Re: block-based parsing?

At 3:02 PM +0100 9/13/05, Philip TAYLOR wrote:

>Maybe, but it's a pretty weak reason for rejecting what appears to
>me to be an excellent and constructive suggestion.  If a browser
>lies about its abilities, then who cares whether the document
>renders as intended in that browser ?  Surely what matters is
>that the document renders as intended /in a browser that doesn't
>tell lies/ ?

    You'd be right, if any browser could ever be trusted to not tell 
lies.  But all of them do and have, intentionally or otherwise.  To 
pick but one example, IE/Win would doubtless claim to support 
'background-position: fixed'; it does not-- and I assure you that 
many, many people care whether a document renders as intended in that 
browser.
    IE/Win is by no means the only culprit, of course.  I have never 
yet seen a browser that didn't botch something it claimed to support. 
And that's what usually sinks these sorts of proposals.  Trust me, 
I'd love to see something like this happen, but only if it could be 
trusted.  Since it would be running inside browsers, it couldn't be 
trusted.

-- 
Eric A. Meyer  (eric@meyerweb.com)
Principal, Complex Spiral Consulting   http://complexspiral.com/
"CSS: The Definitive Guide," "CSS2.0 Programmer's Reference,"
"Eric Meyer on CSS," and more    http://meyerweb.com/eric/books/

Received on Tuesday, 13 September 2005 14:31:05 UTC