[Tagging] tagging capitals

M∡rtin Koppenhoefer dieterdreist at gmail.com
Fri Jan 21 11:58:39 GMT 2011


2011/1/21 Simone Saviolo <simone.saviolo at gmail.com>:
> 2011/1/21 M∡rtin Koppenhoefer <dieterdreist at gmail.com>
> I see a potential problem. Let's say there's a large admin level (say, the
> country) whose capital is *not* the capital of a lesser admin level it is
> in. I'm not sure this actually happens anywhere, as usually, for example,
> Rome is the capital of Italy, Regione Lazio and Provincia di Roma (and of
> course the capital of the Comune di Roma).
> Let's say, though, that city A and city B are both in the province P
> (admin_level=6), part of region R (admin_level=4), and that A is the
> regional capital for R, but the capital of province P is instead B. Using
> capital=4 on A could be ambiguous: does this mean that A is a capital only
> for the admin_level=4 it lies in, or that it is the capital for any admin
> level less than or equal to 4?


I'd say that this problem seems really "potential" and I don't expect
this to happen at all, but who knows. You could check this by looking
if there is another town/city with is tagged capital=6 within the same
admin boundary. If there is not, you would assume that A is also the
capital for the level 6. The same logics could be applied for other
levels. I agree that this is not completely fail safe, as our data in
the area might not be complete hence leading to wrong assumptions (not
to wrong data!), but I would prefer to take this risk in exchange for
easier data evaluation with single values in "capital".

Cheers,
Martin



More information about the Tagging mailing list