W3C home > Mailing lists > Public > public-webapps@w3.org > October to December 2010

Re: DOM3 Events last call comment

From: Charles McCathieNevile <chaals@opera.com>
Date: Fri, 22 Oct 2010 03:10:07 +0200
To: "WebApps WG" <public-webapps@w3.org>
Message-ID: <op.vkydi5t1wxe0ny@widsith.local>
On Fri, 22 Oct 2010 01:38:28 +0200, Doug Schepers <schepers@w3.org> wrote:

> ... I'm saying that when DOMActivate was first specced, in 1999-2000,  
> there wasn't a clean mobile-web model or significant use of inputs other  
> than keyboard and mouse, so click seemed to serve content authors just  
> as well as DOMActivate... they didn't need to think as much about an  
> abstraction that covered keyboard, mouse, touch inputs, voice, and  
> whatever, equally well.
>
> That dynamic has since shifted, and there is more need for an activation  
> event... just not necessarily DOMActivate, because of the other problems  
> with it.

For what it is worth, DOMActivate is largely my fault (although credit for  
good stuff is due to Rich Schwerdtfeger, Al Gilman, Philippe le Hegaret  
and many others).

The idea at the time was to replace the UI events around at the time with  
a set that were based on intentions rather than hardware-specific  
interactions, because I predicted that the existing problems of people  
building interactions that required a specific hardware paradigm would  
only get worse over time. I think that came true...

In the meantime, abstract intention-based events were added in parallel.  
Given the lack of good interoperable implementation and the ability to  
continue doing what they had done and figuring it more or less worked, Web  
developers didn't take them up, and the hardware-based events became more  
and more common.

Meanwhile, browser vendors and others worked to make the hardware-events  
sort of abstract (being able to fire a click in multiple ways, or adding  
extensions that synthesised events from a non-WIMP interface). So the  
abstract events continued to rot.

(Again, that wasn't a surprise, as discussed at the time).

I understand Doug's suggestion (in its strict relationship to this  
comment) to be "give up DOMActivate as a failure, and accept that the  
click event has effectively taken the role, for now". The Web APIs group  
(one of the fore-runners to Web apps) actually resolved that in 2006 at  
its first meeting, and I think we were right at the time and still do.

Meanwhile, there is now momentum to specify a better approach to events,  
and make it work. I think that deserves support as the best way to use our  
energy to get something better.

cheers

Chaals

-- 
Charles McCathieNevile  Opera Software, Standards Group
     je parle français -- hablo español -- jeg lærer norsk
http://my.opera.com/chaals       Try Opera: http://www.opera.com
Received on Friday, 22 October 2010 01:10:48 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 18:49:41 GMT