Web Services Resource Access distributed meeting

 

 

<dug> get a room

<gpilz> Zakim: too much information

<dug> wow - can zakim call everyone?

<Bob> scribe: Katy Warr

<dug> you can't run from the scribemaster forever

<dug> :-)

<asir> indeed Bob is tracking the number of hours :-)

<gpilz> nothing is certain but death, taxes, and scribing

<asir> :-)

AGENDA

No objections to agenda

Approval of minutes

No objections to accepting minutes of feb 10th

WG Admin

<dug> only 4 people will be there so far

<dug> s/4/5/ sorry

Action Item review

<gpilz> ?

<scribe> Chair: Mark Little has created action item list linked to Bugzilla

Bob: steping through the action items in tracker

Ashok: Tag issue may be better discussed at F2F

<scribe> Chair: That is our F2F?

Ashok: correct

<scribe> Chair: Agreed to discuss at our F2f

<scribe> Chair: Action 8 : 6543

Yves: Will complete in 2 weeks

<scribe> Chair: Action 9: 6548

Geoff: Not completed 48-52,
... will do proposals in next 2 weeks

<dug> +1 to Gil

<scribe> Chair: Action 15 Issue 6401

<gpilz> +q

Gil: Not intending to take up action of output ops with BP

Wu: Would like to see this being taken to BP

<scribe> Chair: Will you take this to BP?

Wu: I am not a member of BP

<scribe> Chair: If the issue is going to get to BP, someone should raise it. don't need to be a member of BP

<Geoff> +q

Dug: think we should close this as we have other action items 1. how do we advrt events 2. how bp compliant

<scribe> Chair: will return next week to discuss whether/not to close this

Approval of first public working drafts

<scribe> Chair: any further review comments that have not be incorporated into the documents

UNKNOWN_SPEAKER: and is everyone happy in taking to FPWG draft

Asir: There are some questions still open

<asir> here it is http://lists.w3.org/Archives/Public/public-ws-resource-access/2009Feb/0083.html

<scribe> ACTION: Yves to prepare FPWD boiler plate [recorded in http://www.w3.org/2009/02/17-ws-ra-minutes.html#action01]

<trackbot> Created ACTION-16 - Prepare FPWD boiler plate [on Yves Lafon - due 2009-02-24].

<dug> #6 too I think

<scribe> Chair: This will resolve point 3

Geoff: and 6

<Zakim> asir, you wanted to ask a question

Geoff: All other issues are addressed

<scribe> chair: we will take a look at boiler plate and consider fine tuning for publication next week

Acceptance of New Issues

<scribe> Chair: noticing that we have resolved 6391 but awaiting text for closure

Geoff: Only open issue for resolving refences is addressing only

<dug> RM is an example

<dug> Policy too I think

Geoff: but others need to work through for new standards

<asir> These are good ticklers for the WG!

Geoff: so I wanted a marker to remind us to go through all the other specs (more general issue)

<scribe> Chair: To save time any objection to 68-72 as new issues to address this?

RESOLUTION: No objections to 6568-72 to being opened as new issues

Geoff: Issue 6575

No Objections to opening 6575 and assigning to Geoff

RESOLUTION: 6575 opened

Geoff: Issue 6576

RESOLUTION: 6576 Opened

Geoff: Issue 6577

<dug> asir: makes sense to me

<dug> clearly an oversight

RESOLUTION: 6577 Opened

<asir> But WG members are required to only spend 20% of their time ...

Geoff: Issue 6578

<dug> just the slackers

<asir> we can only spend the time we have, nothing more

RESOLUTION: 6578 Opened

<dug> there are two new issues since the agenda was posted: 6587 and 6588

Geoff: Issue 6579

RESOLUTION: 6579 Opened

<scribe> Chair: putting new issues 6587 and 6588 on next week's agenda

Issue 6424

Geoff: Looking at PDF created (thank you)
... section 3.1 I would like to take an example

<dug> got a URL to the doc?

Geoff: in order to clarify my understanding

pls could someone post URL to doc?

<Bob> looking ...

<Bob> attached to http://lists.w3.org/Archives/Public/public-ws-resource-access/2009Feb/0055.html

thanks: o)

Geoff: section 3.1 says 0 or 1 WS-A EPRS

<dug> isn't the mapping in section 3.2?

Geoff: what does the this map to in the wsdl section 3.2
... First part of section 3.2 has endto tag. How does section 3.1 endto endpoint map to section 3.2 with this tag?
... Endto tag could have nothing inside it

<dug> I'm so lost as to what the confusion is about. [endto endpoint] maps to wse:EndTo

Wu: What is in the endto is endpoint which is of the Endpoint type

<asir> this sounds like a new formal representation on top of xml syntax ... need to better understand Doug

<asir> this is also different from the approach took by XML Protocol and WS-A WGs

Geoff: Concerned about inconsistency between infoset and XML
... The potential could be high

<dug> IMO if there is an inconsistency then the XML should win

Geoff: not that we have to change in 2 spots
... and check that they retain consistant

<asir> the XML Query WG had an automated tool to ensure that both the XML and formal rep stayed in sync

<asir> so did WSDL WG as well, they had a Z machine running to validate the contents

Geoff: people are more familiar with XML representation

Wu: xml is nomartively specified
... infoset specifies the semantics of the elements
... we normatively reference the infoset when used in different standards

Geoff: Is it not possible that the 2 normative references could get out of sync?

