RE: Comments about Network Information API

I would not expect most users to understand that question. Many (e.g. kids) will further have no clue if they are on a metered plan (bucket or capped or whatever is the local term - another issue in how the user is asked).

Metering will also vary significantly if the user is roaming or not, so a response at one time is useless if a roaming state changes. That's why the original SysInfo API provided roaming indication. Without being accurate per the current context, the metered attribute is not very helpful and is potentially harmful.

The metered attribute must reflect the actual current state, or it can't be trusted. Note that this also means it would in essence bring back the wifi/cellular and maybe even roaming awareness, if in a back-door way. I would be OK with that, because I think the privacy concerns on this API have been overstated, especially if the user must provide consent for access to it (which would resolve the whole question IMO).

Thanks,
Bryan Sullivan

-----Original Message-----
From: Mounir Lamouri [mailto:mounir@lamouri.fr] 
Sent: Wednesday, March 28, 2012 5:16 PM
To: public-device-apis@w3.org
Subject: Re: Comments about Network Information API

On 03/22/2012 07:33 AM, Cheng, Diana, Vodafone Group wrote:
> Here, I'm also not sure how 'metered' will be determined by the UA. If the user has to be asked, this will be done for every browsing context I guess and I don't find this ideal. 

In an ideal world, we could work with phone operators to get those
informations. Though, for the moment, we could indeed ask users. But
asking users doesn't mean for each context. That could be as simple as
"are you metered when using a mobile connection?".

--
Mounir

Received on Thursday, 29 March 2012 05:51:02 UTC