I think that's not an error per de, but a 'add as many tag as possible to be
sure to do thevright thing'.
I think data consumers shouldn't encourage this though.
Osm2pgsql users probably don't make a difference between a way and a relation
if is has type=route. It could matter if other tools have to follow to properly
exploit the data, but I don't know.
Yves
Le 17 mai 2018 18:36:29 GMT+02:00, [email protected] a écrit :
>Well, if this particular way of tagging things is generally supported
>by data consumers, then I would consider it a documentation error
>instead of a tagging error.
>
>
>
>But if it’s not used by data consumers in this form (while the same
>information on a relation is), and clearly the people creating these
>tags intended them to be usable, then I would consider a tagging error.
>
>
>
>So, yeah, I would say that “tagging error” IS in a part defined by data
>consumers.
>
>
>
>The “correct” way of tagging something is in the end defined by a
>somewhat nebulous consensus between the way people tag and the way data
>consumers interpret that data.
>
>
>
>From: Mateusz Konieczny <[email protected]>
>Sent: Friday, 18 May 2018 01:51
>To: Tag discussion, strategy and related tools
><[email protected]>
>Subject: Re: [Tagging] type=route tagged on a way?
>
>
>
>17. May 2018 15:11 by [email protected]
><mailto:[email protected]> :
>
>So my question is, is this type of tagging actually supported by data
>consumers, or are these 9000 tagging errors?
>
>
>
>"tagging error" is not defined by whatever something is used by data
>consumers or not
>
>
>
>In Poland I see some route=emergency_access that seems to be OK (though
>there may be
>
>a better way to tag this and type=route in this case is pointless).
Yves
_______________________________________________
Tagging mailing list
[email protected]
https://lists.openstreetmap.org/listinfo/tagging
