OASIS Web Services for Remote Portals (WSRP) TC

WSRP/WSIA Telecon 02/27/2003

Meeting began at 9:06 PST

An attempt to start the meeting at 8:00 AM PST failed due to a lack of a quorum.


The minutes were accepted from the meeting last Thursday, 2/20/03
The Tuesday meeting on 2/25/03 was cancelled due to a lack of a quorum.


Roll Call

Voting Members:

Company

 

Abdelnur, Alejandro

Sun

No

Aickin, Sasha

Plumtree

No

Allamaruaju, Subbu

BEA

Yes

Atkinson, Olin

Novell

Yes

Batra, Atul

Sun

Yes

Blich, Amir

SAP

Yes

Braun, Chris

Novell

Yes

Brooks, Rex

Starbourne

Yes

Cox, T.J.

Novell

No

Cox, William

BEA

No

Dirking, Brian

Stellent

No

Freedman, Michael

Oracle

Yes

Fubini, Ross

Plumtree

Yes

Jacob, Richard

IBM

Yes

Klien, Jon

Reed-Elsivier

Yes

Kramer, Andre

Citrix

Yes

Kropp, Alan

Vignette Corporation

Yes

Leue, Carsten

IBM

Yes

Machak, Dan

Tibco

Yes

Mitsuoka, Madoka

Fujitsu

Yes

Nolan, Adam

Reed-Elsivier

Yes

Randhawa, Sunit

Fujitsu

No

Ratcliffe, Nigel

Factiva

No

Reshef, Eilon

WebCollage

No

Rudnicki, Joe

U.S. Navy

No

Shumaker, Gennady

SAP

Yes

Tamari, Yossi

SAP

No

Tayar, Gil

WebCollage

No

Thompson, Rich

IBM

Yes

van Lydegraf, Eric

Kinzan

No

Wiecha, Charles

IBM

Yes

Young, Michael

Plumtree

No

Total voting members: 32
Voting members in attendance: 19 (59%)
A quorum was present.

Members on Leave of Absence:

 

 

Martin, Monica

Drake Certivo

LOA

Palas, Petr

Moravia IT

LOA

Ramesh, Raj

CommerceOne

LOA

Schaeck, Thomas

IBM

LOA

Smith, Steven

Capitol College

LOA

WSI Members in Attendance (Non-Voting)

 

 

Lucas, Bruce

IBM

Yes


Summary of the substantive changes:

167 - Correct mime types

ACCEPTED

171 - Make RegistrationState field use explicit.

ACCEPTED

172 - explicitly useCachedMarkup

ACCEPTED

142 - Remove performInteraction()?

TIE VOTE

147 - Exposes Registration portType or req registration?:

ACCEPTED

170 - Producer providing UserCategory support for Portlets

ACCEPTED

169 - Is resourceName usage a conformance statement?

ACCEPTED

153 - Make resourceName optional?

ACCEPTED

151 - UserScope definition

ACCEPTED

155 - Resending UserContext semantics

ACCEPTED

174 - Processing a new cacheControl

ACCEPTED

156 - Reset + set properties semantics

ACCEPTED

144 - Meaning of requiresRegistration field

ACCEPTED

158 - ServiceDescription.locales definition

Treated as Change Bar in spec


Small editorial/grammar changes (I do not expect to walk through these unless there are objections):

167 - Correct mime types
Section 5.1.10 / Page 19 / Line 36
Use actual mime types in example rather than common acronyms

171 - Make RegistrationState field use explicit.
Sections 5.1.19 & 5.1.20 / Page 24 / Lines 2 & 21
Title says it all

172 - explicitly useCachedMarkup
Section 6.1.10 / Page 31 / Line 40-42
By being explicit, the conformance statement can stand on its own

Resolution: Accepted without comment


142 - Remove performInteraction()?
Impacts many sections
JSR168 has dropped the need for this operation. Do we still want to keep it? Basic trade-off is the additional functionality vs the additional complexity for understanding, implementing and testing.

