Meeting started at 8:05  PST

 

====================================================================

Roll Call

Voting Members:

---------------------------------

Alejandro Abdelnur     Sun                      no

Sasha Aickin           Plumtree                 no

Subbu Allamaraju       BEA                      yes

Olin Atkinson          Novell                   yes

Atul Batra             Sun                      yes

Amir Blich             SAP                      yes

Michael Bosch          Vignette                 no

Rex Brooks             Starbourne               yes

T.J. Cox               Novell                   no

William Cox            BEA                      no

Brian Dirking          Stellent                 yes

Michael Freedman       Oracle                   yes

Ross Fubini            Plumtree                 yes

Richard Jacob          IBM                      yes

Jon Klein              Reed-Elsevier            no

Andre Kramer           Citrix                   yes

Alan Kropp             Vignette                 yes

Carsten Leue           IBM                      no

Dan Machak             Tibco                    yes

Nigel Ratcliffe        Factiva                  no

Thomas Schaeck         IBM                      no

Gennady Shumaker       SAP                      no

Steven Smith           Capitol College          yes

Yossi Tamari           SAP                      yes

Rich Thompson          IBM                      yes

Charles Wiecha         IBM                      yes

 

 

Total voting members:          26

Voting members in attendance:  16 (62%)

A quorum was present.

 

Members on Leave Of Absence

----------------------------

Chris Braun            Novell                   LOA (for next 45 days)

Sunit Randhawa         Fujitsu                  LOA

Joe Rudnicki           U.S. Navy                LOA

Eric van Lydegraf      Kinzan                   LOA

 

 

Prospective Members (non-voting):

---------------------------------

Winston Damarillo      GlueCode Software        no

Metin Ergener          Vignette                no         

Noah Guyot             Vignette                 yes

David Holladay         Microsoft                yes

Dimitri Jirov          SAP                      yes

David Raphael          Ericsson                no

 

WSIA Members (non-voting):

-----------------------------

Ravi Konuru            IBM                      no

Bruce Lucas            IBM                     no

Graeme Riddell         Bowstreet               no

 

====================================================================

 

Unable to take notes while working with conference call operator to clear up line noise.   Sorry.

 

====================================================================

 

The minutes from the face-to-face (5/12-5/14) were accepted

 

====================================================================

1. SC Status/Issues:


  Interoperability (Richard)

   
  Conformance (Rich)

    [unable to take notes at this time, missed Rich’s report] 
  WSDL (Andre)

    Will become active in July
  PFB (Alan)

    Will become active in July
  Markup (Rex/Andre)

    New charter available for review
  Coordination (Rich)

    Will become active in July
  Interfaces (Mike)

    No activity anticipated until September

 

2. Andre’s “to raise” list

 

-          path encoding URLs using templates (i.e. :/../../ style) requires producer to avoid special chars (/,:,&,%,*)

-          not possible to clone (and set up a portlet session)) and re-direct from performBlockingInteraction

-          setProperties returns a portletHandle (implies clone on write?)

-          WS-Security: How to leverage it for user identification and roles?

-          add performInteraction (non-blocking)

 

3. Outstanding errata (disposition noted in []’s)

 2.1   #16. 5/9/03 (Andre Kramer)
Contact.telecom, Contact.online, Online.uri are not arrays in the IDL.
Should remove the maxOccurs="unbounded" from the WSDL.

Document: WSDL
Section: types
Action: Remove "maxOccurs" attribute from Contact/telecom, Contact/online, Online/uri
[ACCEPTED]


  2.2   #17. 5/9/03 (Andre Kramer)
6.1.17 has UserProfile.bdate while ch 11 has bDate in table. Seems it should be bdate?

Document: Spec
Section: 11 (twice)
Old Text: bDate
New Text: bdate
[ACCEPTED]


  2.3   #18. 5/19/03 (Rich Thompson)
Appendix E (Revision History) had value while the spec was being developed. I don't think it has value in the final spec.

Document: Spec
Section: Appendix E
Action: Delete this section.
[ACCEPTED]


  2.4   #19. 5/13/03 (F2F)
Conformance statement for mode would be better in section 6.8. It should also be reworded for clarity to the portlet developer.

Document: Spec
Section: 6.1.12
Old Text: The Consumer MAY choose to respect this request to change the mode, but since the Portlet cannot depend on that choice it MUST NOT encode this new mode into any of its stateful settings. Rather, the Portlet MUST compute any such impact on stateful settings after the Consumer has actually changed the mode.

New Text: See section 6.8 relative to the processing of such requests.

Document: Spec
Section: 6.8
Old Text: Whether or not such a request is honored is up to the Consumer and often will depend on access control settings for the End-User.

New Text: The Consumer MUST respect requests to change the mode outside of exceptional circumstances (e.g. access control restrictions), but the Portlet must not depend on such a request being respected.
[ACCEPTED]



  2.5   #20. 5/13/03 (F2F)
Conformance statement for window state would be better in section 6.9. It should also be reworded for clarity to the portlet developer.

Document: Spec
Section: 6.1.12
Old Text: The Consumer MAY choose to respect this request to change the window state, but since the Portlet cannot depend on that choice it MUST NOT encode this new window state into any of its stateful settings. Rather, the Portlet MUST compute any such impact on stateful settings after the Consumer has actually changed the window state.

New Text: See section 6.9 relative to the processing of such requests.

Document: Spec
Section: 6.9
Append to first paragraph: Portlets may request window state changes either through parameters on a link that an End-User activates or by returning a newWindowState in a BlockingInteractionResponse. The Consumer SHOULD choose to respect this request to change the window state, but since the Portlet cannot depend on that choice it MUST NOT encode this new window state into any of its stateful settings. Rather, the Portlet MUST compute any such impact on stateful settings after the Consumer has actually changed the window state
[ACCEPTED]



  2.6   #21. 5/20/03 (Alejandro Abdelnur)
The namespacePrefix should be required as it may be used by the portlet to namespace things (ie: javascript thingies) within the generated content. This is independent from templating.

Document: Spec
Section: 6.1.2
Old Text: [O] string namespacePrefix
New Text: [R] string namespacePrefix

Document: WSDL
Section: wsrp_v1_types.xsd
Old Text:
New Text:

Andre:  This is symmetrical with the required Consumer namespace template if Producer rewriting.

Mike F:  This sounds like more of a theoretical issue, prefer to wait and see if experience shows it’s an actual problem.
[DEFERRED TO EMAIL DISCUSSION]


  2.7   #22. 5/21/03 (Alejandro Abdelnur)
Spec doesn't really define the relationship between markup and title. Is title considered part of cache content? If yes, where does it say that in the spec? I see assertions like if MarkupContext.useCachedMarkup is true, then MarkupContext.markupString MUST be NULL. Why doesn't the spec say that MarkupContext.preferredTitle would be NULL too in this case.

Document: Spec
Section: 6.1.10
Old Text: If the value for useCachedMarkup is “true” the markupString and markupBinary fields MUST NOT be returned.

New Text: If the value for useCachedMarkup is “true” the markupString and markupBinary fields MUST NOT be returned. If the value for useCachedMarkup is “true” the preferredTitle field SHOULD NOT be returned and the previous preferredTitle value reused.

Mike F:  Better to leave it undefined.  We’ll need to be get more explicit when we introduce invalidation.

[DEFERRED TO EMAIL DISCUSSION]