Re: [Moderator Action] RE: Straw poll: what options did you review?

As a preface, I really appreciate Larry taking the time to suggest
ways to speed up the protocol review process.  But (I'm sure you
knew there was a "but" coming up soon :-), we need to make sure
that the proverbial baby doesn't get flung out with the proverbial
bath water (:-).  

   From: "Larry Masinter" <LMM@acm.org>

   The criteria for "Proposed Standard" in IETF is described in RFC 2026:

      A Proposed Standard specification is generally stable, has
      resolved known design choices, is believed to be
      well-understood, has received significant community review, and
      appears to enjoy enough community interest to be considered
      valuable.

   I think "significant community review" means "a large number of
   working group participants have reviewed the whole document". I
   think "well-understood" means "most working group members could
   explain what it does" and "enough community interest to be
   considered valuable" means "there are lots of people who plan to
   implement this".

I think that the criteria are quite clear as stated in RF 2026, and
that your interpretation significantly raises the bar beyond what is
stated there.

   I think it's a fairly creative interpretation to translate "significant
   community review" into "at least two people have read each section",
   and not really consistant with my experience of how the IETF works.

I agree that would be a creative interpretation, so I'm sure glad it's
nothing that I ever said (:-).  What I did say was:

"My criteria for sufficient support is that there are two different
implementors planning on using it."

In particular, I was talking about "support" for the current
protocol by the working group, not the level of community review,
and I was talking about plans to use it for an implementation,
not just having read it.

This criteria is derived from the IETF definition of what is required
to transition from proposed standard to draft standard:

 "Draft Standard
 A specification from which at least two independent and interoperable
 implementations from different code bases have been developed, and
 for which sufficient successful operational experience has been
 obtained, may be elevated to the "Draft Standard" level. "

   If you split the document into "core" and "non-core", it's pretty clear
   that the "core" document has been more widely reviewed than "non-core";
   it might be able to make progress as a Proposed Standard because it's
   aimed at a narrower community.

Yes, but the majority of the implementors are planning on implementing
more than core, and therefore we will not have a basis for
interoperation until we have stable Proposed Standard against which we
can code.

   I'll also point out that it seems that progress through area directors,
   IESG review and the RFC editor seem to be proportional to the square
   of the length of the document rather than linear; shorter documents
   get on the queue more quickly because (just as seems to have happened
   with working group participants) reviewers have to find a block of time
   where they can sit down and review the whole thing.

I share your desire to maximize the speed of the review process, but
it doesn't help us to get something quickly through the review process
if it is not something that provides us with the basis for
interoperable implementations.

Cheers,
Geoff

Received on Thursday, 8 February 2001 01:37:12 UTC