On 09/02/2007, at 2:01 PM, Geoff Pack wrote:
David Dorward wrote:
Geoff Pack wrote:
Yes, for now. But wouldn't it be easier for all us if the browsers
just improved their handling of xml, instead of worrying about html5
and xhtml2?
No, since HTML expresses known semantics and random-XML doesn't.
Surely the semantic meaning is in the actual tag names, not just the
fact that they are standardised. It shouldn't matter as long as it's
understandable. Anyway, you can always re-use as many of the HTML tags
as you want, and make up your own when you need to.
While you can style it, there are more clients then those which are
visual.
You can add multiple CSS stylesheets to an XML document, just like
HTML.
Or use can use XSL and transfrom the document into an HTML file with
multiple CSS stylesheets.
I'm not an expert at any of this, btw. What do XHTML2 and HTML5
give us
that we can't do with XML and CSS?
cheers,
Geoff
Semantic meaning is meaning in context, and it's something more
complicated than can be contained in just the dictionary definition
of a word that you use in a tag name. It doesn't make sense to say
that semantics are included in tag names. The grand example of this
is layed out in the recent debate about the hr tag.
But, even if you made the (spurious) assumption that semantic meaning
can be included in a tag name, it would still require a human to
produce the semantics from the tag name. This is okay in one off
applications, but in broader applications like search engines, You
can't make an assumption like a <menu> tag will contain information
about a navigation menu. That <name> will contain a person's name,
etc. Semantics is more than just the individual words. It's the
meaning of the word in a specific context.
XHTML2 and HTML5 give us more than just set of named tags, they give
us a set of agreed upon semantics for those tags, which goes beyond
simply their names. This is essential for broad applications of
machine parsing.
*******************************************************************
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: [EMAIL PROTECTED]
*******************************************************************
