W3C home > Mailing lists > Public > public-html@w3.org > February 2009

Re: ISSUE-4: Versioning, namespace URIs and MIME types ISSUE-60

From: Robert J Burns <rob@robburns.com>
Date: Wed, 18 Feb 2009 11:49:43 -0600
Cc: HTML WG <public-html@w3.org>
Message-Id: <662EE4C3-6FE1-46F2-9770-50C7643B584F@robburns.com>
To: Smylers <Smylers@stripey.com>
HI Smylers,

On Feb 18, 2009, at 11:22 AM, Smylers wrote:

> Boris Zbarsky writes:
>
>> Robert J Burns wrote:
>>
>>> if a DOM/Script author simply lets the UA handle things for them,
>>> then the scripting is not complicated at all. The browser will
>>> determine the proper handling of the element (and 'img' element or
>>> any other element).
>>
>> So what you're saying is that the right way to handle <img> with
>> script for document authors is to create documents that they know are
>> non-conformant in all languages involved (and which might not even be
>> serializable if the language happens to be HTML5, not XHTML5 or
>> XHTML2), and then to assume that UA error handling will "do the right
>> thing"?
>
> I thought that Rob was actually pointing out that browsers can
> distinguish <img> in a valid XHTML 5 document from <img> in a valid
> XHTML 2 document by whether there is an alt attribute or child text --
> and therefore that authors could write either, and browsers could just
> work out what's what and behave accordingly.
>
>> That seems like a highly undesirable situation to me, honestly.
>
> Yeah.

I think you must have missed the earlier part of the thread[1].  The  
discussion started where we're trying to identify irreparable  
collisions between XHTML1 and XHTML2, or XHTML2 and XHTML5, etc.   
Obviously redefining element semantics as HTML5 does is a good example  
of conflicts between XHTML5 and XHTML1, but we're trying to identify  
problems with XHTML2 and XHTML1. There has been frequent rhetoric  
about XHTML2 failing to be backwards compatible, but we haven't really  
found examples even on par with HTML5's redefinition of  
presentationally related elements (note there's no way to determine  
which definition of these elements is in place from context alone).

In terms of the 'img' element, HTML5 should properly specify error  
handling for document that include 'img' elements with content. I  
first reported this oversight over 18 months ago[2]. Now it is being  
raised as a horrible dilemma for the XHTML2 WG even though we haven't  
addressed yet either.

Take care,
Rob

[1]: <http://lists.w3.org/Archives/Public/public-html/2009Feb/0451.html>
[2]: <http://lists.w3.org/Archives/Public/public-html/2007Jul/0761.html>
Received on Wednesday, 18 February 2009 17:50:27 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:16:31 GMT