geographical accuracy

[Graham Seaman]

Andy Armstrong wrote:
> On 21 Feb 2007, at 10:19, Graham Seaman wrote:
>> Can anyone with a GIS background tell me if my assumptions are wrong 
>> here, and more to the point if there's anything obvious I can do to 
>> improve my results?
>
> Geography::NationalGrid::GB uses that NATGRID projection by default 
> and you need WGS84 for Google. From looking at the source it looks as 
> if you can pass a Projection option to the constructor to specify 
> which projection to use. The shift between the two projections is 
> certainly about right to account for your error.
>
I set up a new 'Projection' as below:

 GOOGLEMAP => {
                'scalefactor' => 0.9996012717,
                'phio' => Geography::NationalGrid->deg2rad(49), # 
several km south-east of St Helier, Jersey
                'lambdao' => Geography::NationalGrid->deg2rad(-2),
                'Eo' => 400000,
                'No' => -100000,
                'ellipsoid' => 'wgs84',
                'info' => 'National Grid for Google',
        },

This is identical to the default Projection except that I have changed 
the ellipsoid to wgs84 as advised.
This improves the output but by a tiny amount (typically a change in the 
4th decimal place of the lat/long) which is smaller than the error.  Do 
you know what the correct scalefactor  for wgs84 is? I've tried googling 
with no results so far...

Thanks
Graham
> Is that reasonable?
>
> Disclaimer: I haven't used Geography::NationalGrid::GB but I have my 
> own code that does the same thing.
>
> --Andy Armstrong, hexten.net
>