W3C home > Mailing lists > Public > public-multilingualweb-lt-comments@w3.org > July 2013

Re: Comments on section 7 of ITS 2.0

From: Jirka Kosek <jirka@kosek.cz>
Date: Fri, 05 Jul 2013 13:06:28 +0200
Message-ID: <51D6A8B4.5040203@kosek.cz>
To: Daniel Glazman <daniel.glazman@disruptive-innovations.com>
CC: public-multilingualweb-lt-comments@w3.org
On 5.7.2013 12:21, Daniel Glazman wrote:
>> I think that such document would be non-conforming, see
>> http://www.w3.org/TR/its20/#conformance-class-html5-its, as HTML5 itself
>> prohibits any foreign attributes.
> What about XHTML 1 and 1.1 ? These beasts are not unseen unicorns,
> the world of EPUB2 for instance is full of them.

By XHTML 1 and 1.1 you are talking about specs or document conforming to
respective specs? If you are talking about specs, then they don't allow
any extensions in conforming documents. If you are talking about
documents then HTML5 conformance defined for XHTML documents can be
applied on them.

>> I understand that this will make detection much easier, but not all
>> libraries for serializing XML are giving you enough control over places
>> where namespace declarations will actually appear. So my preference
>> would be to require its:version attribute on the root element if we
>> really want to have something ITS specific on the root element.
> A namespace declaration being an attribute set on an element and
> visible/editable in the OM, I absolutely don't see what you mean.

Reasonable designed XML API will not allow you to set namespace
declarations using methods for attributes because you can easily break
namespace well-formdness then.

> If the libraries you mention cannot control where they set the its
> namespace, they have a serious problem.

I don't care about where namespace declarations will appear. If I add
namespaced attribute down in the tree I don't care whether corresponding
namespace declaration will appear just on this element or whether it
will be propagated to the top of the document tree. Appearance of
namespace declarations is very fragile and I'm against basing any
decision on whether namespace declaration is appearing at some
particular place.

Is it really that expensive in your application to scan for @its:version
attribute? If document uses its:* opposing to its-* it must specify
@its:version somewhere, see:


  Jirka Kosek      e-mail: jirka@kosek.cz      http://xmlguru.cz
       Professional XML consulting and training services
  DocBook customization, custom XSLT/XSL-FO document processing
 OASIS DocBook TC member, W3C Invited Expert, ISO JTC1/SC34 rep.
    Bringing you XML Prague conference    http://xmlprague.cz

Received on Friday, 5 July 2013 11:06:54 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:32:28 UTC