[uf-discuss] geo - accuracy of coordinates

Ryan King ryan at technorati.com
Thu Oct 5 10:57:42 PDT 2006

On Oct 3, 2006, at 1:17 AM, Andy Mabbett wrote:

> In message <CE5C0319-0EFF-46D7-AAF4-8DA0C4339353 at technorati.com>, Ryan
> King <ryan at technorati.com> writes
>> you don't need to have lat/long to get distance. There  are plenty  
>> of services for translating human readable addresses into  machine  
>> readable values.
> But with much less accuracy than lat/ long

Granted, when converting from human readable location information via  
a geocoding service (or any other data source), the information will  
be less accurate and less precise.

But, in the interest of "humans first, machines second" and building  
on existing practices, I think it makes sense to avoid introducing  
artificially precise data when the human readable information is  
sufficient and more likely to be maintained.*


* I consider lat/long data as opaque metadata, which is likely to  
fall victim to the same problems as hidden metadata.

More information about the microformats-discuss mailing list