[OSM-talk] altitude information

Frederik Ramm frederik at remote.org
Thu May 31 00:58:15 BST 2007


Hi,

> Altitude data is already freely available, and GPS precision is really 
> really poor on Z-axis.

I am not an expert here, especially I don't know how good existing 
altitude data is (in terms of lat/lon precision).

I imagine a road going along a ridge; 50m to one side of it you're 
already 500m deeper. Knowing whether the road is on the ridge or in the 
valley is a valuable piece of information (especially, I should imagine, 
for walkers and cyclists).

In this situation, we'd need something like 20-metre lat/lon precision 
on the existing altitude data to determine whether the road is on the 
ridge or in the valley; but we would only need 100-metre altitude 
precision on the GPS track to reliably say where the road is. (I guess 
this is what Rainer meant by high altitude deltas.)

And what about GPSes with built-in pressure altimeters? If we accept 
height data from those, that would revive the old question of how to 
record precision metadata ("this elevation was recorded using an old 
e-trex, don't trust it / this elevation was recorded with a modern super 
duper hillwalking GPS with altimeter and ice crusher...").

Anyway, whatever the answers to these questions are, there should really 
be a little chapter in the Wiki FAQ "why don't you record height 
information" because it is a question that pops up quite often!

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00.09' E008°23.33'




More information about the talk mailing list