Reply on comments on ITS tagset draft

Hello Jirka,

You had given us a while ago comments on the tagset draft which we
discussed this week, see
http://lists.w3.org/Archives/Public/public-i18n-its/2006OctDec/0073.html
and http://www.w3.org/2006/12/20-i18nits-minutes#item04 . We have
prepared some answers, see below your original mail with some replies
marked as >ITSWG. We implemented the changes at
http://www.w3.org/International/its/itstagset/itstagset.html .


Hi Felix,

it was really nice to meet you in Heidelberg. Please find below my
comments to the latest ITS draft from:

http://www.w3.org/International/its/itstagset/itstagset.html

>ITSWG: We replied to your mail taking
http://www.w3.org/TR/2006/CR-its-20061102/ into account, which is more
than 99% identical to what you say at
http://www.w3.org/International/its/itstagset/itstagset.html .

On its:rules version of ITS is specified by its:version attribute. I
think that you should use just version attribute (i.e. no namespace)
because it is very uncommon to use namespaced attributes on elements
from the same namespace. Approach which I propose is already widely
adopted -- for example in XSLT you have both xsl:version (used on
literal elements, see section 2.3 of XSLT spec) and version (used on
xsl:stylesheet) attributes.

>ITSWG: It would be possible to have a version without a namespace at
the its:rules attribute. However, if there is no rules element in a
document, we need a version attribute in a namespace its:version at the
root element of a document. We think that having just one declaration
for the its:version attribute which can be re-used in these two
locations is enough.

ァ2.1: You are talking about style element/attribute from CSS, but those
attributes are presented in *HTML* language to allow specification of CSS.

>ITSWG: Thank you very much for catching this. You will see at
http://www.w3.org/2006/12/20-i18nits-minutes.html#action12 our editor's
AI to make the corrections.

Example 8: In DocBook V5.0 it is mandatory to specify version attribute
on document element. There should be version="5.0" added. You should
also add at least one para element after info element because without it
example document is not a valid DocBook instance.

>ITSWG: Yves made the correction already.

If you are using its:translate attribute it is necessary to specify
its:version? Where exactly version must be specified?
>ITSWG: We think that
http://www.w3.org/TR/2006/CR-its-20061102/#its-version-attribute is
clear about this: The version attribute MUST be specified at the
its:rules element (if that is present). If no rules element is present,
the version attribute MUST be specified at the root element of the document.

The most of
examples specify its:version, but example 10 doesn't have its:version
defined anywhere on ancestors of <title its:translate...> element. I
think that reasonable approach would be to require its:version on
element itself or on one if its ancestors.

>ITSWG: We think that the mechanism described at
http://www.w3.org/TR/2006/CR-its-20061102/#its-version-attribute is
sufficient and makes it easier to search for version information (only
two elements, its:rules or the root element need to be checked for *all*
ITS markup) than the ancestor approach you are suggesting (with that
approach you would need to check ancestors for each local ITS markup and
for rules element).

You are using simple XLink links and you reference XLink 1.0
specification. However XLink 1.0 requires specification of link type
(xlink:type="simple") even for simple links. XLink 1.1 which is under
development removes this constraint by saying that if there is no link
type specified, simple link is assumed. So you probably want to
reference XLink 1.1 instead.

>ITSWG: Thank you for this suggestion, we will change the ODD document
and use xlin:type="simple". We can't reference XLink 1.1 easily since it
is not a Recommendation yet.

Example 4: There is no qterm element in DocBook. Some readers of ITS
specification could be confused by fact, that this example is using
DocBook V4.x which is in no namespace while former DocBook example uses
DocBook V5.x which is in namespace. I would recommend using only one
version of DocBook for consistence.

>ITSWG: Yves made the correction.

ァ6.5.1. You are referencing CSS2.1 which is far from being W3C Rec, at
least AFAIK.

>ITSWG: The reference is only informative, so we don't think this is a
problem.

Appendix E: Schematron schema uses quite strange expressions like
name() = 'its:locNoteRule'. Such test wouldn't work in situations when
user uses different namespace prefix then its. It would be better to
replace this test with self::its:locNoteRule, where prefix is resolved
to namespace via sch:ns declaration

>ITSWG: We made the update.

Thank you very much again,

Felix

Received on Friday, 22 December 2006 06:33:25 UTC