See also: IRC log
Minutes of the meeting of 2006-03-27 were accepted without objection
hugo: headup on change of contact person . I am leaving the WG and W3C by the end of May
Bob: Hugo has been a tremendous help to me and the WG. We will miss him and wish him the best.
Hugo: I will go to yahoo to do some web-servicey "stuff"
Bob: reviewing action items:
1- editors to remove editorial notes, text from hugo accepted...
Bob: we will come back to these later
proposed and new issues
* lc123 - Example Improvement suggestion Owner: ??? Proposal 1: <http://www.w3.org/mid/44299C10.firstname.lastname@example.org>
ideal would be to align examples with WSDL document and use hotel reservations
dhull: describing the example suggestions
bob: any objections to regularizing the examples?
<scribe> ACTION: editors to modify text accordingly [recorded in http://www.w3.org/2006/04/03-ws-addr-minutes.html#action01]
issue: * lc124 - Conformance section Owner: ???
bob: Hugo, is this due to the new document policy?
hugo: W3C had a QA Activity, not really a new rule, but looking from a QA perspective
bob: should we add a conformance section, or change how we use the word "conform" ?
Jonathan: When we talk about
supporting usingAddressing, the use of anonymous or Action is
... for optional features, it is not clear exactly how you specify conformance
bob: there are some general statements in the doc style guide which contain some canned conformance statements
trutt: lots of specs have
"conformance points" which we can consider
... we need to clarify what are the conformance points - are they grouped or separate?
bob: any volunteers for writing a
... do we need a conformance section? We should explore that.
trutt: are the optional points sufficiently clear now, which ones are needed?
Jonathan: it is clear to me, but maybe it is not clear in the spec which individual items are required
bob: can we get an issue owner?
<anish> i see only two occurances of 'conforms to' and they all point to ws-addr (core/soap), i think
Jonathan: I can do it, suggest some clarifying text. If others want to work in parallel, that is fine.
<scribe> ACTION: Jonathan to work on clarifying conformance points [recorded in http://www.w3.org/2006/04/03-ws-addr-minutes.html#action02]
<GlenD> I think that what we did in WSDL was actually a mistake, and that a lot of people are going to be confused when they try to understand how to correctly use extensibility. :(
Jonathan: metadata is an issue - diferent use cases - hard to deal with
glenD: WS-A is used for messages... it cuts out databases, other interesting things... just say a sentence in that direction, may address the issue...?
<dorchard> I tend to agree with Glen. I especially don't like that the conformance seems related to fluffing up an abstract component model...
<anish> is this comment against all the specs or just the wsdl binding?
<dhull> paul, that was me ("WS-A is used for messages")
dhull: is there any product that
this would not apply to?
... if there is no meaningful kind of process it does not apply to, we can say it applies to all..
<bob> class of products
<bob> The generic name for the group of products or services that would implement, for the same purpose, the specification, (i.e., target of the specification). A specification may identify several classes of products.
we can say it is applicable everywhere, not have a canonical list of products it applies to.
bob: we can say: we heard your
comment, but it does not apply because of the following, or we
can address it
... the class of product we can describe here might be as simple as "web services"
jonathan: or a class of systems which consume WSDL
bob: wouldn't the class of products be WSDL consumers?
hugo: we can say WSDL or EPR
consumers. it does not need more than that.
... the conformance item was useful, but this item does not bring a lot of value.
bob: any objections to "WSDL or EPR consumers"?
jonathan: I agree,
bob: no objections,
<scribe> ACTION: bob to craft response respond to submitter; class of consumers is EPR and WSDL consumers [recorded in http://www.w3.org/2006/04/03-ws-addr-minutes.html#action03]
bob: not sure what is the point
<scribe> ACTION: Bob to respond to author. [recorded in http://www.w3.org/2006/04/03-ws-addr-minutes.html#action04]
hugo: possibly use a code font to mark element names
jona: or add a namespace prefix lowercase wsa-w
<hugo> in XMLSpec, <el> can be used to surround element names and <att> can be used to surround attribute names
<scribe> ACTION: editors to investigate ways to typographically clarify how to depict our intent [recorded in http://www.w3.org/2006/04/03-ws-addr-minutes.html#action05]
bob: ways for the application to determine how to use Anon? Is this an implementation issue?
bob: the comment is in end of the
third paragraph of the link posted
... the answer may be: "yes, you are right, but..?"
jonathan: don't quite see the problem with a single endpoint, using non-anonymous, sending it on...
bob: asynchronously with respect to the lifetime of the backchannel?
<anish> i didn't understand what he meant by creating a new binding type. And how would that address his problem
jona: at the application level?
bob: using a non-anonymous
replyTo, you allow a future response to come on that
... can someone craft a response? it is an overall WS-Adderessing question in some ways. Can someone try to take ownership to tease out what Todd means?
Anish: like jonathan, I'm not sure what the issue is...
jonathan: I will respond to him and try to get more detail, and explain our thinking.
<scribe> ACTION: Jonathan to respond to author (Todd) [recorded in http://www.w3.org/2006/04/03-ws-addr-minutes.html#action06]
Bob: We will need two hours for
the meeting next week, things are piling up
... I have opened registration for the F2F
... planning a Wednesday PM activity, possibly at Museum of Fine Arts
... it would need to be fairly early, by 6:30
... adjourning now, prepare for 2 hours next week!
... Paul, thanks for scribing