See also: IRC log
<scribe> scribe: prasad
Wed Morning - Charlton
Wed afternoon - Maryann
Thu Morning - Ashok
Thu Afternoon - Monica
<cferris> RESOLUTION: minutes from May 9 approved http://www.w3.org/2007/05/09-ws-policy-minutes.html
Paul: Information on Wed dinner
on the F2F logistics page. Hosted by Paul
... No meeting following the F2F on May 30th
Dave: Finished Primer, sent to group
Monica and I have a proposal for editorial changes
Paul: We will open an issue and proces it
Dave: Guidelines, no final plan yet. All open AIs may not be covered yet. Will circle back with editorial team
Paul: What have you done?
Dave: I am not sure what state we are in. There have been some late ckins. Will have better idea after editors meeting
Paul: Will have time at the F2F,
to work with Editors Draft given
... Even if editors are reluctant to call it Editors Draft
... one or two of the editors should step through the guidelines doc
... in front of the WG. May allocate 1/4 day or so
ACTION-279 - Still in Progress
ACTION-286 - On going
ACTION-292 - Done (message May 171)
ACTION-294 - Done (issue 4552)
ACTION-295 - Done (msg 152 issue 4556)
ACTION-297 - Done
ACTION-298 - Done (msg 170 issue 4558)
ACTION-299 - Done
Asir: No responses to thread
<PaulC> Welcome Toufic, We are on agenda item 6
Asir: this text can be appended to section 4.5. I provided explanation of why the policies do not intersect
<PaulC> Proposal is for addition to Section 4.5: http://www.w3.org/TR/2007/CR-ws-policy-20070330/#Policy_Intersection
Asir: illustrates two policies whose intersection contain no policy alternatives
A5 and A6
scribe: as the assertions in them are incompatible
<cferris> Therefore, the two assertions are incompatible and hence the two alternatives are incompatible.
chris: suggests changes to the last paragraph. Replace entire paragraph with the above sentance (?)
<PaulC> Chris proposes to replace the last paragraph: The two assertions in alternatives A5 and A6 are incompatible because their nested policy expressions are incompatible. The two alternatives A5 and A6 are incompatible because the assertions within these two alternatives are incompatible. The above two policies are incompatible because there are no compatible alternatives.
<PaulC> With his proposed text.
Asir: If I drop all the "why"s your proposal and what I have will look the same
Monica: If we make it detailed it
might make it difficult on domian specific processing. I
thought therse are just examples
... too much detail in the examples
Ashok: I looked at 4.5. 2nd para, 2nd half of 1st sentce has "determing two alternatives are compatible may involve domain speciifc procesing.". LAter it says approximation. Which is the priority?
Paul: No where you mentioned the example, which is what we are talking about
Ashok: I am asking a higher level q?
Paul: You just restated 2nd para in 4.5
Frederick: I thought yiou had an
empty policy, it matched everything
... why the intersection is empty?
Asir: Last week we discussed why
they should not intersect
... per the FWK they don't intersect
<PaulC> Dan's reply: http://lists.w3.org/Archives/Public/public-ws-policy/2007May/0009.html
Paul: last week people agreed with Dan's explanation but did not want to close the issue until we had concrete examples
<asir> dialing back in
<FrederickHirsch> if you need integrity on a message exchange but do not care how, then you should be able to match
<FrederickHirsch> any assertion that includes nested assertion that says how to do integrity more specifically
<FrederickHirsch> Therefore if empty assertion is not wildcard, need some other wildcard\
Paul: Frederick should review last week's discussion
<FrederickHirsch> need to follow all the turtles all the way down, we are saying every option needs to be explicitly stated by both parties.
<FrederickHirsch> is that what we are saying, no wildcarding?
<Ashok> That's what Asir is saying
Monica: Once we decide they intersect or not, we should probably add a preface on domain independent processing?
Paul: The examples draw on WS-Addrressing
Maryann: I was going to amke just the point Paul made. I.e. No alg. is specified by WS-Addressing. We may need to make that statement and verify with WS-A WG
Chris: 2 different issue. 1st we
wanted an example of empty nested polciy intersection with
non-empty nested policy
... Ashok and Monica have raised a different key point
reg. 2nd 3rd bullets in 2nd para..
when does the processing occur after or prior to domain specific processing
<TRutt__> If a domain spec designs their assertion types (including nesting) to work with the default algorithm, and does not state any domain specific processing rules, the default is used. Addressing designed their nested assertions to work with the default intersection as is
scribe: I would like to get an agreent on the example is related to the other thread and discuss issue 4558 later
Tom: Consensus in WS-A WG was to rely on default alg.
