This is a transcript of the conversation that Steve All and I had in private messages, published with his permission.

I believe that this conversation contains some valuable information for those who want to map bicycle routes and mountain bike routes in countries other than USA.

From broadway_lamb 9 March 2021 at 18:33

Hello,

I can see that you are the one who created the GDMBR route relation.

I was wondering whether that route is somehow marked, and if it’s not, what is the policy regarding mapping non-marked mountain bike routes? I can’t find any info on the wiki.

The thing is that I was going to map a similar mountain bike route in Russia, but the local community members told me that only those routes that are somehow marked should be mapped.

Could you please give me some advice?

From stevea 9 March 2021 at 19:06

On 2021-03-09 18:33:14 UTC broadway_lamb wrote:

I can see that you are the one who created the GDMBR route relation.

Thank you for reaching out to me, that is very polite in OSM!

The route known as GDMBR has a long and storied history in OSM. As you can see by its cycle_network tag (US:ACA), it is a route by ACA, the Adventure Cycle Association, a bicycle advocacy organization in the USA which promotes long-distance bicycle touring. ACA has developed a national-scope bicycle network of about two dozen named (not numbered) routes. These route data are private (proprietary / copyrighted by ACA), so their entry into OSM violates our ODbL. However, some cyclists captured GPS data while riding, so for a very few of their routes (3 or so) pieces of these routes (or in the case of GDMBR, the whole of it) are in OSM, and so far it’s “OK” for us to do this, although we shouldn’t make a habit of it, lest we anger ACA.

I was wondering whether that route is somehow marked, and if it’s not, what is the policy regarding mapping non-marked mountain bike routes? I can’t find any info on the wiki.

None of ACA’s routes are marked with signs, because they are private (see our osm.wiki/United_States/Bicycle_Networks wiki), as ACA charges money for the maps with the route and entering such private routes into OSM violates our ODbL (license). But because the very few which have been entered (without signs) came from people who actually rode the whole route, captured these data in their GPS device as a GPX file, those data belong to them (the bicycle rider who rode and is an OSM Contributor) AND the route data belong to ACA as well. You could say that this is a sort of “truce” between ACA and OSM: they agree not to say OSM is violating their copyright and we agree that OSM will minimize doing this, and we have. (We’ve kept this to “about 3” ACA routes entered into OSM).

Any “non-marked” routes that you find in Russia (or anywhere besides the USA) might use the wiki (which is USA-specific) for anywhere in the world as a guide to make “national biking policy for how to enter routes in OSM” for that country. If the route is marked, it is easy: we can enter these, as they are “on the ground verifiable.” If they are public (a government route, whether signed, not signed or “going to be signed in the future when we get funding to sign,” OSM can enter these, too. When to NOT enter is two-fold: when the route is private, do not enter these and when the route is simply “somebody’s suggested ride” (unsigned), do not enter these, either. Please read the wiki’s notice on “what to map” and “what not to map.” This is best said to be true around the world, not simply in the United States.

The thing is that I was going to map a similar mountain bike route in Russia, but the local community members told me that only those routes that are somehow marked should be mapped.

This is good advice. It is 100% true that you should not map a route which is a private route where the data are copyrighted by a publishing company (or bicycle advocacy organization, like ACA, which definitely wants to protect its copyright). It is also 100% true that you should not map a route which is simply “what some people feel like is a good bicycle ride,” as these are ephemeral and transitory in nature — they are not official and they are not signed. There is good guidance in the wiki, and while it is specific to the USA, the advice about “what to map and not to map in OSM” is true for the whole world. It is perfectly OK to map bike routes which are signed as bike routes. These routes often fit into hierarchies of national, regional and local.

I have been instrumental in helping to harmonize how OSM tags national/regional/local bike routes in the USA (it was not easy and it took years). The wiki above is the result, along with “USBRS” (search for it or click the link in the “See Also” section, it is a related wiki). If I can help you do the same, please use those two wikis as a starting point, and look at other countries and how they do it, also see our cycle_network=* wiki.

Could you please give me some advice?

If this does not help you or fully answer your questions, please ask more specific questions and I will do my best to answer you. Again, thank you for tracking me down and asking me, that’s very OSM-polite!

SteveA

From broadway_lamb 9 March 2021 at 19:28

Thank you for such a detailed answer!

My main question is, if a route is not marked, how do we distinguish between “what some people feel like is a good bicycle ride” and a route that is worth mapping like GDMBR?

For example, if a route similar to GDMBR (but in Russia) has a website, a dedicated team of enthusiasts maintaining it, and is not copyrighted, what other conditions should it satisfy in order to be worth mapping?

From stevea 9 March 2021 at 19:54

On 2021-03-09 19:28:23 UTC broadway_lamb wrote:

Thank you for such a detailed answer!

