W3C home > Mailing lists > Public > public-geolocation@w3.org > March 2012

Re: Firing events in the deviceorientation spec

From: James Graham <jgraham@opera.com>
Date: Thu, 08 Mar 2012 15:36:45 +0100
Message-ID: <4F58C3FD.1050604@opera.com>
To: Lars Erik Bolstad <lbolstad@opera.com>
CC: Andrei Popescu <andreip@google.com>, public-geolocation@w3.org, Anne van Kesteren <annevk@opera.com>
On Thu 08 Mar 2012 03:26:12 PM CET, Lars Erik Bolstad wrote:
> On 08.03.2012 15:17, Anne van Kesteren wrote:
>> On Thu, 08 Mar 2012 15:09:44 +0100, Andrei Popescu <andreip@google.com>
>> wrote:
>>> On Thu, Mar 8, 2012 at 2:05 PM, Anne van Kesteren <annevk@opera.com>
>>> wrote:
>>>> Also, you already depend on HTML, which depends on the latest DOM, 
>>>> so it
>>>> seems really weird to then also depend on some really crappy out of 
>>>> date
>>>> draft.
>>>
>>> I think we could remove that dependency. What do you suggest we do? We
>>> can't keep this WG going just to wait for DOM4.
>>
>> You cannot both dispatch on Window and not depend on HTML. In fact, you
>> need to dispatch events from a task (that is queued) for which you also
>> need to depend on HTML.
>>
>> As for not maintaining standards, that is a bad idea. Just like not
>> maintaining software is.
>>
>>
>
> Anne, I agree that not maintaining standards is a bad idea.
> But we have a charter that expires.
>
> If we introduce the reference to DOM4, we're not going to be able to 
> get DeviceOrientation Events to CR/PR before this working group is 
> dissolved.

Whatever you put in the specification, you depend on DOM4 and HTML5 in 
the sense that a browser that implements DeviceOrientation will build 
it on top of implementations of DOM and HTML, and all modern browsers 
are targeting DOM4 and HTML5. Therefore the question is roughly "should 
the specification lie and pretend that those dependencies don't exist 
in order to potentially make some Process transitions smoother". I 
think it clearly shouldn't; it makes the spec worse and if the problem 
is purely one of Process it should be fixed at that level e.g. by 
asking for permission to advance despite the dependencies (in any sane 
world this son't be hard to get, since all current web platform specs 
have exactly the same issues), or by rechatering the group with the 
very limited remit of maintaining the documents.
Received on Thursday, 8 March 2012 14:37:20 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 22 March 2012 18:13:56 GMT