On Fri, 2011-12-02 at 12:29 +0100, Martin Koppenhoefer wrote: > Maybe I was exaggerating. Generally I don't see a big difference > between routes and ways, and tags that change the overall meaning of > other tags significantly should be avoided in favour of more > fail-proof solutions. If a tag states that there is a route on the > ground (route=bicycle) then there shouldn't be another tag > state=proposed (or say abandoned=yes) that if set says: no there isn't > actually, there might be a route in a while (maybe) or there has been > a route in the past which isn't there any more. > > On the other hand I can follow your arguments: maybe this is not the > best way to denote this, but it is already in widespread use, current > applications are already aware of it, and routes are special interest > tags so most applications won't evaluate them anyway.
I wouldn't say that routes are special interest so much as routes just don't follow a single way, and a single way often has multiple routes. I realize this doesn't tend to be the case so much in Europe, where ways are generally limited to having one route on it, but North America does take this to the far opposite extreme (For example, I can think of one 5-mile loop of road that is collectively a member of three national routes, two interstate routes, a state route and a city route in Tulsa, no leg of which is a member of fewer than 2 routes).
signature.asc
Description: This is a digitally signed message part
_______________________________________________ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
