So, a collection of 12 ncn routes from one national issuer. I imagine that the practical maintenance is done by lower operators?
On a bicycle you can cover a lot of ground in a day, so in Nederland, a local bicycle roundtrip is around one of the larger cities; elsewhere it will become regional vey easily, not because the routes are that long, but because the regions are that small!
Short local routes have almost totally been replaced by the node network system. Regional bicycle routes are also uncommon, this is due to the fact that rcn was hijacked to mean “this is a node network route”. So every regional-ish route was tagged as local or national. Not a problem in our small country, but Germany seems to rely heavily on regional routes.
Like Minh, I’d like to say I am eating popcorn at this excellent dialog. It does remain spicy. I’m listening.
I see a wide plurality of “ways we express our (cycling, cycle routes, cycle routing networks… data in our map” (database). These have been vividly shown with the “visual interpretation-ness” that a map displays (thanks for that tasty sampling, @Peter_Elderson ). We really are at where our very brains (minds, thinking about things…) and ourselves as “each other, other people’s views…)” blend and merge. How we think about things, how we are as people together, how we tag a relational map database.
In Minneapolis (Minnesota, USA) there is a bike route https://www.openstreetmap.org/relation/1749619 nationally-designated (as being so outstanding of an urban cycleway) yet is locally-administrated. It is tagged network=ncn and cycle_net=US:MN:MPRB simultaneously.
I realize things are, um, “less clear” in other places.
That’s an interesting one too, because the overall route belongs to the National Scenic Byway system, a nationwide network, but it consists of seven “districts” with a harmonized set of logos that together form a local network.
For example, one of these local routes is the Chain of Lakes Byway District, tagged network=lcn. We’re currently modeling it as a two-route concurrency, though arguably the National Scenic Byway could be modeled as a route superrelation.
I took a second look at the Grand Rounds Scenic Byway, and I’m no longer sure that it should be tagged as network=ncn. National Scenic Byway is not a coherent national network but rather a nationwide brand or accreditation, just as the National Register of Historic Places is to buildings. Like many NSBs, the GRSB has a distinct on-the-ground identity, so it’s another example of a one-off, network-less route. (For reference, most route=road relations for routes in NSB or America’s Byways don’t mention either program in any tag.)
It’s as if the GRSB is itself a network with its own shield, or perhaps a route superrelation containing the seven “district” routes (not counting distinctions between road, bike, and hiking routes). It was originally mistagged as a U.S. Bicycle Route (cycle_network=US:US), but several years ago it was changed to the current cycle_network=US:MN:MPRB tags, which refers to the Minneapolis Park & Recreation Board but doesn’t specifically identify the GRSB.
Maybe it could be retagged cycle_network=US:MN:MPRB:GRSB, based on how one-off routes are normally tagged. Regardless, it falls into the same quandary of network=lcn versus network=rcn, because a network that spans a park district could be described as either local or regional depending on one’s perspective.
There are also one-off interstate bike routes, such as the 9-11 Trail and Mississippi River Trail. Unfortunately, these routes are tagged cycle_network=USA and cycle_network=US, respectively. These values don’t correspond to any coherent network in reality; they essentially mean not:cycle_network=US:US. Probably cycle_network=US:911 and cycle_network=US:MRT would be more apt. Fortunately, there’s no question about network=ncn.
Once the U.S. community starts experimenting with trail shield rendering in earnest, we’ll likely wage a campaign to rationalize cycle_network=* values for one-off routes, just as we did for one-off road routes a couple years ago.
Looking at EuroVelo, it seems they are created by a private organization: European Cyclists' Federation - Wikipedia
Why would we give the authority to define international routes exclusively to a single organization? This decision seems shortsighted.
Generally, I agree that it can be unclear whether one should e.g. use “iwn” for a hiking route that is regional by size but crosses one or more national borders, or similarly we had discussions about local routes (by extent) that were mapped as regional routes because they crossed regional admin borders.
I don’t agree, with the same reasoning we could stop mapping anything right now and defer the decisions about what to map where and how to external organizations.
I don’t know about authority, but if they create signposted international routes, it classifies as icn. If no other organization or authority creates or operates international cycling routes it’s fine for a community to say that it’s exclusive. I guess that ws (or is) the case. If other signposted European routes appear, they can be mapped as icn too. The requirement of signposting sort of guarantees that they are mappable routes; a hard restriction to one “authority”, even if it’s the only one at the time, would be unwise, I think. You don’t tie an arm behind your back just because you need only one right now.
About crossing borders: that should not be seen as a hard defining criterion. Local and regional routes in e.g. a shared nature reserve easily cross the border, but they are still lcn or rcn in scope. We have a national hiking route (symbol white-over-red) along the most of the DE/NL border, the “Noaber pad” (Neighbour trail). It pops in and out of Germany, but it’s a Dutch operated and maintained ncn trail nonetheless. Thnx, neighbours, for allowing us to put our stickers on your posts, bridges, trees and street furniture!