Fabian: I was also not on the
call last week. So, I have same q? as Frederick
... I share Maryann's concern reg. this impacting SecPolicy
<Zakim> dorchard, you wanted to respond to Frederick on wildcarding and 4558
<FrederickHirsch> +1 David
Dave: Chris mentioned 4558. Also simplicity of creating assertion and policy alternate authors. I agree w/ Chris that we can bundle this all into 4558
Asir: Meat in Dan's example is what can intersect and what can not. It is actually covered in WS-A metadata spec.
Paul: Which part of Dan's msg you want added?
Fabian: 3rd pp
<monica> Here is the text: The thing to recognize is that if Policy 2 is a requester policy then it is incomplete in that it is not acknowledging that the base assertion also reflects support for anonymous responses. The requester determines what response type should be used. So, a client that needs non-anonymous responses will also work with a service that supports all of addressing. The client's policy should reflect that it is compatible with an endpoint th
<monica> Another example of intersection for WS-Addressing utilizes the default intersection algorithm. Consider two input policies:
<monica> As another example of intersection, consider two input policies:
<TRutt__> +1 monica text
<PaulC> Ashok's questions is about: The thing to recognize is that if Policy 2 is a requester policy then it is incomplete in that it is not acknowledging that the base assertion also reflects support for anonymous responses.
<TRutt__> ws addressing empty assertion means no restrictions on use of addressing
<TRutt__> a restriction to non anon is not compatible with no restrictions
Asir: The text is not going into the example
<PaulC> q1: Does the WG agree with Dan's answer in msg 009?
<monica> "without qualification"
<PaulC> q2: Does the WG accept the example for section 4.5 in msg 173?
Tom: The nested assertions are restrictions that need to be enforced. So they are not compatible
<PaulC> q3: Is there need for another example as per msg 009?
<monica> WS-Addressing text: The wsam:Addressing assertion indicates that there are no restrictions on the use of WS-Addressing unless otherwise qualified by assertions in its nested policy expression.
<cferris> question: can one infer from the sentence between the 2nd and 3rd bullet of the intersection algorithm that a domain MAY define domain-specific processing that could state that empty nested policy IS compatible with non-empty nested policy?
<monica> reiterate: nother example of intersection for WS-Addressing utilizes the default intersection algorithm. Consider two input policies:
Paul: Does the example need understand of the WS-A spec
Ashok: No only that no domain specific procesing is needed
<Fabian> I have been on the queue for the last couple of minutes!
<PaulC> Asir points at: If a domain-specific intersection processing algorithm is required this will be known from the QNames of the specific assertion types involved in the policy alternatives.
<TRutt__> no using default has impact on implemtability of policy intersection, which is a strong reason to try to go with the default algorithm
<PaulC> Q1: Does the WG agree with Dan's answer in msg 009? Q2: Does the WG accept the example for section 4.5 in msg 173? Q3: Is there need for another example as per msg 009 (Fabian's point)?
<PaulC> ac asir
Paul: 3 q's in IRC. Under Q2 there were 2 suggestons to change the examples in 173..
and monica's that it is deafult alg. that is being demonstrated
Maryann: What does Q1 mean? Agreement to change
Paul: The question is does it
describe what the current alg. states
... Q2. Anyone can't live
... seems consensus on that
... Q3 need for another example as per Dan's msg.
Frederick: If not the whole example, at least some of the text from that would be useful
Paul: need to move off this agenda item.
<TRutt__> There is an example of a permissive client policy in the wsa metadata spec already
<PaulC> ac mon
Monica: can we separate out the two concerns on the examples
Asir: Will take the text from IRC and post an updated example
Paul: Lots of traffic on the
... Where in this long thread we should spend some time?
