W3C home > Mailing lists > Public > www-style@w3.org > August 2003

RE: q an quote and content properties etc.

From: Jörg Hartmann <jhartmann@aquilacoop.de>
Date: Tue, 5 Aug 2003 13:27:47 +0200
To: "'Chris Lilley'" <chris@w3.org>
Cc: "'Bert Bos'" <bert@w3.org>, <www-style@w3.org>
Message-ID: <00bd01c35b44$a0400600$b97ba8c0@PAQUNB01>

> On Sunday, August 3, 2003, 3:26:10 PM, Jörg wrote:
> 
> JH> 2) to put quotation marks into the presentation part (CSS) instead 
> JH> of the source structure is a really bad idea, since they reflect 
> JH> text structure/semantics (like punctuation in general). <em> is part 
> JH> of the structure, as is <h1> and <p> and <cite> and <q>. That's just 
> JH> the idea of seperating structure from representation. To put 
> JH> quotation marks into a text is by definition _not_ a representation 
> JH> task (while the question of how they are displayed is).

One day later Chris replied:

> I agree completely, but that is what HTML 4 did and thus, CSS 
> had to come up with something to meet this requirement.
> 
> 
> -- 
>  Chris                            mailto:chris@w3.org
> 

JH again:

Sorry, Chris, I guess I don't catch the meaning of your post. What did HTML4
and which requirement had CSS to meet?

My post was about

a) the planned killing of the <q> phrase tag (in XHTML2) which _did_ put
quotation marks into the structure (while <q> would indeed profit of some
improvement for multi-part quotes), when CSS was --correctly-- _only_
responsible for the way typographic type of these marks (their presentation
-> the quotes: property), and

b) the creation of _structural_ properties (like {content:open-quote;} which
does _not_ only define the representation but instead the _placement_ of
quotation itself).

Instead of

a) creating a CSS work-around for a not-that-perfect XHTML functionality
(for inline quotes that either continue over various paragraphes or are
interrupted by, e.g., a mark about the speaker) and

b) killing the whole q tag (in XHTML2) only because IE didn't implement it
(which is true for all language-sensitivity) - Mozilla, e.g., _did_
implement it, and for the little bug in their implementation (only to read
the first value-pair of quotes:) 1) an easy workaround can be put into
stylesheet rules --like q q:lang()-- and 2) a bugfix is in the making.
(Opera, for that matter, _does_ support the <q> tag but falsely implemented
the CSS quotes: property, so it's not very probable that the killing of <q>
and the creation of more stuff in CSS will improve Opera's situation very
much ...)

you should just put that what is missing into the <q> element and ist
attributes. Once more: Quotation marks (and their placement) are structure.
Their display (and only this) is presentation. So instead of inventing CSS's
{content:open-quote;} and the like, you (or the XHTML WG) should improve <q>
with some attributes and you might improve CSS to define how quotation marks
look (or if they render at all) when there are various <q> tags with the
same id and one of them is the last (or _attributed_, _not 'class'ified_, as
being the last), or how a pair of q tags (of the same id) with something in
between should render (dependent on the language, of course).

Well, I will try to propose a detailed solution (for both CSS and XHTML)
when I find the time.

Yours,
Jörg
Received on Tuesday, 5 August 2003 07:27:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:23 GMT