Accept this agenda?

-> http://www.w3.org/XML/XProc/2011/02/17-agenda.html

Accepted.

Accept minutes from the previous meeting?

-> http://www.w3.org/XML/XProc/2011/02/03-minutes.html

Accepted.

Next meeting: telcon, 24 Feb 2011?

Henry gives regrets for 24 Feb; no other regrets heard.

Review of XML processor profiles draft

Norm: Thank you, Henry.

Paul: I like the way it's going, but I'm still not entirely sure what the different classes, A, A', and B are.

Henry: Yes, I agree. When I started, I thought there'd be one class per profile, but that's not the way it worked out. I'll add a gloss.

Paul: My other two things were even more editorial.
... I'm not sure if everyone will understand the distinctions between information items and properties and the classes.

Henry: I'll think about that. I'm not sure an attempt to clarify would help.

Paul: Given that we're thinking of releasing a third edition of the Infoset, I wonder if we should make an edition agnostic reference to the Infoset.

Norm: Vojtech also had some questions, but he's not here.

Henry: I'll try to get John Cowan or Richard Tobin to look carefully at it and see if I've got the class annotations correct. Some of these are not at all obvious.
... In some cases the choices are a little arbitrary.
... They could be wrong. I'm going to wait a week and then try to take a fresh look.
... The question Vojtech asks is a tricky one. I had forgotten I think that references isn't just for ID/IDREF.

Alex: Should processing instruction notation just be X? References is more complicated.

Henry: I really don't like making references implementation dependent. This just gets a little messier. As it says several times, this table uses items/properties to identify things in the document. There are multiple data models.
... I'm going to leave notations and unparsed entities as X, but
... I really don't like making references implementation dependent. This just gets a little messier. As it says several times, this table uses items/properties to identify things in the document. There are multiple data models.
... I'm going to leave notations and unparsed entities as X, but
... I'm going to change processing instruction notation to X, but I'll split id/idref references and entities/entities notation references into two parts.
... Everybody should report the references property with respect to IDs, that's not negotiable.
... Does that sound like it works?

Norm: I think so...
... So, Henry, you're going to prepare another draft for us?

Henry: Yes. I'm afraid this means another WD though.

Norm: Yes, but that's fine.

Namespace documents

Norm: Mohamed pointed out a few more oversights, anyone else see any problems?

Nobody says so.

Norm: Ok, I'll fix the problems Mohamed pointed out and try again.

Any other business?

Alex: Do we have an XProc meeting planned at XML Prague?

Henry: I've agreed to do a 10 minute slot about processor profiles.

Alex: It would be nice to get some feedback.

Adjourned.