Thank you Bill Ricker for the deep, thoughtful and researched background and weigh-in on Connecticut and Rhode Island county status. I'm now leaning in the direction of Greg Troxel that Rhode Island may indeed have counties which are administrative, though I withhold my final judgement (and it is solely mine, which shouldn't influence all of OSM that much) for the present time. Again, it is only Rhode Island which does not have counties tagged admin_level=6, the other 49 states do. (Mass. didn't for a week or so in May during a dispute, that has been remedied).
Thank you Greg Troxel for thoughtful weigh-in, too. I agree it is important to recognize that "the federal government recognizes counties" (for "everything") and uses FIPS codes "for everything," however, I'll also point out that OSM has, for some time, made distinctions between what the federal government does (in the guise of the Census Bureau, for example, and so mentioned on the United States admin_level wiki) and what OSM does. Most of the time (CCCs, Independent Cities...) what OSM does w.r.t. admin_level and what the Census Bureau does more-or-less harmonize. Some of the time (the further division of Alaska's Unorganized Borough into counties, what the Census Bureau does, versus census tracts, as OSM does, how the District of Columbia is not called a "county equivalent" but instead gets admin_level=4 while the contiguous city of Washington gets admin_level=8...) OSM and the federal government do not harmonize. Indeed, as has already been mentioned, the federal government even disagrees with itself: one bureau says USVI has two subdivisions (as county equivalents), another federal department (USGS) says USVI has three (as distinct islands), so "take your pick" (two or three?) Honestly, I'm not trying to stir up muck or make trouble, rather I have striven mightily to get as much understanding and agreement as possible where, when and how it can be. While Mass. and Conn. seem to (now) be "solved issues" (DO include admin_level=6 on county boundaries, much to do with "local support"), Rhode Island continues to remain under discussion (though it seems like it is tipping towards "let's make it 50 out of 50"). As I read Greg's "let's not make our map awkward for our downstream users" (and indeed I have heard this before, for example as part of Mashin's argument for why COGs should be tagged with admin_level, still disagreeable and not done in OSM), I must say how that seems to fly in the face of (disagree with) the notion that we "map accurately what is." As I have said before, I honestly don't think we want to shoehorn something that isn't something into a box that says it is. Especially when we have plastic tagging (we do; boundary=COG is a good example). If this means a bit of extra logic, code or providing for a data exception, reducing the "clean, easy" path of code and data parsing, well, that might seem unfortunate, but it does mean that our important tenet of wanting our map to be "truly data accurate" (or as much so as possible) is closer to reality. I certainly see both sides of this, but there are two sides and I'm not sure which is more important: convenience or accuracy. I lean towards accuracy, that is simply me (and my nature). Others are welcome to disagree, which means some discussion must continue. Honestly, I think the discussion is productive, provided we remain civil, and we're doing a good job so far, imo. SteveA _______________________________________________ Talk-us mailing list Talk-us@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-us
