[whatwg] Re: Rendering Unknown Elements and IE Support

[Dean Edwards]

Matthew Thomas wrote:
> For Internet Explorer, the WhatWG specs will be implemented externally 
> using HTCs.
> 
> Therefore, perhaps, more likely than any of the items on your list is 
> one of the following:
> 
> *   Use a patent lawsuit (whether or not it is ultimately
>     successful) as justification for permanently removing/
>     /reducing support for HTCs. (Precedent: Using the
>     ultimately-unsuccessful Eolas lawsuit as justification
>     for permanently removing support for Netscape-API
>     plug-ins.)
> 
> *   Use a security vulnerability (whether or not it is
>     otherwise fixable) as justification for permanently
>     removing/reducing support for HTCs. (Precedent: Using a
>     security vulnerability as justification for permanently
>     removing support for the Gopher protocol.)
> 
> I don't know whether HTCs are used too much elsewhere to make that 
> infeasible, but it is perhaps something to think about when drafting the 
> Web Apps and Web Controls specs.
> 
microsoft use HTCs on their main web site:
http://www.microsoft.com/
the site uses a behavior called "default.htc" to provide flyout menus.
behaviors are also used on the msdn site:
http://msdn.microsoft.com/library/
a behavior called "deeptree.htc" provides the tree view control. the 
control is quite complex and uses IE's inbuilt XSL transformer to load 
and transform xml documents into leaf nodes.

if we speculate that microsoft will "spoil" whatever IE implementation 
we provide then we may as well not bother. to withdraw support for HTCs 
would be a major climbdown. it would also annoy some large corporations 
who have deployed them on their corporate intranets.

there is a fallback position in light of such a catastrophe. 
implementing WF2 functionality using javascript alone would provide an 
acceptable level of client-side support for this platform.

-dean