[ebxml-cppa] Analysis of assigned issues

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

Here's my take on the assigned issues.

201, Duplicate sections for attributes/elements that reoccur

   Duplicate sections are unnecessary; it would be wise to remove any
   duplications for version 2.0 since duplications have the danger of
   getting out of sync.  A simple section cross-reference should suffice
   when an element or attribute is used in more than one place.  The
   disposition indicates "Ogden will check".  Since I am not sure what to
   look for, I recommend finding out if Peter has checked.

214, Significance of different Service Bindings in Collaboration Role

   This issue appears to be resolved in V 1.10.  There is one
   ServiceBinding under CollaborationRole.  The new text in 8.4.12
   (ThisPartyActionBinding)  covers the use of different endpoints when
   needed.

   There appears to be a typo in 8.4.5.  It states "via the ServiceBinding
   element(s) siblings within this CollaborationRole element."  Presumably,
   it should be changed to "via the ServiceBinding element within..."

221, nonRepudiationOfReceipt attribue concerns business level ack

   The text proposed in the issue description has been added in V 1.10.  I
   recommend closing this issue.

241, scope of cppid uniqueness

   We should state the scope of uniqueness of the CPPId in Version 2.0.
   Unique within the CPPs of a party is sufficient since qualifying the
   CPPId with the PartyId makes the identification of the CPP globally
   unique.

   We should require that a CPPId never be reused by the same party since
   it cannot be guaranteed that an old CPP with the same CPPId does not
   exist.

242, HTTP connection timeout

   It has been my understanding that HTTP implementations generally do not
   provide for control of the connection timeout from outside the HTTP
   implementation.  If this is true, we have no choice but to state that
   this is a local configuration matter. If, in fact, the connection
   timeout can be controlled from outside the HTTP implementation, then I
   recommend including it in the CPP and CPA and making the change for
   Version 2.0.

243, schemaLocation attribute

   Let's please separate this into these issues:

   243a, schemaLocation attribute

      I agree that since the name space URI resolves to the appropriate
      schema, we should eliminate section 8.2. This has been done in V
      1.10.

   243b, hardcoded cross references

      I see that section 8.2 has in fact been eliminated in ver 1.10, so I
      assume that Tony has fixed the symbolic references.  We should root
      out any remaining non-symbolic section references and convert them so
      that we will be free to make other changes that impinge on the cross
      references. This should be done for V 2.0.

244, PackageId for MSH level messages

   This has been added to V 1.10 so we can close the issue.

245, Use of ApplicationCertificateRef

   Indeed, v 1.10 says that ApplicationCertificateRef is for
   application-level digital signatures.  I believe that we should provide
   for application-level encryption, so I agree that the
   ApplicationCertificateRef and ApplicationSecurityDetailsRef section
   should be revised to cover application-level encryption as well as
   application-level signing. This should be done for v 2.0.

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