W3C

WS-Policy teleconference

21 Feb 2007

See also: IRC log

agenda

Attendees

Present
Frederick_Hirsch, Chris_Ferris, Yakov_Sverdlov, Sergey_Beryozkin, Dale_Moberg, Plh, Fabian, Paul, m2, Mark_Little, Prasad_Yendluri, Toufic, Sanka, Abbie_Barbir, Symon, maryann, Charlton, Ashok_Malhotra, Asir, Dave_Orchard, Glen, Umit, Tom_Rutt
Regrets
Felix
Chair
Chris
Scribe
Dave

Contents


Philippe: felix will be unavailable for next 2 weeks

previous minutes

resolution: minutes of Feb 14th approved

July f2f meeting

<paulc> Logistics page: http://lists.w3.org/Archives/Member/member-ws-policy/2007Feb/0040.html

<paulc> Change F2F date proposal: http://lists.w3.org/Archives/Member/member-ws-policy/2007Feb/0038.html

paulc: any objections to moving to july 17-19th

wg: no objections

RESOLUTION: july f2f on July 17th-19th

Paulc: need to update admin page, w3c calendar, and logistics page

ACTION: Philippe to put up logistics page for Dublin F2F [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action02]

<trackbot> Created ACTION-225 - Put up logistics page for Dublin F2F [on Philippe Le Hegaret - due 2007-02-28].

editors report

<cferris> ACTION: Philippe to update member page to reflect the change in F2F dates and to change the cancelled meeting from Aug 1 to Jul 25 [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action03]

<trackbot> Created ACTION-226 - Update member page to reflect the change in F2F dates and to change the cancelled meeting from Aug 1 to Jul 25 [on Philippe Le Hegaret - due 2007-02-28].

toufic: nothing to report

<paulc> ACTION: 189 to Logistics page: http://lists.w3.org/Archives/Member/member-ws-policy/2007Feb/0040.html [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action04]

<prasad> AI-211 done http://lists.w3.org/Archives/Public/public-ws-policy/2007Feb/0106.html

<paulc> ACTION: 211 to 20007Feb/0106.html [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action05]

<paulc> WSDL WG confirmation: http://lists.w3.org/Archives/Public/public-ws-policy-comments/2007Feb/0002.html

<paulc> Proposed feddback to WS-A: http://lists.w3.org/Archives/Public/public-ws-policy/2007Feb/0110.html

feedback to ws-addressing

chrisf: goes throuh his email.

<cferris> ACTION: Chris to add issues for attachments, guidelines and primer related to the changed qname of the ws-addressing metadata assertion [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action06]

<trackbot> Created ACTION-227 - Add issues for attachments, guidelines and primer related to the changed qname of the ws-addressing metadata assertion [on Christopher Ferris - due 2007-02-28].

fabian: fine with recommended approach
... reasoning in points a&b I don't agree with
... fine with points c,d,e

<Fabian> don't agree with: Similarly,

<Fabian> the second alternative makes no statement what-so-ever.

fabian: it does make a statement, that ws-addressing is required.

chrisf: maybe change to say the nested policy assertion makes no statement

umit: if that cleared up in a), would you be ok?

fabian: b) two policies, one for client, one for server, both use optional nested assertions.
... it says "the following two policies are compatible .." and I can't follow what it's saying
... of course they are compatible because it says optional.

chrisf: they are almost recommending client side use optional but not server.

chrisf: this is an abominable use of optional.

fabian: but of course this is what optional is intended for.

chrisf: If I'm offering policy, I only support anonymous or non-anonymous, there's no way to say this.

scribe lost.

umit: problem with optional isn't with optional, it's since the assertion semantic is support, then when you use support with optional.
... the semantic of the nested assertion does not end up helping because of a)

<cferris> if either side decides (for whatever reason) to use the guidance that is in the metadata spec (that wsp:optional be used to circumvent the problem that the corresponding endpoint might not specify its support for the capabilities) is misguided because it COULD lead to the example that we cite

umit: this is a misuse of optional because of the support semantics

chrisf: they are advocating optional .... scribe lost again.
... could have intersection come back and say intersection is found when it actually isn't.

