A few questions on reading the F2F minutes and on items that did not come
up in discussion at the F2F. . Several of these are related to the Cloud
Model and how we address certain things in it (e.g., we dismiss Email
Services as no longer being considered in the overall Semantic Model)
1. Discussion advanced from considering whether to proceed with PPDMAP
to indicating that PPDMAP should advance to Stable level and proceed with
last calls. However, I am unable to find any draft of PPDMAP on the FTP
site.
2. In previous discussions we concluded that operations are directed
to specific services and therefore do not need the service identification in
their name. However, the semantic model structures operations by service
(ServicesOperations => CopyServiceOperations, FaxOutServiceOperations, etc).
This was the reason for the MFD Model specification to use the
<service>ServiceOperation nomenclature. Since the objective is to stress a
consistent set of operations across the different job-oriented imaging
services (with a few omissions), would it be preferable to have a general
ImagingServiceOperations element with specific services referencing back to
it?
3. With the schema on the present form, I think we would need to
continue with the <service>ServiceOperations nomenclature. Ii would then
appear that the Cloud Proxy operations should similarly include a <service>
term. Indeed, the current schema has the proxy operations (for Print only)
as a subset under PrintServiceOperations (e.g., AcknowledgePrintJob). It is
not obvious from the schema that these are separate operations originated by
a "Cloud Proxy" .
4. Which brings up the problem of the term "Proxy" , which already is
used in "ProxyRestriction" (apparently derived from SAML -Security
Assertion Markup Language). Do we have a problem there?
5. At the F2F we also said that the Semantic Model should include the
(Cloud) proxy, but it is unclear how to do this. It could be but need not
be a device subunit. We do not model the User Client, to which the Proxy may
be most analogous. Would equipping the Services with the capability of
accepting an optional set of Proxy originated operation requests adequate?
6. We still have the question of using the same element name for a
binary element indicating support for a capability and the element
indicating the value for that capability. For example, DocumentName as a
Boolean under DocumentDescriptionCapabilities and DocumentName as a string
under DocumentDescription. (There is also a DocumentNameSupported
Boolean) The system does not seem to have problems with this, since they
are members of different complex elements. But it may be confusing to
humans.
7. The semantic model still contains EmailIn and Email out (although,
curiously, there are no EmailIN or EmailOut Operations, whereas there are
Operations for all other Services.) Does SM3 include EmailIn and EmailOut
Services?
Thanks,
Bill Wagner