You are welcome!

My main question is, if a route is not marked, how do we distinguish between “what some people feel like is a good bicycle ride” and a route that is worth mapping like GDMBR?

That is a question that rides the edge between what we (in the USA, by no means “the only correct method”) call simply “what some people feel like a good bicycle route” and what we have in the USA which we call “quasi-national routes.” These are routes which ARE signed, they are thousands (or hundreds) of kilometers long, and so they are so significant in the world of bicycle riding that they “rise to the level of a national route” and we call these quasi-national bicycle routes. However, because the USA has only a single national bicycle network (the numbered USBRS, “national routes,” this system/network is still developing over decades and only about 30% complete), and the three or four routes which we consider to be “quasi-national” are so very major and lengthy (compared to local or regional / statewide) routes, we had to make some decisions.

The decisions are that there is only one national network (numbered, official, signed, government-approved): the USBRS and any other routes which are “so large as to be national in scope” (the three or four quasi-national routes), these cannot be numbered nor use the same kinds of signs, they MUST be signed, and we distinguish them further with the cycle_network key (values are US:US for USBRS routes and US for quasi-national routes, as each of them is “independent” of any other route and are not members of a “system of routes”).

We are quite firm that if a route is not signed, AND it is not published by a government, it falls down to the category of “not a route,” and should not get any relation tagged type=route, route=bicycle and network=* tags (where * is either ncn, rcn or lcn — see our wiki about bicycle and bicycle routes to understand how to use these). The idea in being strict like this is that ONLY government routes and SIGNED routes will get lcn, rcn, ncn tags. Otherwise, everything else is a “free for all” and you would have a crazy “whatever we feel like is a good bike ride” appear in OSM as routes and that isn’t correct.

In the USA, we DO have people who want to put these (“a good bike ride”) somewhere on the Internet, there are web sites like www.RideWithGPS.com that allow you to upload and manage “private” routes that are not “official” (government or signed or both) routes. Please read the wiki I have pointed you towards and you will get a flavor for some of the issues involved and how careful you must be for good rules to be established in a country, otherwise, OSM becomes a “free for all” mess of bike routes which are “unofficial” and “we feel like this a good bike ride.”

For example, if a route similar to GDMBR (but in Russia) has a website, a dedicated team of enthusiasts maintaining it, and is not copyrighted, what other conditions should it satisfy in order to be worth mapping?

This comes close to what we call a quasi-national route, and we DO map these in OSM, we even give them a high-level tag of network=ncn, but ALL of our quasi-national routes in the USA — every single one of them — are SIGNED. THAT is what makes them not official by a government, but “official by somebody who takes the care and expense to put up signs.” If your route has a website, that’s a good start — it means there is a community of people who “feel” the route is “real.” If your route were to put up signs, you could absolutely 100% map it, because OSM has the “on the ground verifiable” rule, and signs are verifiable. But you are at an in-between of “not quite there yet” because of no signs — the website is good, but it isn’t quite enough. In the USA, because of the thousands of such bike rides on RideWithGPS website, we could have these migrate into OSM and we would have a mess. Your route, while it sounds important and like it has many people dedicated to it, still lacks what might be called “verifiability,” except for a web site, which hasn’t really been enough all over the OSM world to justify calling it a route in OSM.

The only way that I could think of justifying putting this route in OSM that is similar to GDMBR is that so many people have ridden it, some are OSM Contributors and “own the GPX data from their GPS device,” then you could say “I and hundreds or thousands of us consider this to be such an important bicycle route, that we assign it a route=bicycle relation.” However, you would then have to decide if it is network= ncn, rcn or lcn and you open the door anybody (in your country, even in OSM worldwide) doing this (at the same network level) to any “similar” bicycle route. Think carefully about doing that, especially about “what important facts about any given bicycle route is required before it is important or relevant enough to enter into OSM at a given network level.” What you are doing is effectively making the rules for the determination of THAT route true for ALL similar routes in your country / region. If you are prepared to “manage” that (must have many, many frequent riders, must have a website, should be signed but it’s OK if it is not — this last rule is different than the rest of the world in OSM bicycle routing — then you might consider entering it as a route.

Also, note that route=bicycle and route=mtb (mountain bike, unpaved route) are tagged differently with different rules. (See our wikis on both).

Happy to answer questions and share my experience, SteveA

From broadway_lamb 9 March 2021 at 22:04

Thank you. I have read the wiki and it was very helpful. However, AFAIU, that wiki only talks about https://wiki.openstreetmap.org/wiki/Tag:route=bicycle, not https://wiki.openstreetmap.org/wiki/Tag:route=mtb. Although GDMBR is mentioned there, it is https://wiki.openstreetmap.org/wiki/Tag:route=mtb. Does it matter in this case?

