[Bug 139] New: WS-ctx conformance - implicit/explict, particular or general

[bugzilla-daemon@arjuna.com]

http://services.arjuna.com/wscaf-issues/show_bug.cgi?id=139

           Summary: WS-ctx conformance - implicit/explict, particular or
                    general
           Product: WS-Context
           Version: 1.0
          Platform: PC
        OS/Version: Windows 2000
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Implementation and interoperability
        AssignedTo: ws-caf@lists.oasis-open.org
        ReportedBy: peter.furniss@choreology.com
         QAContact: mark.little@arjuna.com


There is currently (0.3) no statement of what is expected and required of a "WS-
Context implementation". Whether or not there should be an explicit statement, 
there is clearly a distinction to be made between an implmentation that uses ws-
context exclusively for some set of referencing specifications (e.g. ws-cf + ws-
txm) and one that is intended to be usable by any referencing specification 
(including the example ones that have been suggested on various occasions)

In an implementation of ws-cf or a particular txm protocol, say, the use of WS-
Context might be no more than a particular choice of uri for the namespace 
where the protocol element came from the ws-context schema. Depending on the 
requirements made by referencing specification, there need be no support for 
most of the fields of the context itself, no facility for pass-by-reference or 
dereferencing, and the child-context mechanism might not be used as such.

On the other hand, an implementation that was intended to offer WS-Context "as 
such" would need to provide support for all of the features and mechanisms that 
are open to a ws-context-using specification. Although many fields in the 
context are optional, a general ws-context implementation still needs to 
support them, and have some means (which could be non-interoperable hooks, in 
some cases) of letting implementations of referencing specifications use them.

( An alternative approach would be to say the ws-context implementations are 
always only in relation to a particular referencing specification or set of 
such, and there is no concept of a general ws-context implementation. Howwever, 
this would essentially be to say that ws-context defines a pattern of use - a 
menu of a la carte features - and has little reason to exist as a separate 
specification. )

The specification should make clear that an implementation tied to a specific 
referencing specification is not required to support features of ws-context 
that are not used by that referencing specification (though it can do).



------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.