While I disagree with the "WG Draft" part, partially because I
think we do still derive significant value from cross-area
review, I agree with the rest of this.
FWIW, I am actually quite sympathetic to Ned's argument that the
presence of two levels past proposed and the amount of nonsense
that seems necessary with each is demoralizing and acts as an
obstruction for some people to get to the second level. It
hasn't persuaded me that this proposal is helpful only because,
if the experience with Proposed is extended from a three levels
to two, we will see the nonsense level at the second level
increase on the basis of "last shot, so we better get it right"
reasoning (which I believe is one of the things that has driven
the threshold for Proposed to its current level). If we could
somehow take the combination of a quest for perfection and the
notion that, if the IESG has to review something, ADs have to
find something to complain about or some way to impose their
stamp on the document out of the system, then I'd completely
believe Ned's argument... except that, with the nonsense removed
from the 2nd and 3rd levels, that incentive to remove the third
might be much reduced.
As I indicated in my note last week, the way out of these
problems seems to me to be is much less symbolic tuning of the
standards process and much more having the community express
clear support for the IESG changing the way things are done
--within the existing rules or even closer to their intent-- and
having the IESG do it.
--On Tuesday, September 06, 2011 14:35 -0700 Ted Hardie
I actually have a lot of sympathy with Andrew's formulation,
largely because the document wants you to infer something
rather than making it explicit. Take this text:
2.1. The First Maturity Level: Proposed Standard
The stated requirements for Proposed Standard are not
changed; they remain exactly as specified in RFC 2026 .
No new requirements are introduced; no existing published
requirements are relaxed.
The document doesn't actually say out loud there that the
requirements for Proposed Standard have been considerably
increased by IESG practice over the years, nor does it charge
subsequent IESGs to return to a faithful reading of the actual
text. You can infer it from the refrain ("stated
requirements" and "published requirements"), but I don't think
you could fairly call it explicit. You can certainly get that
from Russ in bar or on a mailing list, but we normally try to
write our documents such that you don't have to have shared a
bar with the author to get their real intent.
If the IESG does not choose to follow-up that inference with
action, we have effectively moved from a one step standards
process pretending to be a three-step standards process to a
one step standards process pretending to have two. That's
hardly worth the electrons which have been spent in this
Ietf mailing list