See also: IRC log
[postponed until next week]
issue 025 done
Hugo Haas to start a mailing list and get concall details for TF. DONE
resolved, will be sent
ACTION: Yin-Leng to send ws-cdl review on behalf of the group
Chair: We haven't had a draft in a while, need to update, when should we publish another draft?
<bob> do the draft, don't dodge it
Marc: Do we need a new working group draft and go through a bunch of process or can we just publish a stable URL?
Chair: Don't want to slow editors down, so if Hugo will do the work....
Hugo: I can handle it, it's not a big deal
Chair: Do people want a week to look at the drafts and then we can vote?
Paco: Is the HTML version up to date?
Gudge: No, the XML is up to date, look at that.
<TomRutt> one week is fine
ACTION: Chair will schedule a vote for next week (1/31) to vote on releasing the editor draft
<plh> ACTION: Chair will schedule a vote for next week (1/31) to vote on releasing the editor draft
Chair: Issue 42, extensibility model, originated by Glen Danniels, anyone want to own this issue?
Chair: I'll volunteer David Orchard
Chair: Issue 43, Marc, do you mind owning?
... Yes, I"ll own
Gudge: There is a security issue here, it should probably be captured
<plh> Gudge: unless you use qnames in content, it's ok to use exc14n
ACTION: Gudge will be very careful and send text for issue 43
Chair: Issue 44 - we talk about the rules for replying to a message but don't specify what they are
Chair: Hugo, will you take this?
Chair: Is the strawman in the proposal a good starting point?
Hugo: It is pretty detailed but it may not have all the details
Chair: Have people looked at this yet?
General Response: No
Chair: I will put this on the agenda for next week
Chair: Issue 45 - Action defaults don't work with URNs
Chair: Talked about this last week and had an action item for Jonathan to make a proposal and he did that
Chair: Put link into IRC
Chair: Have people looked at this?
Chair: Jonathan isn't here.
Paco: This is what Jonathan brought up at the F2F?
Chair: Oops, wrong URI, let's try again.
Chair: Looks like a complete proposal for generating a default action
Chair: Do people want to approve this text now or want more time?
Paco: I'm o.k. with it
Marc: Presumably if we run across any nitty problems we can open an issue? I haven't read it in full detail.
Chair: Does anyone object to taking Jonathan's e-mail proposal that is pasted in IRC as the resolution of issue 45?
RESOLUTION: Issue 45 closed using the text in the e-mail in IRC
Chair: Another issue not on the agenda, Dims has had a long standing action item to talk to HP about modifying EPRs, he was going to make sure this was a new issue
Chair: We haven't heard anything yet
Chair: We will hold off until he gets on the call
Chair: Anish isn't here but we were waiting on the resolution of the operation name requirement in the WSDL WG and the WSDL WG did make some forward motion on that last week.
Chair: Anything else we need on this issue? Should we wait for Anish? Did WSDL remove the requirement so we have nothing to discuss?
Chair: O.k. we will wait for Anish, there might be some non-normative text but we can leave until next week
Chair: Issue raised by Rebecca
Rebecca: I should go through all the discussions on this issue in other contexts and see if I can outline what the attitudes have been on this so we can then discuss it specifically
Chair: My understanding is that we just need to define a non-meta data specific way of including metadata by reference as well as by value, today we only do it by value
Rebecca: Non-metadata specific way?
Chair: I could write an extension that says this is my kind of EPR metadata that my extension invented and then refer to it, by reference, in the EPR. The issue is to say we should have a generic way of doing that instead of making each extension invent their own mechanism.
Rebecca: What is metadata? What's included and not included in that issue?
Chair: How would that affect the spec?
Rebecca: Our proposal, for
example, of having WSDL location. A reference to a service
definition was being categorized as being meta-data.
...: We just side stepped the categorization and figured out how to deal with it, but I never figured out why it was metadata. Is anything not defined in the EPR metadata?
Chair: In some sense everything in an EPR is metadata.
Rebecca: Except the URL?
Chair: The address? This is getting us to the identifyer versus non-identifier issue we side stepped with issue 1. Why don't you summarize your thoughts, you already have an action item for that.
Chair: Multiple actions issues, Gudge had an action item that he did complete, you called it regarding issue 9 on the mailing list by mistake?
Gudge: Yes, sorry about that. I proposed language and people pushed back and asked why we need to say anything? People should be able to read.
Chair: Are you o.k. with not adding any text? Other people?
Paco: Sounds fine to me
Gudge: The editors suggested clarifying text and the group said no
Chair: Do we need more text on this? The simplest proposal would be to close the issue with no action.
Philippe: Let's close it
Chair: Does anyone object to closing issue 25 with no action?
RESOLUTION: Issue 25 closed, the group affirmed that there is only one action
Chair: There hasn't been much discussion on this yet, Steve or Rebecca would you like to referesh our memory?
Steve: We want to have a way for
services with multiple addresses to be able to advertise those
addresses in one structure and apps wanting to talk to that app
can pick up the structure and use the one that works
...: I put in some proposals, a suggestion to augment the EPR with additional port information and the other is to come up with a different structure to contain multiple EPRs. I didn't get a lot of feedback.
... I got contacted by IBM and MS to talk about it and work on it as a proposal but we haven't made much progress. I've been after these guys but we are just now getting to the point where we might start working.
Chair: My concern is that where we are in the schedule if that is something we can do in a short amount of time if there is interest but if it is going to blow out the schedule I'm getting concerned. I haven't seen interest on the list. Are people interested in this?
Paco: We are quite interested in
this and Steve is right that we talked and didn't follow up in
a timely fashion and I would be happy to take a joint action
item with him
...: To bring something quickly to the list
Chair: If you folks can coordinate and start the discussion I'd like to see direction, either a new proposal or direction toward an existing proposal.
ACTION: Steve will refresh the discussion on issue 26
<TomRutt> I would like an explanation why the service element of epr is not good enough
Chair: Hugo, did you find that e-mail?
Hugo: Unfortunately no. What I do remember is that we have the choice of either limiting our contributions to the member organizations or to open it up to other people. Basically when members are going to contribute they should put their contribution under the W3C document license and that gives us the right to do whatever we want with it in terms of republishing but they still own the copyright. I think it gets more complex for people outside the
Chair: If you could keep digging on that
Phillipe: We have a policy on this and the (link included in IRC) will answer all questions
Chair: We will look at that and once we have the test case situation together we can decide on the issue
Chair: Harris, last week at the f2f on issue 41, the resolution was that we needed something more specific on the message exchange patterns and different part of the SOAP stack in use. So we could test this version of WSDL, SOAP with this transport binding and this pattern, etc.
Chair: Recognizing there are other test cases we need to cover in the future we would need this for issue 22. We gave Anish an action item to help you to develop the template but he is on vacation and couldn't complete the item by Friday.
Chair: I'm wondering if someone else who was at the meeting could take an action item to help Harris out with this? Anyone?
Chair: Anish will be out for the entire week so if he isn't back until next week we wouldn't see another go at that for another couple of weeks and not get input for a couple of weeks after that. I'm hoping this is the data that would help the asynch task force with its work.
Chair: Is anyone who was at the F2F willing to help Harris?
Bob: I'll help
ACTION: Bob will help Harris and CC the Chair so he can help as well, will get work on that for next week for issue 41
Chair: O.k. thanks everyone! Talk to you next week