<asir> Here is the conclusion in WS-A Metadata doc - "These two examples show the use of wsp:Optional and wsp:Ignorable, and how they can be used to produce non-empty intersections between client and endpoint policies. "

umit: b) is a consequence of a)

<cferris> right... this is the guidance with which we have an issue

umit: all problems from support semantics.

paulc: need to concentrate on b)'s point isn't explained well enough.

fabian: 3.1.6 paragraph.
... perhaps expand the intents?

<TRutt_> problematic sementics of B ) relates to example 3-11 in metatdata doc

chrisf: can't express the kinds of things they want to express

scribe lost again.

<umit> that is it

paulc: they can't say non anonymous response not supported.

<umit> Non anon response is not supported is not expressable.

<cferris> exactly, you cannot say that you do not support a feature

<umit> it implies resorting to a runtime fault

to say anonymous response but not non-anonymous response not supported.

<TRutt_> +1 to Paul but also relate B ) point to their example 3-11

<monica> no behavior

glen: if they define 2 assertions, and don't use 1 of them, doesn't it say that 1 isn't supported

<asir> Glen, that particular contradicting absence statement is under point A)

<umit> The sections in 3.1.2 and 3.1.3 are exactly the problematic sections

<umit> theyare in A)

<GlenD> gotcha

<umit> I must observe that we clarified the presence/absence of assertions only recently

<umit> also we made a lot of clarification on optionality discussion and intersection recently

<maryann> +1 to anti-practice example

<maryann> in guidelines

<umit> thus, it is possible for WS-A to come up with this approach. I bet they will go for option 2 to preserve their semantics.

dorchard: it seems to me that ws-a really did a mistake.
... and they are really smart folks
... so this is really concerning

trutt: this came about as a result of ws-rx comments, so you could have 2 statements to support ws-a anon and ws-rm make connection,
... we should focus on 3.11

dorchard: I'd almost want this to be an anti-practice in the main doc.

trutt: 4 month back and forth, focusing on policy

<umit> I must observe the use of optional and ignorable is rather new!

trutt: could use non anon and anon responses as policy assertions instead of nesting, that might work.

<umit> not 4 months

trutt: have to understand ws-rm spec

<asir> umit, did the WS-A group debate on the use of optional/ignorable?

trutt: may involve change to ws-rm as well.

<maryann> can we get WS-A to give us feedback on the guidelines doc?

<maryann> it would be good to get their comments on its applicability

umit: might be 4 months of discussion, but on alignment with rm(?)
... some people have been sending feedback on intersection but haven't gotten to this level of review and response yet.

paulc: point about little of 4 month discussion is on using w3c ws-policy

<Zakim> dorchard, you wanted to suggest 3-way meeting.

<asir> ws-a didn't spend sufficient time to discuss the use of ignorable/optional or the absence of assertions in a nested policy expression within addressing assertion

<umit> yes

<umit> +1 to asir

<cferris> +1

<toufic> +1

dorchard: suggest a 3 way meeting

<Yakov> +1 to David

trutt: this was very complex, and didn't end up right.

<umit> there are two issues here, how to handle WS-A specifically and how to illustrate best practices so that others use WS-P in the right way.

paulc(with no chair hat): many of the companies have reps on all 3 groups

paulc: disingenuous to suggest we need a f2f
... rx doesn't even have review of ws-a metadata on the agenda

<umit> Today, support semantics is also prelevant with WS-RX as well with MC!

<TRutt_> ws RX group acked the wsa approach by copying it int their own make connecti policy assertion as a support

paulc: 2 questions: 1) with the amendment that fabian suggests, is the WG ready to lob this.

2) then are we willing to step to the plate to facilitate better communications.