So, to summarize, for a mountain bike route (https://wiki.openstreetmap.org/wiki/Tag:route=mtb) that is:

• mostly unpaved roads
• thousands kilometer long
• non-government
• with no signs

to be considered worth adding to OSM, it should:

• be very well-known
• be ridden by a lot of people
• have a team responsible for keeping it up-to-date
• and have a website.

Does it have to be maintained by some organization (not necessarily a government one), or a group of enthusiasts would suffice? Did I miss something?

GDMBR is not signed, but it is official in the sense that there is an organization responsible for it, and it is a very well-known route, and that is the reason it was added to OSM, am I correct?

Please forgive me if I misunderstood something.

From stevea 9 March 2021 at 23:42

On 2021-03-09 22:04:08 UTC broadway_lamb wrote:

Thank you. I have read the wiki and it was very helpful. However, AFAIU, that wiki only talks about https://wiki.openstreetmap.org/wiki/Tag:route=bicycle, not https://wiki.openstreetmap.org/wiki/Tag:route=mtb. Although GDMBR is mentioned there, it is https://wiki.openstreetmap.org/wiki/Tag:route=mtb. Does it matter in this case?

I am delighted that you make that realization, especially as I prompted you to do this and yes, I believe it IS important in this case, as we are talking about a route=mtb (not a route=bicycle, but when OSM discusses “very large and important” mtb routes, there can be (and has been) some contention and misunderstanding.

Yes, it matters that the route you are talking about is a route=mtb (maybe “major” maybe not). As there are no network=* tags associated with route=mtb, it isn’t an issue whether it is a lcn, rcn or ncn route — though, it can be helpful to think about how mtb routes (especially “major” ones) “fit into” the larger scheme of ALL bicycle routes in a nation / area (“major” ones might be considers “on a national scale,” for example, but they still don’t get a network=* tag, as mtb routes simply do not).

In the case of GDMBR, this was difficult. We had already established (after 2014, when I spoke on this at SOTM-US in Washington, DC: click the 29-minute video link at the bottom of the wiki to watch my presentation on how ACA and I did this together over years) that all ACA routes should not be in OSM at all, as doing so violates OSM’s license. But “a few” had been entered (GDMBR not included, as of yet) and we discovered that these were entered as I described earlier (these were earlier days of OSM): riders of the route who were also OSM Contributors claimed “these GPX data came from my GPS device as I rode the route, so they are mine, even if they also belong to ACA and are copyrighted, therefore I can enter them into OSM.” It was very tricky to say “no” to that, so we didn’t, allowing these “few routes” to remain, but with the wiki admonishing not to do any more of this in the future, as ACA data really are private, they really are copyrighted, they really are an important revenue source to ACA and their entry into OSM literally starves them. In English, we say these data were “grandfathered in” (like when a law is passed, but certain things must remain the same because they were already that way before the law was passed).

So, when GDMBR was first entered, it was entered as an international route (route=bicycle, network=icn). But this is wrong, as GDMBR is ~70% unpaved, clearly a route=mtb, not a route=bicycle, and with no network=* tag. The author did not like me making these changes, as “the renderer he preferred” did not display the route the same way, I simply said “your tagging was wrong, I made it correct and anyway you would be guilty of tagging for the renderer, a no-no in OSM.” This got a little “heated” (contentious), but eventually, “correctness” won the day.

What you are asking is whether a “major” (I think so, you likely think so, too) route=mtb “should” be entered into OSM, even though it is not official (government-published) and has no signs. This would make me and most other OSM bicycle / mtb route - savvy people say “no.” However, you have an enthusiastic group who publish a website about the route, and that is certainly something. If it is a government route, it should go into OSM, even if signage is very light or perhaps even non-existent. If it is a “private” route, as this appears to be, such routes CAN be signed in OSM, but only if they are signed-on-the-ground. Your website is good, as I say, but having signs-on-the-ground would truly make it “real.” Even one sign at or near one terminus and a second sign at or neat the other end of the route. I would consider that “minimal,” but sufficient to enter into OSM. This shouldn’t be terribly expensive, and if you have money for a website and volunteer energy enthusiastic about this mtb route, you can likely design, manufacture and erect a couple of signs, with plans for maybe a dozen or score more along the route at strategic places. Look at this as a way to grow wider recognition about the route. The fact that you have to likely get permission from somebody (the government? I don’t know how it works there) to erect a sign along a dirt bike track gives your mtb route even more legitimacy, in addition to “the nod of assent” from OSM.

