Re: Clarifications to definition of "active element"

On Tue, 30 Jan 2001, Ian Jacobs wrote:

  Charles McCathieNevile wrote:
  > Comments at CMN

  >   >Content always determines what constitutes an active element. For
  >   >instance, the HTML 4 [HTML4] specification defines a number of
  >   >active elements: links, image maps, form controls, element
  >   >instances with a value for the "longdesc" attribute, and element
  >   >instances with scripts (event handlers) explicitly associated
  >   >with them (e.g., through the various "on" attributes). The role
  >   >of an element as an active element is subject to applicability.
  >
  > CMN I found the second paragraph extremely confusing. As far as I can tell
  > the User Agent in fact is the ultimate determinnat of whether there is a
  > behaviour associated with an element.

  But does the user agent add any behaviors? Or does it simply
  implement or not implement the behaviors specified by the author?

  In the same vein, the user agent ultimately determines what the
  user views, but the user agent doesn't just make up content.
  I don't think it makes up behaviors either.

  In this sense, active elements come from content.

   - Ian

Well, in this sense active elements come from the authors understanind of
HTML, but that isn't the definitive characteristic. The definitive
characteristic of an active element is the fact that it has behaviour
associated with it. There may be a constraint that this does not include
behaviour that is generated by the User agent, but I think this is wrong.

Example:
A user agent constructs an outline view of a page, by extracting the headers.
This is represented as a seperate page, that can be navigated to with a
standard command. (Lynx does this all the time. Amaya does it too, but always
opens a new window. Mozilla composer does it in the same window. The
representation, not the outline view, is what I mean is implemented). In
addition, the user agent adds linking behaviour to these items, which links
them to the coresponding point in the original document. I would argue that
no behaviour is specified by the content, and that the user agent has
attached the behaviour. As a technique for implemnenting a requirement of
UAAG.

I agree that we do not expect user agents to just randomly add behaviour to
content elements. I just disagree that this is therefore what defines active
element.

I took an offline action to propose a definition. I am going to do that using
the wai-xtech list, so that the people looking at the glossary can get it -
this term will I presume go into the glossary in whatever way UA defines it
in the first place, but there are a few other people that may have helpful
insights if needed. (Like "shut up charles, you're talking rubbish", or
something... <grin/>). My proposal will of course come back to the UAAG
group at large, I hope in the next few days.

cheers

Charles

-- 
Charles McCathieNevile    http://www.w3.org/People/Charles  phone: +61 409 134 136
W3C Web Accessibility Initiative     http://www.w3.org/WAI    fax: +1 617 258 5999
Location: I-cubed, 110 Victoria Street, Carlton VIC 3053, Australia
(or W3C INRIA, Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, France)

Received on Wednesday, 31 January 2001 00:39:48 UTC