<TRutt_> we should provide PR comment on WS RX make connection spec

<umit> I know Tom, that is why I am concerned. It is again another alignment issue.

umit: Is the guideline or anti-pattern and solution really clear? there are 2 solutions listed.

paulc: when you say option 1), what do you mean?

umit: the first solution in the task force email.
... problem also happening with ws-rx.

trutt: ws-rm was aware of what ws-a did, and copied their mistake.

paulc: if we reviewed ws-rm, we'd have the same comment.
... we should review ws-rm policy as well?
... now understanding daveo's point about the collateral damage from 3-way, and understand the strength of the point even more.
... 1) want to get agreement today.
... 2) we should also send a message to rx pointing to message to rm saying you should be interested in this.

<TRutt_> It is the wsrx make connection spec not the wsrm policy spec

paulc: 3) figure out how to get together with other 2 wgs.

<monica> can we do this real time?

plh: In parallel, I could invite the WS-RX chairs to WS-CG meeting to help the dialogue. Next meeting is March 6.

paulc: need to talk with them really fast
... we might get critical mass for next f2f.

<asir> david owns 'anti practice' within the policy wg

umit: what do we need to do within wg?

trutt: clarify that it's mc policy spec, not rm policy spec

<TRutt_> http://www.oasis-open.org/apps/org/workgroup/ws-rx/email/archives/200702/msg00046.html

<TRutt_> The link I posted is the public review notice for three wsrx specs

asir: we can review their metadata docs

<TRutt_> It is wsmc spec which has the troublesome policy asserttion definition

<TRutt_> pr for wsrx spec closes 27 feb

paulc: let's not send a preliminary email to them.

<cferris> ACTION: Chris and Umit to revise WG draft comments to WS-A WG [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action07]

<trackbot> Created ACTION-228 - And Umit to revise WG draft comments to WS-A WG [on Christopher Ferris - due 2007-02-28].

<asir> [on the public mailing list]

<cferris> ACTION: Fabian and other WG members to reply in timely manner to the revised draft from TF [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action08]

<trackbot> Created ACTION-229 - And other WG members to reply in timely manner to the revised draft from TF [on Fabian Ritzmann - due 2007-02-28].

<cferris> ACTION: Chris and Paul to send resulting material to the WS-A WG by Feb 23 after reaching consensus within the WG [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action09]

<trackbot> Created ACTION-230 - And Paul to send resulting material to the WS-A WG by Feb 23 after reaching consensus within the WG [on Christopher Ferris - due 2007-02-28].

<cferris> ACTION: Chris and Paul to inform the WS-RX TC of our comments on WS-A Metadata [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action10]

<trackbot> Created ACTION-231 - And Paul to inform the WS-RX TC of our comments on WS-A Metadata [on Christopher Ferris - due 2007-02-28].

<TRutt_> It might be good to send a PR comment to wsrx stating that the make connection policy assertion definition is problematic, with a link to our CR comment on WSA - PR closes feb 27

<cferris> ACTION: Chris and Paul to open dialog with WS-A and WS-RX TC chairs to find a way forward to address the problems between the three groups [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action11]

<trackbot> Created ACTION-232 - And Paul to open dialog with WS-A and WS-RX TC chairs to find a way forward to address the problems between the three groups [on Christopher Ferris - due 2007-02-28].

4232

CR interop scenarios and testing

<scribe> ACTION: asir to update features and scenarios table in document [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action12]

<trackbot> Created ACTION-233 - Update features and scenarios table in document [on Asir Vedamuthu - due 2007-02-28].

paulc: what should we be doing at the march f2f?

chrisf: we have a long list of guidelines and primer issues..

paulc: interop..

asir: high level overview of scenarios and explain.
... then what we can realistically achieve..