Arguments for removing: reduction in complexity
Arguments against removing: already defined. Semantics are useful.

Resolution: Tie vote 9 for removing, 9 for keeping, 1 abstain

Vote outcome:

Subbu Allamaraju

BEA

yes

Olin Atkinson

Novell

abstain

Atul Batra

Sun

yes

Amir Blich

SAP

no

Chris Braun

Novell

yes

Rex Brooks

Starbourne

no

Michael Freedman

Oracle

no

Ross Fubini

Plumtree

yes

Richard Jacob

IBM

yes

Jon Klein

Reed-Elsevier

no

Andre Kramer

Citrix

no

Alan Kropp

Vignette Corporation

no

Carsten Leue

IBM

yes

Dan Machak

Tibco

yes

Madoka Mitsuoka

Fujitsu

no

Adam Nolan

Reed-Elsevier

no

Gennady Shumaker

SAP

no

Rich Thompson

IBM

yes

Charles Wiecha

IBM

yes

A yes vote means that performInteraction would be removed.


147 - Exposes Registration portType or requires registration?
Section 3.5 / Page 11 / Line 34-35
Proposal is to change a conformance statement from exposing the in-band technique for registering to the use of the requiresRegistration flag.

Resolution: Accepted - Will drop sentence entirely


170 - Producer providing UserCategory support for Portlets Section 5.1.1 / Page 20 / Line 4-5
Proposal is to drop the conformance level language from the note that the Producer may actually provide this support on behalf of the Portlet.

Resolution: Accepted - Will drop sentence entirely


169 - Is resourceName usage a conformance statement?
Section 5.1.5 / Page 17 / Line 19-20
Question is the value of making the use of resourceName as an indirection in a ResouceList a conformance statement. Proposal is to keep the content, but not as a conformance statement.

Resolution: Accepted - Will drop conformance language.


153 - Make resourceName optional?
Section 5.1.5 / Page 17 / Line 23
Proposal is to have semantics for not supplying a resourceName (namely that values for other locales are not supplied).

Resolution: Accepted


151 - UserScope definition
Section 5.1.11 / Page 19 / Line 19-20 + others
Proposal change UserScope from a tri-value to an open-ended set with 2 predefined values. This drops the needs to use the extensions field just to carry the name of other semantics.

Resolution: Accepted: UserScope field changed to type String.
Define Strings for the 2 pre-defined scopes. This will be modelled more like modes and window-states.


155 - Resending UserContext semantics
Section 5.1.11 / Page 20 / Line 29
Proposal is to clarify the semantics of new information being supplied for either UserContext or Templates.

Resolution: Approved. Semantics should be that new info replaces the old.


174 - Processing a new cacheControl
Section 6.1.10 / Page 31 / Line 40-42
Be explicit when useCachedMarkup is "true" and a new CacheControl is supplied, it replaces the previous one.

Resolution: Approved. Semantics should be that new info replaces the old.


156 - Reset + set properties semantics
Section 5.1.14 / Page 21 / Line 37
Proposal is for clear semantics when a property is included in both the reset and set arrays of a PropertyList.

First question is whether it is necessary to define any semantics on how to handle properties in both arrays, or to enforce the two arrays to be mutually exclusive.

Resolution: Approved. Producer will throw a fault if the same property shows up in both arrays.


144 - Meaning of requiresRegistration field
Section 5.1.18 / Page 23 / Line 17
Proposal is for a new conformance statement for when requiresRegistration='true'.

Resolution: Approved.


158 - ServiceDescription.locales definition
Section 5.1.18 / Page 23 / Line 29
Proposal seeks to clarify that not all LocalizedStrings will have values for all locales.

Resolution: Rich will update the language.


There will be 2 teleconferences next week:

Tuesday, March 4
9:00 am PST / 12:00 am EST / 6:00 pm CET
duration: 1 hour

Thursday March 6
8:00 am PST / 11:00 am EST / 5:00 pm CET
duration: 2 hours


Meeting adjournded at 10:05 PST