RE: pointerID shouldn't have restrictions (was: Re: Last Call comments)

Hey Sangwhan and Konstantinov,

We reconsidered this and agree with you. It also seems to contradict the note below it that says essentially pointerId's should be randomd and developers can't assume any particular selection scheme.  So I've updated the spec [1] to remove this sentence per the resolution in our WG meeting this week [2]. Let me know if this change meets your expectations.

Thanks!

-Jacob

[1] https://dvcs.w3.org/hg/pointerevents/rev/1adb598e1d71

[2] http://www.w3.org/2013/04/16-pointerevents-minutes.html#item02


-----Original Message-----
From: S. Moon [mailto:innodb@gmail.com] On Behalf Of Sangwhan Moon
Sent: Thursday, April 11, 2013 10:22 AM
To: public-pointer-events@w3.org
Subject: pointerID shouldn't have restrictions (was: Re: Last Call comments)

On Friday, April 12, 2013 at 2:11 AM, Scott González wrote:
> On Thu, Apr 11, 2013 at 12:33 PM, Matt Brubeck <mbrubeck@mozilla.com (mailto:mbrubeck@mozilla.com)> wrote:
> > I haven't seen any justification for the pointerID == 1 requirement for mouse input. I agree with Konstantinov that it seems to serve no purpose, and I agree with Sangwhan that it provides a redundant and less-clear way to handle an already-covered use case. I'm also worried it encourages a misconception that other pointerIDs might have meaning other than as opaque identifiers.
> >  
> > Are there any objections to removing this sentence from section 3.1?
> > > "If the device producing the event is a mouse, then the pointerId must be 1. Device types other than mouse must not have a pointerId of 1." I have no objection. I believe there were others who didn't object the last time this was discussed, but I don't think there's a record of who was included in that group.  
> >  
>  

I for one would like to see this limitation go, as it gives the users a false impression that the the pointer IDs have some sort of meaning. I highly suspect that this will be not be the case, and quite possibly in every implementation will be completely different _unless_ the spec defines a pre-allocated range for each pointer type. (which I think is a even worse idea)

Sangwhan

P.S. I've changed the subject as this "Last Call comments" thread is branching into to way to many diverged discussions. The W3C mailer will retain the thread ID so it'll be in the same thread, but at least you'll know what this branch is about before you click on the mail.

Received on Thursday, 18 April 2013 23:46:32 UTC