XML Processing Model WG

Meeting 193, 05 May 2011


See also: IRC log


Norm, Jim, Henry, Alex, Murray, Mohamed (via IRC)
Paul, Vojtech


Accept this agenda?

-> http://www.w3.org/XML/XProc/2011/05/05-agenda.html


Accept minutes from the previous meeting?

-> http://www.w3.org/XML/XProc/2011/04/21-minutes.html


Next meeting: telcon, 26 May 2011?

Norm proposes to cancel 12 and 19 May

Processor profiles discussion

Norm: I haven't got anything new, but given that I'm canceling two week's meetings I thought we should have a meeting.

Murray: I'd like to observe a logical inconsistency that keeps happening. Some time ago, XML was developed. Lots of folks are using angle bracket stuff that isn't really XML and calling it XML.
... You could fork on the standalone attribute. It seems to me that since this is the XML Processing Model WG, we should have a profile that does XML.

Some discussion of what exactly is out there (HTML excepted) that's called XML that isn't well formed.

Norm: Ok, then I think a profile that does DTD validation would satisfy you, yes?

Murray: I think so.

Henry: We talked about DTD validation several months ago and decided against it. I'm not immediately persuaded to go that way now.
... There was some sentiment that we shouldn't be encouraging DTDs.

Murray: But we're the XML Processing Model WG. We should either go back and say that XML should change or we should support validation. All I'm saying is that as we traveld own this road, we keep dropping things. There are probably lots of reasons to do that, but it just occurs to me that this is XML processing and one of the things you can do is validation.

Norm: I suppose having a profile that does validation is a logical profile.

Henry: I think there's a sort of cross product. I'm not sure that validation could be added to several of these profiles.
... The profiles as they stand are necessarily nested within each other. If we define a further one, which is one of them plus validation, the question arises what about adding it to the others.

Norm: Henry, if you're willing to look at the spec costs, that would be good.
... It probably only makes sense to add validation to the profile that reads the external subset.

Henry: Why? There are documents that carry their DTDs in their internal subset.

Norm: Hmm. Yeah.

Murray asks about profiles that have optional steps.

Henry: We don't really have those, that would have a high cost.

Alex: If you're using DTDs, you're in a particular world. I don't think we can necessarily satisfy everyone out there.

Alex: Maybe we could try to satisfy just a subset of that world: DTDs without namespaces, for example.

<scribe> ACTION: Henry to review the spec to see what the cost would be of adding validation. [recorded in http://www.w3.org/2011/05/05-xproc-minutes.html#action01]

Norm mentions that XInclude is no longer in the profiles per the 21 April discussion.

Murray: Hey, I need XInclude for GRRDL.
... I wonder whether the name of the profile could be more descriptive of what it includes. So "browser's in 2011", "xml valid", etc.

Norm: Yeah, Recommended was probably a mistake.

Alex: I agree. I think we really should have a profile that's intended for web browsers.
... that has the word "Browser" in the title.

Some discussion of XML in browsers...

Some discussion of whether there should be a profile for what each of the browsers do.

Murray: Possibly in non-normative appendixes.
... I'd like a chance to think about this offline.

Norm: Let's take this up again at the next meeting.

Murray: And getting XInclude back.

Henry: We also need to reply to comments.

Murray: We have WG meetings in Toronto preceding Balisage. Would anyone like to have a f2f in between?

Henry: Not sure I could go.

Any other business?

None heard.


Summary of Action Items

[NEW] ACTION: Henry to review the spec to see what the cost would be of adding validation. [recorded in http://www.w3.org/2011/05/05-xproc-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2011/05/05 14:50:22 $