Let me restate for clarity - ADs aren't necessarily more technically
astute than *all* the rest of us. That is, we need to be careful that
technical input from ADs isn't automatically assigned extra weight or
control (veto power).
Which is why I suggest ADs provide technical input in open mailing lists
during last calls, to make sure their technical input is on the same
footing as everyone else's technical input. I agree that the IESG's job
is to ensure correctness, completeness, etc. That feedback should be
provided earlier, in an open forum.
If a particular AD doesn't have cycles to monitor every last call - and,
as the AD has to review the doc and to a technical review, anyway, I'm
not sure monitoring last call discussions would consume many more cycles
than the current process - the AD could call on an expert or a
directorate to participate in the open process.
But, I'm indulging in generating solutions rather than identifying
problems. I guess the problem I see is separating technical analysis
from process management...
- Ralph
On Fri, 2005-04-29 at 02:01 -0400, Keith Moore wrote:
I don't see anything wrong with that. It's the ADs' job to push back
on documents with technical flaws. They're supposed to use their
judgments as technical experts, not just be conduits of information
supplied by others.
I disagree that the ADs are necessarily that much more technically
astute than the rest of us.
1. ADs usually _are_ more technically astute than the average IETF
participant (perhaps not you, but the average participant), because ADs
are selected for their expertise while IETF participants are
self-selecting. (Maybe some are selected by their employers, but I
don't think it's generally the practice of most employers to send their
best technical people to standards committees. My impression is that
many employers - not all by any means - would rather send people who
are expendable, and/or who will represent the company's official
position rather than their own best judgment.)
2. ADs also tend to have a broader perspective than the average IETF
participant, because ADs are exposed to everything that IETF does while
most participants' activity is confined to a narrow topic area. That's
not to say that a broad perspective is inherently less valuable than a
narrow perspective - they're both valuable, but for different reasons.
But both of those are less relevant than the fact that it's IESG's
_job_ to make judgments about the quality of specifications and
protocols, and that this is a _necessary_ job. WGs are too frequently
insular and too frequently want to make unacceptable compromises -
somebody has to serve as a check against that. Somebody has to resolve
conflicts between competing concerns. Somebody has to make sure that
the specifications are complete. etc.
I would actually feel more comfortable with ADs providing their
technical judgment with the rest of us, through the same mechanism: WG
or IETF last call. And that technical judgment should be expressed
openly, in an archived WG mailing list, where everyone's technical
input can be reviewed and everyone who provides technical input can be
held accountable.
I have mixed feelings about this. I believe that often it is the case
that ADs providing technical input through the WG mailing list, and
participating in discussions on that list, is the most effective way to
resolve the differences. On the other hand, mailing lists that are
focused on a narrow topic are not good places for resolution of issues
that involve concerns outside the WG's scope - the WG tends to dismiss
those concerns out-of-hand even though they are valid. Also it is
impractical for every AD to participate in lengthy discussions with
every WG whose work is commented on - where each message to a WG from
an AD could elicit several responses, each expecting a response from
the AD who is reviewing dozens of documents every week.
I can imagine a process that encourages the responsible AD and document
author (and/or chair) to go over the DISCUSS comments together and to
identify those which are noncontroversial and those which need further
discussion. The proposed resolution of the noncontroversial issues and
the list of controversial issues should certainly be presented to the
WG, and perhaps the WG should be encouraged to identify potential
compromises on the controversial issues. If the WG comes up with
reasonable compromises, that's great. But in general I don't think we
can afford to insist that such issues be resolved on the WG mailing
list in an discussion with the ADs. Just like a design team within a
WG might need to work out compromises within itself (to be ratified by
the larger group), so might the chair, authors, and ADs need to work
out compromises at that level.
I don't think that it's feasible to insist that all issues raised by
ADs be raised in Last Call, because ADs also have to make judgments
about the importance and validity of Last Call comments, and they may
even have to reconcile differences between conflicting Last Call
comments. What this means is that there are always going to be some
issues raised after Last Call - and it's not clear that having two
separate AD review phases in the process would an improvement (in my
experience, the more often I read a document, the harder it became to
notice the effects of changes to that document). But of course I do
favor ADs (and others) bringing issues to a WG's attention as early as
possible. Ideally those issues should be raised long before Last Call
time, and long before the WG thinks the design is frozen.
Keith
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf