OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

wsia message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Subject: [wsia] Updated Minutes for 11/6 (Website Version)

Title: Updated Minutes for 11/6 (Website Version)
Wednesday, 11/6/02
Voting Members    Company
Stephen Drye (on leave) Art Technology Group   
William Cox     BEA     y
Adrian Fletcher        BEA    
Gino Filicetti  Bowstreet       y
Steven Smith   Capitol College y
Andre Kramer   Citrix  y
Monica Martin  Drake Certivo   y
Raj Ramesh     CommerceOne     y
Timothy N. Jones       CrossWeave      y
Alan Kropp     Epicentric      y
Nigel Ratcliffe        Factiva
Madoka Mitsuoka (on leave)      Fujitsu
Sunit Randhawa  Fujitsu
Richard Cieply  IBM     y
Carsten Leue   IBM     y
Thomas Schaeck, chair  IBM     y
Rich Thompson  IBM     y
Charles Wiecha IBM     y
Eric van Lydegraf      Kinzan  y
Jon Klein      Reed-Elsivier   y
Adam Nolen     Reed-Elsivier   y
Petr Palas     Moravia IT     
Olin Atkinson   Novell 
Chris Braun     Novell  y
T.J. Cox       Novell  y
Michael Freedman       Oracle  y
Mike Hillerman Peoplesoft     
Art Machado     Peoplesoft     
Ken Pugsley     Peoplesoft     
Sasha Aickin    Plumtree        y
Jane Dynin     Plumtree       
Joseph Stanko   Plumtree       
Michael Young   Plumtree       
Amir Blich      SAP    
Gennady Shumaker        SAP    
Yossi Tamari    SAP     y
Rex Brooks     Starbourne      y
Brian Dirking  Stellent       
Alejandro Abdelnur      Sun Microsystems        y
Dave Clegg     Sybase 
Joe Rudnicki    U.S. Navy       y
Eilon Reshef   WebCollage      y
Gil Tayar      WebCollage      y
Prospective Members (non-voting)              
Dan Machak      Tibco  
WSIA Members
Graeme Riddell  Bowstreet      
Bruce Lucas     IBM     y
Ravi Konuru    IBM    
Dan Gisolfi     IBM    

