Re: [Curdle] [saag] Time for SSH3?

I wouldn't compare SSH2 -> SSH3 to IPv4 -> IPv6. IPv6 has been taking
forever because it requires changing every router on the path. A better
comparison would be TLS 1.2 -> TLS 1.3, or HTTP/2 -> HTTP/3, as those only
required modifying the endpoints (for a specific definition of
endpoint...). And that transition happened quite quickly. I do think it
makes sense to consider rearchitecting SSH like we did for TLS, but I agree
with Stephen that we'll need the developers of popular SSH stacks to be
enthusiastic about such an effort for it to have any chance of success.

David

On Wed, Dec 20, 2023 at 8:37 AM Stephen Farrell <stephen.farrell@cs.tcd.ie>
wrote:

>
>
> On 20/12/2023 16:35, Theodore Ts'o wrote:
> > Moreover, if IETF tries to standardize a completely incompatible
> > protocol rewrite without close coperation with development team(s) of
> > the dominant implementation(s), the precedent of IPv6 of taking
> > **decades** to be fully rolled out may be the more relevant
> > comparison.
>
> +1 - if the main developers of SSH implementations were up
> for starting work on an SSH3, then it'd be a good plan. If
> they're not, it'd likely be a bad plan.
>
> Cheers,
> S.
> _______________________________________________
> Curdle mailing list
> Curdle@ietf.org
> https://www.ietf.org/mailman/listinfo/curdle
>