On Mon, Jul 19, 2010 at 6:29 PM, Richard Mann <richard.mann.westoxf...@googlemail.com> wrote: > I think surface started as a binary paved/unpaved for roads (with > paved assumed by default, and paved meaning tarmac), and has got > extended to cover cobbled roads, and (subsequently) as a way of adding > more info for tracks/paths. > > So for most purposes, the principal distinction is between paved and > not, and that can perfectly well be determined by checking for the > presence of the tag, and whether it's value is "paved".
No, because there is: surface=paved surface=asphalt surface=concrete surface=cement And how about: surface=sidewalk surface=concrete_plates surface=plates surface=tar_mac surface=concrete_slab surface=Betonplatten (?) surface=beton (French for concrete) surface=tar mac And since "paved" is usually shorthand for "can be driven on safely at full speed without getting the car dirty", you might include surface=metal, surface=paving_stones... Not so straightforward really. And that's just my analysis. Presumably every renderer or other tool that cares will do their own. > There's a bit of a grey area for well-maintained unsealed paths/roads, > but the binary paved=yes/no doesn't really help. I tend to use other > clues - that it's got a higher road classification, or is flagged as > being part of a cycle route. It wouldn't hurt if there was a value for > "well-maintained unsealed", perhaps surface=graded? I guess it depends on the purpose. Some hire companies won't allow cars to be taken off sealed roads. Most city drivers would drive slower on unsealed roads than on sealed. I'd say it's a pretty common binary distinction that most people make when thinking about or discussing roads: there are "sealed" roads, and there is everything else. Steve _______________________________________________ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging