WS Policy WG
28 Mar 2007

<paulc> Agenda: http://lists.w3.org/Archives/Public/public-ws-policy/2007Mar/0117.html

Roll call and selection of secretary, Chair

agenda v3 at http://lists.w3.org/Archives/Public/public-ws-policy/2007Mar/0117.html

Review and approval of WG minutes, Chair

http://www.w3.org/2007/03/13-ws-policy-minutes.html

http://www.w3.org/2007/03/14-ws-policy-minutes.html

RESOLUTION: minutes adopted

Future WG meetings

Wed-Fri May 23-25 F2F Ottowa, 17-19 F2F meeting in Dublin

Paul: Ashok wanted to confirm the meeting plans for Ottowa
... plan is to meet for 3 days.
... we will use all 3 days for interop. We might finish the WG meeting earlier
... if we have a shorter meeting for the WG, that meeting should be the first 2 full days. Interop would still be 3 days

ashok: if possibly for 1/2 days? last time we ended at lunch of the second day

paul: I propose 2 full days.
... for WG, and 3 days for interop. Meeting time would be 9-5
... abbie, is that fine?

Abbie: ok

paul: we will still need the interop room for friday

<whenry> dinner wed +1

paul: I will move the WG dinner from Thursday to Wednesday

<scribe> ACTION: Felix to update the logistics and admin page with the new Ottowa meeting schedule [recorded in http://www.w3.org/2007/03/28-ws-policy-minutes.html#action01]

<trackbot> Created ACTION-261 - Update the logistics and admin page with the new Ottowa meeting schedule [on Felix Sasaki - due 2007-04-04].

Editorial team report

Asir: sent out diff versions with some false alarms, Felix made a stylesheet to fix these
... sent out the updated interop scenarios as well
... third item: negative test cases
... we created a subdirectory called negative test under round4
... two actions for editors: make the archive available, we created a sticky tag
... another one: editors to double check that the cross-references from attachment doc to "wsdl 11 ei" doc are up to date

paul: we already have approval to publish wsdl11ei draft
... no objection to re-publish framework, attachment, primer and guidelines drafts?

RESOLUTION: WG agrees completely to re-publish framework, attachment, primer and guidelines drafts together with wsdl11ei drafts

<asir> Thank you Felix for all the great work!!

<cferris> +1

the WG thanks the editors and Felix for the work on preparing the documents

<scribe> ACTION: pcotton2 and Felix to send a note to the chairs list (esp. the usual liaison WGs) about the re-publication of the framework and attachment docs [recorded in http://www.w3.org/2007/03/28-ws-policy-minutes.html#action05]

<trackbot> Created ACTION-262 - And Felix to send a note to the chairs list (esp. the usual liaison WGs) about the re-publication of the framework and attachment docs [on Paul Cotton - due 2007-04-04].

action items

ACTION-242 Ashok to provide expected outcome docs for Feature 23 test cases

<scribe> done

ACTION-243 Asir to produce a list of change areas in the framework and attachments primer and guidelines specs related to a change in the reference to the oasis ws-sec policy spec to reference the latest available published committee draft

<scribe> pending

ACTION-248 David Orchard start e-mails on 4300

moot

dorchard: I respond to Umit explaining what we did and mark the action as closed

ACTION-250 Asir to have timestamps on document and stable packages for release

<scribe> done

ACTION-251 Editors to work out a process for making the interop scenario/test cases available as a "package" so that we have an ED and WD versions

<scribe> done

ACTION-252 Fabian to open new follow-up issue to 4370

<scribe> pending

<cferris> pending

ACTION-253 Chris Ferris to convey the disposition of WSDL WG comments to the WSDL WG

ACTION-255 Charlton Barreto to open new primer issue to relate to the discussion of 4292

<scribe> done

ACTION-256 Monica to create new primer issue related to 4262 Monica Martin 2007-03-21

<scribe> done

ACTION-257 Felix to publish revised WSDL1.1 Identifiers document as WD

paul: Felix will close this after publication

ACTION-258 Editors to double check that the republished Attachments document reflects the latest WSDL1.1 Identifiers draft

<scribe> done

ACTION-259 Maryann Hondo to provide a revised proposal for Issue 3978.

<scribe> done

ACTION-260 Maryann Hondo to provide revised proposal for 4074

<scribe> pending

Liaison items

paul: ongoing discussion in addressing wg
... no need to do s.t. about 4374, 4375, and 4376 today

Maryann: status in addressing wg AFAIK:
... WG will vote on proposals on Monday and plans to republish

<cferris> I spoke with Bob Freund... he plans on ensuring a vote on monday on accepting one of option F or G

paul: this is a last call, correct? They will have to have another last call?

