[OSM-talk] Unmapped areas and map confidence

Ben Robbins ben_robbins_ at hotmail.com
Mon Feb 19 19:26:51 GMT 2007


>In addition to what you're saying, it is also of interest to know as much 
>as possible about the accuracy of data. If I drive along an already mapped 
>road, twice, and find my track to be 10 metres south of what is mapped on 
>each journey, then I tend to move the existing way - assuming it must have 
>been done with older hardware or on a bad day. However if there was some 
>tag telling me that the old way was professionally measured using full WAAS 
>correction and aided by inertial navigation systems, then I'd probably 
>stand back and leave the way alone ;-)

This is true.  If I have a handful of gps tracks goign down a road where the 
signal is poor, the actual path of the road may not be just the average.  
Often its better if I use a slower means of transport, or If there are not 
leaves on the trees and hedges.   To get over this problem though wouldn't 
there need to be some way of labelling sections of a gpx file as different 
grades of acuracy, so as to be able to have them display differently in the 
editing program?

>Sometimes I use the source tag to confer a level of confidence (typical: 
>"landuse=forest, source=landsat"

I have been very appriecative when people have done that and I have come 
across the road to find gps data now exsists for it.  I think it should be 
encorraged.

Ben

_________________________________________________________________
MSN Hotmail is evolving – check out the new Windows Live Mail 
http://ideas.live.com





More information about the talk mailing list