So, to summarize, for a mountain bike route (https://wiki.openstreetmap.org/wiki/Tag:route=mtb) that is:

• mostly unpaved roads
• thousands kilometer long
• non-government
• with no signs

Yes, “mostly” is a good word, let’s simply be more vague and say “major in scope” rather than “1000s of km long” and now you are talking about a mtb that could possibly go into OSM, even without a sign. But if either is true that 1) it IS government (not non-government) and 2) that it DOES have signs, well, now you are 100% OK to enter it into OSM. That’s the sort of guidance I’d use, and I’m not nobody on this subject, but I certainly do keep my listening ears open to see what others say. I actually got a talking to by two members of the DWG at that 2014 SOTM-US conference that I shouldn’t be doing this, as the routes we were entering were “proposed,” but it was a friendly lunch, I got a letter to the OSM-US Secretary from AASHTO (our national highways coordinating body in Washington, who numbers the routes in the USBRS) and it smoothed things over. But you can see how some people are very sensitive to “things which really should NOT be entered into OSM really SHOULD NOT be entered into OSM!” It seems you are feeling that pretty strong right now.

to be considered worth adding to OSM, it should:

• be very well-known
• be ridden by a lot of people
• have a team responsible for keeping it up-to-date
• and have a website.

Those are all an excellent start, but might not be completely sufficient in some people’s opinions. Either one of “government-approved” (in whatever manner they might offer their imprimatur of acceptance, again, I don’t know how such things work — in the USA it took over 35 years for the USBRS to begin to gain real traction and began to substantially grow in robust fashion) OR if there are even a whisper of signs (one at one end, a second at the other…) might be “just enough.”

Does it have to be maintained by some organization (not necessarily a government one), or a group of enthusiasts would suffice? Did I miss something?

As ACA, who “maintain” (publish, they don’t actually “clean the trail…”) GDMBR is a private organization (at a national scope), it doesn’t HAVE TO be governmental. For example, two of our quasi-national routes (different than ACA routes) are by other private organizations, and while these are route=bicycle, network=ncn (not route=mtb, no network=*), these are absolutely not governmental organizations. However, these are very well-signed routes. That’s a big deal in OSM, as it is “on the ground” verifiable.

GDMBR is not signed, but it is official in the sense that there is an organization responsible for it, and it is a very well-known route, and that is the reason it was added to OSM, am I correct?

The story is complicated because OSM was simultaneously trying to not offend ACA, which “owns copyright” of the route data and OSM was violating its own license (ODbL) in some sense by allowing entry of GDMBR route data. (Remember, ACA makes its “bread and butter” money to pay the bills by charging money for these copyrighted data by those who wish to use ACA’s excellent bike maps to ride their routes). However, at the insistence of the “riders” of the route, who insisted they were using / importing “their” data from their GPX tracks from their GPS device, OSM reached a fragile consensus that allowed GDMBR data to be entered for that strict reason.

Please forgive me if I misunderstood something.

You are doing fantastic and seem to be absorbing everything wonderfully! Thank you also for your excellent English, it seems flawless.

Again, I am happy to answer any additional questions if I have left something unsaid or unanswered; I am trying to be complete, but for a complex topic such as GDMBR and how something like it might happen in another country, it is a tedious topic and maybe there is some back-and-forth to get all the questions asked and answered. More if we need to!

SteveA

From broadway_lamb 9 March 2021 at 23:56

Thank you for your excellent answer and for the time you’ve taken to explain everything to me. Things are now very clear for me!

Do you mind if I share our messages in my OSM diary? I believe they are a very valuable source of information not only for the Russian OSM community, but for other OSM communities around the world.

From stevea 10 March 2021 at 00:01

On 2021-03-09 23:56:19 UTC broadway_lamb wrote:

Thank you for your excellent answer and for the time you’ve taken to explain everything to me. Things are now very clear for me!

Do you mind if I share our messages in my OSM diary? I believe they are a very valuable source of information not only for the Russian OSM community, but for other OSM communities around the world.

I would be honored if you were to share these with the wider OSM community. I found two or three misspellings in them — and not a single misspelling in your English to me! Though, likely, you will translate them to Russian or another language and “clean them up.” If you use quotation marks and quote me (use my name), I do ask for a faithful translation; thank you.

Finally, it would be nice to have a link to wherever this translation ends up: these topics, especially at a national level as I have experienced them, quite often do “make the news” and people find them both informative, “news-worthy” and downright interesting!

A wonderful outcome might be a national-level wiki similar to the one we have in the USA, but for your part of the world. I know for a fact that it is responsible for very crisp and accurate bicycle route tagging in the USA, and of course, we want that all over the world in OSM: wiki are an excellent method to do this.

It really was enjoyable to have this discussion with you.

Best regards, Steve All, Santa Cruz, California, USA