W3C home > Mailing lists > Public > www-style@w3.org > September 2005

Re: Defining Browser Compatibility Without Relying on the Browser

From: Richard York <richy@smilingsouls.net>
Date: Sun, 18 Sep 2005 12:05:35 -0500
Message-ID: <432D9E5F.7080403@smilingsouls.net>
To: robmientjes@gmail.com
CC: www-style@w3.org

Rob Mientjes wrote:

>>Disclaimer: I don't see much benefit in this feature personally.  I
>>believe server-side processing to be more appropriate.  I'm merely
>>pointing out that it is possible to deploy such a feature without
>>relying on the browser to tell the truth about what it supports.

> This was recently discussed here, IIRC, and the point is, IE does
> support "fixed", for example, but it supports it wrongly. That's a
> problem you have to consider before thinking this is a safeguard
> solution.

It doesn't support it incorrectly, it supports it partially.

Admittedly, the example I use doesn't make much sense or contain a 
practical application of such a feature, which is pretty much due to the 
fact that I don't really get how such a feature could be useful.  Many 
developers seems to want to be able to conditionally include or exclude 
rules based on support of a particular feature though, and I'm just 
exploring one possible solution to that.  The central argument against 
that feature seemed to be that the browser cannot be trusted to tell the 
truth about what it supports, and given the imperfection of any 
implementation, many, many different scenarios can exist where a given 
feature could fail, or behave contradictory to the specification where 
it was beleived before to be supported perfectly.

Additional scenarios can be introduced for incorrect support, e.g.
background-attachment: fixed !incorrect;

The gapping hole in this solution though (and I'm quite sure there are 
far more than I'm even thinking of now), is how do you use existing CSS 
in the way I have here to say that something like multiple background 
images are supported, or other cases where the property's syntax is 
changed or expanded, and that's something I haven't given alot of 
thought to.

Whatever the case, this is just my 0.02 on the discussion.

Regards,
Richard York

-- 
Mail_IMAP: A PHP/C-client/PEAR solution for webmail
Author: Beginning CSS: Cascading Style Sheets for Web Design
http://www.richard-york.com
Received on Sunday, 18 September 2005 17:05:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:40 GMT