W3C home > Mailing lists > Public > www-multimodal@w3.org > October 2013

RE: Apparent inconsistency between W3C Pointer Event spec and EMMA 1.1 spec [Honeywell Internal]

From: JOHNSTON, MICHAEL J (MICHAEL J) <johnston@research.att.com>
Date: Fri, 11 Oct 2013 16:15:59 -0400
To: "www-multimodal@w3.org" <www-multimodal@w3.org>
Message-ID: <8BB06FCB-7743-47A5-86C9-61A4FBFAA43D@research.att.com>
Dear Steve,

Thanks for your thoughtful and detailed comments on the EMMA specification and
its relation to DOM2/DOM3 events and Pointer Events. We are discussing these 
issues now in the EMMA subgroup of the Multimodal Interaction activity and will
address them as needed in our ongoing work on future versions of the EMMA
specification. 

Some initial comments:

First regarding the relationship between DOM2/DOM3 events and EMMA. EMMA is an
XML markup language for representing user inputs and the various stages of
their processing. As such it plays a very different role from the eventing
mechanisms (DOM2/DOM3/Pointer), which operate within a page. The two might
well operate together in concert in certain applications, for example in a multimodal
 application one might have a handler on the page which captures e.g. a Pointer event, and
constructs an EMMA document which is posted to a multimodal 
interaction manager or dialog manager operating on a remote server. That server might 
then integrate that input with a spoken command and pass back an EMMA
document to the page with the combined interpretation of the touch
event with speech. I don't see EMMA being used within the page though as
a substitute for the eventing mechanisms for capture of input from
mouse/pen/touch etc within the browser.

Secondly, regarding the set of values emma:device-type. This a new annotation
being considered for EMMA 1.1, and by design, like emma:mode the set of values is 
intended to be open. The space of available devices is growing rapidly and we 
did not want emma:device-type to be limited to a predefined set of device types. 
Given that the set of values is open a more general 'touch' device type could certainly be 
added to the classification. We initially chose 'touchscreen' and  'touchpad' to 
differentiate input on e.g. touch sensitive screen on phone/tablet from 
touch pad on a laptop. 

Thanks also for all of your editorial comments. We are tracking those
and will incorporate then into the next draft of the specification. 

best
Michael Johnston
AT&T
Received on Friday, 11 October 2013 20:16:43 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:06:38 UTC