On 1/21/2016 11:18 AM, Shumon Huque wrote:
Lastly, new protocols take a long time to get deployed. Look at IPv6
- I'm speaking from experience, having first deployed it in
production in 2002. And it's still largely undeployed. Shutting down
the DANE working group while the protocol is still in its infancy,
and while there is still potential work in the queue, sends the wrong
message in my opinion.

First... +100 to what Shumon said!  We are at the stage of DANE deployment 
where, while the protocol may be well-developed, the deployment is only now 
beginning... and I think it would send a wrong message to kill of the group now.

But to answer Dave...

On Feb 4, 2016, at 11:06 AM, Dave Crocker 
<[email protected]<mailto:[email protected]>> wrote:

If there is work to do, then where is the draft charter text describing the 
problem and nature of work to be done, timeline for completion, indication of 
who is asking for the work, indication of who will do the work, and indication 
of who will adopt (use) the work?

Good point.  Perhaps the action here is to update the DANE WG Charter with the 
deliverables such as the ones Shumon indicates.   All of the milestones on the 
existing charter ( https://datatracker.ietf.org/wg/dane/charter/ ) have been 
completed and so by the process it would make sense to declare victory and shut 
the group down.

In terms of overseeing adoption of an IETF specification, that's not the job of 
the IETF.  That's the job of whatever Internet constituency(ies) want to see 
the adoption.  The IETF work is a segment of an industry process.  
Simplistically, it divides into :  1) formulate the needs, 2) design a 
solution, 3) deploy the solution.  The IETF is (only) step #2.

Yes, BUT...  the reality is that during the deployment (step 3) of the solution 
there are very frequently issues discovered that were not anticipated in the 
original development of the protocol.  Changes that need to be made to the 
protocol are identified.  Additional uses cases are often found.  Ways to 
extend a protocol are discovered. New guidance is developed.  All of those are 
pieces of information that can be fed back into the IETF process to wind up 
with a better solution (step #2).

We've in fact *already* seen this within DANE with some of the work that Viktor 
and Wes have fed back into the group based on their work with DANE in SMTP.

The question is - if the DANE working group shuts down, where does that 
feedback go so that DANE could be improved?

Right now, I'm seeing a good bit of interest in DANE out in the wider industry. 
 I'd like to make the process for incorporating that feedback during deployment 
as efficient and fast as possible.  To me, keeping the DANE WG around for a bit 
more would be one way to make sure that the feedback loop is out there.

Alternatively, for IPv6, there is the V6OPS WG where these kind of 
operations/deployment issues can be brought.  For "DNS" in general there is 
DNSOP.  Could DANE issues be brought there?  I guess so if the charter were 
updated... but that then throws even more into an already packed WG.

Or... we could close the DANE WG and wait around until there were enough issues 
to merit opening up a new DANE-related WG (similar to how EPPEXT was eventually 
formed after PROVREG shut down).

I agree that "waiting for feedback from deployment" is NOT a reason to keep a 
WG on the books ALONE, but given points by Shumon and others I *do* think there 
is enough DANE-related work to keep the WG going... which then also can be that 
potential feedback mechanism.

My 2 cents,
Dan
--
Dan York
Senior Content Strategist, Internet Society
[email protected]<mailto:[email protected]>   +1-802-735-1624
Jabber: [email protected]<mailto:[email protected]>
Skype: danyork   http://twitter.com/danyork

http://www.internetsociety.org/




_______________________________________________
dane mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dane

Reply via email to