RE: HTML is a formatting/UI language

Håkon Wium Lie wrote:

>  > 	How would you render this in HTML+CSS, and maintain the
> 'semantic' content?
>
> Here's one example:
>
>   <div class="person" xmlns="urn:hl7" UID=".." SSN="...">
>  	<div class="n">
>  	   <span class="FirstName">John</SPAN>
>  	   <span class="LastName">Smith</SPAN>
>  	</div>
>  	<div class="a" type="home"><SPAN CLASS="Address1">... ...</DIV>
>  	<span class="DOB" dt:dt="iso...">...
>  	<span class="insurance" ID="...">
>  		<span class="business" role="Insuror">...</SPAN>
>  		...
>  	</span>
>  	<div class="person" role="Provider">...
>  	<div class="person" role="EmergencyContact"> ...
>  	<div class="diagnosis" CPT="...">...
>  	<div class="medication" ...>
>
> I understand if you don't want to use this format internally, but when
> shipping across the Web, my example can be shown in 100 million
> browsers -- including aural ones -- yours can't. And they both contain
> your original semantics. You can programatically convert my document
> back to yours.

	This is already starting to look painful and this version is simplified. A
'full fidelity' representation (yes this does mean validating) would be even
more painful, and I picked a simplified example to start. To me, there
exists some sort of factor which is roughly the ratio of semantic
content/number of tokens.

	Yes this can be 'displayed' in a browser, but not the way I want it to be
displayed. I am replacing custom client application with dynamic browser
applications, using the well known model-view-controller paradigm, the
document is the model, the view is the transformed HTML and the controller
is the javascript/XSL containing frame.

>
> There is a small cost involved in the server-side conversion, yes. But
> you can still do the client side processing since all your data is
> still there.
>
> What you gain is accessibility.
>

	That's a short term detail. Assuming all 100 million browsers understood
XSL, I can transform XML into HTML+CSS on the client so the only argument
against client side validation is that not everyone can do it. With that
logic we'd still be using VT-100's (and many of us are!) We need to be in
this for the long term correct way to do things and separating content from
presentation remains a good basic rule.

Jonathan Borden
http://jabr.ne.mediaone.net


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list