architecture causes some unfixable issues. So in a way, XMPP being bad justifies Matrix's existence. If they didn't hate on XMPP, they'd have to admit the whole new protocol is a wasted effort.
But it's false.
There are a bunch of people building on XMPP, and a few days ago I've seen Tigase publish their implementation of MIX.
ICYMI, MIX is a replacement for MUC, but based on pubsub and aiming for a more Discord/Slack/Matrix-style experience (with history and all that).
Also, some of popular proprietary chat apps (forgot which) use XMPP internally.
Both of these show that XMPP is a viable base to build a modern chat on.
Your other argument - convincing faimly, user-friendliness, etc - are more about the clients and top-level features, not really about the core protocol.
And you're right, XMPP's ecosystem with the mess of extension sucks.
But that doesn't mean Matrix couldn't use XMPP core and build their extensions and clients on top, achieveing the same result without making the latency skyrocket
@mewmew @allison @luna @cadence @lnxw37a2
Ok, so they have a chance to end up with a decent, scalable protocol. And after a few trials and errors they'll probably have a performant and reliable implementation.
I think by that time MIX will be widely implemented, possibly with nice clients with Matrix/Discord-like UX.
Does that mean Matrix was wasted effort? Not necessarily.
But I don't think Matrix is significantly better than XMPP overall. It wins in some areas, loses in others.
The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!