On 11/12/2022 13.10, Sven Geggus wrote:
> This seems to be somewhat similar to the recent discussion about using
> site-relations for camp-sites.
> > Some people think that what I do in OpenCampingMap currently (using
site
> relations) is an abuse of the "One feature, one OSM element" principle.
That's an interesting observation, thanks. I agree that those proposals
have some similarities :-)
> Thus inventing another relation type for such loose bindings of distinct
> geographical features would make more sense than using a site relation.>
> Let us call this relation, type=feature for now.
>
> A native english speaker will probably come up with a better name.
Additionally, "feature" in the context of GIS has a specific meaning for
many people, but however it might be called, I see your idea :-)
> The tagging would then be (In my case):
> type=feature
> feature=tourism:camp_site
>
> and in your case:
> type=feature
> feature=piste:type:sled
>
> The **feature** tag should then make clear that **only one element**
tagged
> tourism=camp_site or piste:type=sled is valid in such a relation.
>
> This would then mark where all the distinct objects belong to.
>
> Would this be a viable path?
This would be a question to more OSM experienced people than me but to
better understand your idea:
"type=feature" would mean that "this is a collection of related things
and metadata where their relation cannot be deduced otherwise". And
depending on the value of the additional "feature key" (like
"feature=tourism:camp_site" or "feature=piste:type:sled") additionally
allowed elements and their meaning would be defined (especially the
"role" of relation members)?
Best regards,
Philipp
_______________________________________________
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging
