Re: [Rfced-future] AUTH48 and editing before approval (was: Re: Welcome to the RFC Editor Future Development Program)

[Russ Housley <[e-mail]>]

> On Apr 2, 2020, at 12:12 PM, Ted Lemon <[e-mail]> wrote:
> 
> On Apr 2, 2020, at 11:52 AM, Russ Housley <[e-mail] <mailto:[e-mail]>> wrote:
>> The IESG approval process almost always leads to changes based in DISCUSS or COMMENT.  Sometimes these are technical changes.  We should not do copy edits until the technical document is stable.  That is exactly where the copy editing is in the current process.
> 
> One thing to bear in mind is that it is not uncommon to discover a technical issue in AUTH48, just as we do with errata.  I don’t think the right thing to do is to say “don’t fix it, since it’s AUTH48.”  I also don’t think the right thing to do is to say “just fix it and don’t check with the WG.”  However, in at least one case, we had some serious problems that were caught in AUTH48 after the WG had disbanded.  We fixed these, and I think did an IETF last call, although this was ten years ago and my recollection is imperfect.
> 
> I say this not to propose a solution, but simply to point out that this is a nasty problem.  I don’t think there’s an easy solution, but I’m sure that we could define a formal process if there were energy to do this.  I agree with other commenters who say that this is out of scope for the current effort, since this is really an IETF-track problem, not an all-track problem.

This is very rare, but it does happen.  When it does, the AD needs to decide if the fix is obvious or if the fix need to go back to a WG or IETF Last Call.

Russ