W3C home > Mailing lists > Public > public-web-intents@w3.org > December 2011

Re: Web Intents - Scenario: TV System (part 6)

From: Greg Billock <gbillock@google.com>
Date: Wed, 7 Dec 2011 11:12:17 -0800
Message-ID: <CAAxVY9cr15ve-QDsD8-rsYrfnJOMFHVBqiHfMLc7HUQavpZw7A@mail.gmail.com>
To: WebIntents <public-web-intents@w3.org>
On Wed, Dec 7, 2011 at 10:03 AM, Bjartur Thorlacius
<svartman95@gmail.com> wrote:
> What information does the web page have to better make decisions as to what
> verb to use that the user and UA do not, and why can the page not
> declaratively communicate that information to the UA? Why is there a web
> page involved at all?

There could be some other mechanism to control these devices (perhaps
even one the UA controls). My understanding of this discussion is that
it's geared specifically about whether the Web Intents spec as
proposed (or as altered) would be appropriate to the use cases of web
pages interacting with this equipment. Clearly, in this case, the UA
needs to provide some sort of hook such that web content can reach
through the UA and get a hold of local gear. The question before us,
then, for the purposes of this API, is whether and how it would be
sufficient to expose to web content whatever (much more complex!)
interaction there is going on between the UA or local-resident
controller and the consumer electronics gear that is being controlled.

Use cases, in my mind, are more like "play this video stream on this
particular TV" and "grab this local content off my media server to
upload" and "mute that amplifier." That is, they are particular
"launches" of actions that can be bound by the UA to a potentially
very complicated interaction going on behind the scenes. That is, even
if we decide exactly how Web Intents will work to surface those use
cases to web content, there will still be a great deal of work and
thinking to do in figuring out how UAs ought to discover and interact
with HN gear at the local network level.

If the Web Intents API is being suggested as being applied more
broadly -- that is, something that wouldn't be anchored to web content
-- that's something I really haven't considered. I think we should
focus on applications to the web ecosystem, and that broadening like
this won't be fruitful. If it turns out that there are clever ways for
the UA to take advantage of the work to provide functionality to
non-web-content apps or whatever, that's fine, but I think our primary
responsibility ought to be to web content developers and users.

Received on Wednesday, 7 December 2011 19:12:46 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:14:45 UTC