It sounds very useful. But would such a session token provide functionality not already provided in other specifications such as WS-CAF (or more specifically WS-Context)?
Jeff Bohren
BMC
-----Original Message-----
From: Gary P Cole [mailto:]
Sent: Mon 6/27/2005 1:54 PM
To: Bohren, Jeffrey
Cc: PSTC
Subject: Session context token?
Jeff,
Kent Spaulding, the developer working on Sun's SPML2 implementation for
the Interop, points out that our previous provider implementation
returns an opaque token (in our case, a "sessionID") that represents the
logical session (or perhaps session context). If the requestor passes
this token as part of subsequent requests, it saves a great deal of
processing on the provider's side. It may also assist with audit
logging, which you've indicated you expect will become important.
I believe that this feature (i.e., an opaque and optional session
context element returned by the provider and passed by the requestor as
part of subsequent requests) may be important enough and general enough
to recognize as part of the specification. I'd like to know what other
members and participants think.
Would you be willing to consider this issue?
