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

Re: First stab at System Info & Events API

From: Max Froumentin <maxfro@opera.com>
Date: Mon, 26 Oct 2009 13:58:34 +0100
Message-ID: <4AE59CFA.3030902@opera.com>
To: "Tran, Dzung D" <dzung.d.tran@intel.com>
CC: "public-device-apis@w3.org" <public-device-apis@w3.org>
On 21/10/2009 05:24, Tran, Dzung D wrote:

> http://dev.w3.org/2009/dap/system-info/Overview.html

Impressive! A few comments (sorry if I've missed some text that voids 
some of them)

- I still don't see why Compass should be in there. It overlaps too much
with W3C's geolocation API and introduces oddities such as latitude and 
longitude attributes being part of the CompassPosition object. Other 
attributes such as declination, if found useful, should be submitted to 
geolocation v2.

- The IDL should have a '?' for optional nullable attributes
http://dev.w3.org/2006/webapi/WebIDL/#idl-nullable-type

- numLogicalProcessors: it seems normal to provide that information, yet 
it is not consistent with the fact that we don't indicate if there are 
more than 1 battery, 1 or more compasses, etc.

- colorDepthChangedCallback: is that really useful? If it's about 
displays being added or removed, then I would argue for having instead
a displayChangedCallback function, that returns the new dimension, 
orientation and depth

- objects like Power expose their attributes directly (and keep them up 
to date, I suppose). That's countrary to objects like Compass (or 
geolocation's Position) which can only be returned from a callback. For 
the sake of consistency I would recommend changing CPU, say, to:

interface Cpu {
     // attributes like usage, replaced with:
     void getCpuUsage(successCallback, [Optional] in CpuUsageOptions);

     long watchCpuUsage (in CpuUsageCallback successCallback, [Optional] 
in CpuUsageErrorCallback errorCallback, [Optional] in CpuUsageOptions 
options);
     void clearWatch (in int watchId);
};

- There should be a way to open the lists of possible cpu architectures, 
network connection types, etc. (some people
still write browsers for Motorola CPUs, for instance)

Cheers,
Max.
Received on Monday, 26 October 2009 12:59:10 UTC

This archive was generated by hypermail 2.3.1 : Monday, 23 October 2017 14:53:39 UTC