<asir> can we have a link to proposal 4
<monica> Here is Ashok's post: http://lists.w3.org/Archives/Public/public-ws-policy/2007May/0192.html
<PaulC> Chris's response to David's questions on option 2:
Discussion on Dave's Q2 reg. Is it ok to omit Assertions marked Ignorable="true"
<toufic> thanks, rummy
<dorchard> cferris *LOVES* rummy :-)
<toufic> clearly, trying to emulate his idol
<Zakim> dorchard, you wanted to ask for a yes or no answer. Can a provider omit an ignorable assertion who's behaviour it will apply.
Dave: I like to ask an y or no answer, Can a provider omit an ignorable assertion who's behaviour it will apply.
Chris; Yes. This behavior applies to the policy processing fwk
scribe: important to talk about wht is not in the policy but what is in the policy
<maryann> ok so i don't understand the subtle variations
Ashok: Simple q? - Is what u r proposing what is in the spec?
Chris: Some subtle differences ...
Ashok: So now we are taklking of client and provider perspectives
<Zakim> dorchard, you wanted to ask a yes or no answer. Can a provider omit an optional="true" assertion from alternative who's behaviour it will apply.
Dave: Tthis seems to open a can of worms, not applying behavior not in the policy. Confusion on what the scoping of these thiungs are.
Paul: You seem to be ignoring the application of intersection. Chris's proposal talks of provider policy and intersection
<dmoberg> Is this a positive formulation of Chris F's msg 181 formulation? A policy requester can engage the behaviors of a > compatible policy alternative when the policy alternative is part of the > intersection result. Although more than one policy alternative may exist > in an intersection result, only one policy alternative of all the expected > behaviors is engaged. Additional behaviors could be engaged if doing so > does not interfere with satisfying the
Paul: people should look at and understand what other people have proposed.
<cferris> dale, your statement was incomplete, but I think I see where you are going
<cferris> and I think I agree
<cferris> would like to see the remainder
Paul: Chris, is the note from Tony in the same direction as yours?
<maryann> do we have a link to tony's note?
Paul: How should we go fwd on
this at the F2F. Should we focus on Tony & Chris
... If Fwk does not say something, it is out of scope
... FWKonly tells what hapens when you intersect two policies
Ashok: There was an earlier
discussion on what you should not do, not in the policy
... are we tsill there or have moved away from that
Paul: is it same as Dave's #3?
Ashok: I am not speaking of Dave's options. I am speaking of Chris's closed world option
<dmoberg> Here is my question about msg 181 formulation
Paul: That was #1. We have agreed that is not the case
<dmoberg> A policy requester can engage the behaviors of a compatible policy alternative
<dmoberg> when the policy alternative is part of the intersection result. Although more
<dmoberg> than one policy alternative may exist in an intersection result, only one
<dmoberg> policy alternative of all the expected behaviors is engaged. Additional behaviors
<dmoberg> could be engaged if doing so does not interfere with satisfying the policy assertions
<dmoberg> in the policy alternative of interest.
Chris: I am done :) If someone comes up with a counter proposal, I will respond but, I am done with trying to come up with a middle ground
Paul: People should come up with
concrete counter proposal
... just asking questions is easy.
This is scribe.perl Revision: 1.128 of Date: 2007/02/23 21:38:13 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/MAy/of May/ Succeeded: s/authors/processing/ FAILED: s/.LAter/ Later/ Succeeded: s/abiout/about/ Succeeded: s/2nsd/2nd/ Succeeded: s/draws/draw/ Succeeded: s/1at/1st/ Succeeded: s/on WS-A/in WS-A WG/ Succeeded: s/metadat/metadata/ Succeeded: s/Pauil/Paul/ Succeeded: s/;/:/ Succeeded: s/exaple/example/ Succeeded: s/ok to ignore/ok to omit/ Found Scribe: prasad Inferring ScribeNick: prasad Present: Fabian ArnaudM Chris_Ferris Plh Charlton Prasad_Yendluri Frederick_Hirsch monica Tom_Rutt Yakov_Sverdlov asir Ashok_Malhotra Maryann Sergey_Beryozkin Dale_Moberg Paul_Cotton Symon Dave_Orchard GlenD Toufic WARNING: No meeting title found! You should specify the meeting title like this: <dbooth> Meeting: Weekly Baking Club Meeting WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Got date from IRC log name: 16 May 2007 Guessing minutes URL: http://www.w3.org/2007/05/16-ws-policy-minutes.html People with action items:[End of scribe.perl diagnostic output]