[Tagging] Stop the large feature madness (was: Tag for a plateau or tableland?)

Joseph Eisenberg joseph.eisenberg at gmail.com
Wed Apr 17 13:02:33 UTC 2019

I did not intend to encourage mappers to start adding giant
multipolygons for the Tibetan plateau or the Colorado Plateau. In fact
I'm doing my best to discourage mappers from adding non-verifiable,
huge areas to the database: see

I asked about an appropriate tag for tablelands / mesas / plateaus
after finding that several place=locality nodes in Andorra are used to
map features named "Pla de *” (x2) and "Planell(s) *" (x3), both of
which are likely types of plateau, if I'm reading the Catalan
dictionary correctly. It would be nice if many of the place=locality
nodes could be updated to a more specific tag, and there didn't seem
to be anything that fit other than natural=plateau.

Personally, I don't plan to add any closed ways, let alone
multipolgyons, to map large plateaus, though I will check if the Table
Rocks near Medford OR are tagged correctly:
https://en.wikipedia.org/wiki/Upper_and_Lower_Table_Rock - tablelands
about 1 and 2 square miles in size with clearly defined borders
(cliffs) on all sides. These would actually be an example of a feature
that does have a verifiable border, and could therefore be mapped as
an area by following the top of the cliff all the way around, but I
don't see any great benefit to doing all that work to copy the path of
the natural=cliff, when a node at the center of the feature will do.

I believe many people are using natural=peak to add the name of
plateaus / mesas / tablelands. This may be acceptable for buttes,
where the flat top of the hill is small, but for a 1 or 2 kilometer
width plateau there may be several topographical peaks, while the name
may refer to the whole flat topped mountain.

(For most mountains natural=ridge is an verifiable alternative when
the name is not actually associated with a particular peak, but some
tablelands are flat enough that a mapper could not be expected to
identify a ridge or a particular peak)

I planned to document the use of natural=plateau - I will suggest that
multipolgyons be avoided and that the tag not be used to map
non-verifiable geometries, like the Columbia plateau, the Altiplano,
or any of the other huge plateaus listed here:
https://en.wikipedia.org/wiki/Plateau#Large_plateaus - these all are
vaguely bordered by hills, mountains and lowlands, quite different
from a small tableland surrounded by cliffs on all sides.

On 4/17/19, Christoph Hormann <osm at imagico.de> wrote:
> On Wednesday 17 April 2019, Frederik Ramm wrote:
>> [...]
>> The way OSM usually works is someone stumbles over something in
>> reality, with a discernible name or property, and adds it to OSM. We
>> are, first and foremost, surveyors.
>> The larger a feature becomes, the less suitable OSM is for mapping
>> it. And in the case of the several large-scale objects I have
>> mentioned, most contributors don't even have surveying in mind, but
>> just writing down existing conventions.
> Indeed.  We should always keep in mind that OSM is fundamentally about
> collecting local knowledge of the geography.  'local' is key here.  If
> you try to map some geometry for the Altiplano or the Tibet Plateau
> that is not local knowledge.
> As a rule of thumb i'd say something that can at least coarsely be
> surveyed on the ground by a single mapper during a single day is
> usually suitable to be mapped as a distinct named feature, provided it
> is otherwise verifiable of course.  For larger things mapping should
> focus on locally mapping locally surveyable constituent parts or
> aspects of the feature but i would be very careful with creating
> features for them as a whole because this very often drifts from the
> OSM idea of mapping local knowledge to a Wikipedia-like recording of
> social conventions.
> Some of the things Joseph mentioned (like buttes) are certainly mappable
> in OSM under this rule - but i'd suggest creating specific well defined
> tags with a precise and tight definition for them and not a generic tag
> for any elevated region.
> In any case i think the most valuable thing to map of any of such is the
> constituent elements and aspects of it like natural=cliff,
> natural=arete, natural=peak, natural=bare_rock, natural=scree etc.
> --
> Christoph Hormann
> http://www.imagico.de/
> _______________________________________________
> Tagging mailing list
> Tagging at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging

More information about the Tagging mailing list