WS Policy WG
18 Apr 2007

Minutes Review

http://www.w3.org/2007/04/11-ws-policy-minutes.html

<scribe> chair: Minutes approved without objection

<cferris> RESOLUTION: minutes from april 11 approved

Future meetings

<scribe> chair: Confirmed meeting schedule in agenda

Editorial Team report

<monica> chris - please add me to the roll

Frederick: Still being worked.
... Action 277 is still being worked.

<cferris> acl pa

Frederick: Dashboard changes have been made.

paulc: Requests that editors should summarize changes to a batch of changes.

<paulc> and to identify any new best practises added especially if they are not in the original IBM/MST proposal

Action Item Review

Action 243 pending.

Asir: plan to send this material out in time for the next WG call on 25 April

<asir> URI to the latest SP is http://docs.oasis-open.org/ws-sx/ws-securitypolicy/200702

paulc: related to this, security policy is moving to committee spec

cferris: 274 summary is correct. Requests that Toufic please check his results.

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

<monica> scribe is asking a question

paulc: 0 for 0 is correct from HP.

cferris: HP all say TBD.

<cferris> ACTION: Prasad and Asir to scrutinize the HP results for UDDI Round 4 and resolve the discrepancies in results count [recorded in http://www.w3.org/2007/04/18-ws-policy-minutes.html#action01]

<trackbot> Created ACTION-281 - And Asir to scrutinize the HP results for UDDI Round 4 and resolve the discrepancies in results count [on Prasad Yendluri - due 2007-04-25].

cferris: action 274 done.

<cferris> http://lists.w3.org/Archives/Public/public-ws-policy/2007Apr/0060.html

<asir> Doug's response is at http://lists.w3.org/Archives/Public/public-ws-policy/2007Apr/0063.html

cferris: Action 274 will be discussed next week when Abbie is present.
... Action 277 pending.
... Action 278 done.
... Action 279 left open, but Chris and Paul have already engaged Bob Freund

<cferris> http://lists.w3.org/Archives/Public/public-ws-policy/2007Apr/0057.html

cferris: Action 280 done, on agenda.

<cferris> http://lists.w3.org/Archives/Public/public-ws-policy/2007Apr/0058.html

WS-Policy Call for Implementations, Chair

Abbie to check on fixed IP range visible for external use

paulc: Sent out a reminder to Abbie

<cferris> AI 276 http://lists.w3.org/Archives/Public/public-ws-policy/2007Apr/0060.html

cferris: relates to the negative cases; how they are detected
... no objections to closing this issue with the following resolution

<cferris> ACTION: Asir to open and close new issue on test scenarios related to negative testing and the proposal here http://lists.w3.org/Archives/Public/public-ws-policy/2007Apr/0060.html [recorded in http://www.w3.org/2007/04/18-ws-policy-minutes.html#action02]

<trackbot> Created ACTION-282 - Open and close new issue on test scenarios related to negative testing and the proposal here http://lists.w3.org/Archives/Public/public-ws-policy/2007Apr/0060.html [on Asir Vedamuthu - due 2007-04-25].

<cferris> RESOLUTION: Issue XXXX closed with resolution in http://lists.w3.org/Archives/Public/public-ws-policy/2007Apr/0060.html

CR Issues

Issue 4318 currently on hold.

paulc: WSSX making progress.

Liaison items

<cferris> http://lists.w3.org/Archives/Public/public-ws-policy/2007Apr/0058.html

<Zakim> asir, you wanted to say that Layer 7 should be 28 of 28 and to

asir: Action 271. Answer Ashok's question.
... discussed in some detail in the last WG meeting.
... Action 271, done.

http://lists.w3.org/Archives/Public/public-ws-policy/2007Apr/0017.html

<Symon> Symon Chang

ashok: likes what is written, but would like to see something a little more precise
... are you speaking about the vocabulary in normal or compact form?

asir: normal form
... current wording is at the data model level

<monica> queue please

??: At the data model level the XML structure is not discussed

<prasad> Definition: A policy vocabulary is the set of all policy assertion types of the assertions contained *directly* within a policy.

<prasad> http://lists.w3.org/Archives/Public/public-ws-policy/2007Apr/att-0058/action-270.html

<monica> use 'immediate'

<monica> or 'immediately contained'

monica: I think I understand the point being by Ashok.
... "immediate" or "immediately contained" could be helpful

cferris: there is agreement that four vocabularies are appropriate, but the document doesn't quite capture what we agreed on

<prasad> Immediate is more confusing than direct

cferris: "immediate" may not have a crisp definition either

tom: a policy vocabulary applies at the given level, independent of the parent potentially
... nested policies have there own independent vocabulary

maryann: why no defintion of nested policy vocabulary definition?

<TRutt__> the nested policy expression is its own policy which has its own policy vocabulary, indpenedent from that of its parent assertion's vocabulary

<TRutt__> s/indepenedent/independent/

<monica> +1

<monica> separate independently defined vs dependent in context

asir: Understands, and will add a sentence to clarify

<cferris> RESOLUTION: add a definition of nested policy vocabulary

<monica> +1 to definition

<maryann> i would like a DEFINITION

<maryann> i disagree

<maryann> its not a repetition

<maryann> it is a definition in context

<monica> if it was duplication we wouldn't be 'discussing' it and we wouldn't have confusion in other groups

<maryann> that's what makes it different

<maryann> that's why it needs its own defintion

<monica> queue please

dan: multiple definitions can lead to problems. Context plays a role, but we must make it clear that the definitions are equivalent

monica: Context plays a role in interpreting or understanding a policy vocabulary. We need clarification.

