Re: PositionOptions.timeout & UserAgent permission

Yeah, I think it must be clarified, as I think that starting the  
timeout once the user accepts does not match the wording of the spec.

I'm afraid I still think that timeout, the old lastPosition, and  
maximumAge are still muddly. These are all implementation details. All  
the developer wants is a position. Developers may basically either  
want a precise position or a lazy position. And when I think of  
developers, I do no think many are going to want the lazy position at  
all. Why not just give them less options instead of more that they  
will always set to precise?

If a precise position can be given, it will be. If not (and that is up  
to the UA and the backing implementation), it will give a lazy  
position if it can. If not, return an error callback. Simplified API,  
simplified user code, and simplified UA implementations.

-- Greg

On Mar 14, 2009, at 8:27 PM, Doug Turner wrote:

> Hi Greg,
> I have interpreted the PositionOptions timeout starting point when  
> the user accepts the geolocation request.  if the user never clicks  
> "allows", there will be no callbacks.  Sounds like we need a bit of  
> clarification here.
> From the user's perspective, if the timeout happens at the same time  
> the user grants access, they might be confused at this behavior.   
> However, from the developers point of view, not getting a timeout  
> failure in this case might require them to hook up a separate timeout.
> Doug
> On Mar 14, 2009, at 7:26 PM, Greg Bolsinga wrote:
>> Hello --
>> I have a implementation detail question about the PositionOptions  
>> timeout.
>> What if the UA has to put up a UI that asks the user to allow  
>> Geolocation, and then what if the user takes longer to respond than  
>> the timeout? If the user allows it, and the timeout has expired,  
>> which wins? Does the page get the position, or the timeout error?
>> In my implementation, by the time the user is requested for  
>> permission, a location is extremely likely to be known, so I can  
>> stop the timeout then, but I just want to be sure.
>> Thanks,
>> -- Greg

Received on Sunday, 15 March 2009 19:33:22 UTC