[Ietf-not43] any more comments for the requirements

[Edward Lewis]

I think you need a much more specifically worded requirement.  EPP 
(as an element of PROVREG) is chartered to handle more than domain 
names - although all of the work to date has been towards that goal. 
Also, EPP has yet to appear in RFC's and when it will it will be a 
proposed standard and hence subject to more revision.  And finally, 
it is important to make the distinction between EPP and CRISP plain 
to see, the two bodies of work are trying to solve issues for 
different sets of people.

EPP is designed to be the interface between registrars and 
registries, a kind of "in office" communication device.  Besides (the 
anomaly that is) the check command, EPP will see a net data flow from 
client to server. If I recall correctly, CRISP is exposed to the open 
Internet, with a net flow from the server to the clients.  I think 
that that is just the beginning of the differences.

EPP is currently before the IESG.  Besides the non-determinism of the 
passage of the set of documents, some implementation experience is 
being gained that will need to be included in the next round of 
documents.  It would be irresponsible to try and guess when other 
protocol specifications should be able to rely on EPP's definition.

Finally, at the chartering there was a direction to expand the early 
EPP proposal to cover more than domain names.  Up to now, the 
expertise to do so hasn't been participating in the process (it's not 
like we asked and they snubbed us) so all you see in the documents 
are domain names.  This may change in the future, so CRISP shouldn't 
expect to represent *every*thing that's in EPP.

At 12:04PM -0700 9/23/02, Rick Wesson wrote:
>andy,
>
>sure that sounds like a reasonable requirement.
>
>-rick
>
>On Mon, 23 Sep 2002, Andrew Newton wrote:
>
>>  So, something like:
>>
>>  Domains in this service MUST be capable of representing attributes and
>>  values of the domain objects in EPP.  Contacts in this service MUST be
>>  capable of representing attributes and values of the contact objects in
>>  EPP.  The status values indicating the state of a domain MUST be an
>>  equivalent set or superset to the status values indicating the state of
>>  a domain in EPP.
>>
>>  Perhaps a little more massaged, but is that towards what you are after?

-- 
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Edward Lewis                                          +1-703-227-9854
ARIN Research Engineer