--On Wednesday, August 31, 2011 11:08 -0400 Keith Moore
<moore(_at_)network-heretics(_dot_)com> wrote:
On Aug 31, 2011, at 10:42 AM, John C Klensin wrote:
We ought to, IMO, be permitting
publication of PS documents at the second level as long as
there are no _obvious_ ambiguities that cannot be figured out
(the same way) by people of good will acting in good faith
and with help from WG lists and as long as there are no other
"known [technical] defects".
That would be fine with me if we could somehow effectively
discourage deployment of proposed standards in products. But
that's a lost cause, IMO.
Keith,
IMO, there are two possibilities here. At this point, sadly,
both involve a chicken-and-egg problem. Such is life.
(1) We proceed as if Proposed Standards are what 2026 (and the
earlier culture) claims they are and work on ways to reinforce
that notion in the community. If that is our goal, than getting
documents out sooner and having them look a bit rough as well as
being a bit rough is A Good Thing and reinforces the "deploying
at PS, much less at I-D, is taking a big risk". The idea isn't
foreign to the industry: to take a handy example, we saw
products identified as 801.11-draft-N floating around for years.
Every vendor who shipped one (and their more sophisticated
customers) knew that there could be serious incompatibilities
between their approximations to the drafts and the real 802.11n.
(2) We accept, and effectively encourage, deployment of proposed
standards in products, either because it is a lost cause or
because we think it is a good idea. As part of that, we move
further down the path of ensuring that PS documents are complete
and polished (my group (1)) and of guaranteeing that there will
be no significant changes in the future (either in going to DS
or in grade). If that is really our position, then we better
stop grumbling about how long it takes to get PS documents out
(and the partially-consequential deployment of technologies from
I-Ds), accept the fact that the SDOs who used to be very
concerned about how much faster the IETF was than they were have
now won and become faster than us (in large measure because we
have slowed down so much). And we should stop wasting time
trying to figure out how many maturity levels we need or what
those criteria should be because the answer is "one level is
what we get and pretending anything else, e.g., by trying to
fine-tune procedures, is an embarrassing public act of
self-delusion" (stronger language occurs to me but would
probably get the message killed by spam filters).
Remember that, while ignoring procedures and category
definitions that we don't follow is not desirable, "fixing" them
to reflect a model that doesn't (and won't) exist either is a
public demonstration that we are disconnected from reality. I'd
much rather leave that distinction to some other SDOs than join
them. YMMD.
john
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf