- From: Felix Sasaki <fsasaki@w3.org>
- Date: Thu, 02 Feb 2006 13:27:15 +0900
- To: "public-i18n-its@w3.org" <public-i18n-its@w3.org>
Hi all,
The minutes of yesterday's call are at
http://www.w3.org/2006/02/01-i18nits-minutes.html
and below as text.
Cheers,
Felix
[1]W3C
[1] http://www.w3.org/
i18n ITS working group
1 Feb 2006
[2]Agenda
[2]
http://lists.w3.org/Archives/Member/member-i18n-its/2006JanMar/0052.html
Attendees
Present
Andrzej, Christian, Felix, Sebastian, Yves
Regrets
Diane, Richard
Chair
Yves
Scribe
Felix
Contents
* [3]Topics
1. [4]Action items
2. [5]selection mechanism
3. [6]other topics
* [7]Summary of Action Items
_________________________________________________________
Action items
<scribe> ACTION: Felix to put up an agenda for the tech plenary f2f.
(PENDING) [recorded in
[8]http://www.w3.org/2006/02/01-i18nits-minutes.html#action01]
<scribe> ACTION: Yves to write examples of qualified and
non-qualified vocabularies for bug
[9]http://www.w3.org/Bugs/Public/show_bug.cgi?id=2719 (DONE)
[recorded in
[10]http://www.w3.org/2006/02/01-i18nits-minutes.html#action02]
[9] http://www.w3.org/Bugs/Public/show_bug.cgi?id=2719
<YvesS> [11]http://www.w3.org/Bugs/Public/show_bug.cgi?id=2719
[11] http://www.w3.org/Bugs/Public/show_bug.cgi?id=2719
Yves: the problem is:
... if we have dislocated ITS
... the XPath expression may or may not have prefixes
... depending on the prefixes, the XPath expression may work or not
felix: a proposal:
[12]http://lists.w3.org/Archives/Member/member-i18n-its/2006JanMar/0
057.html
[12]
http://lists.w3.org/Archives/Member/member-i18n-its/2006JanMar/0057.html
<its:ns its:prefix="db"
its:uri="[13]http://docbook.org/ns/docbook"">
[13] http://docbook.org/ns/docbook
felix: part of its:documentRules element
... you can have several elements of these, zero or more
sr: we have to translate XPath first
... we have to export the namespace declaration
<YvesS> ... SR: possible conflic issues.
felix: what would you do if several documents
sr: with XSLT, you could use the ns element as well
Yves: what happens if you have a document without namespaces?
felix: a possible solution: its:translateSelector="//db:p | //p
yves: I would like to tackle both e.g. DocBook with namespaces and
without
SR: are you talking about processing a "read only" document?
... you might be forced to rewrite the document to one which has the
document namespace (i.e. docbook 5)
... you would be mad to have to sets of rules for different versions
of a format
yves: we might have the problem with other formats which have
sometimes namespace and sometimes not
sr: what you can do in XSLT 2:
... specifying a default prefix
... one rule which defines "db" to be nothing, and one which says
"db" is a namespace
<scribe> ACTION: SR, Yves and Felix to test the usage of the <ns>
element, to see if it is usefull for the next draft [recorded in
[14]http://www.w3.org/2006/02/01-i18nits-minutes.html#action03]
<scribe> ACTION: Yves to explore constraint support in CSS and
possibly contact a CSS guru. (PENDING) [recorded in
[15]http://www.w3.org/2006/02/01-i18nits-minutes.html#action04]
<scribe> ACTION: All to look at parameters for the f2f around www
conference or xtech. (DONE) [recorded in
[16]http://www.w3.org/2006/02/01-i18nits-minutes.html#action05]
<scribe> ACTION: All to think about f2f at tech plenary and see if
they can make it or not. (DONE) [recorded in
[17]http://www.w3.org/2006/02/01-i18nits-minutes.html#action06]
Christian: I might not be able to come
<scribe> ACTION: Andrzej to start an article about the relations
between various standards. (ONGOING) [recorded in
[18]http://www.w3.org/2006/02/01-i18nits-minutes.html#action07]
<scribe> ACTION: Christian and Felix to rephrase the conformance
section (ONGOING) [recorded in
[19]http://www.w3.org/2006/02/01-i18nits-minutes.html#action08]
[20]http://lists.w3.org/Archives/Public/public-i18n-its/2006JanMar/0
091.html
[20]
http://lists.w3.org/Archives/Public/public-i18n-its/2006JanMar/0091.html
Christian: We made some progress
... we said "we want to say s.t. about schema conformance"
... on the other hand we want to say s.t. about processing
application
... we agreed on that
... what I don't understand:
... I argued for a more fine grained distinction for schema
conformance
... a more fine grained distinction might it easier to talk about
specific types of conformance
... I looked at Ruby
felix: I found out that ruby has no test suite
christian: we may want to talk about two different types of
processors
... non-interpreting and interpreting
... the former can read or write ITS markup
... the latter would act upon the semantics of ITS markup
... the motivation for this distinction is:
... it may become if we talk about very different types of
applications
felix: I would propose to have tests for our conformance levels
... it might be difficult if we abstract too much from the features
to test the featues
christian: how about support in an editor?
