Yes, but compatibility with existing internet standards is also important. For example you can't have end-to-end encryption if you use a non-standard protocol. VC startups like Matrix only increase fragmentation of the ecosystem.
TIL that XMPP is defined in an RFC. You're correct, I wasn't aware of that. I really don't understand why the IETF take such a decision though. I don't know why these guys are defining high-level protocols for things like messaging at all.
But back to your earlier points:
For example you can't have end-to-end encryption if you use a non-standard protocol
This doesn't make any sense whatsoever. Matrix has E2EE while using a "non-standard" protocol. So does Signal, in fact, it created the strongest E2EE protocol out there.
VC startups like Matrix only increase fragmentation of the ecosystem
Every new project that is created increases fragmentation. So does Revolt, Discord, Skype, WhatsApp, Signal, Telegram, etc. These all use "non-standard" protocols.
Also, the author of RFC 6120 is a Cisco employee, how is a multinational corporation better than a VC-funded startup? XMPP is an open standard, just like the Matrix protocol. It doesn't matter who created it.
There are lots of high-level standards for communication. You might have heard of email, its protocols are also defined by the IETF. For instant messaging it is XMPP.
This doesn’t make any sense whatsoever. Matrix has E2EE while using a “non-standard” protocol.
You can only encrypt messages when the recipient happens to be a Matrix user too. If they use another protocol it's not possible. That's why we need standards and that's why building on existing internet standards is important as opposed to everyone cooking up their own IM protocol like Matrix does.
Every new project that is created increases fragmentation. So does Revolt, Discord, Skype, WhatsApp, Signal, Telegram, etc. These all use “non-standard” protocols.
Yes, they all increase fragmentation as they do not interoperate with one another. A standard would solve this problem.
Lot's of standards are written by people who work at large corporations. Also multiple experts from multiple corporations work together. Wasn't the original author of the XMPP protocol was hired after the fact by Cisco, precisely because he wrote XMPP and the first server implementation?
The IETF still has a much better track record than any single corporation or VC funded start-up.