#97 Caching proposals
2 proposals.  1.  Schema in 0.7 and 0.8  2.  Modified proposal from Yossi that treates items from current draft as scopes and introduces InvalidationKeys.
Gil:  Common web app caching mechanism is not expiry, but invalidation based.
Carsten:  That's why the proposals contain cache hints.  The "input space" is too large to calculate a key.
Gil:  Hinting is a good approach, though there is no certain algorithm that's going to cover.
Sasha:  Actually expiry caching is used extensively.  (if not modified eTag).  This introduces a "round-trip" for validation.
Mike F:  Yossi's proposal is for invalidation-based caching.  
Sasha:  Haven't figured out how invalidation-based scheme intersects with validation/expiry mechanisms.
Mike F:  HierarchyŠexpiry ý hints ý state (scope) based ý "catch-all" action-based expiry (reset all).  The problem with hints/state levels of this hierarchy is that it forces the Consumer to "know" details about the caches.
Sasha:  Review Yossi's proposal?
Mike F:  MarkupResponse includes markup scope (User, Producer, refHandle, Registration).  Invalidation tag prefix returned by Producer in interactionResponse to invalidate content.  Consumer matches prefix "slice" into cached content.
See Yossi's email of 9/30 for details
Gil:  This is an attempt to mirror http-based caching we have today, plus adds invalidation.
Andre:  Concern that the invalidation tag prefix mechanism will be difficult for developers to code to.
Mike F:  When used properly, it's very performant.
Sasha:  If Producer uses the invalidation scheme, the Consumer MUST adhere to it.  That's going to be complicated.
Rich:  That's actually true of any optional (i.e. beyond expiry) caching scheme the Producer is usingŠConsumer MUST adhere to it.
Eilon:  Registration meta-data will help to inform Consumer what it has to do.
Thomas:  No micro-flags for fine-grained cache control.  All or nothing.  Should simplify Producer implementations.
Mike F:  So what's more elemental (in terms of industry usage today) is pure expiry, or expiry + validation (eTags).  For a Producer that wants to be extremely performant (i.e., a "home page" portlet), an invalidation mechanism would be nice to have, although that's certainly a more complicated scheme for the Producer to implement.
Rich:  Could a Consumer that wants to do invalidation-keying signal its capability to the Producer?
Gil:  Invalidation seems like a v2.0 consideration.
#45 Signal state change in InteractionResponse
How "sticky" is navigationalState?  Or, how long does the Consumer need to remember navState? 
Mike F:  Understand maintaining state within the context of the aggregated unit.
This is related to the question of the Consumer MUST remember the last navState, in the event no navState gets returned in an InteractionResponse.  Does this potentially imply Consumer session?
Gil:  No, null navState is an "alias" for "give me what I sent last time".
The other question is what the meaning of null navState in InteractionResponse means?  Must it always mean the Consumer manages the navState between invocations?
Mike F:  There are useful scenarios for this to actually mean "null" or no state.
Eilon:  Make it a required field, so "null" is no longer available.  So, return "empty" to signify no state.
#100 Entity property model
Slide depicting various property structures.
Gil:  The "reset" attribute on Property doesn't make sense for getProperty()?
Bruce:  May just want a separate interface to reset properties to default.
Gil:  for getEntityProperties, the signature includes a propertyList.  No reason to include property values in this call.
Charlie:  Don't recall why the working group chose this, but it should be ok to revert to a list of names.
Looking at proposal for <wsrp:localization/>
Bill:  Why are we inventing a WSRP version of resource localization?  Are there other standards that we should be waiting on? 
Charlie:  How can we go out without support for internationalization in 1.0?
Mike F:  Vendors already have internationalized products, so for us to propose spec that doesn't permit portlets to make use of this capability is a non-starter.
Rich:  Summary of the open questions, and we'll vote later in the week: 
1.      Indicate desired locale of requested descriptive resources, with localization URL that points to a structure containing full localization information?  Or include full locale data for all supported locales inline when getting a descriptive resource.
2.     Do we invent a new localization scheme for WSRP, or do we use xml-lang?
#91 setEntityProperties atomic?
Atomic..but not transactional.
Carsten:  Difference between atomic and transactional?
Bruce:  Use case for atomicity of property changes?  All-or-nothing truly required?
Rich:  Consumer-managed properties, especially if they're a hierarchy, is one.
Mike F:  So, "synchronized" may be a better term. 
Rich:  Validated and synchronized update, not required to be transactional.
#120 PropertiesDescription not in entityDescription?
Gil:  It's potentially too large, not all Consumers need/want property description whenever they get an entity description.
Bill:  Maybe we should consider the general problem of specifying (micro flags?) what should be included in the response to any request.
Rich:  Our approach has been that unless TC folks indicate something really is mandatory in a given resposne, then it gets separated (with separate API to retrieve it).
#96 Review well-known URL param names
Mike F:  Don't understand 'wsia:refHandle'.  The spec already mandates that the Consumer has to return the refHandle on all requests, so why does it belong in the Producer-rewritten URL?
Andre:  Wondered why this wasn't entityHandle?
Mike F:  It's not specific enough.
Rich:  Consumer needs to supply the indirection of refHandle, since it can change on a getMarkup. 
#35 Is 'recommend' strong enough for URL-encoding statement
Consumer needs to know what the character encoding of the parameters is.
Mike F:  Suggested solution is to mandate that encoding, UTF-8.  Consumers will therefore always know what the answer is.
Gil:  This isn't important for Consumer URL rewriting?
Mike F:  Not always true, there are places where the Consumer requires the Producer to do URL encoding.
2 proposed mechanisms:
1.       Mandate:  UTF-8
2.       URL itself carries the information.  Add a token, 'charset', that the Producer re-writes with the encoding.
Sasha:  Dislike #2.  You need to parse the URL to find the charset.  Not normal behavior that off-the-shelf parsers will take advantage of. 
Gil:  So #2 could really be viewed as a Boolean.  Either the markup is returned in UTF-8, or it wasn't.
Mike F:  Correct, although you lose the particular charset in use. 
Vote:  1: 17  2:  1  Abstain:  5
#138 URL templates for performInteraction?
Mike F:  Think it's natural in markupParams.
Resolved:  Yes
#129 Double-encoding of wsia:url
urlType==resource, so it's an embedded URL.  Problem is conflicting levels of encoding.
Rich:  If we change verbiage to mandate "strict" URL encoding, then double-encoding isn't needed.
Resolved:  Producer does strict encoding (i.e. make it a valid query string value Š encode '&', '=', '?', ':', '/' and '?').
Mike F:  For both Producer and Consumer URL's?
Rich:  Since it's the template that's being replaced, the Producer must always do this.
#146 Require urlType token be at beginning of URL sequence
Efficient parsing.
Sasha:  Enforcing an ordering could be a problem for debuggability.
Mike F:  Not a debugging issue, since if it's missing it's easily logged/found.
Rich:  Suggest:  /wisa:rewrite?urlType&wsia:urlŠ
#131 wsia:requestParameters should be double-encoded
Gil, Sasha:  What is requestParameters again?  Isn't this navigationalState?
Rich:  It's where the Producer puts any "extra" query string parameters.
Gil:  So these shouldn't be double-encoded.
Resolved:  Change 'doubly' to 'strictly'.
#37 & 128  Using predefined namespace prefixes is unconventional
Switch to 'XXX-'
What XXX will be is for Friday.
#63 Entities / Window states
Lingering issue:  What are the semantics of the Consumer refusing to change window state?
Mike F:  Some confusion over window statesŠis it a state?  Or is it a style?
Rich:  It's a state that implies a style.
Mike F:  Consumer refuses, then it should immediately call getMarkup.
Gil:  How typical would it be for the Consumer to refuse a state?
Mike F:  If the Consumer receives a request for a state it doesn't support.
Sasha:  This seems very unlikely to occur.  Leads to complexity on both Producer and Consumer.
#102 Difference between Edit and Config
Resolved:  Drop Config
#114 Should custom roles, modes, window states be namespaced?
URI's are encouraged.
Gil:  The standard roles, modes, window states aren't URI's.  We're not eating our own dog food.
Resolved:  Leave as is.
#48 Should an entity state that it is cloneable?
Sept F2F said No.  We've also changed cloneEntity to take an entityHandle, and not a refHandle.
Mike F:  So, should we recast this into "should an entity's transient state be cloneable"?
Rich:  Separate issue.
Resolved:  No, per 105 below.
Pending resolutions
105 Add releaseRefHandle(refHandle, registrationContext) to Markup portType
Mike F:  Should it take an array of refHandles?
Resolved.  ReleaseRefHandles(refHandle[], registrationContext)
Vote:  Yes:  11  No:  8  Abstain:  5
Not-present voters will have a chance to vote next week.
Eilon:  Not well named.  It's the underlying entity that gets cloned, and a new refHandle gets created to reference it.
Discussion regarding the use of cloning transient state.  Consumer manages property changes, user edits an entity, and entity gets cloned: should also "clone" (really, transfer) the refHandle to the new clone so that the user's current runtime state is preserved.
Resolved:  Not in this version.
115 GET Forms
1.  Consumers SHOULD support method=get
2.   entity metadata usesMethodGet and URL template structure:  www.consumer.com/_{wsia:requestParameters}/
Mike F:  What are the use cases for supporting GET at all?
Sasha:  The fact that it does have different behavior then POST, and it's heavily used today (e.g. Google).
Yossi:  We will encourage Producers to use POST? 
Rich:  Yes.
98 Should operations (other than getProperty) return Entity property values?
99 Should property operations optionally take a user context?
#150 Use a SOAP fault to indicate a redirect
Mike F:  Redirect means ignore every other param in the response.  Unconventional.
Andre:  .NET doesn't honor faults, or at least not very well.  It would not be usual to carry useful data like the redirect URL in a fault.
Rich:  The suggestion really boils down to dividing the InteractionResponse.
Mike F:  Proposed SOAP fault only as a means of illustrating the issue.
Sasha:  You're also supposed to send back content that could be displayed.
Resolved:  Divide InteractionResponse into a choice:  normal semantics or redirectURL.
(Does anyone really need new mode, new window state, etc., along with a redirectURL?)
#111 getMarkup called for MINIMIZED
Language suggestion:  [see slide]
#153 Remove "extension" fields from ServiceDescription and RegistrationData
userProfileExtensions, consumerExtensions, registrationProperties in RegistrationData
registrationProperties in ServiceDescription.
Mike F:  These could just live as standard extensions, and the fields should be removed.
Yossi:  Makes life a bit more difficult, since Consumer now has to hunt for entity's extended user profile attributes (for example) in the extensions.
Rich:  consumerExtensions was brought up long ago, not sure anyone has found a real use for it.
Charlie:  registrationProperties appear to be very close to wsdl endpoint properties.  We need to be clear that they are not.

