W3C home > Mailing lists > Public > www-dom@w3.org > January to March 2011

Re: PFWG comments on DOM 3 Events [WebApps-ISSUE-175]

From: Doug Schepers <schepers@w3.org>
Date: Wed, 05 Jan 2011 19:11:39 -0500
Message-ID: <4D2508BB.1080502@w3.org>
To: Michael Cooper <cooper@w3.org>
CC: www-dom@w3.org, List WAI Liaison <wai-liaison@w3.org>, List WAI PF <w3c-wai-pf@w3.org>
Hi, PFWG-

Thanks for your comment.  We understand and empathize with your concerns 
regarding the deprecation of both mutation events and the 'DOMActivate' 
event.

Regarding mutation events, we have at least two possible solutions for a 
more performant replacement.  The first was proposed by Microsoft, and 
is the basis for the Asynchronous DOM Mutation Notification deliverable 
listed in the WebApps WG charter [1].  An alternate, more generalized 
solution is the NodeList proposal by John Resig [2], if John is amenable 
to standardizing that at W3C.  In either event, we are committed to 
publishing a first public working draft of some solution within the same 
timescale as DOM3 Events is in Last Call, or in Candidate Recommendation 
phase at the very latest.

Regarding the 'DOMActivate' event type, while the Web Events WG may 
include a wider set of higher-level intentional events, the DOM3 Events 
specification does not depend upon the deliverables of that group.  As 
stated in DOM3 Events, the current and foreseeable best practice, and 
the solution most commonly used by Web developers, is the 'click' event 
type; this event type, despite the unfortunate name, has been 
implemented and post-facto defined to meet the majority of use cases, 
including accessibility use cases, for which 'DOMActivate' was first 
defined, and has worked cross-platform and cross-device for several 
years, and will continue to do so regardless of any other high-level 
event types, because it ensures backwards compatibility with existing 
Web content, and because content authors will continue to use it.  I 
want to be perfectly clear that, effectively, 'click' *is* the most 
accessible solution at this point, and to be pragmatic, accessibility 
advocates should understand and propagate that fact.  That said, there 
are use cases that having a higher-level 'activate' event type might 
solve better, and the Web Events WG will certainly be exploring those 
going forward.


In acknowledgment of your concerns, I have noted this as ISSUE-175 in 
our issue tracker, and have marked it as a Last Call comment, to ensure 
that your concerns are addressed formally in the Director's review of 
our transition requests.

[1] http://www.w3.org/2010/webapps/charter/Overview.html#mutation
[2] https://github.com/jeresig/nodelist#readme
[3] http://www.w3.org/2008/webapps/track/issues/175

Regards-
-Doug Schepers
W3C Team Contact, SVG, WebApps, and Web Events WGs


Michael Cooper wrote (on 1/5/11 12:48 PM):
> The Protocols and Formats Working Group is submitting this comment on
> the DOM Level 3 Events Last Call Working Draft of 7 September 2010
> [http://www.w3.org/TR/2010/WD-DOM-Level-3-Events-20100907/].
> Authorization to send this as a formal comment of the PFWG is archived
> at http://www.w3.org/2011/01/05-pf-minutes.html#item07. We know you are
> aware of our concerns and we have coordinated with you extensively
> outside of the public comments channel to arrive at appropriate
> solutions. We decided that it was important also to register our
> concerns on the public record as formal public comments; hence this message.
>
> The PFWG has two concerns about DOM Level 3 Events:
>
>     * The deprecation of mutation events;
>     * The deprecation of the "DOMActivate" event and, more generally,
>       the need for expanded device-independent events.
>
> Regarding mutation events, we understand that they are not well
> supported as defined and there are plans to meet the use case in the
> future with an improved solution. Further, we acknowledge that
> "deprecated" does not mean "not supported". Therefore we do not wish to
> register an objection to this. But we do note that the availability of
> an improved solution in the future is critical and anticipation of this
> is our reason for non-objection to the current state.
>
> In discussion with you about device independent events, there was
> consensus between both the PFWG and the Web Applications WG that there
> is a need for a new class of device-independent events. This is critical
> to our ability to produce accessible solutions on mobile devices that
> make require gestures to command and control applications. As an
> outgrowth of our process with you and other processes, a Web Events
> Working Group has been chartered to take on work that we believe will
> meet the requirement. Our potential objection to issues in DOM 3 Events
> is addressed by, and contingent on, this work.
>
> In summary, potential PFWG concerns about DOM 3 Events are addressed by
> work planned to take place in the future. While we believe all relevant
> parties intend in good faith to execute this work, the outcome of future
> work cannot be guaranteed. Therefore it is important for us to register,
> as a formal comment on DOM 3 Events, that should the above work fail to
> achieve the expected deliverables, our concerns with DOM 3 Events would
> be renewed.
>
> We do not request any edits to the specification, merely recording of
> this comment in the issue disposition report when you advance to the
> next stage of the W3C Recommendation process.
>
> On behalf of the PFWG,
> Michael Cooper
> --
>
> Michael Cooper
> Web Accessibility Specialist
> World Wide Web Consortium, Web Accessibility Initiative
> E-mail cooper@w3.org <mailto:cooper@w3.org>
> Information Page <http://www.w3.org/People/cooper/>
>
Received on Thursday, 6 January 2011 00:11:45 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:14:07 GMT