<cferris> I will confirm with Bob whether they plan a LC

<scribe> ACTION: cferris to contact addressing WG to find out their schedule regarding policy assertion [recorded in http://www.w3.org/2007/03/28-ws-policy-minutes.html#action06]

<trackbot> Created ACTION-263 - Contact addressing WG to find out their schedule regarding policy assertion [on Christopher Ferris - due 2007-04-04].

paul: would help to know their schedule and how it intersects with ours for May
... one if the issue 4374, 4375, and 4376 is an issue for the policy specs, so we can't go out of CR before they are resolved

WS-Policy Call for Implementations

<paulcotton> dashboard: http://dev.w3.org/cvsweb/~checkout~/2006/ws/policy/interop/results/dashboard-summary.html?only_with_tag=Release-1-20070322

(paul summarizes the results)

paul: there will the negative tests be in the dashboard

asir: will be a separate column in round 4

<scribe> ACTION: asir to update the dashboard with negative tests [recorded in http://www.w3.org/2007/03/28-ws-policy-minutes.html#action07]

<trackbot> Created ACTION-264 - Update the dashboard with negative tests [on Asir Vedamuthu - due 2007-04-04].

<scribe> ACTION: cferris to update the dashboard summary with negative tests [recorded in http://www.w3.org/2007/03/28-ws-policy-minutes.html#action08]

<trackbot> Created ACTION-265 - Update the dashboard summary with negative tests [on Christopher Ferris - due 2007-04-04].

<cferris> I haven't had a chance to look deeply into this

paul: on the output of the dashboard summary: e.g. in round 3 WSDL 1.1, there are different total numers of tests. Could Chris look into this?

<asir> I think the details page is upto date - http://dev.w3.org/cvsweb/~checkout~/2006/ws/policy/interop/results/dashboard.html

<cferris> the results SHOULD be reported in terms of N out of M completed (reported) tests... a test result can have N/A in it and that does not count against (like SAT results)

<scribe> ACTION: cferris to check the dashboard summary representation to make sure they are properly corrected [recorded in http://www.w3.org/2007/03/28-ws-policy-minutes.html#action09]

<trackbot> Created ACTION-266 - Check the dashboard summary representation to make sure they are properly corrected [on Christopher Ferris - due 2007-04-04].

<cferris> what I will have to look into is whether the results would be mis-reported if the complete set of tests were not reflected (e.g. those unattempted tests were simply omitted, rather than reported as n/a)

paul: on the call for implementation: we are expecting 2 or 3 more WG members and also outside participants to step forward
... want to make sure that we can move forward even without a lot of WSDL 2.0 implementations. That's one reason why I expected interop in May to last for 3 days

CR issues

Issue 4318: [All docs] Which WS-SecurityPolicy version should be used as a reference

paul: issue is on hold

ACTION-252 Fabian to open new follow-up issue to 4370

paul: this is open

fabian: will do this by the next call
... can promise Tuesday night

4394 Type of the WSDL 2.0 {policy} property?

Asirs reply at http://lists.w3.org/Archives/Public/public-ws-policy/2007Mar/0116.html

Asir: have not seen a reply from the commentor to my mail

paul: Asir, do you propose any changes in the CR docs?

Asir: no

paul: we could close this without any changes to the attachemnt doc.. Any objections?

RESOLUTION: 4394 resolved without any changes in the document and the clarification in http://lists.w3.org/Archives/Public/public-ws-policy/2007Mar/0116.html

<scribe> ACTION: Asir to convey resolution of 4394 to the commentor [recorded in http://www.w3.org/2007/03/28-ws-policy-minutes.html#action10]

<trackbot> Created ACTION-267 - Convey resolution of 4394 to the commentor [on Asir Vedamuthu - due 2007-04-04].

paul: Asir, please also update Bugzilla and ask the person to respond if they object

comment from chris

paul: chris made a comment saying "the could say 'the type of the policy propert' instead of 'the value of the policy property'

asir: we should use the same terminology as wsdl 2.0 They use the word "value"

<Fabian> I will need to leave in 10 minutes.

chris: we should also say that our text is aligned with WSDL 2.0 and we should say that to the commentor as well

paul: OK

Primer issues

<charlton> http://lists.w3.org/Archives/Public/public-ws-policy/2007Mar/0124.html

paul: 4393 should be processed together with 4262, Monica proposed

http://lists.w3.org/Archives/Public/public-ws-policy/2007Mar/0124 , introduced by Charlton

frederick: how should this be changed

charlton: I made a proposal to the whole first part of that section

frederick: not sure why the provider may intend what you describe

charlton: provider can't force the consumer to have a mode, but if there is a failure, a consumer could apply different intersection modes in order to pinpoint the failure

frederick: announcing what should happen is a different issue than the possibility to use different modes

charlton: the fact that alternatives are not provided because the provider has specific settings

paul: frederick, did you have time to look at this proposal

chris: +1 to fredericks comment
... don't think this has anything to do with ignorable

charlton: the paragraphs I am proposing to not discuss ignorable

<fhirsch3> so maybe this generic material does not belong in the ignorable section.

charlton: they address the more generic case and touch on Monicas inital action item

<fhirsch3> Likewise agree with Chris that provider should not expect consumer to use certain modes

chris: the client can decide what modes to choose
... the provider can make no assumption about that