<danroth> What context are we talking about here?

maryann: nesting policy does provide context, but dependent and independent still presents challenges

<TRutt__> I agree with Monica that context of nested policy assertion types is important. Alternative F of WS addr resolution assumes that ws addressing without nested assertions means no replies type, with nested assetions to indicate support for particular response types, while alternative G has empty addressing assertion implying full response support, with nested policy assertion types for excluding some response types. The context decides what an empty (n

cferris: haven't found a resolution to this, so let's move this to e-mail to hammer out a proposal

<TRutt__> The context decides what an empty (no nested policy) ws addressing assertion means

cferris: Agreement that something crisp would be appropriate, and an example using Adressing would be helpful
... and we should also add the statement on the top-level, direct, immediate clarification

<TRutt__> The context of a top level assertion type is global, while the context of a nested policy assertion type is scoped within its parent assertion type. This is a significant difference.

asir: Second issue, the role of policy and vocabulary.
... the proposal is to clarify in the intersection section
... added a clarifying statement about the union of two compatible vocabularies added to section 4.5
... discusses only "will not be applied", not "negation"

Tom: nothing is not compatible with anything

asir: there are no semantics of the empty policy to the best of my knowledge

<TRutt__> ws addressing alternative G proposal has defined a semantic of empty ws addressing assertion meaning no restrictions on response types.

<asir> Great explanation Chris!

<cferris> maryann suggests that the example I gave would help elucidate the text proposed

<cferris> ACTION: Chris to send email with the example he gave discussing the role of policy vocabulary after intersection [recorded in http://www.w3.org/2007/04/18-ws-policy-minutes.html#action03]

<trackbot> Created ACTION-283 - Send email with the example he gave discussing the role of policy vocabulary after intersection [on Christopher Ferris - due 2007-04-25].

dale: what would happen to the intersection if we were to drop negation?

cferris: yes, you correct. The negation doesn't affect the intersection itself. It does affect the application of the compatible policy.

<monica> differentiate 'is not done' vs 'restricted from doing so'

<cferris> 1. new proposal(s) for policy vocabulary and policy alternative vocabulary

<cferris> 2. proposal(s) for definition of nested policy vocabulary

<monica> i'm here

<cferris> 3. monica raised question of context

<cferris> 4. example using the addressing policy pointing out the four vocabularies

<cferris> 5. example of post intersection policy vocabulary

<cferris> 6. significance of empty nested policy

<maryann> this is the text from the email:

<maryann> Note: The consensus of our group was that "absence means negation"

<maryann> convention should be abandoned.

<maryann> [Relates to CWA - see 2] Need to explore to see if this could affect

<maryann> wsp:Optional and may

<maryann> require that policy assertions are explicitly stated. Doing away with

<maryann> this assumption may result in the least change needed. This would

<maryann> necessitate that

<maryann> authors define what empty means if empty is allowed. Eliminating this

<cferris> 04 016. significance of empty nested policy must be defined in its parent's definition

<maryann> assumption implies

<maryann> that authors define that any policy instance must consist either of

<maryann> mutually

<maryann> exclusive alternatives or an empty alternative and they explain the

<maryann> semantics of the empty

<maryann> assertion.

<monica> can we have a mini-task force?

<monica> add monica and moberg

<maryann> i will volunteer for 5 or 6

<monica> i'll take 3 and combine with 5 and 6

<cferris> ACTION: Asir to start email thread to develop proposal for def'ns of policy vocabulary and policy alternative vocabulary [recorded in http://www.w3.org/2007/04/18-ws-policy-minutes.html#action04]

<trackbot> Created ACTION-284 - Start email thread to develop proposal for def\'ns of policy vocabulary and policy alternative vocabulary [on Asir Vedamuthu - due 2007-04-25].

<asir> 1 and 2 go together

<cferris> ACTION: Monica to initiate thread related to the role of context for nested policy [recorded in http://www.w3.org/2007/04/18-ws-policy-minutes.html#action05]

<trackbot> Created ACTION-285 - Initiate thread related to the role of context for nested policy [on Monica Martin - due 2007-04-25].

<asir> 1, 2 and 4 go together

<cferris> ACTION: Maryann to work with Tom and Monica and anyone else interested in developing assertion author guidelines related to nested policy vis a vis this new understanding of nested policy vocabulary [recorded in http://www.w3.org/2007/04/18-ws-policy-minutes.html#action06]

<trackbot> Created ACTION-286 - Work with Tom and Monica and anyone else interested in developing assertion author guidelines related to nested policy vis a vis this new understanding of nested policy vocabulary [on Maryann Hondo - due 2007-04-25].

<monica> yes please

<monica> thank you

<monica> sorry in oasis symposium

paulc: Action 280 will be put at the front on next week's agenda