Re: text/html for xml extensions of XHTML

On Wed, 2 May 2001, William F. Hammond wrote:
>
> Let's try again:
>
> -----
> An item served as "text/html" should be handled as an XML version of
> "html" if
>
> a.  It begins, apart from white space, with the string "<?xml " .
>     (See comment on this in regard to Appendix C of the XHTML spec
>     above.)

That's implementable (although I still disagree with doing it).


> OR
>
> b.  The document begins with zero or more comments and processing
>     instructions that conform to the XML specification [...]

That's *not*. Not without writing a complex parser. Don't forget that the
point in time where you want to know what to do with the data stream is
before you've launched any (heavy weight) parsers -- in the network
library, in the lightweight piece of code implementing the MIME type
distribution. If you can't write the autodetection in 4 lines or so, then
forget it.


>   2. XHTML is current html.

I don't understand what that means. XHTML isn't HTML 4.01, nor is it tag
soup, nor is XHTML + MathML compatible with tag soup.

-- 
Ian Hickson                                     )\     _. - ._.)       fL
Netscape, Standards Compliance QA              /. `- '  (  `--'
+1 650 937 6593                                `- , ) -  > ) \
irc.mozilla.org:Hixie _________________________  (.' \) (.' -' __________