RE: [ietf-provreg] EPP Document Updates

[Edward Lewis <Ed.Lewis@Neustar.biz>]

Just to be set the table, I recognize the desire to make some changes 
to the specifications.  What I am saying is that I don't know that an 
IETF WG is the way to go.

I recognize that many (amorphous) organizations listen to what the 
IETF says.  Unfortunately, "listening" is not the best way to 
interact with the IETF.  "Interacting" is.  As long as the IETF is a 
volunteer organization, the IETF is hard pressed get organizations 
that listen-only to interact.  From this observation, I claim that a 
WG is not a "sure-fire" answer.

What leads to a successful IETF WG?  You need a compelling problem 
that will draw volunteers.  You need an identifiable goal with a 
clear benefit of being reached.  Even if the work is clear, such as 
"getting EPP to Draft Standard" you need to have a real motivation to 
get it done.  If there is no motivation to get it done, then the need 
for DS becomes a "reason to live" and find other little distractions 
to study alone they way.  (Look at DNSSEC to see what I mean.  Yes we 
want to sign the DNS, along the way we've been tinkering all over the 
protocol.)

Dropping into my opinion now, instead of an IETF WG, I'd like see 
more interactions in venues attuned to registration activities. 
Maybe the some venues are broken, maybe some are too new.  Perhaps 
though that something like this (EPP) will be reason to fix venues 
and to mature others.  I don't mean to name names as accusation, but 
as examples.  I've already mentioned CENTR and ICANN.  What about 
APTLD?  I believe I once heard of an African ccTLD organizational 
effort.  I don't mean to omit any organization - I don't know of many.

I will say that I do see this as bigger than just EPP.  Internet 
registration also includes DNS, IRIS, WhoIs, billing, etc., maybe 
even the issuance of certificates.  Some of what registries 
"requires" coordination - such as DNS, EPP, IRIS.  Some of what 
registries do can be shared, and some is best left to competitive 
practices.  Within the IETF, the only coordination of these issues is 
through the informal mutual attendance pattern - there is a lot of 
overlap in the CRISP (IRIS), DNSEXT, DNSOP, ENUM, etc. groups even 
though they are in different IETF Areas.  For instance, questions 
about whether EPP ought to get extensions for routing came up in 
CRISP, and DNSSEC extensions for EPP (RFC 4TBD) was reviewed in DNSOP.

Okay, I've strayed far away from the small edits we have been talking 
about.  What you are seeing is about the third attempt at a reply, 
each time I do from the problems at hand into the crevices I see 
needing epoxy.  To close this off now - we need to keep in mind, what 
are short term needs and what might we need in the coming years.  EPP 
lives in a context of a few protocols, yet it is its own beast.

-- 
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Edward Lewis                                                +1-571-434-5468
NeuStar

If you knew what I was thinking, you'd understand what I was saying.