ietf
[Top] [All Lists]

Re: The Non-IETF Informational RFC Publication Fiction

2000-06-27 11:30:01
In the case of RFC-2188, the RFC Editor did *nothing* and just waited
for the IESG for more than 7 months. That is well documented.

A bit of explanation is in order.

The IESG takes its review responsibilities seriously, and cannot
always in good conscience approve the publication of a document.

In such cases the IESG tries to give feedback to the author as to 
what changes to the specification might cause it to (in the opinion
of IESG) merit publication.  Sometimes providing this feedback is 
quite difficult, especially when the specification is poorly written,
or when the author lacks knowledge about how to design a protocol 
that works well over the Internet.  (Lack of understanding about
security - including DoS attacks, and congestion avoidance, are 
common problems)

In most cases the author and IESG work out a compromise that allows
the document to be published.  But when the author stubbornly insists 
that he has the *right* to have his document published without significant 
changes, it becomes nearly impossible for the author and IESG to compromise. 

In this case IESG, knowing that the RFC Editor might choose to publish
the document over its objections, attempts to write an IESG Note that
will at least explain to readers what IESG thinks is wrong with the 
document.  This can also be quite difficult and time consuming, because
it's not always possible to rectify major problems with a brief note.

(Note that in the case of at least one of Mr. Banan's documents, the 
RFC Editor further negotated with IESG to soften the wording of the 
original IESG Note.)

The problem is not that it took 7 months for the IESG to produce a
note about one of Mr. Banan's documents, but that the review of Mr. Banan's 
documents consumed precious IESG resources that would have been better 
applied to documents (both individual submissions and working group documents)
which were more deserving of publication.  This resulted in significant 
delays in publication of the more deserving documents.

If I were to suggest any change in the RFC review and publication process,
it would be to give IESG the power to say "no" to publication of individual
submissions.  (perhaps with the possibility of formal appeal to IAB)
I do not believe that IESG would do this capriciously, and I believe 
that this would make better use of IESG resources.  And such a change would 
make it quite clear that individuals authors do NOT have the right to have 
their documents published as RFCs.

Keith (no longer on IESG, and not speaking for them)