Re: [CSS21] response to issue 115 (and 44)

On Thursday, February 19, 2004, 12:23:12 AM, Ernest wrote:

>> From: Chris Lilley <chris@w3.org>
>> Because most stylesheets out there are in what? Most are in US-ASCII,
>> I would guess, since the entire syntax of CSS uses US-ASCII. The only
>> opportunities to have anything else are replaced content in:before and
>> :after, which is not too common in practice since it doesn't work in
>> MSIE/Win.
>>
>> So, if most stylesheets are US-ASCII then a default of UTF-8 would
>> work pretty well.

EC> Better hope that the author hasn't used any classes or IDs that don't use
EC> non-ASCII characters as well. For non-English documents, I'd say that's
EC> an unreasonable assumption, since its likely the author will be using
EC> names that are meaningful to him in his native language.

Certainly, although the fact that URIs are limited to US-ASCII means
that pointers to such IDs are difficult to construct so they are used
less than you might think. For a solution, see IRIs.

Your point about classes is well made, however.  Added to the point
about comments, made by Boris, its clear that a significant percentage
of stylesheets worldwide wil use an encoding that goes beyond
US-ASCII. Some stats on this would help, of course.

Its also clear that those stylesheets are currently relying on luck to
be interpreted correctly.

EC> Most US stylesheets /= Most stylesheets.

<sarcasm>
No, really?
</sarcasm>

EC> Think globally. Act locally!

<irony>
Thanks for the note to consider internationalization sometime, I must
remember to do that.
</irony>

Actually the entire reason I waded in on this discussion, and the
reason I alerted people like the I18N Activity lead to the discussion
also, was a concern that the character encoding be correctly and
robustly discoverable in a manner consistent with the rest of the web
architecture.

My observation that the syntax characters of CSS are exclusively drawn
from US-ASCII (which is true) does not equate to a belief that
US-ASCII is suitable for all Web content, or that what works for the
US works for the world. As anyone that knows me will attest.

[ Actually US-ASCII does not work for the US either, since for example
Spanish, Vietnamese, Chinese, and other languages which are the native
languages of sizeable percentages of US citizens need more than
US-ASCII. ]


-- 
 Chris Lilley                    mailto:chris@w3.org
 Chair, W3C SVG Working Group
 Member, W3C Technical Architecture Group

Received on Wednesday, 18 February 2004 18:42:16 UTC