... e.g. in the open office suite?
... that allows authors to mark a piece of content
... with the ITS translate="no" flag
<scribe> ACTION: Christian and Felix to provide examples for tests,
and see if we can come up with a common set of conformance levels
[recorded in
[21]http://www.w3.org/2006/02/01-i18nits-minutes.html#action09]
<scribe> ACTION: Christian to write a non-normative section on
selection, at the beginning of the scope section. (ONGOING)
[recorded in
[22]http://www.w3.org/2006/02/01-i18nits-minutes.html#action10]
Yves: you may want to wait, since we want to change things e.g. with
in situ
selection mechanism
Yves: we had a problem with conformance, e.g. an in situ translate
attribute without translateSelector
... not having a selector would simplify a lot of things
... it makes it though a little bit less flexible
... e.g. if you want to override attribute information
... it would reduce the complexity
Christian: an experience:
... we worked with an editor
... adding an additional attribute would be straightforward
... the implementation of ITS information which does not use in situ
approach would be more complex?
... if you have distinctions between the different p elements, you
end up with a lot of different documentRule elements
... impact on conformance: if we work with in situ
... we may see more users
Yves: Christian is talking about people using an editor, right?
Christian: right
Yves: it is more on the user side of ITS?
... another issue I was thinking of: some documents may not allow
attributes
Felix: I am afraid that in situ selectors are not testable as a
feature
Christian: I compared this to CSS
... if I compare this to HTML: if there would be only a style
element in HTML?
Yves: you still would have the selector at the dislocated version
Felix: in CSS, you have a section with a lot of selectors
... in the document, you use CSS without selectors
Sebastian: It would not be possible to override an attribute in the
instance
... but that is bad style anyway
... the implementation gets simpler without in situ use of selectors
felix: selector attributes could be locally defined at the
documentRule element, which is better for validation
Sebastian: you can do anything with the dislocated selector
... if you have a rule at the top "translate every p"
... you don't want to say "translate no" at the backmatter
felix: you could express that by the order of the documentRule
elements
sebastian: the fact that you can't do selectors in situ with CSS
... it could be a good selling point for us, if we can make it
clear.
yves: it has another advantage: XPath expressions which go to a
parent element are bad, but could happen in the in situ case
Andrzej: translate=no is a rare case
... the picture for XML based meta data files is:
... you get situation were you don't want to translate in given
instances
... meta data configurations are straight forward
... in text based files, if you have to not translate s.t.
