[OSM-talk] stop deleting abandoned railroads

[moltonel 3x Combo]

On 23/08/2015, mick <bareman at tpg.com.au> wrote:
> On Sun, 23 Aug 2015 00:09:43 +0100
> moltonel 3x Combo <moltonel at gmail.com> wrote:
>> Do people actually do this ? It sounds like a strawman argument to me.
>> I do a fair bit of walking and cycling, and when planing a trip I look
>> at the global topographic data but it never occured to me to look for
>> railroads. Why use the local railroad hint when you've got the global
>> DEM data ?
>
> How fine is the granularity of the DEM data?

Between 30 and 90m depending on location, when you look at the most
common publicly-available data. Most of the world is at 30m now, and
most of the really bad artefacts have been fixed.

30m is plenty of granularity when you're planning a walk or a cycle.
It can miss a cutting or an embankment, but those areas are normaly
flat enough to begin with that you wouldn't have been put off by the
DEM data alone. I'm sure there are extreme cases where this isn't
true, but you still want to mainly look at DEM most of the time.

Tags that I actually look at when choosing an osm path/footway/track
is surface, tracktype, and sac_scale.

http://www2.jpl.nasa.gov/srtm/
https://en.wikipedia.org/wiki/Shuttle_Radar_Topography_Mission
https://en.wikipedia.org/wiki/Advanced_Spaceborne_Thermal_Emission_and_Reflection_Radiometer