W3C home > Mailing lists > Public > www-html@w3.org > February 2000

Re: inline CSS (was: is anyone interested in XHTML?)

From: Jan Roland Eriksson <jrexon@newsguy.com>
Date: Mon, 21 Feb 2000 22:57:26 +0100
To: Murray Altheim <altheim@eng.sun.com>
Cc: www-html@w3.org
Message-ID: <asb3bsgku7iehdt0oi5vkadcvd38tfo77j@4ax.com>
On Mon, 21 Feb 2000 03:38:48 -0800, Murray Altheim <altheim@eng.sun.com>

>Jan Roland Eriksson wrote:
>> Just for the record of "correctness"
>> Priorities...
>>   CSS1: ua -> user -> author -> user!important -> author!important
>>   css2: ua -> user -> author -> author!important -> user!important
>> Effectively 1: author!important styling is redundant in CSS2.

>Just for the record of what I read in the CSS1 and CSS2 specs, you
>might want to quote where you find this. I was quoting verbatim from
>CSS1 and CSS2 (I didn't make it up!)

I was not implying that you did make it up...only that you may have
missed a few lines in that behemoth called the CSS2 spec...
Heck I have missed a lot of lines in there...

>so your statement contradicts the specs.

Not really...

<URL:http://www.w3.org/TR/REC-CSS2/changes.html> Appendix B3.

The appendix is said to be "informative" only, which is required not to
break other statements saying that all of CSS1 is also supposed to be
valid CSS2.

Otoh, the only two decent browser implementations I know of right now
and that are well on their way, both take this change serious.

The third one I don't care about any more, I'm leaving WinNT for good to
go Linux on the very same day that Opera releases a first public beta
for Linux.

>> Effectively 2: the C in CSS is a shaky business anyway, mostly
>>                due to the CDATA type content allowed for the
>>                CLASS attribute.
>>               (CLASS attribute values can't be validated)
>It wouldn't be 'validated' by the XML parser anyway.

Is there really _any_ attribute value that could be validated by an xml
parser? I have always been thinking that the DTD'less possibility of xml
rules that out?

>One of the things I'm lobbying to change is that we use notations
>where appropriate, and the <style> and <script> element types
>should probably be using a notation identifier rather than a
>CDATA wrapper (although the CDATA wrapper shouldn't keep it from
>being parsed, since the code doing the parsing of ECMAscript or
>CSS isn't an XML parser anyway.

Yup, that's the way I understood it too, it's like the parser "calls a
separate function(...)" to deal with that part, right?

Jan Roland Eriksson <jrexon@newsguy.com>
Received on Monday, 21 February 2000 16:52:38 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:05:53 UTC