Re: [profile] A new TF for the TV Profile

On Thu, 19 Jan 2012 21:42:46 +0100, Scott Wilson  
<> wrote:

> On 19 Jan 2012, at 18:23, Giuseppe Pascale wrote:
>> Hi all,
>> given the interest in working on the TV profile as announced in [1], is  
>> probably good to start a new TF in charge of it.
>> I'll work in the following days to create a wiki page for such new TF.
>> If someone is willing to chair this new TF, please let me know.
>> About phone calls, the idea was to start with bi-weekly calls, let me  
>> know if there are any objections to this.
>> Finally, feedbacks on the profile document (that meanwhile got some  
>> more text) are welcome
>> The document contains some open question you may want to share your  
>> opinion on.
>> Other open question where contribution are welcome are:
>> - which technologies are fundamental for TV deployments and should be  
>> included in the profile?
> I think there needs to be a section on web application packaging and  
> metadata referencing W3C Widgets: Packaging and Configuration [1]; an  
> issue identified in several submissions to the Future of Offline Web  
> Applications meeting was competing non-interoperable packaging formats  
> affecting TV/STB web app deployment. Some vendors are already using  
> similar packaging formats (e.g Samsung's smart TV profile); there is  
> also work taking place in the Webinos[2] project on extending W3C  
> Widgets for TV applications.

Good suggestion. W3C widgets is now W3C Recommendation and is referenced  
by several TV specs, for example OIPF specs.

I added a section about the application model and about client side  

For now I only added a reference to the packaging spec, but we need to  
consider other specs of the widgets family

>> - how do we address evaluation of performances?
>> - what kind of input methods should be considered?
> I think traditional Remote Control input methods are a basic requirement  
> for TV-based apps,

There are two issues here:
- remote controls are evolving and devices addressed by this specification  
are not only TV sets/STB. So we need to consider the option that a  
"traditional" remote control is not available.
- historically different specifications have used slightly different  
approaches to handling of rc inputs.
When those specs were written, DOM didn't have specific support for RC  
control input.

Now there is some text in the DOM3 spec

but AFAIK nobody has implemented it yet. We should discuss how to proceed  
in this area, if deprecate old approaches in favour of the DOM3 approach  
or if we should talk to the group working on the DOM3 document and propose  
an alternative approach.

> as are virtual keyboards (most TV interfaces use both in combination).
not sure
> Further down the line, there is Kinect-type camera input coming from  
> Samsung.
What would be needed to be standardized here? I expect that for general  
purpose application these "devices" will just emulate traditional remote.
I mean, even if such device recognize a "gestures" to do play/pause/stop  
such gestures will probably be mapped on "equivalent" RC keys.

Of course more complex use cases (gaming) may require something more, but  
this is outside the scope of this profile I believe; could be an  
interesting topic for the web events WG.


>> Feel free to share your opinion but also to point to existing work in  
>> these areas.
>> /g
>> --
>> Giuseppe Pascale
>> TV & Connected Devices
>> Opera Software
> [1]
> [2]

Giuseppe Pascale
TV & Connected Devices
Opera Software

Received on Friday, 20 January 2012 12:21:18 UTC