[talk-au] Mapping "off track" hiking routes

Andrew Harvey andrew.harvey4 at gmail.com
Fri Oct 23 03:41:01 UTC 2020


On Fri, 23 Oct 2020 at 11:28, Brendan Barnes <brenbarnes at gmail.com> wrote:

> Thanks Phil, I appreciate your insights on this.
>
> On the side issue, yes I definitely agree "informal path" track-tracing
> can have an effect on human use in the local, often wilderness,
> environments where they are recorded. However I have a preference to better
> tag these tracks eg abandoned=path or trail_visibility=no etc. Deleting a
> way on OSM loses the context of why it should no longer be used - it's just
> gone. A good example of this is
> https://www.openstreetmap.org/way/635206779 - by having access=no,
> removed=yes etc and a clear note on why not to use the way is useful to
> mappers, and it doesn't display on the popular renderers. If the way was
> deleted, in the future there'd be nothing to stop a well-intentioned
> armchair mapper creating a new way based on what they see on aerial
> imagery, who doesn't know about the no entry signage surveyed on the ground.
>

Agreed.

https://www.openstreetmap.org/way/688547036
<https://www.openstreetmap.org/way/688547036#map=17/-43.20539/147.77497> is
another example, but using the lifecyle prefix tagging
https://wiki.openstreetmap.org/wiki/Lifecycle_prefix so
abandoned:highway=path.

Simply tagging removed=yes doesn't tell data consumers what type of feature
was removed here. We don't tag historical features with no evidence on the
ground, but if there's still something on the ground we should try and tag
with the lifecyle prefix or the other style (highway=removed +
removed=path) so we know what kind of thing was removed.


> My question was more about tagging a hiking *route*, not a *track*. I'd
> like to produce AAWT route data from OSM. In real life, the AAWT route
> passes through Dairymans Creek area and I'd like OSM to convey that as it
> does form part of the one long overall AAWT hiking route. Comparing what I
> surveyed on the ground (GPS trace, and noting NPWS signage at wilderness
> boundary) to the official map (not licenced for use -
> https://theaustralianalps.files.wordpress.com/2013/11/map-15.pdf) all
> sources do mark the Dairymans Creek segment as forming part of the overall
> AAWT *route*.
>
> Perhaps Way 813749214 could be widened to an area covering the width of
> the plain? I'm open to ideas. My preference is for "something" to go here
> so OSM matches the context of the one overall AAWT hiking route, rather
> than "nothing" and leave a gap in the AAWT relation.
>

A route is usually a collection of ways, but in this case there is no
actual way for the section which should be navigated off track.

The fuzzy tag was previously proposed
https://wiki.openstreetmap.org/wiki/Proposed_features/Fuzzy as a way to
draw a line on the map with a buffer saying the feature is roughly this
area, but not exactly defined by the line. So while you could use an area
like you suggest, personally I'd opt for a roughly drawn way with fuzzy=yes
or something like that, you don't even really need any tags on that way if
it's a member of the route relation.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-au/attachments/20201023/336d37a2/attachment.htm>


More information about the Talk-au mailing list