On 2020-08-13 14:07, dktue wrote:
> I think that it's quite hard for data consumers (again: think of an
> overpass-query to find all mid-stations) to determine which role a station
> has. Like Martin said: Why not just solve the (huge!) special case of
> mountain aerialways where we really have one bottom_station, zero or more
> mid_station and one upper_station?
Because cases that don't obviously fit that model will remain untagged,
or get subjectively/wrongly tagged with this model or some other tags.
There are many cases of arialways that start and finish at roughly the
same level. Do they have two bottom stations, or two top stations, or do
you force someone to make a judgement call as to which is slightly
higher than the other?
If you want to find all mid-stations, do you want to find ALL
mid-stations or are you happy with MOST or MANY or SOME?
Modelling reality is all about what to leave out. If top-stations,
mid-stations and bottom-stations can be reliably derived from elevation
and topology then there is no need to add a more explicit tag.
_______________________________________________
Tagging mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/tagging
