Am Fr., 7. Feb. 2020 um 11:26 Uhr schrieb Christoph Hormann <o...@imagico.de
>:

> I currently tend towards a broader solution of dropping rendering of all
> barrier tags on polygons.



great, this would make it very clear that there is indeed some problem with
the tagging. Although I guess carto would get a lot of flak for a decision
like this.


This would
> open a path for the various solutions already discussed - like
> introducing a new tagging scheme for indicating a polygon to
> be 'enclosed by a barrier'



there is for example fenced=yes, currently used 4841 times (even applicable
to nodes)



> or by strictly adhering to 'one feature, one
> OSM element' without implicit tagging of barriers.



adhering to 'one feature, one OSM element' and supporting fenced=yes on
polygons isn't a contradiction. Implicit tagging is ok, it's explicit
tagging of more than one thing on the same element, which creates the
problem. "fenced=yes" doesn't say this is a fence, it says this has a fence
around it. The problem with this approach is that it slows down the adding
of details, because if you want to add the height of the fence you would
either have to make it its own object at this point, or resort to new tags
like fence:height=x
_______________________________________________
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Reply via email to