Re: [irsg] New document shepherd writeup

[Carsten Bormann <[e-mail]>]

On 2022-05-04, at 14:46, Joel Halpern <[e-mail]> wrote:
> 
> The important existing requirement was that the authors (and by some interpretations the named contributors) confirmed explicitly that all known IPR believed to be relevant has been disclosed. That requires explicit responses from those people.

Requirement for what?

I’m not aware of such a requirement.

(I’m aware of an obligation to disclose, buried somewhere in BCP78/89.)

I’m aware about a checklist item that a shepherd can answer based on their knowledge.  If I’m shepherding a document, and I can’t get that confirmation (e.g., because the author died or because of a more mundane reason), then I’ll write this up and send the shepherd report forward.
[Note that a typical responsible AD response, asking the WG chairs to demote that non-responding author, does nothing at all to solve the underlying problem.]

I think that we want to have two things here:

— A general “did you know that you have to/SHOULD disclose” broadcast reminder (unreliably) to the constituency (WG, …).  The shepherd could report that this broadcast reminder has been sent.

— the desire for an explicit confirmation round with a select set.  Note that we make no requirements on the formality of this confirmation (is it a signed email?  How did you verify the key?), not even that it is uttered in the public.
Whether the contents of the RFC 7322 contributor section belong to that set or not depends on how the authors are using that contributor section — “Peter Smith proposed alternative 3 that clarified the choices but after much deliberation was not included in this specification” vs. more actual authors than 5 so we shove some into the contributors section because of RFC 7322.

As an author, I wouldn’t want to be in a position that I need lie about someone’s contribution just because they have a burnout or are otherwise hard to reach.  This confirmation is good practice, but not a requirement.

Grüße, Carsten