<gpilz> +q

Wu: The XML is seen as the normative reference for the specification

Geoff: I would like to see something to say specifically that the XML is the norm reference

Gil: Possible that infoset and XML out of sync
... isn't that a bug in our spec that we'd need to fix?

Geoff: Next point is interop
... if we only test the XML, are we really testing the infoset

Gil: Infoset is abstract property

Geoff: But we are not testing correctness of infoset

Gil: We can only test interop on things that come to the wire - i.e. XML
... Don't need to test other protocols which infoset mapping can map to

Geoff: ok
... Next issue: conformance
... what do we do to claim conformance if another protocol (e.g. carrier pigeon) maps to this infoset

Wu: Must map to concrete cases

<asir2> will the abstract representation be normative or non-normative?

Wu: and we added paragraph to clarify that you would have to define conformance for each case mapping to the infoset

<scribe> Chair: Are there any objections to accepting this proposal?

Geoff: I require another week to consider

Asir: Is the WG considering applying such changes to all the specs or just to eventing?
... this is large changes that need ot be carefully considered/reviewed

<scribe> Chair: Are you objecting to this idea due to the number of changes?

Asir: Not necessarily, I'd like to understand whether current thinking is one spec or all

<scribe> Chair: Wu?

Wu: I think this proposal will add value in terms of making the specs more applicable
... in particular eventing allows binding to different specs

<Geoff> +1 on that

Dug: I think we should retain consistency across specs - all or nothing
... actually, seeing what Wu has done with eventing, it's not a scarey as I expected

<scribe> Chair: Anyone hate the idea of this across all specs?

Asir: would like to think about it more

Geoff: agrees with Doug - all in or nothing

<scribe> Chair: Everyone agree in consitency?

No objections - all agree

<scribe> Chair: will revisit this issue next week

Issue 6426

<Bob> proposal at http://lists.w3.org/Archives/Public/public-ws-resource-access/2009Feb/0065.html

Li: This has gone back an forth on public list
... made some changes to the delivery element and schema
... had many discussions

<scribe> Chair: Any objections to proposal linked to link contained in comment 6?

RESOLUTION: No objections to Resolving Issue-6426 with the contents of Comment #6

<dug> 6404 might be ok

Issue 6428

<gpilz> +q

Li: Proposal that hasn't been integrated

<Geoff> +q

Li: yet that we could use element rather than attribute

Gil: really haven't agreed whether wrapped events should be covered
... however, I am happy to agree on this so long as we separate the concept of wrapped

Geoff: agreed: something like wrapped could be in a format section

<Geoff> +q

Doug: Wu and I perhaps go back and look at element vs attribute discussions

Geoff: please include me in that

Issue 6431

<dug> 6431 will not be quick

Li: Acked to add pause and resume to ws-eventing
... tried to retain current ws-eventing style
... key issue was how many msgs to maintain when pausing
... so we used an attribute in order to indicate how many messages you need to maintain

<Geoff> +q

Li: quite long because we tried to be very complete wrt examples

Doug: when retaining messages, is it expected that the event source will retain over system restart?

<gpilz> +q

Doug: i.e. is there gaurantee that messages wont be lost

Li: I would think that this was out of the scope of the eventing spec
... the gaurantee would be dependent on the implementation

Doug: We should put words to clarify this

Li: Agreed

Geoff: I am concerned about the complexity that this is added to the specification

Gil: I would like more time to think about this
... Need to ensure that proposal says something about pause and expiration time
... in particular, would expect expiration to cause msgs saved during pause to be lost
... also Doug's point is not specific to this issue

Doug: Overall I agree

<dug> Wu - do you have a URL to what you just referenced?

<asir2> what is 206?

<gpilz> a real URL would be appreciated

<gpilz> or at least enough to perform a decent search

<dug> http://www.ibm.com/developerworks/library/specification/ws-wsdmmgmt/

<dug> I tihnk

Wu: See feature for (link) about re: roadmap 206 web paper
... we need to standardise on this as it's part of roadmap and folk are already wanting to implement

<scribe> Chair: Are you Wu and Li clear on what's required to take this proposal to list?

Li and Wu to refine

Issue 6472

<dug> http://www.w3.org/Bugs/Public/show_bug.cgi?id=6472

RESOLUTION: Resolve Issue-6472 with Katy's suggestion 1 contained in bugzilla without objection