Re: QUIC ossification

[Spencer Dawkins at IETF <[e-mail]>]

Speaking as an individual who is trying to be helpful :-)

On Fri, Feb 15, 2019 at 1:33 AM Mikkel Fahnøe Jørgensen <[e-mail]>
wrote:

> It would be useful with a private version range that can be deployed in a
> limited context without involving IANA.
>
>
> ------------------------------
> *Fra:* QUIC <[e-mail]> på vegne af Martin Thomson <
> [e-mail]>
> *Sendt:* fredag, februar 15, 2019 3:43 AM
> *Til:* [e-mail]
> *Emne:* Re: QUIC ossification
>
> On Fri, Feb 15, 2019, at 12:22, Ian Swett wrote:
> > I believe the concern is that a middlebox/etc will conclude that a
> > certain range of versions are IETF standard versions and should be
> > allowed and others are fine to block because they're experimental,
> > thereby making the majority of versions largely undeployable.
>
> That only suggests to me that we not segregate the space. The proposed
> variations won't all happen in that space.
>

 So, yes, I get the interest in being able to dork around without
immediately having to involve IANA, and I think you're both exactly right
to point to the tension/tussle between people wanting to know what they're
admitting and people wanting to experiment.

ISTM (keeping in mind that TSV ADs are stuckees for appointing services and
port number registry expert reviewers) that this is EXACTLY analogous to
transport-level port numbers, except (for extra credit) all the transport
stuff in QUIC is encrypted, so anyone who cares about what they're
admitting has to look at something else, and if anyone has a better
suggestion than looking at version numbers, you should probably tell the
working group, which will immediately encrypt it :D

So - if you don't want people to squat on version numbers, which has been
problematic for port numbers since the 1980s, you probably want a
segregated version number space for experiments, and if you think (or
observe) that people who care what traffic they are admitting are blocking
experimental version numbers, you probably want that space to be so small
that it really can't be used for wide-scale, long-term deployments.

I'll stop typing now, and let people like Lars, who has forgotten more
about port number reviews than I've ever known, tell me why this is a bad
plan ;-) ...

Spencer, again, as individual