Or, we can decide what we generally view as an international route, and decide for each particular route if we view it as international. The real-world definition (if it exists) is not changed by the OSM-view.
icn is just a classification of networks, plural, not a single exclusive network. North America has its own international network (an informal coordination among multiple entities), but this doesn’t preempt any network in Europe or, for that matter, any other overlapping international network that might arise in North America. If there is a one-to-one relationship between an international geography and an international route network, it’s purely by coincidence. If anyone wants more clarity than that, they need something beyond network=**n.
Anyways, the public or private nature of a designating entity matters less than the ephemerality of its routes. The USBRS is as official as it gets for an ncn over here, but its designating authority is an NGO. Many long-distance hiking routes and networks rely on blazes and signs posted by volunteer organizations rather than governments, but the existence of these signs also says governments tolerate or even endorse the routes. It’d be more effective to exclude routes that exist only on paper (or only on the Web).
This is what the proposed importance=international was supposed to be about. This proposal was split out from the network=* proposal back in 2009 based on a realization that some mappers were putting an individual feature’s geographical scope in network=* instead of anything to do with a network of features. Here we are some 15 years later.
This really has been going on for 15+ years, I can say personally (and our data trail backs me up). The topic and how we’ve got a particular history and even “bent” towards “ways of doing and tagging things” in different parts of the world in a way that has its frictions elsewhere. This has its difficulties. They seem or are surmountable to me. Complicated, OK, but tease-out-able like unsnarling hair.
Yes, it’s a matter of balance. We need to make decisions where the real world is ambiguous, but we cannot decide to map the world how we’d like it to be… And this implies being very attentive to how organizations such as ECF work.
There is something in my computer science-oriented mind (sometimes) about the difference between “starting at 0 or 1?” Where OSM defines “international” seems like a good place to say “consensus begins here.”
I’ll (painting a chalk line) say “smartly, crisply define ‘international’ for cycle routes.” And that is a starting line. Maybe things slow or even pause as routes enter one-at-a-time (or as a group, say, an initial international network of routes).
Things might need to be “teased out one-at-a-time” (for a while) and then get “rolling along” with noticeable patterns. Agglutination, merging of tags…it is a future we might foresee.
I don’t know, but that (above, how I mention giving this a push) is a start ahead. I continue to listen.
Edit: Postscript; for me in the USA (North America), network=icn means “crosses an international border.” That’s a starting place that works for me, and because of the definition of the word ‘international,’ likely works in a LOT of circumstances.
I believe I once met a text that said “continental” rather than “international” and I must admit that it appealed to me even though I daily tag regional cross-border routes as icn or iwn.
Waymarked trails uses “Continental” as title for the infopanel section of i*n routes .
Appealing, yes. OTOH, there are long routes that span a few (european) countries, but not the continent.
our national hiking federation has a class system that is more or less close to iwn/nwn/rwn/lwn, but with a different approach to how classes are defined.
The key to their class system is interest. A route is of national interest, local interest, etc. It captures both who are the stakeholders (operators, governing bodies) and how “marketing” of the routes is organized.
Continuing great stuff to read here! (For example, “stakeholders AND marketing”).
I don’t want to put words in the mouth of Sarah (Lonvia, author of waymarkedtrails.org) as I don’t know what prompted her to choose “Continental” instead of “International.” I also find appeal in the way that her choice broadens the scope to include “routes which span international borders” along with “routes which span the geography of a continent.” However, it is exactly these sorts of “stretch-to-fit semantics” that can help us move our tagging and (wiki-) documenting of how we do things in a forward (more agreeable around the world) direction.
The OSM classes don’t prescribe exactly what aspects the communities should weigh in, or weigh most, when assigning **n to routes. “Interest” is fine; if the mappers to run with this, no problem. Problems arise when trying to unify the world under one idea, because no single definition fits all countries around the world.
Same as with the highway classes. A road can be a trunk in one country, while the same physical road in another country would classify as a minor unclassified road. No one set of exact criteria covers that.
So, if I understand @Peter_Elderson correctly, OSM is “doomed to the status quo” of how differing approaches (to *cn or even **n routing) have evolved in countries / regions / continents, and “what we have is what there will be?” In short, we’re stuck with what we have now?
It is easier to “give up” and accept the status quo but it seems to make 100% moot (subject to uncertainty, having no practical relevance because of that uncertainty) the topic of *cn routes. I’d say “aww, too bad” or “darn, another attempt at harmonization foiled again.” (UK English-ers, I know you spell this harmonisation).
Maybe this entire endeavor is simply too (socially, technically…) difficult.
Regardless, it falls into the same quandary of 