Resolved:  ServiceDescription.registrationProperties of type ModelDescription
Drop consumerExtensions.
Mike F:  Rename userProfileExtensions to remove the "extension".  (customUserProfileData?)

#118 RegistrationData.userProfileExtensions[] needs description sub-element
Mike F:  Human-readable description?  This should be an extension.
Resolved:  Leave as String[].
#130 minimal encoding of URL rewriting params
Add ":"
#101 Valid values for MarkupParam fields in spec examples/references.
See 0.8 draft, email suggested changes to Rich.
#117 Markup interface should be first described in spec
Keep as is
#137 Require "DefaultTemplate"
Two different ones (secure/unsecured)
For case where Producer rewriting requires templates, but Consumer didn't pass any.
Mike F:  Doesn't seem useful for 1.0 timeframe.
Rich:  Either Consumer sends defaults, or the full list of templates (MUST)
 #151 Clear list of Consumer and Producer requirements
Resolved:  Separate document to provide guidance to implementers.
#139 Delete unused glossary terms
Keep 'Action' as a synonym for 'Interaction'.
Mike F:  We also need a term for "portlet instance" based on yesterday's decision to optionally carry a portlet ID.
#134 use of the GUEST role
If there's no user ID, what's the value of carrying the GUEST role?
Mike F:  This is a role that applies to this user, regardless of the portlet he uses.
Alej:  Can a profile key be shared by multiple users?



Rex Brooks
Starbourne Communications Design
1361-A Addison, Berkeley, CA 94702 *510-849-2309
http://www.starbourne.com * rexb@starbourne.com

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Powered by eList eXpress LLC