RE: [wsdm] MUWS Comments part 2

["Vambenepe, William N" <vbp@hp.com>]

I made a bunch of corrections from your annotations in the doc.

 

See <wv></wv> below for more.

 

William

---

From: Homayoun Pourheidari [mailto:homayoun@bea.com]
Sent: Monday, November 29, 2004 5:37 PM
To: wsdm@lists.oasis-open.org
Subject: Re: [wsdm] MUWS Comments part 2

William,

My comments on part 2.  Most of them are typos or cosmetic changes.  However, I think state and status capabilities need more work.

3.2.4. I don’t understand what you’re suggesting here?  We have a capability that has no properties in it?  What about the 2 that you just defined – state and statetransition?  I agree that the state values should be defined by the managed resource owner.  But how would the manager know where to go to get the state information if it doesn’t know how to ask for it?
 

<wv>the idea is that people are free to use any property that conveys the correct semantic (like "CurrentState", "DesiredState", "StateAtTheSameTimeLastYear") but the consumer will be able to recognize either the element or type used as being the one we define so they know that this property contains the description of a state. What this state correponds to it model-specific.</wv>

 
3.3.1. I disagree with the language that the operational status is "independent of the specific state model used".  I think it would be dependent but it is not a direct mapping.
 

<wv>I tried to clarify this way: "The operational status capability defines a simple representation of the availability of the resource. This is expressed in terms (defined by MUWS) independent from the specific state model (defined by domain experts) for the resource."</wv> 

 
Cheers,
H.
--

Zhili Zhang wrote:

Hi William,

 

Here are my comments on document wd-wsdm-muws-part2-1.0-20041126.doc.

 

(1) In the WSDL, the following resource property documents from namespace "http://docs.oasis-open.org/wsdm/2004/12/muws/wd-wsdm-muws-part2-1.0.xsd" are referenced in various portTypes. However, they are not defined in that namespace.

 

muws-p2-xs:DescriptionProperties
muws-p2-xs:MetricsProperties
muws-p2-xs:RelationshipsProperties
muws-p2-xs:RelationshipResourceProperties

 

(2) QName "xml:lang" is used in schema. Strictly speaking, the XSD should have the following import statement:

 

import namespace = "http://www.w3.org/XML/1998/namespace"

 

Otherwise, several schema validators will give the following error:
#src-resolve.4 QName reference "xml:lang" can't be resolved as namespace "http://www.w3.org/XML/1998/namespace" wasn't imported. (see http://www.w3.org/TR/xmlschema-1/#src-resolve)

 

Other validators like .NET may work well with it. However, a simple fix like this will make WSDM schemas solid.

 

(3) The namespace table in Introduction section (page 5) lists namespaces for ServiceGroup (wssg), etc. Couple of others are missing from the table, such as wsrf-rp, wstop, wsa.

 

(4) Should we insert OASIS legal disclaimer into normative XSDs, WSDLs, topic instances? Schemas and WSDLs from other groups like WSN, WSRF have it.

 

Thanks

Zhili

 

 

 

-----Original Message-----
From: Heather Kreger [mailto:kreger@us.ibm.com]
Sent: Monday, November 29, 2004 12:24 PM
To: vbp@hp.com
Cc: wsdm@lists.oasis-open.org
Subject: [wsdm] MUWS Comments part 2

Here are my comments on MUWS Part 2.



My biggest issues are:

1. In Platform Properties - addresses addressing more that properties, split out into two sections and make more normative references to resource properties specs.
2. Need Topics in the events section.
3. Configuration is missing half its section
4. Advertising is missing advertisement by registration

Heather Kreger
STSM, Web Services Lead Architect for SWG Emerging Technologies
Author of "Java and JMX: Building Manageable Systems"
kreger@us.ibm.com
919-543-3211 (t/l 441)  cell:919-496-9572