Re: Proposal: version at-rule

On Wed, 24 Mar 2004 09:34:19 -0500 (EST), Craig Francis  
<craig@bbmcarlson.com> wrote:
..

> Sorry, I think Alan Plum had a point.
>
> Origionally CSS worked well with
>
>
>      <style type="text/css">
>      </style>
>
>
> then V2 came out with a method to hide this new CSS from old "V1 only"  
> browsers, which was
>
>
>      <style type="text/css">
> 	@import "file.css";
>      </style>

CSS1 already contained the @import syntax. Netscape 4 just didn't support  
it. Which was the first 'css hack' used to hide (parts of) stylesheets  
 from browser by exploiting non-support of features and known bugs.

..

> Granted it does not ensure that all the browsers will have full  
> compliance, but thats another story. At least by putting in a version  
> number it means that future browsers wont bother trying to understand a  
> CSS rule that it just cant do.

They will not bother anyway, this works quite well already. The problem is  
more in browsers that support a property or value only partially - and  
this is exactly the most likely case where developer proud, or marketing  
pressure
, will prevent browsers from admitting "we don't support this". What would  
be useful is some way to group some rules together that should be applied  
all together or none at all. But any official capability detection to  
accomplish this will be very error prone, for the reason stated above.

-- 
The Web is a procrastination apparatus:    |  Rijk van Geijtenbeek
It can absorb as much time as              |   Documentation & QA
is required to ensure that you             |   Opera Software ASA
won't get any real work done.  - J.Nielsen |  mailto:rijk@opera.com M

Received on Thursday, 25 March 2004 05:00:06 UTC