[Imports] Vermont Town boundaries from VCGI

Frederik Ramm frederik at remote.org
Sun May 4 09:25:16 UTC 2014


On 04.05.2014 07:08, Andrew Guertin wrote:
> As far as I can tell that addresses all the comments from previously.
> Any more?

I have not followed the discussion in detail so I'm sorry if I repeat
something already said, but be aware that place=* tags tend to be on
nodes not multipolygons, with the node placed roughly at the population
center, the central square of a city or whatever.

With place=* on multipolygons, renderers will try their best and place a
label at the center of the polygon, if they label it at all, which will
end up with a town label for "Duxbury" in the middle of the woods.

When I see "boundary=administrative, admin_level 8" then I know this is
just a boundary and doesn't say much about whether people live there at
all, and how many, and where. But when I see "place=town" then I expect
a proper town with (usually) a pub and a church.

If I were doing this import, I would probably take the time and separate
the place=* tags into nodes of their own, manually placing them at
something that looks like a populated place, and remove place=* from the
multipolyongs; or if I didn't have the time for that sort of work, I'd
not add any place=* at all.

But I'm not very familiar with the place=* usage in the US and if this
is the done thing over there, feel free to ignore me.


Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"

More information about the Imports mailing list