Re: [ws-caf-editors] context model

[Greg Pavlik <greg.pavlik@oracle.com>]

right, I wasn't there to object ;-)

I'm not saying the UML isn't useful or good, I just think it's insufficient to gain the clarity that we're trying to achieve.

Mark Little wrote:

Greg, I forgot that you weren't on the telecon. yesterday. However, that's 
where the idea of producing UML came from. I think that's what the TC is 
expecting now.

Mark.

  

===== Original Message From Greg Pavlik <greg.pavlik@oracle.com> =====
My comment in private (I suppose this is still private) to Mark was that
in my opinion a formal model and a UML diagram are very different, and
we really need to focus on the former. At best, the diagram aids someone
in understanding the model description. In this case, I think the text
and diagram should be in the same document, since the diagram shows
cardinality of relationships, but the relationships themselves are
spelled out with their meaning and semantic implications in the text.

The activity identifier in the text may be my blunder -- I'll have to
double check -- the ContextType has a context-identifier, which I have
always assumed maps to the identity of the activitiy. If that's correct,
perhaps we can just say so?

For refreshers, here's the context service schema:

   <xs:complexType name="ContextType">
       <xs:sequence>
           <xs:element name="context-identifier" type="xs:anyURI"/>
           <xs:element name="activity-service" type="xs:anyURI"
minOccurs="0"/>
           <xs:element name="type" type="xs:anyURI" minOccurs="0"/>
           <xs:element name="activity-list" minOccurs="0">
               <xs:complexType>
                   <xs:sequence>
                       <xs:element name="service" type="xs:anyURI"
minOccurs="0" maxOccurs="unbounded"/>
                   </xs:sequence>
                   <xs:attribute name="mustUnderstand"
type="xs:boolean" use="optional" default="false"/>
                   <xs:attribute name="mustPropagate" type="xs:boolean"
use="optional" default="false"/>
               </xs:complexType>
           </xs:element>
           <xs:element name="child-contexts" minOccurs="0">
               <xs:complexType>
                   <xs:sequence>
                       <xs:element name="child-context"
type="tns:ContextType" maxOccurs="unbounded"/>
                   </xs:sequence>
               </xs:complexType>
           </xs:element>
           <xs:any namespace="##other" processContents="lax"
minOccurs="0" maxOccurs="unbounded"/>
       </xs:sequence>
       <xs:attribute name="timeout" type="xs:int" use="optional"/>
   </xs:complexType>


Greg
(ps it's amazing how unhelpful UML when you try to say talk about
things that aren't classes, objects or actors).

Martin Chapman wrote:

    

what is the relationship between activity identifier and context? the
text implies a realtionship yet its not on the diagram. also I sugest
that rather having a note saying that activities can be nested, that
there is an self refernceing association on  activity.

Martin.



      

-----Original Message-----
From: Mark Little [mailto:Mark.Little@arjuna.com]
Sent: 09 March 2004 22:16
To: ws-caf-editors@lists.oasis-open.org
Cc: mark.little@arjuna.com
Subject: [ws-caf-editors] context model


Here's the context model we seem to have agreement on so far.
If you can take
a quick look and let me know of anything that you think needs
qualifying, I'll
send it to the main group. I want to then follow up with
email explaining the
diagram (basically the text we've already got). From there we
can persue some
of the other issues (such as removing status types, context
as one block in
the SOAP header or one entry per ALS, ...)

Mark.

----
Mark Little,
Chief Architect, Transactions,
Arjuna Technologies Ltd.



        

www.arjuna.com