[Bug 5656] Defining interoperability for user-defined reference schemes

http://www.w3.org/Bugs/Public/show_bug.cgi?id=5656

           Summary: Defining interoperability for user-defined reference
                    schemes
           Product: SML
           Version: LC
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Interchange Format
        AssignedTo: cmsmcq@w3.org
        ReportedBy: kirk.wilson@ca.com
         QAContact: public-sml@w3.org


SML defines three conditions for defining reference schemes (section 4.5). 
These conditions constitute an extension mechanism for defining user-defined
reference schemes.  SML itself uses these conditions to define the SML URI
Reference Scheme. SML-IF adds substantially to the definition of the SMI URI
Reference Scheme in section 5.3.4 regarding how SML URI reference scheme
instance are processed in SML-IF documents.  However, it is never discussed how
user-defined reference scheme are supported, or what needs to be done to
support these schemes, in SML-IF.  The text merely states, "Resolution of URI
references in category #3 is defined in their respective scheme definitions." 
but nowhere is explicitly stated as a requirement of user-defined reference
schemes.

In particular, SML-IF needs to recognize that to address the interoperability
of user-defined references there needs to be discussions of:
1.  An algorithm for generating URI aliases from the instances of the reference
scheme so that targeted documents can be identified within the SML-IF document.
2. The need to define the process by which a user-defined reference schema
instance is resolved in terms of the target document in SML-IF (i.e., there
needs to be explicit recognition of the fact that a user-defined reference
scheme that aims for interoperability must spell out something like what
section 5.3.4 does for the SML URI reference scheme.

(These issues have emerged with the writing of the "EPR Note"--I wish these
issues would have been more explicitly spelled out in SML-IF.)

Received on Tuesday, 22 April 2008 20:08:26 UTC