W3C home > Mailing lists > Public > www-style@w3.org > June 2008

Re: [css3-mediaqueries]

From: Anne van Kesteren <annevk@opera.com>
Date: Wed, 18 Jun 2008 13:24:34 +0200
To: "L. David Baron" <dbaron@dbaron.org>, www-style@w3.org
Cc: "Ian Hickson" <ian@hixie.ch>
Message-ID: <op.ucxza8bo64w2qv@annevk-t60.oslo.opera.com>

On Wed, 18 Jun 2008 08:23:50 +0200, L. David Baron <dbaron@dbaron.org>  
wrote:
> http://www.w3.org/TR/2007/CR-css3-mediaqueries-20070606/#syntax
> says:
>   # Expressions involving unknown media features or unknown/illegal
>   # values are always false.

I read "always false" as being different from just "false", fwiw, and am  
therefore not really convinced that Acid3 as it stands is correct or that  
I introduced a change other than not preserving the media query during  
parsing.


> This is changed in http://dev.w3.org/csswg/css3-mediaqueries/ to
> read:
>   # Unknown media features. User agents are to ignore a media query
>   # when one of the specified media features is unknown.
>   [...]
>   # Unknown media feature values. As with unknown media features,
>   # user agents are to ignore a media query when one of the
>   # specified media feature values is unknown.
>
> I'm not sure I'm happy about this change.  There's a tradeoff here:
> whether to preserve that either one of "(query)" or "not (query)" is
> true (the CR's way), or whether to be precise and always turn
> unknown into false (the way in the dev.w3.org draft).  I can see
> arguments on either side (although a supported or query-supported
> keyword, prefix-able by not, could potentially get the benefits of
> both sides).  (I'd hoped to write a little more about this, but
> haven't had the chance.)
>
> My bigger worry, however, is that the old wording is interoperably
> implemented and tested in Acid3 (http://acid3.acidtests.org/).

Sigh.

It seems more sensible and in line with CSS syntax parsing to simply drop  
the media query in question. This also happens with malformed or unknown  
declarations within a declaration block. Everything expressed in CSS  
syntax that is unknown or malformed usually never ends up in the final  
model the renderer looks at. I'm not sure why this should be different for  
media queries.

Though we already made the first step with media types, I suppose, which I  
now see also says "always false" so I'm not sure the change I made on your  
request was correct per the CR draft.


-- 
Anne van Kesteren
<http://annevankesteren.nl/>
<http://www.opera.com/>
Received on Wednesday, 18 June 2008 11:24:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:55:07 GMT