SM> Semantic Model Teleconference - August 30

SM> Semantic Model Teleconference - August 30

SM> Semantic Model Teleconference - August 30

Zehler, Peter Peter.Zehler at
Thu Aug 17 15:14:54 EDT 2006

On Wednesday August 30 there will be a Semantic Model Teleconference.

August 30 2006
1 pm - 2 pm eastern
Tele-Con Number:1-866-365-4406
PassCode: 2635888
WebEx Meeting Details ( 
Meeting Number: 747275324
Meeting Passcode: PwgSm
Meeting Host: Peter Zehler
Join Instructions for Instant Net Conference:
1. Join the meeting now:
2. Enter the required fields.
3. Indicate that you have read the Privacy Policy.
4. Click on Proceed. If you have problems with the above link, go to , select "join an event" and use the meeting
number and passcode from above. 

Agenda: 	1) Approve previous meeting minutes
		2) Agenda bashing
		3) Walk through the updated schema (Link below) 
			a) Recent changes
			b) Proposed direction
		4) Class structure of Semantic Model
		5) Chameleon schema issues
		6) Logical division of PWG schema
		7) incorporation of SM v1 printer elements into
		8) Discuss next steps and dates

I have just uploaded an update to the Semantic Model 

I am well aware that this does not pass the W3C's XSV but XMLSPY can
handle it.  The crux of the issue is that this is a "mid-stream" edit
that I would like to discuss.  If you want the details on the problem
see below.  

Specifically the discussion I would like to have is how far we want to
go with a class hierarchy in the PWG SM v2.  Take a look at System
Status and any of the service's status.  You will see there is a base
class that is used across all of them.  There are ramification in
element naming.

Another aspect of the model that was changed was the use of multiple
namespaces.  Note that the copyservice is now in its own namespace while
the print service remains in the sm: namespace.  The elements specific
to WIMS are also in their own namespace.

I also removed the legacy PWG SM v1 PrinterStatus and included all the
elements in the print service extension to the status class

For the convenience of those without XMLSPY I also included an HTML
version of the schema that might be useful for browsing.  (warning: it's
about 12M)


Explanation of problem:
  The actual problem is that the Chameleon schema has an unfortunate
limitation.  You can only reference (via an explicit ref or through a
type declaration) entities that are bound to a namespace (i.e. schema
file with a target namespace).  That is easily fixed by importing the
appropriate sub-schema.  The side effect occurs when the chameleon
schema is included in a schema file from the same namespace ad the
chameleon's imported sub-schema.  
In other words a schema file(A) with a target namespace includes a
chameleon schema file (B) without a namespace.  The effect of that is
that the chameleon schema (B)takes on the target namespace of (A).  If
the Chameleon schema (B) IMPORTS a schema file (C) that has the same
namespace as (A), and through the chameleon effect (B), that is not
allowed.  There are some tools that are forgiving.  An import with the
same namespace as the imported is functionally an include. 

Peter Zehler
Xerox Innovation Group
Email: Peter.Zehler at
Voice: (585) 265-8755
FAX: (585) 265-7441
US Mail: Peter Zehler
Xerox Corp.
800 Phillips Rd.
M/S 128-25E
Webster NY, 14580-9701 

More information about the Sm mailing list