Re: Expand of WG’s Target Input Devic es - related new chater

On Tue, Nov 1, 2011 at 5:37 AM, João Eiras <joaoe@opera.com> wrote:
> On Mon, 31 Oct 2011 23:53:26 +0100, 전종홍 <hollobit@etri.re.kr> wrote:
>> 2) Gamepad [related :
>> http://dvcs.w3.org/hg/webevents/raw-file/tip/gamepad.html ]
>>
>
> I don't see any special needs for gamepad specific events. These can be
> easily supported using standard key events plus the joystick API. Plus, the
> keyIdentifier property of keyboard events can return the typical key names
> (A, B, C, D, Select, Enter, L1, L2, R1, R2, etc).

Incorrect.  Modern games and controllers use analog buttons, while key
events assume digital buttons.  One could potentially extend normal
key events to return analog data, but unless we expect standard
keyboards to actually support that sort of thing, it's probably better
to handle it specially within the Gamepad API.


> Plus, it is hardware bias and as such, will break if I want to use a
> keyboard instead of a gamepad, so the user agent will have to go through
> extra hurdles just to map the keyboard into gamepad events. Vice versa also
> applies. The user might want to use a game pad to play a game that just uses
> regular key events, and the gamepad specific API makes sure that won't
> happen.

It seems completely reasonable to provide an API for mapping portions
of a keyboard into a "virtual gamepad".  I don't know if this needs to
be in v1 of the Gamepad API spec, but it's still a reasonable request.


>> 3) Trackball
>>
>
> This is the exact same thing as a mouse. Simple two dimensional pointing
> device with a click button. Don't understand the extra features it needs
> that aren't present in mouse events.

Agreed.  (I'm using a trackball right now - it's just an upside down mouse.)

~TJ

Received on Tuesday, 1 November 2011 20:37:36 UTC