Accept this agenda?

-> http://www.w3.org/XML/XProc/2013/08/21-agenda

Accepted, as ammended

Accept minutes from the previous meeting?

-> http://www.w3.org/XML/XProc/2013/07/24-minutes

Accepted.

Next meeting: 28 August 2013 or at the face-to-face

Next meeting: 28 August 2013

Review of open action items

A-217-03 completed

A-231-02 completed

A-228-01 overtaken by events

Face to face prep, updated requirements and use cases document

-> http://www.w3.org/XML/XProc/docs/requirements-v2-jim.html

Norm walks through section 4

Jim: I added the ????-mark items in section 4.10 from other lists I had

Some discussion of p:empty and whether or not it's needed.

<jf_2013> Alex orig email on empty http://lists.w3.org/Archives/Public/public-xml-processing-model-wg/2013Aug/0000.html

It's used in parameter input ports, but those are going away

Alex: I also use it in p:document
... Now that I've been reminded why we need it, I guess it's ok.

Norm: All of the 4.x items are marked must/should except for 4.9; let's mark 4.9 must
... Let's move "allow multiple p:catch" up to its own top-level item, marked "should"

Norm; So. Is that everything. More to the point, if that's all we did, would it be a success?

Alex: Let's pull AVTs out as its own "must" thing. All the other syntactic simplifications are "shoulds"

Norm: Yes, I agree.
... Off topic, but for the minutes, wrt loading extension functions, the spec currently says "thou shalt not", I think we may want to soften that.

Some discussion of importing function libraries

Jim: I think the goal is to allow for reuse of existing libraries.

Norm: Well, I think it's also about allowing pipeline authors to write their own functions in p:when test expressions.

<scribe> ACTION: A-235-01 Alex to provide a use case for the extension library item [recorded in http://www.w3.org/2013/08/21-xproc-minutes.html#action01]

Vojtech: Someone once asked about using XProc steps as XPath functions.
... Maybe we could allow functions to be created from XProc pipelines.

Alex: Certainly, in my particular use case, I can do what I want with steps. But it's heavyweight.

Norm: I think mapping steps to functions is an interesting idea.

Alex: I wonder if we can find the message where that was requested.
... With respect to the extension library, if loading them was implementation defined, it would be nice if there was some way to make the dependency explicit, in metadata, perhaps.

<jfuller_2013> Norm's cx:import reference http://lists.w3.org/Archives/Public/public-xml-processing-model-wg/2013Aug/0031.html

Norm: Back to face-to-face planning
... It sounds like we're happy with these requirements
... Coming out of the f2f, I want a publishable requirements documents

Alex: I'd like to be done before TPAC

<jfuller_2013> another link for ext funcs http://norman.walsh.name/2013/08/20/extensionFunctions

Some discussion of XML processor profiles

Jim: Would it help if we matched each profile with a concrete example?

Alex: Maybe, but let's review the comments more carefully before we add work to our plate

Vojtech: Right, but if we add more perspective, then we run the risk of introducing new areas of contention

Alex: It would be nice to arrive at the f2f with a concrete list of issues to address.

Any other business?

None heard.

Adjourned.