Re: [wsn] Draft Minutes, 13 June 2005

[David Hull <dmh@tibco.com>]

That's fine with me.  I don't think I did a great job with that section.

Peter Niblett wrote:

>David
>
>Thank you for taking these minutes. I have just one comment. In the
>exchange on the URL rename, I think William's "Good luck with that" was in
>response to my question about whether we could get the OASIS web site to
>set the MIME type. I suggest adding my comment as follows...
>
>Peter: Observation:  As a result of new URLs and location, locations will
>not have .wsdl or .xsd.  So you may have to rename to make your tooling
>happy.  RF is happy with this.  Are we happy?
>Sgg: Rational(TM) can handle this if MIME type is present.
>Peter:Could we get the OASIS site to set the required MIME type?
>William: Good luck with that.
>Peter: We're OK with this, right?
>William: Alignment with RF is good.
>
>Peter Niblett
>
>
>
>
>                                                                           
>             David Hull                                                    
>             <dmh@tibco.com>                                               
>                                                                        To 
>             13/06/2005 19:17          wsn@lists.oasis-open.org            
>                                                                        cc 
>                                                                           
>                                                                   Subject 
>                                       [wsn] Draft Minutes, 13 June 2005   
>                                                                           
>                                                                           
>                                                                           
>                                                                           
>                                                                           
>                                                                           
>
>
>
>
>Roll Call:
>      We are quorate (60%)
>Minute taker:
>      David Hull
>Approval of minutes from 6/6:
>Peter: Question from Tom re: AI 97.
>
>Peter has posted email clarifying. Pointer to clarification will be
>incorporated into minutes. Approved subject to this modification.
>Action items review
>AI 56 - Sid/Igor: Fix typos in WS-Topics
>Neither is present.  No update.
>
>AI 74 - BaseN editors to correct problems in implementation of issue 2.34
>Bryan believes WSDL and schema are now in sync.  Closed.
>
>
>AI 79 - Lily to propose new issue and solution outside BaseN to replace
>2.26
>Still open.
>
>
>AI 81 - Chairs to recruit editors for the primer
>Not a burning issue.  Still open.
>
>
>AI 82 - Spec editors to incorporate content from the white paper
>Done for BaseN.
>Topics editors not available.
>Lily: Partially done for Brokered, need to double-check.
>Peter: Brokered looks done.  Should check as per issue verification
>process.
>Done for BaseN, Brokered.  Still open for WS-Topics.
>
>
>AI 85 - WS-Topics editors to respond to questions raised in Kato-san's
>email regarding use of terms TopicPath, TopicExpression
>TopicPathExpression, and the names of the related Faults.
>Editors not here.
>
>
>AI 93 WS-Topics editors: Edits/improvements to WS-Topics raised by Ian
>Springer.
>Editors not here.
>
>
>AI 96 - Chairs to create calendar entry for september F2F
>Closed.
>
>
>AI 97 - Editors to update the status text in all specs. See process doc
>BaseN and Brokered copied in WSRF text, but this is not quite what's in
>process doc ("that" vs. "the").  Specs will be updated.  AI remains open.
>
>
>AI 99 - Editors to move to new format for namespace URIs
>Done for BaseN, Brokered.  Topics editors not here.
>
>
>AI 103 - Chairs to send list of acknowledgements and ask if people want to
>be added/removed.
>Progressing.  Remains open.
>
>
>AI 104 - Chairs to produce list of reviewers for CDs
>Remains open (need to do brokered?)
>
>
>AI 106 - BaseN editors to delete Version f line 522 sentence "element MUST
>accept all NotificationMessages "
>Bryan: Done.
>Peter: Done as AI, still needs to be verified.  AI can be closed.
>
>
>AI 109 - Bryan to clarify new para "The NP MUST respond with ..." in the
>description of wsnt:Filter/wsnt:TopicExpression/@Dialect.
>Bryan: Made some changes, still needs to be verified.
>Peter: AI to be closed (issue to be verified).
>
>
>AI 110 - Peter to email DaveH about DaveH's concern w/ incomplete
>separation between BaseN and Topics.
>Done.
>Dmh: would like to discuss separation a bit more.
>
>
>AI 111 - Peter to email DaveH about DaveH's concern w/ cardinality of
>useRaw
>Done.
>Dmh: fine with that.
>
>
