[uf-discuss] geo - accuracy of coordinates

[Chris Casciano]

On Oct 2, 2006, at 5:41 PM, Colin Barrett wrote:

> On Oct 2, 2006, at 10:58 AM, Andy Mabbett wrote:
>
>> Or the capacity to describe a polygon...
>
> I call the 80/20 rule into effect here. A radius is Good Enough. We  
> don't need to re-implement KML here.

I think somehow we jumped right from defining the accuracy of a  
'point' into defining an area.

Since I've started using flickr's geo tagging I often finding myself  
wanting to say something "somewhere over here" when tacking the  
proverbial thumbtack onto the map and marking something as "all of  
this sort of rectangular Central Park"

To the earlier discussion I don't think precision in decimal points  
cuts it either as a legitimate solution because unless you're pulling  
data from a GPS unit at time of capture the fuzzyness isn't in the  
tools measurement capabilities its in the ability of humans to be  
precise in their description or recollection (along with plenty of  
other issues of convention and usage on the web not necessarily  
caring about significant digits in that manner).

So whats the solution? not sure... I'd love to see sites like flickr  
allow a user to specify how precise a marking is so you don't get a  
mish-mosh of very specific points and sorta-near-here points, but I  
don't know what that means for the geo mf, if anything.

-- 
[ Chris Casciano ]
[ chris at placenamehere.com ] [ http://placenamehere.com ]