On 2007-01-08 12:03, Adrian Farrel wrote:
Brian,
The I-D tracker provides a handy button for the DISCUSSing AD
to forward the DISCUSS to parties outside the IESG - normally
by default it's the WG Chairs.
Brian, I am not suggesting that IESG has to do anything different. Let
them continue to raise their DISCUSSes through the I-D tracker. It is
the right tool for the job.
But note that the current version of the tracker does not raise the
DISCUSS with anyone. It simply logs it.
Let us at least start by having all DISCUSSes and COMMENTS automatically
forwarded to the WG chairs. These notes arrive in the tracker at
pseudo-random times and it must be a very diligent chair that spots them
all.
Indeed. And that's what I was saying - the AD registering a DISCUSS has
a button to press that does exactly that (+/- some UI details). I was under
the impression that all ADs do that, except for very unusual cases.
If we don't do this then they simply are not DISCUSSes. They are just
post-it notes.
But regardless of this, I am concerned that the resolution of a DISCUSS
is not archived anywhere. If you want to restrict the DISCUSS from
reaching the WG unless the WG chair decides, then you MUST log the
resolution (not just the fact of reslution) of each DISCUSS in the I-D
tracker.
Well, there are really only three main ways a DISCUSS gets resolved:
1. The AD withdraws it
2. The I-D gets updated
3. A Note to the RFC Editor is inserted in the tracker
(There are certainly corner cases beyond those, but those are
the large majority). These are all tracked events. What we don't
have is a comment added to the DISCUSS saying "resolved by
version -17" or whatever. Is that needed?
I'm not convinced personally
that sending the raw DISCUSS to the whole WG is the correct answer.
Sometimes it can be quickly resolved (for example if it's a technical
typo, or a simple misconception by the discussing AD).
In which case, no damage done by sending it to the WG?
A slight increase in traffic on the mailing list. So what?
I think you will find a variety of opinion on that.
And, do you assume that the WG chair is the best person to resolve these
simple issues? The chair may also suffer from the same simple
misconception resulting in the wrong thing happening to the I-D.
But the WG Chair is the PROTO shepherd and does have responsibility.
Other times,
it definitely does need WG discussion. I think it's best to leave
this in the hands of the WG chairs to decide case-by-case.
Well, assuming that the DISCUSS arrives at the WG chair as an email,
this might be a reasonable compromise. But it seems like make-work to me.
Harald said...
As Spencer has noted, a DISCUSS often passes through several
iterations from the time a concern is raised to the time it's clear
what has to be discussed with the WG. I think it would make the
IESG's work more difficult if every iteration of such DISCUSSes were
copied to the WG.
Hmmm. Would this work be more difficult because of the very large flood
of email responses that would be generated? (I don't think this would
happen with any of the WGs I follow!)
Or is there some other reason? It can't be a matter of politics or fear
of being out-spoken because the DISCUSS is public domain anyway.
If formulated as a notification at a convenient place in the
procedure, for instance "1 day after a telechat, the current status
of all ballots discussed that still have DISCUSSes get copied to the
WG", I think it would be more useful than "the WG gets copies of
every iteration".
As a completely random example, draft-ietf-ipv6-over-ppp-v2-02
currently has 2 DISCUSSes. The tracker log shows 4 entered DISCUSSes
(some revisions) and 6 COMMENTs - 10 mails to the WG mailing list
seems excessive, while 1 mail seems more likely to be seen as useful.
I completely disagree! The WG needs to know that it is completing last
call on I-Ds that are not making it through review smoothly. How else
will the WG improve its output?
Looking at this particular I-D, the I-D tracker seems to be being used
for conversations between IESG members. A bit odd, perhaps?
Well, it puts that conversation on the public record.
Brian
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf