Re: [weirds] requirements/success was - Re: I-D Action: draft-ietf-...

[Edward Lewis <[e-mail]>]

On Dec 3, 2012, at 13:13, SM wrote:

> At 07:44 03-12-2012, Edward Lewis wrote:
>> There isn't a distinct "names community" that is separate from a "numbers community."  There are about a dozen organizations that do both (i.e., the NIRs and IANA, hence ICANN).
> 
> How many individuals affiliated with IANA have reviewed the drafts?
> 
> How many individuals affiliated with ICANN have reviewed the drafts?
> 
> How many individuals affiliated with NIRs have reviewed the drafts?
> 

I have tried to get input from NIRs, but my attempt to contact one particular person hasn't been successful.

There is a rat hole discussion that could happen here.  First, the IETF shies away from recognizing people having affiliations.  The IETF doesn't have members, no voting mechanism before approving a document.  The IETF relies on consensus being volunteered from participants.  That leaves open the possibility that the result can be highly biased "the wrong way" - in the sense that the entirety of the problem was underestimated or represented just a fraction of the operator communitys' concerns.  The best the IETF can do is try to beat the bushes to get as much input as it can, not sit back and expect volunteers to come calling.

A lesson from the PROVREG WG and EPP.  EPP was first published around 2003 and eventually elevated to full standard by 2009.  The elevation followed the full IETF process.  However, the operator community, which pretty much came of age around 2004-2005 (estimates vary) was unaware of the IETF's mailing list for EPP.  So in 2009 there was a lot of skepticism in the operator community over the elevation of EPP to full standard when it seemed that the protocol had so many flaws.

Between October 2009 and May 2010 a few things happened (the bookend dates here correspond to CENTR Tech meetings).  The operator community and the PROVREG mail list were introduced to each other.  It seems that while those on the PROVREG list were aware of the state of the art of registration, those who came on later never heard of the list (because there was no IETF WG to gateway them in).  After this happened, the angst over the protocol subsided and issues were worked out ceasing the desire to replace EPP with something different.  I'm keeping this short - the point is that there really needs to be an effort from the IETF outwards to make sure that what gets into IETF documents is relevant, and not rely on reviews trickling in on time.

And, returning to where my comments began, the operator community in this case is not so neatly divided between names and numbers.  Perhaps in some sense is seems that way because the number of numbers people on the list is disproportionately higher than names people (if you are to count the number of RIRs as 5 registries and the domain names as having around 300).  NIRs though...they expand the RIR's 5 number registries by about double that.

I dunno, this is a rat hole.  I expect as much though as we are talking about splitting input into "names" and "numbers" - attaching affiliations when the process does not recognize that.

>> There seems to be an assumption that "just because the IETF is working on it, people are
> 
> The IETF is not working on it.  The editors of the drafts are working on it.  People who have posted reviews to this mailing list are working on it.


In the sense that the IETF is anyone participating on a mail list or document or face to face meeting, yes, the "IETF is working on it."  I've seen other organizations refer to things the "IETF is working on" in that vein.

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Edward Lewis             
NeuStar                    You can leave a voice message at [phone]

There are no answers - just tradeoffs, decisions, and responses.