Re: Draft Charter of Emergency Information TF (was: Re: [DRAFT] Disaster Prevention and Response TF charter

On Oct 4, 2011, at 10:33 PM, Giuseppe Pascale wrote:
> On Tue, 04 Oct 2011 14:13:36 +0200, Silvia Pfeiffer <silviapfeiffer1@gmail.com> wrote:
> 
>> 2011/10/4 Giuseppe Pascale <giuseppep@opera.com>:
>>> My understanding of this TF work is not to create new standards/protocols in
>>> the area of emergency notification,
>>> but to review the existing ones and see if anything needs to be done to
>>> support such notifications in a web browsers.
>>> 
>>> One of the outcome could also be: nothing needs to be done.
>>> 
>>> My personal feeling on this is that notifications should not be handled at
>>> web application level but at a UA level (for end user notifications, I
>>> mean).
>> 
>> I wonder how this would work. Are you suggesting that every browser
>> (UA) when it goes online registers with a national notification
>> service from which it would get emergency notifications if there are
>> any to be delivered?Seeing as the Web is fundamentally a
>> pull-information based infrastructure, pushing information can only
>> work if the UA allows it (e.g. RSS feed style). However, as soon as
>> you make the information-push UA-dependent and not user-dependent, you
>> run into all sorts of privacy issues.
>> 
>> For example, if all UAs in the US had to register with a US agency as
>> soon as they go online, that single agency would know everything about
>> when everyone in the US is going online, their IP addresses and their
>> devices.
>> 
>> Why not just go with a Web application, such as an RSS feed to which
>> you can subscribe that gives you emergency notifications to those
>> channels that you usually communicate on (could, e.g. be twitter,
>> facebook, google+, email, RSS reader etc)?
>> 
>> 
> 
> What I was trying to highlight is that if emergency notification is something that can save your life, having to rely on the application to timely show you a message or having it only for some type of content (e.g. video) doesn't feel safe.
> 
> To be honest, I'm wondering if this is something related to UA at all. What we are actually saying here is that when there is an emergency all possible channels (including Internet) should be used to notify about the emergency. Connection to a network (being it the Internet or any other network) is something in control of the OS. Also, visibility of application is something in control of the OS (you don't want to miss your earthquake notification just because your were watching a movie and not looking at (or not using at all) your browser.
> I also think that most devices are connected to internet also when not browsing.
> 
> So it feels to me this is something to be handled at OS level, if it needs to be robust. And at a network protocol level, to make sure all networks you may connect to (broadband, DVB, others) provide this kind of functionalities (many already do).

I see your point, but I'm not sure for now that which layer is the best place to implement specific functions related to emergency information or disaster response. I think it would help us to clarify and classify use cases and requirements before discussing the layers. There may be something you and I were not aware in this area, and I am looking forward to hearing or finding something new I've not yet noticed even thought I'm an expert on Emergency Warning System through broadcasting (DVB, ISDB, etc).

My two cents.

Although the IG has currently strong focus on HTML5, HTML.next, or web browsers, its scope is not restricted to them as we described in the charter. Many other activities in W3C such as SemanticWeb, Voice Browser, MMI, eGov, Off-line Web, etc may be able to get benefit from the deliverables of the IG.

Yosuke


> /g
> 
> 
> 
> -- 
> Giuseppe Pascale
> TV & Connected Devices
> Opera Software
> 

--
Yosuke Funahashi
co-Chair, W3C Web and TV Interest Group
Researcher, Keio University Research Institute at SFC
Board Director, Tomo-Digi Corporation

Received on Wednesday, 5 October 2011 02:39:23 UTC