Andrew, I loaded your data into JOSM and its really something! Good work. I think if you poke around, you'll see that many of the place=* places have already been imported by the GNIS import. Most of them will be (perhaps incorrectly) place=hamlet, unless another editor upgraded them.
For instance, the town of Northfield (where I briefly lived) has a node with the proper attributes: https://www.openstreetmap.org/node/158851899 Now, I just noticed in your data the names include the census place type, like Northfield *Village, *Barton *Village, *etc. I believe those are place types and not the actual administrative name. in the census data, they spell the types with lowercase to indicate this (I believe). Head to http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml and type in "Northfield vi" and you'll see what I mean. Perhaps you want to look into that. Kindly, Elliott On Mon, May 5, 2014 at 1:53 PM, Andrew Guertin <[email protected]>wrote: > I've removed place=* from all elements. > http://www.uvm.edu/~aguertin/osm/vermont-towns-counties-villages4.osm > > > Having seen everyone's objections and thought about it more, I still think > that within the context of a Vermont-only map, it would make most sense to > keep things labelled with place, but to better match how they're used > elsewhere I'll leave them out. > > > --Andrew > > _______________________________________________ > Imports mailing list > [email protected] > https://lists.openstreetmap.org/listinfo/imports > -- Elliott Plack http://about.me/elliottp
_______________________________________________ Imports mailing list [email protected] https://lists.openstreetmap.org/listinfo/imports
