[saag] Re: post quantum guidance draft

Nico Williams <[email protected]> Mon, 1 Dec 2025 10:46:45 -0600
Newsgroups gmane.ietf.saag
Message-ID <aS3GdVE8H5rrcmdn@ubby>
On Mon, Dec 01, 2025 at 08:16:41AM -0800, Eric Rescorla wrote:
> To the extent to which you just want to address the problem of lack of
> guidance about PQ algorithms, I would suggest that a better way to do
> that would be to try to get to some consensus about whether there
> is in fact a problem that needs to be solved and what that problem is,
> separate from the merits of this particular draft.

We probably could not achieve IETF rough consensus on any proposition
regarding non-hybrid PQ algorithms (whether to work on them, whether to
make them recommended/MIT, have a moratorium on them, etc.).

This is why doing things at the WG level is the only way to make
progress, except then we have the same problem: that we might not be
able to reach WG rough consensus either.  And so here we are, with
constant appeals of WG consensus calls.

TBF I no longer think that an I-D like Stephen's can help for this
reason, that we would be deadlocked.  Except maybe that we could get
consensus on a _short_ moratorium, like say one year, but probably not
either (one author wrote to me bitterly off-list about this being
his/her long-time work we're talking about [as if hybrids would somehow
cause their work to not be used!]).  I expect that those who wouldn't
agree to stick to hybrids for now wouldn't agree to any specific length
of time for a non-hybrid PQ moratorium out of fear that one year will
turn to five, and then five into even longer, thus I figure no consensus
can be achieved on any proposition.  And with controversial calls of WG
consensus are more easily defended than controversial calls of IETF
consensus, I understand why we'll end up w/o a cross-area advice
document here -- but ideally we'd seek IETF consensus because DJB's
concerns might really not be remote.

Nico
-- 

_______________________________________________
saag mailing list -- [email protected]
To unsubscribe send an email to [email protected]