W3C home > Mailing lists > Public > public-device-apis@w3.org > May 2009

Re: Draft Charter for a Device API and Security WG

From: Robin Berjon <robin@berjon.com>
Date: Thu, 14 May 2009 11:12:07 +0200
Cc: David Rogers <david.rogers@omtp.org>, Arthur Barstow <Art.Barstow@nokia.com>, Marcin Hanclik <Marcin.Hanclik@access-company.com>, public-device-apis@w3.org, Philipp Hoschka <ph@w3.org>, Nick Allott <nick.allott@omtp.org>
Message-Id: <D272CA20-65E9-42DA-9A6B-48966DD1DA86@berjon.com>
To: Dominique Hazael-Massieux <dom@w3.org>
On May 14, 2009, at 08:51 , Dominique Hazael-Massieux wrote:
> Le mercredi 13 mai 2009 à 21:38 +0100, David Rogers a écrit :
>> In general we are very pleased with the charter, so here are some
>> small changes from us:
>>
>> 1) Title: "Device APIs and Security Working Group Charter"
>
> I'm not sure about adding "and Security" to the name of the group:
> * it makes the name quite long and mouthful
> * it reads awkwardly: it won't work on device security, nor on  
> security
> in general, but on the securing access to device APIs which that title
> doesn't convey
>
> I don't feel very strongly about it, so I can add it back if others  
> do.

Neither do I, just some suggestions in case they help consensus:

   Securely Exposed Device APIs
   Securely Accessible Device Interfaces
   Device APIs and their Security Model

Dave: the content of the charter and the names of the specifications  
matter a lot more than the name of the WG.

>> * User Interaction API
>> 	- A set of APIs that gives a widget or website far better control  
>> of how it manifests itself on different platforms. This would  
>> include minimise/maximise 	functions, window size, alerting  
>> mechanisms etc.
>
> Hmm... Is that really a "device service" API? It feels more like
> something for the WebApps Working Group. Is it so strongly related to
> the others APIs that it would need to be developed in this group?

What exactly falls under WebApps and this WG has been rather fuzzy  
from the start — that's something that's likely to crop up during the  
WGs' lifetimes and that we'll have to deal with. I think that the way  
one could draw the line for this one is that access to core window  
manager functions is rather low-level (from a web point of view) and  
more "device". If WebApps were to define something for UI, it would be  
in-browser UI (à la XUL), whereas this is access to platform UI*.

Initially that API is there because it's in BONDI, and it's in BONDI  
because there was a strong consensus that there was a need for it (or  
at least for a limited version of it) and only a small minority push- 
back.

There's a case to be made for it. One of the things it could provide  
is a DI access to the menu, which is very useful on mobile devices  
(that tend to have two contextual menus) and could usefully replace  
the device-dependent ones that keep cropping up. That being said the  
way it is currently phrased seems like an invitation for runaway scope  
creep, though that can probably be handled within the WG.

* not to get into a naming rat-hole, but maybe this should be Platform  
rather than Device? Platform Access Framework?

-- 
Robin Berjon - http://berjon.com/
     Feel like hiring me? Go to http://robineko.com/
Received on Thursday, 14 May 2009 09:12:48 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:13:59 GMT