- From: Rick Jelliffe <ricko@gate.sinica.edu.tw>
- Date: Wed, 14 Jul 1999 12:50:00 +0800
- To: <jelks@jelks.nu>, <html-tidy@w3.org>
From: Jelks Cabaniss <jelks@jelks.nu>
> If you want to see something that uses non-standard stuff that doesn't
validate
> as *anything*, take a look at the output from MS Word 2000. Here is a
sample:
>
> http://cpcug.org/user/jelks/demo/word2k.htm
>
> Interesting (?) porridge composed of XHTML, XML, HTML, and CSS
"extensions" ...
> :)
It is certainly not well-formed XML: no literal delimiters (" or ') on the
attribute values. Also, there are empty elements in the head that do not
use empty tags.
It is not SGML or XML, because it uses comment delimiters to hide the script
code.
So it is "HTML with namespaces".
But it uses the schema name as a namespace URI. It then becomes a small step
to requiring the namespace name be a schema name (as Microsoft's BizTalk
does) which in turn leads to "data capture": I have an article "How to
Promote Organic Plurality on the WWW" about this issue, if anyone is
interested, at http://www.ascc.net/xml/en/utf-8/monolith.html
Rick Jelliffe
Received on Wednesday, 14 July 1999 00:50:00 UTC