See also: IRC log
Chair: DaveH & MikeC
Scribe: DaveH suggests having occasional "trout pond"
discussions, perhaps organized around "position papers".
... ACTION: to W3C team and co-chairs to encourage have such trout pond discussions and papers stating a position.
<dbooth> Teleconference bridge: http://www.w3.org/2002/ws/desc/2/04/f2fNovLogistics.html#Bridge
Scribe: The presentation bye Glen Daniels is at http://lists.w3.org/Archives/Public/www-ws-arch/2002Nov/0085.html
<ATrout> rrsagent where am i?
<DaveH> trout topic : are SOAP feature URIs really the same as namespaces? Should they be?
... I think so, but probably because I think of namespaces different than lots of others...but I am an author of it
<chrisf> a feature is an abstract resource, it has identity, we use URIs to identify resources
<DaveH> trout topic: should we, the wsa, recommend
that we close the loop between URI as names and access to
<dougb> dave, should the trout be served to the TAG -- the name / access issue seems to be on their table already. DaveO?
<DaveO> I agree that name/access issue is a tag issue.
<DaveH> do we need to help them understand our
perspective and urgency/non-urgency?
... It may be a set of semantics specific to web services - descriptions and the like
<dbooth> Glen's presentation in HTML: http://www.w3.org/2002/ws/arch/2/11/glend-features/glen-features_clean.htm
<DaveH> is a module spec of the more generic class
"description" that we have identified?
... Isn't that the same thing schema is talking about as NUNs?
<DonM> In the case of wsdl, wouldn't a more generic
<wsdl:featureBinding> be better, and more general than
... I like using schema to restrict the value space of properties -- so having them typed is a good thing.
... +1 to using features throughout web services architecture
<dougb> ? has the idea of wsdl:feature come up in the Description WG?
<DonM> Not yet -- wsd wg still getting head around the specific case of soap. Would like to see move to general mechanism.
<DaveH> chairs wish to thank glen for his efforts to educate, motivate and "do the right thing"
Scribe: discussion around whether there should be feature registries, etc.
<DonM> Thanks again to Glen for putting the presentation together. TIBCO supports. I have to drop off now.
Scribe: glen thinks the basic architecture should be pretty simple, largely the feature stuff.
<DaveO> DaveO suggestion: Split soap 1.2 part 1 into 2 parts, and soap 1.2 doesn't have to go back to last call.
Scribe: Chrisf thinks it would be relatively easy to
partition part 1.
... The chairs asked for volunteers for a task foce to factor out the feature stuff from soap part 1 and to liase with the XMLP group to solicit such changes.
... The volunteers are GlenD, ChrisF, Hugo and DaveO.
... Martin is posting a URL for section 3.2 text.
<MartinC> draft pub sub write up http://lists.w3.org/Archives/Public/www-archive/2002Nov/0048.html
Scribe: ACTION: Editors to incorporate the text from
into section 3.2.
... hugo presents text on security...
... ACTION: Editors to incorporate Hugo's (modified) text -- Hugo will post the URL for the text.
<hugo> Security text: http://lists.w3.org/Archives/Public/www-archive/2002Nov/0051.html
<DaveH> http://www.w3.org/2002/11/chor-proposal - just FYI
<dbooth> Photos from the F2F: http://lists.w3.org/Archives/Member/w3c-ws-arch/2002Nov/0054.html
Scribe: Frank's presentation is at http://lists.w3.org/Archives/Public/www-archive/2002Nov/0050.html
<MartinC> is there a meta model for semantics?
<dbooth> DaveH, for another description of "semantics" see section 2.7.1 in the WSDL Primer draft: http://dev.w3.org/cvsweb/~checkout~//2002/ws/desc/wsdl12/wsdl12-primer.html?rev=1.3&content-type=text/html#N1011C
<MartinC> I'm sure Betrand Russell could have answer this one
<dbooth> Roughly speaking, when semantics are
"grounded" it means that they have been defined in terms that you
already understand by some other mechanism.
Scribe: there is a difference between knowing what your contracts are and what a contract is -- Frank
<DaveH> franks slide 7/8 is much more defined and constrained than the para
Scribe: ACTION: Frank and DaveH to take slides 3, 7, 8
into paragraphs for section 3.2.
... Hugo describes his planned AC talk on web services, including choreography.
<MartinC> there is consensus on trouts
Scribe: DavidB talks about text for choreography to be
included in section 3.2
... MarkJ -- choreography is about linkages among web service *operations*, not web services per se
... this captures conversational uses within a web service as well as relationships among operations in different web services
... WSDL has proved very useful for describing a single service. Currently complex natural language descriptions outlining the obligations of the participants and detailing how to use a service (sequencing, state management, etc.) have to accompany a WSDL description. The next step is to partially replace these somewhat imprecise instructions with precise language. This will simplify the daunting task companies now face when trying to use Web services to integrate th
<geoff_a> my personal position (obsession?) is that standardized choreography is necessary (but not, I think, sufficient) for observability
Scribe: we shouldn't lose the choreography text from the
charter -- it complements the text that DavidB provided
... DaveidB will finish wordsmithing and re-present.
... DaveH presenting proposed text for 3.2.
... on "why we need semantics"
<geoff_a> WSDL is like a shrink-wrap contract - if you use it, you agree to it.
<TomCarrol> When is a contract a contract
<Trout> when lawyers are involved
<TomCarrol> A contract is binding upon acceptance
by two parties
... So there is a difference between a offered contract and a binding contract
<dougb> of course, acceptance can still be implicit
in use -- my credit card agreements change rapidly and are
explicitly (in the updated agreement) accepted when I next use the
... WSDL essentially works in this way though prior negotiation could happen offline to create the WSDL document
<Trout> or agreements as you negotiate usage via some super duper uddi
<dbooth> The Web Service architecture enables the semantics of a Web Service to be published.
<dougb> suggestion: consequences
<dbooth> "The Web Service architecture enables
semantics to be associated with Web Service descriptions."
... "The Web Service architecture enables semantics to be referenced from Web Service descriptions."
Scribe: ACTION: Editors will incorporate text that DaveH
sends on semantics to www-archive
... DaveH please post URL when it is available.
<DaveH> what email address?....just a moment
... Semantics semantics semantics trouts semantics
<hugo> www-archive should hold all the wordings
<DaveH> Action 4 completed.
<dougb> do we need action item to incorporate Dave Burdett's text?
<Trout> I'm logging. I don't understand 'where am i', Trout. Try /msg RRSAgent help
<chrisf> ACTION: DavidBu to send
choreography text to list
... ACTION: editors to incorporate the extended feature descriptions into revised WSA and update the version in CVS
<davidbu> Choreography text for section 3.2 is at http://lists.w3.org/Archives/Public/www-archive/2002Nov/0052.html
<DaveH> time for unstructured conversation about
what to call web services at different levels of conformance
... level 1?
Scribe: end of scribing ...