Alternative to defining a bunch of MP-specific frames

[martinthomson]

With the new explicit path identifier, maybe I can suggest an improvement.

Look at Victor's https://datatracker.ietf.org/doc/draft-vvv-quic-namespaces/ It strikes me that the trick he is using with the NS frame there might be useful for MP QUIC. Only path-specific frames would be affected. You can even start with a default value of the current path.

[marten-seemann]

Not treating frames as separate entities breaks a ton of assumption I have in my code. I think this massively complicates implementations.

[qdeconinck]

Actually, this was already discussed a bit as part of #214 (#214 (comment)); I believe frames should be fully self-contained, and not doing so would complicate at lot implementations.

[mirjak]

Given this was discussed in #214 already, can we close this issue?