[Imports] Washington State county boundaries

[Toby Murray]

I'm working on license cleaning state/county borders in the U.S. Most
of them have been pretty easy but Washington state has a lot more
license taint so I'm looking at re-importing some of the ways.
Fortunately the relations are clean so I'm only looking replacing
(some of) the geometry. The best source I found was the Washington
State Department of Ecology. Their copyright page seemed clear at
first but then pnorman on IRC muddied the waters and suggested I get
some input from a wider audience. The page in question is here:
http://www.ecy.wa.gov/copyright.html

Seems like a basic attribution requirement. But of course that gets
complicated with derived works and whatnot. I guess I'll send someone
at the department a quick email too and see what they say.

As for the rest of the import, the data can be viewed here. It is the
"Counties" data set and the "County_arc.shp" file inside of that.
http://www.ecy.wa.gov/services/gis/data/data.htm#c

my initial ogr2osm translation file can be seen here:
https://github.com/ToeBee/ogr2osm-translations/blob/master/wa_counties.py

Since I'm only using the geometry, it is pretty much braindead. I will
have to manually add the generated ways into the appropriate boundary
relations.

I simplified the ways in JOSM with a max error setting of 5 meters.
This took the total node count from just under 50,000 to 18,500
although I won't be using all of them since some of the ways are
already relicensable. The geometry matches up very well with what is
already in OSM. Where it diverges it seems that the new data is better
(smoother curves, follows visible features more closely)

Paul suggested removing the source:ref tag from the ways and putting
it in the changeset instead.

Thoughts?

Toby