[Tagging] Simplify building:part areas
cmue81 at gmx.de
Sat Aug 19 18:36:50 UTC 2017
> Sent: Fri, 18 Aug 2017 20:06:29 +0100
> From: "Javier Sánchez Portero" <javiersanp at gmail.com>
> To: "Tag discussion, strategy and related tools" <tagging at openstreetmap.org>
> Subject: Re: [Tagging] Simplify building:part areas
> Josm validation will raises a warning for duplicated ways (way1 and way2).
> If I use the open ways + MP relations schema mentioned by Christian,
> the situation is almost the same. I will end up with three MP relations
> instead of closed ways and Josm validation will raises a warning for
> relations with the same members.
Well, it's a warning, not an error. Its purpose is to raise awareness,
i.e. you should double check what you did and check if the matter of
those warnings is what you intended to do.
You should also be aware that while josm devs try hard to stick to what
can be made out as a group consensus, its not always congruent with the
wiki description or vastly employed tagging methods. Most of the time
the validator tries to reaffirm what may be deducted from taginfo.
Naturally, the usefulness of the validator decreases in "areas" where
a good data scheme is still in flux or worked on or discussed by the
After all, you might not have raised questions about building:part
tags, if this was a solid and well documented (read beginner-proof)
part of OSMs map features . Doing 3d stuff right is still subject
of discussion and well, mailing lists.
If you want to work with a current, least common denominator, you
should imho check the result rendering of new data entered with
- main map
- kendzi3d plugin
and maybe other data consumers listed in . But again, if you
feel that all of these data consumers mistreat a certain aspect
of modeled 3d data that you think is a correct or good way to do
it, then please try to raise this issue here, in the forums or
relevant bug trackers. Do not try to break a good modeling ap-
proach in favor of broken or incompatible renderers.
More information about the Tagging