>AI 112 - Peter, Tom, Bryan to review and suggest alternatives via emails
>for section 2 (terminology and concepts) of BaseN
>Comments forthcoming.  Still open.
>
>
>AI 113 - Peter to add AI82 to verification list
>Done.
>
>
>AI 114 - Peter to update process document w/ correct format for doc ID and
>URLs
>Done.
>
>
>AI 115 - Editors to remove references to WS-Notif primer
>Bryan: Done for BaseN.
>Lily: Done for Brokered.
>Topics editors not present.
>
>
>
>Separation of Topics and Base:
>William(?): Aren't there already issues open for this?
>Peter: Issues 2.1, 2.52.  Two questions: Technically, are they now
>separate?  Structural: Should the docs be separated?
>Dmh: Separating structural makes the first question very easy to answer.
>Peter: If an NP is-a resource, it has to have FixedTopicSet.  You can use
>that to tell if topics are enabled.
>Dmh: This is broadening the meaning of FixedTopicSet, unequal basis with
>other extensions.
>Peter & Steve: Discussion of default values.
>Dmh: How would this sort of thing apply to other extensions?
>Peter: Topics isn't just another extension.
>Dmh: How is optional feature different from extension?
>Sgg: Extensions aren't specified at all.
>Dmh: Point.
>Peter: Could FixedTopicSet be optional (minOccurs=0, default is true)?
>Dmh: So it will never be undefined (always true or false).
>Peter, dmh, sgg: This is in line with the behavior for extensions/options
>in general.
>Peter: Making FTS optional seems good, even if it doesn't resolve the whole
>issue of separation.
>Sgg: Should be made part of resolution to 2.1.
>AI Sanjay: Update resolution of 2.1 to including making FixedTopicSet
>optional with default of true.
>Peter: Larger issue: Moving topic stuff to WS-Topics.
>Sgg: Splits port type across two specs.  Not keen on it.
>Dmh: Would make BaseN look more like WSE.
>William: Would need two port types.
>Bryan: Could leave GetCurrentMessage in BaseN.  Without topic it means
>"last message produced by NP", as for subscription with no filters.
>William: Doesn't GCM apply to other filters?
>Peter: No mechanism for it.
>Bryan: GSM is associated with a subscription.
>Others: No.
>Bryan: So with no topics, you just get the last message for that NP (as per
>a subscription with no filter).
>Peter: GCM is independent of any subscription.  You specify a topic.  You
>get back the last notification for that topic.
>Dmh: So one alternative would be to broaden to include any filter.
>William, Sgg: Would be hard to implement/limit scope (would NP have to
>remember all previous messages?)
>Bryan(?): Same issue exists with topics.
>Dmh: Yep.
>Peter: Topic gives nice finite limitation.
>Sgg: NP doesn't have to cache if it doesn't want to.
>Dmh: That would apply generally as well.
>Peter: NP knows topics ahead of time.  Not so with arbitrary filters.
>Dmh: So NP can refuse to answer.
>William: If we don't change GCM, then GCM should move (if we move).
>Peter: Problems: port type split, close to the wire.
>Dmh: Sympathetic to that.
>Peter: Might do split during public review.
>William: That's permissible.
>Dmh: Wanted explicit discussion of the issue, what it entails, feeling for
>whether it's bad, good but not worth it, or definely worth doing.
>William: Do you feel you have that.
>Dmh: Pretty much.
>(discussion of consensus)
>Consensus: Go to CD as is, consider split as a possibility after CD.
>
>
>URL changes in line with WSRF
>Peter: Observation:  As a result of new URLs and location, locations will
>not have .wsdl or .xsd.  So you may have to rename to make your tooling
>happy.  RF is happy with this.  Are we happy?
>Sgg: Rational(TM) can handle this if MIME type is present.
>William: Good luck with that.
>Peter: We're OK with this, right?
>William: Alignment with RF is good.
>
>
>Editors' updates from BaseN and Brokered.
>BaseN
>Bryan: Copied in Steve's proposal.  Should have addressed all issues.
>Sgg: There has been feedback from reviewers.
>Peter: 4 verified as correct.  5 have led to comments from reviewers.
>Bryan: Still minor changes, but all the issues have been addressed.
>Bryan, Sgg: Hold off version (k) until reviews are in.
>Sgg: Could chair gently remind reviewers of this.
>Peter: Volunteers for 1.4 review?  Not really.  Will draw straws.
>William: My email at paypal.com is ...
>
>
>Brokered
>Lily: Version (f) uploaded.  All issues should be addressed, except 1
>missing bit in terminology part.  Will upload revision.
>Peter: Can we assign reviewers?
>Lily: Believe so.
>Peter: Chairs will assign reviewers.
>
>
>Other Business
>William: Other business, other than imploring reviewers to review?  Want to
>finish public review before F2F?
>Peter: Will we have the full time for the next call?
>Omnes: Yes.
>
>Adjourned
>
>
>
>
>
>
>- Status of editors
>
>- Status of reviewers
>
>
>  
>