I attempted to contact at least some of the authors of "bicycle routes" in the Fort Worth area (and waited the requisite two weeks), alas, to no avail.
So I'll say this here: when tagging for bicycles in OSM, there are two "levels" at which this is appropriate: 1) is infrastructure tagging, 2) is route tagging. In Fort Worth, a great deal of bicycle infrastructure (tagging individual ways with tags like highway=cycleway or cycleway=track, "Class I," cycleway=lane, "Class II," and bicycle=yes "Class III" is extant and maybe complete. (I agree that those "Class" designations are California-flavored, but that's where I'm from and other states use these, too. I don't know if Texas does or not). Much of Fort Worth's tagging of this sort is extensive and appears accurate in an OSM sense, I don't have a problem with it. However, where Fort Worth's bicycle tagging in OSM is problematic is route tagging. Currently, there are two relations in the area tagged route=bicycle, one is network=lcn (https://www.osm.org/relation/7193738) and is unnamed (it might be something intended to represent the "Greater Fort Worth/Tarrant County local bicycle network"), another is network=rcn (https://www.osm.org/relation/7213397), with the name "Trinity Trails." The problem with both is that they are "gigantic agglomerations of hundreds of segments of bicycle infrastructure" (often discontiguous) instead of discretely numbered (or named) individual contiguous routes. I'm pretty sure this isn't what the City of Fort Worth (or the County of Tarrant, I'm not sure who is the bicycle protocol numbering authority who designates these as bicycle routes actually) has in mind with bicycle routing, nor is this how OSM tags these across the world. We tag individual, contiguous routes with these tags, as they are part of a comprehensive network of routes at a particular political-hierarchy level with the network key's values of ncn (USBRs) rcn (state-level bicycle routes) and lcn (local/county/city bicycle routes). We use the cycle_network tag (see https://wiki.osm.org/wiki/Key:cycle_network) to identify the specific numbering / naming authority of that network. Without local knowledge to do so, I have "broken out" a couple of the more obvious individual routes (these two are wholly disconnected from other parts of the network and make obvious choices to do this) into "OSM proper" bicycle routes tagged network=lcn, but without only "good guess" ref=* or name=* tags (like "3rd" on 3rd Street). As there are hundreds of segments of infrastructure in each of these "gigantic glom relations," someone with local knowledge of the individual routes is DEEPLY encouraged to reduce these gigantic glom relations to zero and "trade places" of their (again, properly-tagged) individual elements into smaller, contiguous, sensible local bicycle routes. (The two example "seeds" I offer are https://www.osm.org/relation/8845474 and https://www.osm.org/relation/8845475). Helpful wiki is at https://wiki.osm.org/wiki/United_States/Bicycle_Networks#Local and excellent "better example" bicycle relations from which proper route structure can be learned are found in nearby Plano and Dallas, Texas. Thank you for making OSM (and its OpenCycleMap renderer) one of the most comprehensive (and widely used) bicycle route mapping platforms in the world. SteveA California _______________________________________________ Talk-us mailing list Talk-us@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk-us
