See also: IRC log
<SRT> Agenda addition above
<MSM> http://www.w3.org/2005/06/07-ws-cg-minutes.html
Minutes are approved
Steve: what about the August break?
Mark: it looks like that for Addressing, but it isn't official yet
Steve: we'll do one for Choreography
<SRT> 18/19 Addressing in Bedford Mass
<SRT> 20/21 Description in Bedford Mass
Hugo: we'll have 18-19 July for Addressing and 20-21 for Description, in Bedford, MA, USA, at Sonic
<scribe> ACTION: Hugo to (possibly) work on an updated glossary and bring it for review [PENDING] [recorded in http://www.w3.org/2005/06/28-ws-cg-irc]
Hugo: I really want to get to it, but can't seem to...
Mark: with regards to MEP, we haven't made any progress
Jonathan: the Addressing WG has been working on looking at using WS-Addressing for doing async interactions
... lots of different approaches proposed
... one item has emerged as consensus in the Async TF: the lack of one-way MEP/binding in SOAP
... this can help for async scenarios, using 2 one-way MEPs in the WSDL description
... in addition, if we had such a one-way binding, we could default to it in the SOAP 1.2 binding in WSDL 2.0
... the WSDWG would like the CG to help make this happen, as this is out of scope for us
Hugo: note that the timing issue for WSDL 2.0; I thing that we should have something before we go to CR to do this change
Mark: from the POV of Addressing, everybody agreed that this is good work and that it should happen, but that it is likely to be orthogonal to requirements that we may have
... the general feeling is that we should encourage this work, but we're likely to come with more requirements
Jonathan: the contradiction is that an open scope goes against WSDL's need for something fast
Mike: XMLP wants a very clear scope, based on previous discussions
Jonathan: DaveO has a binding which may do one-way, async, and more
... so it's not clear how it relates to the simple one-way work
Philippe: I wonder what would be the cost of adapting the one-way binding from the WS-I BP for SOAP 1.2
Mark: which one-way MEP?
Hugo: the 202 Accepted HTTP binding clarification
Jonathan: SOAP 1.2 would need more work, as it would need an MEP
Mark: we could indeed declare victory with the simple case
Hugo: the interesting question is: will this help us do our ReplyTo testing at the CR stage in Addressing? I think the answer is yes
Mark: I think that the async TF is indeed after optimization
Jonathan: we may not need a standard solution to do the CR testing
Mark: a number of people want a standard solution
Philippe: we need a Rec somewhere
Jonathan: I think that we reference it in the test suite
Philippe: we could do a Note at the time of the test suite, and maybe do a Rec later on
Jonathan: everybody hopes that the XMLP WG is the right home; I'm not sure if it's feasible
Mike: it seems that the existing wording in the charter could cover that
Jonathan: are you under the PP?
Philippe: no, they're not; they can't do a Rec track doc without a charter change
... they the XMLP WG can produce a WG Note
Mark: how do we go to Rec then?
Philippe: we could recharter
Steve: and the Note could serve as the scope
Mike: there was pushback against just a MEP scope
... everybody is expecting XMLP to shut down
Jonathan: so the suggestion is to have XMLP work immediately on a Note, and on parallel explore getting the Note on the Rec track
Mike: I'm not absolutely confident that the WG will be happy about this plan
... I need more information about how to get to Rec track
Jonathan: the CG can work on a draft charter
Mike: you think that it can go in parrallel with XMLP's work?
Jonathan: do we want to be pro-active about going to Rec?
Mike: if it's going to be a Note, why doesn't the Async TF do it?
Mark: they can't publish anything on their own
Philippe: but they can publish via the Addressing WG
Steve: it boils down to who needs it and who wants to do it
... and it isn't necessarily XMLP's job
Hugo: I think DaveO had a draft; if this can help us go through CR in Addressing and be used in WSDL 2.0, then it may be a high priority item for the TF
Mark: I am concerned about the ability to deliver of the TF at this point
Jonathan: we could refocus the TF
Philippe: would DaveO be happy to push the one-way stuff as a WG Note?
Jonathan: I'd be happy to do that from the WSDWG POV
... however, I have doubts about defaulting to something defined in a Note in the WSDL 2.0 Rec
Philippe: but the Note could be a good starting point for a Rec
Mike: why would we do a Rec?
Philippe: interest
Jonathan: IP, review, etc.
<MDChapman> by the time we finishe discussing this we could have charted a new fast track WG:-)
Jonathan: the highest decision we need to make is whether this needs to be a Rec or not
... if we want a Rec, starting work on a note seems not the best thing to do
<MDChapman> +1
Jonathan: or if we do, we should have XMLP start the work and recharter them
Mike: I think XMLP would be more motivated for a Rec
Steve: I'd rather it'd be a Rec track
Philippe: so let's ask the XMLP WG, and if they don't want to, we can look at doing a Note
<MDChapman> what about the other option of a new WG?
Hugo: for a new WG: we could ask the async TF if they want to morph into a new WG for this purpose
... in case XMLP says no
<MDChapman> thats what i was thinking
Philippe: in the first step, we should look for something simple
Mike: and mimic the WS-I stuff?
Hugo: it's already saying that in the Desc WG request
Mike: it's just a hint, though
Jonathan: the WS-I stuff is pretty simple
<MDChapman> i dont think it wud take that much effort to write a very focussed charted drawing togther the threads from wsd/xmpl and ws-a
Mike: how does DaveO's work relate to that?
Jonathan: I think he took that into account
Mike: I'm happy to take this back to XMLP, with Desc's reqs and DaveO's draft
... we would need to move to the PP
Philippe: it would basically be a rechartering
Mike: I may need a week to get all this straight to the WG
Mark: a head-up: DaveO and I will be away next week
<scribe> ACTION: Mike to present one-way MEP/binding plan to XMLP WG [recorded in http://www.w3.org/2005/06/28-ws-cg-irc]
5 July 2005 (next week)
and then returning to our normal schedule