Re: [quicwg/base-drafts] per-CID PN space to avoid PN leak becoming a privacy issue? (#1317)

[janaiyengar <[e-mail]>]

@kazuho : Thank you for writing up this issue. This is exactly the alternative design I had thought of in London. As @ianswett notes, if we were to go here, I suspect this bridges most of the gap from where we are to multipath. I agree that while it's not in scope for v1, if this (i) can be implemented simply for the non-multipath case, and (ii) can be extended by implementations to experiment with multipath, then we're headed in the right direction.

Thinking about this some, I think this is plausible. The current approach of using PNE or whatever, as you note, has exactly the same requirements as this mechanism does, in that a single loss recovery context with exceptions can handle both. If extended however, which we don't need to specify in v1, then multipath is achievable.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/1317#issuecomment-383745711