<Fabian> http://dev.w3.org/cvsweb/~checkout~/2006/ws/policy/interop/WS-Policy-Scenarios.pdf

<asir> link to interop directory on CVS is http://dev.w3.org/cvsweb/~checkout~/2006/ws/policy/interop/

asir: table outlines features and how covered.
... round 1 unit test cases, normalize, merge, includes expected outcomes.

<paulc> Web Services Policy 1.5 Features and Interop Scenarios: http://lists.w3.org/Archives/Public/public-ws-policy/2007Jan/att-0143/ws-policy-features-01-15-2006.pdf maps features to the various interop rounds.

asir: round 3 tests: interop using security policy assertions, wsdl 1.1 and 2.0 attachment
... round 4 tests: external policy attachment, uddi, application/xml+ws-policy

paulc: interop test should be rnd 1 and rnd 2 before f2f, then round 3 at f2f.

<prasad> webMethods expects to be ready for doing all the UDDI scenarios

chrisf: sounds reasonable

paulc: ashok, you asked question about what are we really going to do.

<prasad> Prasad: Any objections doing UDDI scenarios at March F2F?

<prasad> Paul: No. I am trying set a mim bar

asir: msft will be ready to round 3 testing

symon: bea will be ready for rnds 1-3

<prasad> Prasad: As long as there are no objections to beyond round 3 happening, that is good

maryann: ibm will ready for 1-3

ashok: oracle should be ready for 1-3, not quite sure

toufic: layer-7 pretty ready for rnd 4, not sure about 3.

<asir> Toufic, that is great!

<prasad> Paul: Prasad would be happy about round 4

<prasad> Toufic: Prasad and I had been talking offline about round 4 (UDDI)

fabian: sun attending in person, rnd 3 100%, not sure about self-test 1-2 yet

<sanka> WSO2 is expecting to do Round3 testing remotely ..

<cferris> ACTION: Asir to provide format for recording interop results [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action13]

<scribe> ACTION: asir to provide results matrix [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action14]

<trackbot> Created ACTION-234 - Provide format for recording interop results [on Asir Vedamuthu - due 2007-02-28].

<trackbot> Created ACTION-235 - Provide results matrix [on Asir Vedamuthu - due 2007-02-28].

paulc: what about xml:id, xml:base, ?
... we have 5 concrete proposals we should put on the agenda.

Summary of Action Items

[NEW] ACTION: 189 to Logistics page: http://lists.w3.org/Archives/Member/member-ws-policy/2007Feb/0040.html [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action04]
[NEW] ACTION: 211 to 20007Feb/0106.html [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action05]
[NEW] ACTION: Asir to provide format for recording interop results [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action13]
[NEW] ACTION: asir to provide results matrix [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action14]
[NEW] ACTION: asir to update features and scenarios table in document [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action12]
[NEW] ACTION: Chris and Paul to inform the WS-RX TC of our comments on WS-A Metadata [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action10]
[NEW] ACTION: Chris and Paul to open dialog with WS-A and WS-RX TC chairs to find a way forward to address the problems between the three groups [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action11]
[NEW] ACTION: Chris and Paul to send resulting material to the WS-A WG by Feb 23 after reaching consensus within the WG [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action09]
[NEW] ACTION: Chris and Umit to revise WG draft comments to WS-A WG [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action07]
[NEW] ACTION: Chris to add issues for attachments, guidelines and primer related to the changed qname of the ws-addressing metadata assertion [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action06]
[NEW] ACTION: Fabian and other WG members to reply in timely manner to the revised draft from TF [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action08]
[NEW] ACTION: Philippe to put up logistics page for Dublin F2F [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action02]
[NEW] ACTION: Philippe to update member page to reflect the change in F2F dates and to change the cancelled meeting from Aug 1 to Jul 25 [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action03]
[NEW] ACTION: PLH to put up logistics page for Dublin F2F [recorded in http://www.w3.org/2007/02/21-ws-policy-minutes.html#action01]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.127 (CVS log)
$Date: 2007/02/21 23:20:16 $