ietf
[Top] [All Lists]

Re: Purpose of IESG Review

2013-04-11 17:08:18
l(_dot_)wood(_at_)surrey(_dot_)ac(_dot_)uk 
<l(_dot_)wood(_at_)surrey(_dot_)ac(_dot_)uk> wrote:

+1 to Joe's comment.

Example: the existence of the extensibility bit in multipath tcp,
which i understand came out of a review by the iesg member responsible
for security.

   I assume you're talking RFC 6824. I recommend reading the Narrative
Minutes of September 13th:

http://www.ietf.org/iesg/minutes/2012/narrative-minutes-2012-09-13.html

There were DISCUSSes from Stephen Farrell (Security), Barry Leiba
(Applications), Robert Sparks (RAI), and Sean Turner (Security).
Stephen Farrell did complain about a negotiation scheme that only allowed
seven security algorithms; and asked "how you could practically extend
this design for stronger cryptographic security".

In that context, that would be outside the scope of any security
review,

   Hardly! Those are exactly what I would hope for in a Security review.

and the comments weren't raised in a personal capacity years earlier
on the relevant mailing list.

   I'm not going to research that; but it seems hardly relevant...

Sure, getting past iesg only cost multipath tcp a bit.

   (which, BTW, I strongly endorse!)

But iesg members exceeding their bounds as reviewers and leaving a
personal mark seems commonplace.

   Perception is easily mistaken for reality. :^(

   But if you look at the datatracker history:

http://datatracker.ietf.org/doc/draft-ietf-mptcp-multiaddressed/history/

you'll see that all four DISCUSSes were cleared by October 22.

   Considering that this document started life in June of 2010, and
was a major enhancement of TCP, 40 days doesn't seem excessive, IMHO.

iesg members are there for expertise in their area and to provide that
expertise in focused reviews,

   Note that there's really a lot of overlap between areas: so "focused"
may not be the right criterion.

not to block until a protocol is redesigned to suit their personal
tastes.

   I am told this used to happen. I have not experienced it in the five
years I have been scribing.

====

   I really don't know how to change the perception -- but I strongly
recommend referring to the Narrative Minutes. Hopefully that history
will be preserved "forever".

--
John Leslie <john(_at_)jlc(_dot_)net>

<Prev in Thread] Current Thread [Next in Thread>