W3C home > Mailing lists > Public > www-style@w3.org > July 2001

RE: @version rule

From: Jeffrey Yasskin <jyasskin@hotmail.com>
Date: Sun, 29 Jul 2001 08:45:57 -0500
To: <www-style@w3.org>
Message-ID: <000001c11834$cbc75990$a601a8c0@bluejay>
Let me see if I understand the forward compatible parsing rules. Any
style rule with an invalid selector (according to the UA) is ignored.
Any invalid property is ignored. Any invalid value causes its property
to be ignored. Am I correct? What if a stylesheet author wants two
different properties from different levels of CSS to either both be
applied or both fail? What if the meanings of various properties in CSS
change as happened at
(http://www.w3.org/TR/REC-CSS2/changes.html#changes-from-css1) and in
various places in CSS 3 and will happen again in future versions?

The reason I believe more version information is needed than is
currently present is simply that many stylesheet authors use dodges like
@media rules in order to target their stylesheets to browsers which
support a specific version of CSS. I believe that @media rules and other
non-standard ways of selecting the CSS version are more bug-prone, less
intuitive, and generally "worse" than a standard @version rule.

Authors are already using workarounds to get at @version functionality.
Why not make it easier on them?

Jeffrey Yasskin

-----Original Message-----
From: www-style-request@w3.org [mailto:www-style-request@w3.org] On
Behalf Of Jan Roland Eriksson
Sent: Sunday, July 29, 2001 6:15 AM
To: www-style@w3.org
Subject: Re: @version rule

On Sat, 28 Jul 2001 17:52:20 -0500, you wrote:

>>From: Jan Roland Eriksson <jrexon@newsguy.com>
>>To: www-style@w3.org
>>Subject: Re: @version rule
>>Date: Sat, 28 Jul 2001 21:26:50 +0200
>> >>From: Ian Hickson <ian@hixie.ch>
>>[wrote some sense]
>> >>To: Jeffrey Yasskin <jyasskin@hotmail.com>
>>who obviously does not understand shit about the concept at hand.
>This is possible. Please teach me to understand. What's wrong with each

>of my points?

All of them; The original CSS1 spec already describes the concept of
"forward compatible parsing", if extensions to CSS are designed
according to those original rules of CSS, no version info at all need to
be introduced.

CSS support is _not_ a compulsory component of a user agent, any
programmer can decide for him/her self about what level of CSS (or what
mix of CSS versions) to code support for.

As long as the resulting code adheres to the concept of "forward
compatible parsing" as originally described, all is Ok and the idea of a
@version can be killed off.

Your error (and quite a few of "experts" error too) is that you seem to
think "backward" instead of "forward", give those two words a serious
though on how they connect to CSS parsing, and maybe the tokens will
fall down the hatch.

Received on Sunday, 29 July 2001 09:45:57 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:26:58 UTC