[dnsext] Re: ASSET and NEGATIVE RRTYPE requests

[Andrew Sullivan <ajs@shinkuro.com>]

On Thu, Dec 18, 2008 at 12:35:45PM -0500, Matt Larson wrote:
> Andrew,
> 
> On Thu, 18 Dec 2008, Andrew Sullivan wrote:
> > After due consideration under the procedures of RFC 5395, the
> > request for the assignment of the ASSET RRTYPE is rejected.
> 
> On Thu, 18 Dec 2008, Andrew Sullivan wrote:
> > After due consideration under the terms of RFC 5395, the request for
> > the RRTYPE assignment NEGATIVE has been rejected.
> 
> Who were the designated Experts for these reviews?

Sorry, I likely should have covered this in my notes, but I thought
I'd mentioned it when we started those evaluations.  It's been a long
time, however, so better to repeat.  We initiated both of those while
the RFC was still 2929bis (i.e. it hadn't been finalized yet).  

The IESG appointed two experts: Roy Arends and Frederico Neves.  They
also appointed me chair of the panel of experts, which means that I'm
officially an expert too.  (You all may stop laughing now ;-)

In order to get as much experience with the procedures as possible,
and in order to check each other's work, we decided we should all
evaluate these cases, in order to see that things are working.  (We
are in fact taking the same strategy with the open reviews of requests
from Jim Reid.)  For paperwork purposes, I think this means that I'm
the designated expert, because I handled the communication with IANA.
But in fact we all reviewed them.  I anticipate in future that IANA
will pick an expert who will not be me, and I'll just be around to
ensure the smooth functioning of things.  (Depending on the volume of
requests we get, we may also go looking for additional experts for the
pool.)  
 
> While Section 3.1.1 of RFC 5395 does not require it, I think the
> Working Group would benefit from understanding the reasoning for
> rejecting these two requests.  This clarification will help future
> requesters to understand the criteria.

In the NEGATIVE case, during the namedroppers discussion period,
someone suggested a different strategy which did not depend on the
assignment.  Stuart Cheshire asked us to put the request "on hold"
while he investigated that option, but he tells me he hasn't had time
to look at it properly.  The six weeks maximum is long since past, so
for administrative reasons we have to reject the request.  A similar
request could be made in future, however, and we would not treat this
rejection as a precedent.

In the ASSET case, there were some more serious problems.  Of special
concern is that the existing I-D that defines the RRTYPE has a
fallback mechanism to TXT.  This seemed to violate RFC 5395 section
3.1.2, rule 2.  We had some questions as well about some assumptions
in the I-D and whether the documentation was complete enough.  I
asked for some clarification some time ago, but I haven't received
anything.  Partly because I've been delinquent, this has been hanging
on, but at IANA's nudging I had to conclude that we weren't going to
hear anything back, so the request has to be rejected.

Does this clear things up?

Best,

Andrew

-- 
Andrew Sullivan
ajs@shinkuro.com
Shinkuro, Inc.

--
to unsubscribe send a message to namedroppers-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/namedroppers/>