On 01/05/13 21:05, Brian E Carpenter wrote:
On 02/05/2013 05:59, Dave Crocker wrote:
The blog nicely classes the problem as being too heavy-weight during
final stages. The quick discussion thread seems focused on adding a
moment at which the draft specification is considered 'baked'.
I think that's still too late.
What, you agree with your younger self?
http://tools.ietf.org/html/draft-carpenter-icar-sirs-01
Apart from the non-diverse acronym, I still think that proposal
was a good one.
Brian
Rereading the background document that lead to the SIRS proposal is also
worthwhile:
http://tools.ietf.org/html/draft-ietf-problem-issue-statement-00
I think the views in sections A.2.6. to A.2.9 are still pretty much
applicable; judging from recent comments some people still think the
situation with the IESG is a 'battle of wills'.
I guess one way of summarising this whole problem might be:
We set on a group of people (a WG) who by definition are supposed to
work in 'common mode' on a particular problem in what is mostly a closed
bubble. It's perhaps not entirely surprising that what comes out at IETF
Last Call time has more than its fair share of 'common mode failures',
symptoms of groupthink and blinkered views of the overall requirements
for effective operation in the wider Internet.
The comments in the 'problem' draft and the SIRS proposal reflect an
attempt to suggest how the closed bubble can be kept in contact with the
wider world. However, getting the relationship between outside
reviewers and the WG teams right would be non-trivial - not to mention a
significant organisational effort and a load on the reviewers. To be
honest, AFAIK we have never attempted to quantify how much extra effort
(if any) a SIRS-like exercise would require as compared with the
existing directorate reviews.
Personally, I don't believe that we should attempt to categorise some
I-Ds as more significant than others. This requires 20-20 hindsight
with which we are certainly not equipped.
/Elwyn
BTW
Re the diversity discussion: take a look at the editorial panel on the
problem draft - ok, its only along one of the dimensions.
Jari might also remember a contribution he made to 'problem':
o WG process is too slow, because of feeping creaturism, deadlocked
conflicts or lack of worker bandwidth (Jari Arkko)
I can certainly agree on the feeping creaturism!
/E