[microformats-discuss] When we say "XHTML" do we mean...

Tantek Ç elik tantek at cs.stanford.edu
Mon Oct 17 10:45:14 PDT 2005


On 10/17/05 10:00 AM, "Ryan King" <ryan at technorati.com> wrote:

> On Oct 17, 2005, at 8:37 AM, Robert Bachmann wrote:
> 
>> Ryan Tomayko wrote:
>> 
>>> ... application/xhtml+xml type XHTML, or
>>> XHTML-looking-markup-that-people-will-serve-as-text/html-and-will-
>>> not-contain-namespace-information-90%-of-the-time
>>> type XHTML? I'm just wondering how liberal implementors are
>>> planning on
>>> making their parsers and whether XHTML is going to be considered by
>>> implementors as a MUST or a SHOULD.
>> As IE6 doesn't support application/xhtml+xml most of the XHTML pages I
>> know are served with the MIME type text/html. So it wouldn't be very
>> wise to require "application/xhtml+xml".
>> 
>> IMO it should be enough to require well-formed XHTML which is sent as
>> text/html OR application/xhtml+xml.
> 
> I know that in X2V, Brian has just used tidy to cleanup any invalid
> markup.
> 
> My general take on XHTML is this: produce valid, wellformed xhtml;
> send it as text/html ('cause I have to). When consuming, I'll
> typically use tidy or TagSoup to clean things up before I touch them.

The question of 'When we say "XHTML" do we mean...' is something that I've
often heard.

In my opinion:

For many reasons applications/xhtml+xml is not practical today.  As
specified, I'm not sure if it will ever be practical (there's plenty of
offlist discussion of this, so I won't go into details other than to say
incremental rendering, scripting, events).

However, I *still* find utility in both the term "XHTML" and the spec XHTML
1.0.

Hence I define use of "XHTML" as *valid* use of *strict* HTML4.x, or XHTML
1.0 that conforms to Appendix C Compatibility Guidelines, and omits use of
any presentational elements/attributes.

I've made this statement with more details in my presentations on "The
Elements of Meaningful XHTML":

 http://tantek.com/presentations/2005/09/elements-of-xhtml/
 http://we05.com/podcast/mp3/we05-2-tantek-celik.mp3

Like I said, this is my pragmatic opinion of what I think "XHTML" means
today, others may (will likely) have other opinions.

Thanks,

Tantek





More information about the microformats-discuss mailing list