Re: XHTML 2.0 User Agent Conformance

David Woolley wrote:

>> What happens with an errata which fixes an error in the original 
>> XHTML 2.0
>> specification, UA's pre-Errata would handle a document different from 
>> UA's
>
>
> The errata would have to be against XML itself, as browsers are not
> required to read the DTD or validate against it, and there is a strong
> lobby that says that browsers should not even have internal knowledge
> of the DTD.

  Why shouldn't UAs have knowledge of the XHTML DTD?  If a DOCTYPE is 
specified, then isn't it a requirement of XML that the document 
validates against it (With a Validating Processor[1])?  For any other 
XML document with a DOCTYPE, UAs (incl. Netscape and IE) stop and 
display error messages for invalid documents; Why should this be any 
different for XHTML?

  If a UA has no knowledge of the DTD, then an author could write 
something like:

<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en">
<head>
     <title>Invalid Document</title>
</head>
<body>
    <p>This is some annoying <blink>blinking</blink> text!</p>
    <!-- Invalid Blink Element -->
</body>
</html>

  And because of current UAs tendencies to just blindly process HTML, 
this document, even with the invalid <blink> element, will be rendered 
without error.  Isn't this exactly the kind of poor use of (X)HTML W3 is 
trying to stop by developing standards?

  IMO, this poor use of XHTML would be gradually stopped if UAs like 
Netscape, IE, etc. started to become Validating Processors (especially 
for XHTML2).
  It's also interesting to note that Netscape runs in "Standards 
Compliance Mode" for the above sample document, depite the invalid 
<blink> element.

[1] http://www.w3.org/TR/REC-xml#sec-conformance

CYA
...Lachy

Received on Saturday, 1 November 2003 06:42:32 UTC