W3C home > Mailing lists > Public > www-tag@w3.org > September 2002

Re: TAG Comments on XHTML 2.0 and HLink

From: Paul Prescod <paul@prescod.net>
Date: Thu, 26 Sep 2002 13:50:57 -0800
Message-ID: <3D938141.4080404@prescod.net>
To: Svgdeveloper@aol.com, www-tag@w3.org
CC: Steven.Pemberton@cwi.nl

Svgdeveloper@aol.com wrote:
>...
> 
> If there is a sound technical case to be made lucidly and succinctly for the 
> HTML WG's (and XForms WG's) opposition to XLink please feel free to make it 
> known.

When an XHTML author embeds a picture, video or formula into their 
otherwise prose document, they are clear in their minds that they are 
shifting into a different semantic mode and dealing with a different 
data type. What makes the linking situation different is that people are 
not switching mental models when they want to make a link. The link is 
just one attribute in the middle of a sea of HTML. It isn't a shift, it 
is simply a feature of HTML.

They shouldn't be forced to absorb any of:

  a) the XLink namespace
  b) the XLink data model
  c) the XLink attribute names and element types

I believe it to be a fundamentally bad user interface decision to force 
XHTML developers to think about XLink at all.

Furthermore, the techniques that would allow XHTML to be mapped to XLink 
semantics through annotation are well-known. Of course it always takes 
less engineering effort to make the binding between two components 
simple and let the user deal with the disconnect. As an analogy, we 
could force people to click on buttons to bring up SVG pictures in a 
totally different browser. But seamless is better and is achievable.

  Paul Prescod
Received on Thursday, 26 September 2002 16:51:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:11 GMT