<fhirsch3> +1 to Chris

chris: ignorable has nothing to do with what the provider does about the intersection modes
... some implementations may choose just to implement strict mode

<prasad> +1 to Chris

<whenry> +1 to Chris

asir: first point "a provider may influence the mode choice of a requester"

charlton: I'm not saying that

asir: opening statement in the proposal is not providing the context

frederick: the text is not in the right section

<cferris> +1 to frederick

frederick: we might want to have a section in general about "what happens if s.t. goes wrong?", and we may want to have this material in that section

<cferris> 06notes that this is implementation detail... we could get really hand-wavy about this, but would prefer not to make any specific suggestions, even in the primer01

Monica: you have the description of mode and the treatment of ignorable in one section
... if the mode is only with respect to ignorable, we should consider restructuring the section accordingly

<charlton> i would rather make a suggestion, which is rather broad and open-ended, to how to deal wiith situations in particular dealing with such 'conflicts'

Monica: the question is: is mode only relevant for ignorable?

<charlton> this is in itself 'hand-wavy'

paul: currently we don't have the proposal for resolving this today

<fhirsch3> I think the question is whether we need material on diagnosing unexpected results in primer, guidelines

paul: charlton, please reply with a revised proposal using the feedback from today

<charlton> there are two questions

paul: using the march http://lists.w3.org/Archives/Public/public-ws-policy/2007Mar/0124 track

Monica: should we still address http://lists.w3.org/Archives/Public/public-ws-policy/2007Mar/0125 ?
... 125 gives the reference to the action item

<charlton> how to diagnose unexpected results, which are in particular exposed by 'conflicts' in intersection mode

paul: keep both actions linked together and make progress

<charlton> intersection mode 'conflicts' highlight a scenario where results that are unexpected are particularly difficult to diagnose

paul: going on to http://www.w3.org/Bugs/Public/show_bug.cgi?id=4414
... Maryann introduces the issue

<whenry> Whover just jointed needs to mute

dorchard: trying to find the differences, need more time to look at this

asir: a diff would be good to see changes
... Maryann talked about two parts. Part one is ok
... on part two: (sorry, scribed missed)

Maryann: if you have a versioning assertion, if you change the value of the assertion, that is one use case
... if it is not deployed yet, you introduce a new assertion. We have to be more explicit about this example

asir: the assertion here is hypothetic

Maryann: I want to know what the use case is
... if it is a valid use case, we should add it to the test case

dorchard: I don't understand why it is unclear and why the hypothetic assertion is confusing
... don't see the relation between versioning scenario and test cases

asir: my third point: I see a contradiction between various sentences

Maryann: again the use case was not clear to me
... the use case(s) don't give an example

paul: I think we should take this to mail
... Maryann, diff markup would help e.g. in an attached PDF

dorchard: if it is easy to do, that would be good

Maryann: will resent the proposal with diff visualization

Asir: I will post my comments as well

Guidelines issues, Part 1

<scribe> postponed, we do 11a) first

Guideline issues, Part 2 (agenda item 11a)

<paulcotton> http://lists.w3.org/Archives/Public/public-ws-policy/2007Mar/0069.html

paul: have seen no support for this except IBM and MS
... discussion: does the WG want to do this revision?

dorchard: I support this

frederick: supportive, seems to be a lot of work

paul: anybody against the proposal?

<abbie> +1

paul: now second step: how to do this?
... many good practices correlate directly with material in the document
... possible way to go forward: distribute among WG participants material to be rewritten as good practices sections
... but this is a lot of work, as Frederick noted, so any comments on how to get this done?

Asir: at the f2f we had 4 BP in the guidelines doc
... the 4 BP are highlighted in the changes section

paul: we could defer this to the editors
... but we need a work plan on how to let this happen

<sanka> seems that didnnoticed the time difference

<sanka> seems that I didn't notice the time difference ..

Frederick: will the WG go through the BP in the suplied PDF before we put them in the doc?

paul: I thought people could take 5-6 of the BPs and get going
... if the editor's see that this does not work they can come back to us

<scribe> ACTION: editors to think about proposal from paul: taking a subset of the BPs and see how long it will take them to give revised sections back [recorded in http://www.w3.org/2007/03/28-ws-policy-minutes.html#action11]

<trackbot> Created ACTION-268 - Think about proposal from paul: taking a subset of the BPs and see how long it will take them to give revised sections back [on Editors - due 2007-04-04].

Guidelines issues, Part 1

paul: AI 260 for issue 4047 is pending, on issue 3978 there was no response so far

asir: 3978 is a fairly long proposal, could Maryann do a diff document?

Maryann: yes

Guideline issues, Part 2 (issue 3988)

paul: will be done after progress on 3989

summary of todays outcome

paul: agreement to publish 5 drafts and f2f dates
... resolved one issue
... decided to have diff documents for others

Felix: thanks to paul / MS to provide the bridge today!