RE: XHTML 2.0 considered harmful

At 2003-01-15T14:03-0500, Peter Foti (PeterF) wrote:-

> Company ABC has some data to share with Company XYZ, for example, a news
> article that will be included on a web page.  A simplified version of this
> data might look something like this:
>
> <div>
>    <div class="articletitle">Why XHTML 2.0 is Considered Harmful</div>
>    <div id="whyxhtmlisharmful">Blah blah blah...</div>
> </div>
>
> Company ABC is concerned that the data must be presented the way that they
> intended it to be, but they don't have access to the <head> of Company XYZ's
> web page.  So how then does Company XYZ no how to "style" the data it is
> given?

Since ABC and XYZ are apparently cooperating here, XYZ can presumably put
ABC's style rules into their own style sheet, adding a .ABC or whatever to
the selectors. Or they could just put the article in a separate document,
referencing ABC's style sheet, and embed it using object.

The style attribute contains only declarations, not selectors, effectively
having an implicit ID selector... So using the style attribute for this is
likely to involve repeating quite a few declarations on many elements,
because the same style attribute will (more or less) need to be specified
on every element which the selector in the original style sheet would have
matched. I don't think I really want that sort of bloat.

And then there's the issue of specifying the style sheet language for the
style attribute. I don't much like the current way, i.e. the hateful <meta
http-equiv=... /> with a pretend HTTP header. (I notice that the XHTML 2.0
draft maintains the fiction that "HTTP servers use this attribute to
gather information for HTTP response message headers." Are there any
servers that do this?)

It's easy to think up all sorts of ways to improve on the HTML 4 style
attribute, which could address these issues, but I must confess that I
personally have not seen a compelling case for keeping it in any form.


Tim Bagot

Received on Wednesday, 15 January 2003 16:13:43 UTC