Re: Invariants draft

[Mikkel Fahnøe Jørgensen <[e-mail]>]

> More generally: I find it unfortunate that UDP is made an invariant, anf
> possibly even IP.

That's intentional. For instance, QUIC over DCCP would be something
else entirely and not covered by this promise. The same applies to
layering QUIC directly on top of IP, or any imaginable substrate (like
your embedded device bus example). Nothing we do here would prevent
those things. Though each - as an entirely new thing - would have to
overcome its own challenges.


That is a fair point. For example it might make sense to remove UDP but
extend the type field with a routing class identifier for some networks or
use cases and such a change could still build on the UDP invariants by
stating where it differs. The same could be said about multiplexing.
Perhaps my concern is more monopolising QUIC, the name, for UDP. Thus one
could say QUIC/UDP invariants. But I also see you have to make some
boundaries for a protocol.