Martin,
A draft reflects WG consensus when it has the name draft-ietf-wg-.... (that's the definition of that name). Before then, it's just a proposal from one or more individuals.
It's the document editor and WG chair's jobs to make sure that WG drafts do in fact reflect WG status. If a WG participant disagrees with the editor and the chair that the draft reflects WG consensus, they can appeal to an AD or the IESG as a whole.
Content consensus of a draft document is judged formally only during WGLC. The consensus to rebrand the draft document to a WG draft is mere housekeeping that has no meaning outside of the WG and does not necessarily affect the way the draft is edited and updated. WG drafts can and do fail of achieving WG content consensus.
Any ID is just a proposal from one or more individuals, or a
scratch pad for rough thoughts, or stupid pranks. Making them WG
drafts does not change that.
In my experience, the WG does NOT gain control of the document
(and I have several worked examples in the DNSOP WG). In some
cases, if it did, I might be more successful at getting fixes
adopted against the author's will, but I think making the WG
directly in control of the content of a given document prior to
the WG submitting it for publication is generally a bad idea.
IDs are working documents. Period. Regardless of whether they are named draft-person... draft-ietf... draft-wgname they have exactly the same standing within the community as any other ID. They are explicitly NOT standards, although they may be (draft) specifications. RFC7221 purports to give WG drafts a special status as WG owned, but I will note that that document is not a BCP and the IESG discussion (https://datatracker.ietf.org/doc/rfc7221/writeup/) was clear that 7221 does not create a new step in the process of publication of RFCs.
https://datatracker.ietf.org/doc/draft-carpenter-gendispatch-rfc7221bis/01/ might change things, but as of now, the above holds true.
Later, Mike
This is from RFC 2418:
6.3. Document Editor
Most IETF working groups focus their efforts on a document, or set of
documents, that capture the results of the group's work. A working
group generally designates a person or persons to serve as the Editor
for a particular document. The Document Editor is responsible for
ensuring that the contents of the document accurately reflect the
decisions that have been made by the working group.
As a general practice, the Working Group Chair and Document Editor
positions are filled by different individuals to help ensure that the
resulting documents accurately reflect the consensus of the working
group and that all processes are followed.
Cheers,Andy
On Tue, Apr 27, 2021 at 12:07 PM Martin Vigoureux <martin.vigoureux@xxxxxxxxx> wrote:
Andy,
quick question, if I may:
Le 2021-04-27 à 17:06, Andrew G. Malis a écrit :
> Keith,
>
> I disagree. WGs have charters, which result in RFCs. During that
> process, they have consensus-based working drafts that are refined to
> meet their charter goals. That's an "adopted" draft. But it doesn't have
> to be based on a single individual draft, a working draft can be the
> result of merging earlier individual drafts, or can even originate as a
> WG draft without a preceding individual draft or drafts. But yes,
> working drafts do reflect WG consensus, and they have formal standing as
> such.
At which point in time to do they reflect WG consensus, according to you?
As examples to illustrate my ask: From day 1 or only at "Publication
Requested" time, or some other time, if any specific one?
-m
>
> Cheers,
> Andy
>
>
> On Tue, Apr 27, 2021 at 10:27 AM Keith Moore <moore@xxxxxxxxxxxxxxxxxxxx
> <mailto:moore@xxxxxxxxxxxxxxxxxxxx>> wrote:
>
> On 4/27/21 10:17 AM, Salz, Rich wrote:
>
>>> There was also a suggestion to add something to the boilerplate text of individual I-Ds along the lines of "anyone can submit an I-D; they have no formal standing until they are adopted by a group in the IETF or IRTF". Would that provide additional clarification?
>> Oh yes, PLEASE!
>
> concur. Except get rid of the "adopted" bit, because even assuming
> that "adoption" of a draft by a WG is useful, it doesn't imply any
> kind of broad support from the organization. Just say that the
> existence of a draft does not mean it has any formal standing with
> IETF or any other organization. Documents with formal standing in
> IETF are published as RFCs.
>
> Keith
>
>
