Re: [dnsext] DNSEXT closing down soon

[Andrew Sullivan <[e-mail]>]

Dear colleagues,

On Sat, Dec 03, 2011 at 01:19:48PM -0500, Edward Lewis wrote:
> However, the WG really hasn't
> shown much in the way of progress, even when viewed over the long
> haul.  DNSEXT was formed in about 1999, with the conclusion of two
> other DNS related WGs.  In the 12 year run how many documents made
> it to Full Standard or even Draft Standard?  (Not very many.)

> If there's work to do, it's not engineering, it's document clean up.
> There seems little appetite for that, which bodes poorly for a
> working group, especially one consisting of volunteers.

Speaking only personally, I think the above two observations by Ed are
the key: the WG has not worked very hard to complete documents that
are in train, and the main job at this point is document clean-up.  We
could go further and note that the last big push in the WG was to get
DNSSEC -- a real extension -- done.  It would not be hard to argue
that the WG should properly have been wound down then, and we didn't
do it.

With my co-chair hat on but without having discussed this with Olafur,
I will say that, when asked by our AD about this in the recent past,
it has been my view that the WG does not want to tackle those document
clean-up matters, and that if we want a WG that has as its main task
something for the DNS that was roughly like YAM, we should charter a
new WG for that.  Chartering such a WG would be an opportunity to
guage whether there is real interest in doing the work (for, to begin
with, we'd need some Internet Drafts to discuss during a chartering
discussion).

DNSEXT actually made an effort to undertake this documentation work,
and it didn't go anywhere: volunteers just didn't want to work on that
topic.  People all seemed to agree that it was important _someone_ do
it, but none of them had the time, the expertise, or both.

>  Will the mail list continue?

I see no reason for it not to.
 
>  If so, how will current non-subscribers learn of the list?

Apart from the IETF list of such lists, I suppose informally.

>  Who will be the focal point for questions from other WGs about DNS?

No hat, personal opinion: I am not sure DNSEXT is any more anyway.
There is a basic problem that DNSEXT doesn't do review: not of its own
documents, and not of others' documents either.  DNS64 got hardly any
review by people who really knew the DNS, even though some of us asked
repeatedly for such reviews.  A review came from the DNS Directorate
at the very last minute; it uncovered several serious issues with the
document.  MIF recently did a WGLC that was echoed in DNSEXT, and
while there was a lot of discussion on the list, there were not that
many complete reviews that I saw come from here.  There was a great
deal of discussion about whether MIF's starting premise was one we
wanted to accept, but that did not seem to me to extend so far as to
reading the document and suggesting text changes that would say, "This
is a bad idea, but if you are determined to do it. . . "  

Moreover, if we think that there are issues in other WGs that need DNS
clue, we can do that by participating in those other WGs.  
 
>  Are all instructions to IANA complete and understood?

Hat back on.  No.  One of the major problems we in fact have has to do
with RRTYPE code point assignment, which is (from IANA's point of
view) broken, and which also does not happen according to the
published process.  I believe that needs to be fixed.  I would like to
say I think we can do it by the spring, but history does not make me
optimistic.

Best regards,

A

-- 
Andrew Sullivan
[e-mail]