minutes of today's conference call

[Martin W Sachs/Watson/IBM <mwsachs@us.ibm.com>]

Attending:  Tony Weida, Trish Rogers, Stefano Pogliani, Nikola Stojanovic,
Dale Moberg, Scott Hinkelman, Marty Sachs.

Terminology:  I announced that we are trying to schedule a terminology BOF
on the Monday morning of the Tokyo meeting following the formal program. I
said that it is essential that we close on the terminology issues early in
the Tokyo meeting so we can start developing our CPP-CPA specification.

Stefano suggested including a discussion in Tokyo on has Business Service
Interface proposal.

I pointed out that we will need to deal with the term "Service Interface".
If TRP adopts "service interface" to denote its abstract interface to the
higher level (application layer or application services layer), we will
need a new term for the abstract partner to partner services interface in
the CPP and CPA.  Note that the term "service interface is commonly used in
protocol stacks to denote the interface between one layer and the one
above. The term "collaboration interface" was suggested for the CPP/CPA.

Discussion of the new CPP-CPA specification (formerly the IBM tpaML
proposal):

   Dale suggested starting by discussing the areas in the CPP-CPA spec that
   Marty will be flagging with "TEAM NOTEs".

   Scott suggested considering separating the CPP into components that can
   be linked together and including a "thin" option in the CPP as well as
   in the CPA.  As we discussed at the October F2F, "thin" means that a lot
   of the content in the CPP is actually in separate documents.  He
   suggested allowing a business to register its abstract services but not
   the technology definition;  the technology definition could be provided
   to an interested party by the business after authentication.  Dale
   pointed out that there is a security question in accessing the
   information at a business.  Marty observed that this is more a registry
   issue than a CPP/CPA issue.  Scott suggested providing a URI which links
   to the information at the business.  Stefano suggested using ACLs in the
   repository.  This is harder to implement but more secure and avoids the
   need to show links to information that may be withheld (avoids HTTP 4xx
   errors). Stefano cautioned that if an interested party has to follow too
   many URLs, they might just give up and take their business elsewhere.

   Stefano pointed out that it is not clear whether the choreographic
   details will be in the CPP or in some other document linked to the CPP.
   Marty suggested first working through the choreographic details and then
   deciding where the choreography belongs.  Scott observed that this
   information may well fit into the BP space because it has to mimic what
   is in the BP model.  Stefano pointed out that there is not a consensus
   on what the modeling language for choreography should be.  Stefano will
   include a summary of the recent choreography thread in his upcoming
   update of his Business Service Interface white paper.

The next TP conference call will take place after the Tokyo meeting on a
day to be decided at that meeting.

Regards,
Marty


*************************************************************************************

Martin W. Sachs
IBM T. J. Watson Research Center
P. O. B. 704
Yorktown Hts, NY 10598
914-784-7287;  IBM tie line 863-7287
Notes address:  Martin W Sachs/Watson/IBM
Internet address:  mwsachs @ us.ibm.com
*************************************************************************************