... in a given instance
... you would only do that in an instance basis
Felix: XPath is testable dislocated better than in situ
... the draft is not difficult to change
<scribe> ACTION: I will post a mail to say "that is the direction
without selectors in situ we go" [recorded in
[23]http://www.w3.org/2006/02/01-i18nits-minutes.html#action11]
<scribe> ACTION: SR, Yves and Felix to change their implementations
to work only without in situ selectors [recorded in
[24]http://www.w3.org/2006/02/01-i18nits-minutes.html#action12]
<scribe> ACTION: Everybody to follow new syntax (DONE) [recorded in
[25]http://www.w3.org/2006/02/01-i18nits-minutes.html#action13]
<scribe> ACTION: Everybody to read the spec and comment on the spec
until the publication date (ONGOING) [recorded in
[26]http://www.w3.org/2006/02/01-i18nits-minutes.html#action14]
<scribe> ACTION: Owners of the modularizations to keep working on
them. (ONGOING) [recorded in
[27]http://www.w3.org/2006/02/01-i18nits-minutes.html#action15]
This is some start for xmlspec:
[28]http://www.w3.org/International/its/itstagset/itstagset.html#xml
spec
[28]
http://www.w3.org/International/its/itstagset/itstagset.html#xmlspec
other topics
Yves: the whole inline segmentation thing has to be discussed
further
<YvesS> That was a productive meeting. Thanks all.
<YvesS> bye.
Summary of Action Items
[NEW] ACTION: Christian and Felix to provide examples for tests, and
see if we can come up with a common set of conformance levels
[recorded in
[29]http://www.w3.org/2006/02/01-i18nits-minutes.html#action09]
[NEW] ACTION: I will post a mail to say "that is the direction
without selectors in situ we go" [recorded in
[30]http://www.w3.org/2006/02/01-i18nits-minutes.html#action11]
[NEW] ACTION: SR, Yves and Felix to change their implementations to
work only without in situ selectors [recorded in
[31]http://www.w3.org/2006/02/01-i18nits-minutes.html#action12]
[NEW] ACTION: SR, Yves and Felix to test the usage of the element,
to see if it is usefull for the next draft [recorded in
[32]http://www.w3.org/2006/02/01-i18nits-minutes.html#action03]
[PENDING] ACTION: Andrzej to start an article about the relations
between various standards. [recorded in
[33]http://www.w3.org/2006/02/01-i18nits-minutes.html#action07]
[PENDING] ACTION: Christian to write a non-normative section on
selection, at the beginning of the scope section. [recorded in
[34]http://www.w3.org/2006/02/01-i18nits-minutes.html#action10]
[PENDING] ACTION: Everybody to read the spec and comment on the spec
until the publication date [recorded in
[35]http://www.w3.org/2006/02/01-i18nits-minutes.html#action14]
[PENDING] ACTION: Felix to put up an agenda for the tech plenary
f2f. [recorded in
[36]http://www.w3.org/2006/02/01-i18nits-minutes.html#action01]
[PENDING] ACTION: Owners of the modularizations to keep working on
them. [recorded in
[37]http://www.w3.org/2006/02/01-i18nits-minutes.html#action15]
[PENDING] ACTION: Yves to explore constraint support in CSS and
possibly contact a CSS guru. [recorded in
[38]http://www.w3.org/2006/02/01-i18nits-minutes.html#action04]
[DONE] ACTION: All to look at parameters for the f2f around www
conference or xtech. [recorded in
[39]http://www.w3.org/2006/02/01-i18nits-minutes.html#action05]
[DONE] ACTION: All to think about f2f at tech plenary and see if
they can make it or not. [recorded in
[40]http://www.w3.org/2006/02/01-i18nits-minutes.html#action06]
[DONE] ACTION: Everybody to follow new syntax [recorded in
[41]http://www.w3.org/2006/02/01-i18nits-minutes.html#action13]
[DONE] ACTION: Yves to write examples of qualified and non-qualified
vocabularies for bug
[42]http://www.w3.org/Bugs/Public/show_bug.cgi?id=2719 recorded in
[43]http://www.w3.org/2006/02/01-i18nits-minutes.html#action02]
[End of minutes]
_________________________________________________________
[42] http://www.w3.org/Bugs/Public/show_bug.cgi?id=2719
Minutes formatted by David Booth's [44]scribe.perl version 1.126
([45]CVS log)
$Date: 2006/02/02 04:22:45 $
[44] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm
[45] http://dev.w3.org/cvsweb/2002/scribe/
Received on Thursday, 2 February 2006 04:27:24 UTC