osm.wiki/Ohio_River

The previous version of this boundary was based on the USGS topo layer which, while not completely up-to-date with the ’90s court rulings, is much more accurate than the centerline.

It looks like this changeset merged the OH–KY boundary with the Ohio River centerline. That’s incorrect: KY and WV own most of the Ohio River (at the expense of OH, IN, and IL). See for example: http://www.nytimes.com/1981/10/21/us/kentucky-indiana-and-ohio-end-river-boundary-dispute.html
http://articles.chicagotribune.com/1991-05-29/news/9102170953_1_wardens-illinois-residents-kentucky

Never mind, found it: <http://gis3.oit.ohio.gov/LBRS/_downloads/docs/White%20Paper-LBRS_2011.pdf>. We should make a tile layer of it to trace with.

OGRIP’s LBRS is public domain? That’s awesome – where does it say that?

…the bypassed road, that is.

How recent was the change? When Wilmington got its bypass, I think it took a little while for ODOT to take all the signs down and hand the road over to the city and township.

These place nodes shouldn’t have been promoted to place=town. As documented at <osm.wiki/Key:place>, the rule of thumb is to use place=town where the population is 10,000 or more. I’ve also been promoting county seats with only a few thousand residents to place=town since they’re just as significant. But promoting every place name to a town makes the map cluttered and inaccurate – Yankee Town doesn’t even seem to be more than a place=isolated_dwelling.

Thanks for updating this interchange!

Yes, that would be best. Thanks!

Also, the map currently shows Congregation Ohav Shalom as the only building on the Schilling schoolgrounds. However, from the aerial imagery, it seems like the synagogue is one wing of the building and the school is the other, and both share the same property. I think it would be clearer to have the overall property labeled “Congregation Ohav Shalom;Schilling School for Gifted Children” and each wing of the building labeled separately.

Please avoid copying verbatim from CAGIS datasets at this time. They’re free of charge but it isn’t clear whether we can use them in OSM yet. We’re trying to secure permission from them to license their data under terms compatible with OSM’s license. Until then, it’s fine to cite CAGIS as a source to back up information you have otherwise (such as through local knowledge), but please don’t copy. Thanks for understanding.

From humble beginnings…

Alternatively, if a strip mall looks like a bunch of buildings smooshed together, it’s also fine to draw each business as a separate building area, but make sure to join the buildings together.

A single retail building can have multiple tenants: draw the building as one area; then, for each tenant, either add a point or draw an area overlapping part of the building (but not marked as a building).

Oh, wrong link. Here’s the soccer field: <http://tile.openstreetmap.fr/?zoom=19&lat=39.38808&lon=-84.43523>.

No need to name the soccer field “Soccer Field”. Choosing the “Soccer field” option in iD adds a “sport=soccer” tag to the area. Although the main map doesn’t do anything special with “sport=soccer”, various alternative maps do. For example, once it updates to include your edit, the French OSM map at <http://tile.openstreetmap.fr/?zoom=18&lat=39.38774&lon=-84.42123> will superimpose the layout of a soccer field.

By the way, you can often find names of creeks and lakes on the USGS Topographic Maps layer in either Potlatch or iD. (Though years out of date, it’s also really handy and accurate for county/township/city limits.)

Hi Nate, a couple of these `highway=tracks` are named roads, even if they’re unpaved. I’m thinking it might be more useful to tag them as `highway=residential` and add `surface=unpaved` (or gravel or dirt or whatever), since they could appear in addresses.

(That, and I’m happy to finally have a way to comment on changesets!)