[Tagging] Power transmission refinement proposal

Martin Koppenhoefer dieterdreist at gmail.com
Mon Feb 11 17:44:23 UTC 2013

2013/2/11 François Lacombe <francois.lacombe at telecom-bretagne.eu>:
> Hi folks,
> Last night, FK270673 grants me ownership of his power transmission
> refinement proposal
> http://wiki.openstreetmap.org/wiki/Proposed_features/Power_transmission_refinement
> As the Power generation refinement one, it would be great to get it accepted
> in few month.
> More time may be needed to improve and complete it, instead of the Power
> generation proposal.

reading this I am happy that you chose a way of tagging that is
somehow compatible with what we already have (i.e. you don't change
meaning of in-use tags), with the exception of power=generator (and
subtags). Unfortunately power=generator is in heavy use (124233
times). Looking at taginfo it seems as if the proposed generator:type
is already used (not very much):

Personally I think we could live with the redefinition of
power=generator as long as there is a broad consensus.

As far as I see there remains some ambiguity from the proposed
changes: should generator:source (the by far most used subtag) also be
deprecated? Or would it not be logical anymore to tag the power source
to a generator (e.g. this could also be "vapor"? Maybe this should
become power:source=foo tagged on the proposed power=plant?

You suggest to tag all kind of transformers always as nodes which
doesn't seem reasonable being them sometimes quite big (the one in the
example picture seems to be at least 10-20 sqm). There is really no
need to reduce the use case to nodes, as there isn't a need to require
an area for a substation.

According to the proposal we would then need to know the exact voltage
in order to distinguish between the various sizes of substations, i.e.
a non-expert (one that doesn't know the voltage) mapper would tag
these two the same:
(and you would have to be a real expert in power distribution, you
would even have to look up the locally used voltage when mapping in a
different country, etc.)
Also when evaluating the data you would have to know the used tensions
in order to make a world map of power distribution. IMHO there could
be a way to make a rough distinction (e.g. by using subclasses which
describe the role in the distribution network like local distribution,
regional distribution).


More information about the Tagging mailing list