RE: Architectural Consistency Requirements for Forms

Hello,
 
I agree that templates that are instantiated dynamically based on the
model are a more powerful feature that repeating. For instance,
indefinitely deep nesting (e.g., tree widgets [1]) is not possible with
repeating, which requires to explicitely state the level of nesting in
the UI markup. Also, supporting different types of model items is easier
with templates.

Of course, templates require the use of some kind of a datamodel, which
in context of WF2 could be the run-time DOM. One option would be to use
a subset of XBL.

-mikko

[1] For tree widgets, I proposed and implemented a tree construct for
XForms
http://lists.w3.org/Archives/Public/www-forms/2006May/0056.html . This
is not as general as templates, though.

________________________________

	From: public-html-request@w3.org
[mailto:public-html-request@w3.org] On Behalf Of ext David Hyatt
	Sent: 04 May, 2007 03:22
	To: John Boyer
	Cc: HTML WG; public-forms@w3.org; Maciej Stachowiak
	Subject: Re: Architectural Consistency Requirements for Forms
	
	
	Here is a link to a bunch of documents describing XUL templates.
Note that the model back end that XUL templates use is RDF, and I am not
advocating that at all.  Imagine the feature reformulated to operate on
an XML DOM back end instead of an RDF graph. 

	I think there is overlap here too with the <datagrid>.  There
should be architectural consistency in how it binds to a model and
functions with that model and any more generic feature that also wants
to bind to a model.

	http://www.mozilla.org/docs/xul/xulnotes/template-primer.html

	dave
	(hyatt@apple.com)

Received on Friday, 4 May 2007 07:45:26 UTC