[whatwg] about rich internat applications

Hi Dean,

> now that we are all chatting let's move the discussion on to *how* we
> are going to do this. rather than *why*. i'm not saying it isn't
> interesting though ;-)
> i'm kind of interested in what technologies will be involved, timescales
> etc. there isn't much information on the web site apart from the spec.
> is it too early for this kind of discussion?
> -dean
> 

I agree. I guess for the moment the only open source framework we can modify
is Mozilla, the other browsers are closed source. I have nothing against
that but only that I cannot do anything but wait for these vendors to do
something. After more than 10 years I still have nothing in hands except
notably from Microsoft with behaviors.

Here are the choices:

a) Implement something totally new in Mozilla and let it be alone in the
woods to support its new feature.
b) Embrace and expand. Implement behaviors in Mozilla and let people create
components on both Mozilla and IE.

Premisses:
a) It is very unlikely that Microsoft will improve its browser. They see
Avalon and XAML as the rich internet application platform.
b) There exist some available components (i.e. behaviors) available on the
market place. So behavior has some advantage here. 

Technical issues:
It implies that the CSS engine in Mozilla has to be modified to
expand/update not only the Gecko visual object but also the code attached to
it. Basically to add new prototype properties to elements.


What is important is that after code is bind to an element that we can
access its code as follow

<table id="sortTable" .....>
......
</table>

var sortTable = getElementById("sortTable");
var name = getElementById("name");
sortTable.sortBy("name");

Also that ECMAScript code can be written "object oriented" style like in the
source code bind to an element.

object.method = function() {
.....
} 

I am scanning Mozilla code to understand how I can do that....

Cheers
Didier PH Martin

Received on Monday, 7 June 2004 18:53:03 UTC