See also: IRC log
-> http://www.w3.org/XML/XProc/2010/07/08-agenda
Henry: I'd like to talk about an addition to XProc for 1.1.
Norm: Ok.
-> http://www.w3.org/XML/XProc/2010/07/01-minutes
Accepted.
Henry gives regrets for 15 July through 5 August.
Norm: I don't think there have
been any new comments and I failed to follow up with Bjoern, so
I don't think there's anything here this week.
... Henry, how goes the draft wrt the whitespace issue
(invariants)
Henry: No progress so far.
Carine: I'm not comfortable with
four profiles. There seems to be little justification for that
level of granularity.
... It ends up being a do-it-yourself profile with whatever
bits you want. I understand that some of the concern came from
the browsers not doing much.
... It's probably ok to have a very small browser profile, do
we need three more?
... It seems a bit useless to have that many.
Henry: I think that's a good point. I guess I remain moderately hopeful that looking at the different invariants will give us a reason for making these distinctions.
Norm: I guess we need three and if we have only three then there will be one obviously missing.
Henry: We went ahead and added them when people suggested them and before we publish, or in the next draft, we might call out explicitly the question of whether we need them all and see what people say.
Norm: Sounds good to me.
... But I'm waffling about it.
<scribe> ACTION: Henry to add a note to the draft raising the question of whether we really need *four* profiles. [recorded in http://www.w3.org/2010/07/08-xproc-minutes.html#action01]
Carine: That's fine by me. It
just looks like the 'pizza menu problem' in the US. There are
all those names.
... I think some note that suggested how each profile might be
used would be helpful.
Norm: That sounds like a good idea to me, we should have suggested uses for the profiles before we're done.
Henry: I keep ending up writing
vastly more structure than I need to in order to build a
fragment with a bit of content from the context.
... It would be useful to have a single template rule.
... Maybe it would be sufficient in XProc to allow a version of
inline in which XSLT curly brace rules apply. Or more likely
XQuery curly brace rules apply.
Norm: Yes, I've been there. Figuring out how to get that into XProc is what's been a stumbling block for me.
None heard.
Adjourned.