Dear Margaret,
I want to be clear first: I certainly consider one or two cases. But
no ad hominem is intended. To the contrary I try to take advantage
from experience to see how a slight procedure description improvement
changing nothing (cf. Brian) could help avoiding a problem I see no
reason for.
At 13:58 22/08/2005, Margaret Wasserman wrote:
Hi Jefsey,
At 11:05 AM +0200 8/18/05, JFC (Jefsey) Morfin wrote:
http://www.ietf.org/internet-drafts/draft-ietf-proto-wgchair-doc-shepherding-05.txt
I agree with Brian that this procedures does not change the
standards process and/or the official role of the WG chair. In
fact, the ideas in this document were largely adapted from things
that WG chairs were already doing in some areas or in some
groups. The idea is to ask the WG chairs to take more
responsibility for the quality of WG output, hopefully reducing the
number of documents that come to the IESG will serious quality issues.
Same point of view. I just want still to help decreasing the number
of that documents, in helping decreasing the chances of appeal.
However, I do think it would make sense to consider/discuss your
specific suggestions:
1. The first addition is that the proposed write-ups are presented
for quick comments to the WG.
The "ballot write-up" portion of the questionnaire does become
public, and I suppose that it could be reviewed by the WG. However,
I am not sure what benefit that would provide...
Just to help making sure everything (including disagreements) are
reported the way everyone sees them and reduce the number of
documents. Let assume you have a disagreement: it is better that
every party agrees on the way their own positions will be presented
than the IESG to receive a flood of mails. This way IESG at least
knows that each position has been reviewed by its supporter.
Before the PROTO questionnaire was in use, ADs would provide the
ballot write-ups with no input from anyone.
This permitted comments on the WG Chair. I am confident quite the
same information may be offered by the WG Chair himself, with the
help/support of transparency.
In general, they consist of the abstract of the document (perhaps
with a bit more text to say why the specification is needed), a
summary of the WG consensus process and a statement about who has
reviewed the document in the later stages of review/approval. These
statements do not end-up in the RFC or in any archival location...
What is your concern about these statements? And, what do you hope
that a review will accomplish?
There are probably may cases. Both of trust and distrust. The
write-up should be received as a comment by the WG through its
WG-Chair on the accomplished work. This seems important at least in
three cases I experienced:
1. the charter has actually not been considered because a leading
group has a predefined vision of it.
2. the charter could not be fulfilled for good reasons
3. the charter is only partly adequate and had to be upgraded
I think that we should be careful about adding any more steps to the
standards publication process, so I will personally tend to push
back on any steps that do not, IMO, add significant value.
Full agreement. I measured that from comments. This can be addressed
in another way (see below).
2. two questions more are added, one on the way the Charter has
been respected, one on the care given not to favor one technical
vision over others (one might refer to RFC 3869). I suppose
competition in a WG is not between propositions but for the best
user needs support?
Today, we do not have an explicit check that a WG work item that has
been submitted for publication matches a WG charter milestone or is
otherwise within the WG charter. There is an implicit check during
AD review, perhaps, but not an explicit one.
I would like to see such an explicit check added, so I personally
agree that it would be a good addition to the PROTO questionnaire
for the WG chair to state what WG milestone is represented by a
particular document and/or otherwise explain how the document is
in-charter for the WG. I think that we should consider this
addition if/when the PROTO process is updated.
There is an implicit implicit check. I would suggest there is an
implicit explicit check with possible comments. This would only
translate into a wording like "if the Shepherding Chair does not
commit the Charter has been technically inclusively fulfilled the
part I.x must document it."
The WG Chair already de facto commits there is a rough consensus.
This would add he also commits the charter has been considered and
fulfilled on a technically inclusive basis. This follows what Brian
has reminded. Otherwise he has a standard place to mention and
possibly explain it.
Others may disagree, of course.
The second portion of your suggestion is, IMO, already represented
in the questionnaire, as we ask about the WG consensus process,
contentious issues, other options discussed by the WG and the
strength of WG consensus on the document. It is the role of the WG
chair to determine WG consensus, and the questionnaire asks the WG
chair to report on the state of that consensus.
I sense that there is some concern underlying your questions
regarding the choices being made by a particular WG... The PROTO
process isn't intended to, and really won't, affect the authority or
the responsibility of WG chairs to guide the WG through difficult
decision-making processes and to make a determination regarding when
rough consensus has been reached on a particular choice or approach.
This is the core role of the WG chair. The PROTO document does give
the WG chair an explicit opportunity to share any concerns that
he/she might have about a particular consensus call and/or selection
process with the AD, but it doesn't change the nature of that process.
I do not want anything else than to help the WG Chair. If the Chair
knows that he will have to report on the way the WG has considered
the Charter, this might help him considering better the Charter while
conducting his WG. We are not here to dispute but to achieve the best
work in common: the WG community results from the adherence to the
Charter. If Members see the Charter is not considered enough
(even in a different way) they vote with their feet. To common detriment.
If you are concerned that an unfair selection process is being
followed in a particular WG (or has been in the recent past), you
should talk to the WG chairs about it. If you are not satisfied
with the WG chairs' response, you should raise the issue with the
responsible AD listed on the charter page.
In one particular current case I certainly consider, this has been
done. The PROTO target is to reduce the IESG load and paperwork. I
try to help reducing them in helping a better common understanding
from the very beginning when a situation developed where a few
immediate DISCUSS or even the wish to avoid them could prevent a
difficult IETF LC and appeals. But my point also and mainly concern
the cases where a WG uncovers new possibilities not covered by the
Charter: we are here in an innovation area we all agree we want to do better.
I suggest that a flexible way for a WG - through standard procedure -
to comment its own Charter and to explain some resulting positions
and choices, can only be of help to everyone and to foster innovation.
All the best.
jfc
_______________________________________________
Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf