DRINKS WG Minutes for the 77th IETF Meeting 
==========================================


WEDNESDAY, March 24, 2010
0900-1130  Morning Session I+II, Room "Palos Verdes" 
Data for Reachability of Inter/tra-NetworK SIP (drinks)

Chairs:
     Richard Shockey <richard@shockey.us>
     Alexander Mayrhofer <alexander.mayrhofer@enum.at>

 Real-time Applications and Infrastructure Area Directors:
     Robert Sparks <rjsparks@nostrum.com>
     Cullen Jennings <fluffy@cisco.com>

 Real-time Applications and Infrastructure Area Advisor:
     Cullen Jennings <fluffy@cisco.com>

 Mailing Lists:
     General Discussion: drinks@ietf.org
     To Subscribe:       https://www.ietf.org/mailman/listinfo/drinks
     Archive:            http://www.ietf.org/mail-archive/web/drinks/current/maillist.html

MinutesTaker 
       Richard Shockey <richard@shockey.us.
Jabber Scribe
       Adam Uzelac [adam.uzelac@gmail.com]

1. Agenda bashing..

Approval of Agenda with addition of a second discussion of use cases by David Schwartz

2. Discussion of Interim meeting possibly in May on the East Coast of the US

3. Alex Mayrhofer: Discussion of current milestones and revision timetable 

Gonzalo acting as incoming Area director indicated approval for an aggressive time table for completion

4. Discussion :  "DRINKS Use cases and Protocol Requirements" (Sumanth Channabasappa)   http://tools.ietf.org/wg/drinks/draft-ietf-drinks-usecases-requirements/

Gonzallo:  What is Session Establishment Data its not clear what SED really is?

Jean Francois Mule:   Drinks terminology comes from the Speermint terminology drafts.

Cullen Jennings:  The confusion over terminology still exists.

Jon Peterson : The intent was that only the data to enable ENUM was within scope.

AlexMayrhofer (Chair) : Is a Informational document necessary

Jean Franswoir Mule we should describe on the list what the SED data is and a informational document is not necessary

David Schwartz: Much more discussion about SED is on going in the context of source base routing.

Richard Shockey(Chair):  A simple reminder that  DRINKS Is the provisioning not the query mechanism.

5. Discussion David Schwartz discussion of additional use cases slides on authority to add modify or delete a public identity. ( see slides )
 
Jean Francois Mule:   Provisioning protocols should not contain business rules in the use case where you have to define the authority for a public identity.
  
Ken Cartwright:   This is continuing to create confusion.  In defining what authority is publishing a change to the public identity.

Manjul Maharishi: The regulatory body is the defining body that defines the target on how the session is routed.

Alex Mayrhofer: (Chair): The provisioning mechanism should keep business rule decisions out of the data model.

Siad Ali: These issues are implementation based 

Jean Francois Mule:   Term Validating Entity is the better term for regulatory body. View of carrier of record is a North American centric view we don’t not want to scope creep… here 

Ken Cartwright: This adds levels of complexity into the operations that we don’t need at this stage.

Richard Shockey (Chair): We are trying to solve one problem first. 

David Schwartz: LUF vs LRF based is the target a URN  URI binary data or the route.

Jean Francois Mule: We need more check and balance of the requirements.

Ken Cartwright:  So there is still confusion on what is and what is not the LUF.

6. Discussion of the protocol document.<slides> "Session Peering Provisioning Protocol" (Jean-Francois Mule)
      http://tools.ietf.org/id/draft-mule-drinks-proto-02.txt

Discussion of changes in draft 02

Jon Peterson : Is the target domain and the terminology on are the distinctions between target domain and URI artificial? 
 
Ken Cartwright: Is the target domain part of the route groups or the other way around?

David Schwartz. We have had problems on what is the LUF.
  
Ken Cartwright:  do we know what the target domain actually IS?

Jean Francois Mule:  3263 we know what the target domain is.

Ken Cartwright… NO use case for a text record..???

AlexMayrhofer (chair) : there is a hole in the SPEERMINT documents etc.

Jon Peterson: URI have the ability to LUF + LEF  what about the distinction in the services apply all use the NAPTR model in the LUF vs the LRF

Hadriel Kaplan:  What was the model for this?

Jean Francois Mule: My preferred offering is on entry point to the network.

Jean Francois Mule:  Move the priority of a route?  The question is clarity on a more generic data object vs one that is NAPTR specific.

David Schwartz: How do you express opaque strings in URN here? ( SPID)

Jean Francois Mule: We need clarity on how those identifiers are expressed

Richard Shockey.  Lets take this to the list.

Jon Peterson:   We should offer service providers the tools they need to route service. Is LUF is Target Domain.?

Hadriel Kaplan Yes LUF is target domain.
 
Jean Francois Mule:   IF you want to FQDN use URI if you want to to use SPID a URN define SIPD’s as URN? The domain portion is where you do the SRV look up.
 
Jon Peterson:  URI are the way to go since this is essentially about ENUM provisioning. "target domain" 
does not neccessarily mean "fqdn", could be "administrative domain" as well.

Jean Francois Mule: Are we on the wrong track here.?

David Schwartz: Flexibility here is the key?
7. Chairs HUMM ON the adoption of the protocol document as a WG item .. Substantial HUMM in favor no one opposed. Will confirm the decision on the list.

8. Discussion Ken ssp over soap "SPPP Over SOAP and HTTP" (Ken Cartwright) http://tools.ietf.org/id/draft-cartwright-drinks-sppp-over-soap-01.txt

Cullen Jennings : SOAP has good mechanism for firewall transversal.   Authentication ?  The approach in this draft not really not soap etc.

Ken Cartwright : The generic WSDL approach was used in NET Conf.
 
Hennes  :  This idea was discussed in keyproff ( key provisioning) and there a presentation on this idea 

Ken Cartwright:  The rationale for the use of SOAP was to encourage acceptability in the industry. This was he driving factor for using SOAP at this point.

9. Chairs HUMM on the adoption of SPPP over SOAP as a working group item. Result: More in favor than opposed will take to the list for confirmation.