There was a long discussion in the telegram group about which routes should be in OSM and which should not be.
IIRC the decision was that if the route has an operator that is responsible for this route and would make sure this route is maintained then it should be in OSM - assuming it has fact on the ground (makings).
If the above is not the case it shouldn’t be in OSM.
The problem at that time is that there was no good alternative to OSM.
There was a request from the community to facilitate this need to share routes.
Some activities, such as hiking can use websites like Nakeb to suggest routes to the website’s authors, but this wasn’t a solution that could work for the entire community.
At mapeak.com, we have developed a way to manage these kinds of collections and allow specific OSM users to publish routes to the general public.
This is to allow removing those routes from OSM.
This capability is currently free of charge in mapeak.com and is aiming at bridging this gap. mapeak.com, the global successor of IHM has all the capabilities needed to solve this including route planning based on OSM ways, difficulty setting, type of activity etc.
It would be great to see if this fits the community requirements and if not, what is still missing.
Ways are usually backed by facts on the ground, so we wouldn’t want to remove those. In some rare cases ways are representing a single route, but usually you need a relation to “collect” a few ways together to form a route. So this is mostly around relations. In cases of ways that represent an “unofficial” route, we’d want to remove some tags from it.
I am having a hard time with a blanket deletion rule. Especially for cycling relations as mentioned above. These serve well the local bicycling community. Hopefully also tourists one day.
I am very happy with the new capability in Mapeak. Also happy to contribute to it for example here: Mapeak
Note that deleting because Mapeak has a solution is “mapping for the renderer”. A big OSM no no.
To my recollection there was no clear OSM guideline about deleting such relations anyway.
Let’s delete relations on a case by case examination. If the route is no longer useful and doesn’t have a managing father, sure.
I wasn’t suggesting a bulk removal. I suggested an alternative place for things that don’t have any facts on the ground, which from OSM perspective should not be in OSM, as far as I understand the guidelines.
Here’s a relation for example that I think should not be in OSM:
This is just one example of people abusing OSM because there wasn’t a good alternative.
So this thread is about informing people there is a good alternative now.
I think you are missing the point here.
The person who added it to OSM wanted people to find it. Deleting it without allowing it to move to a new home won’t solve anything.
I’m looking to understand if we can rally the community around this. The license question is a good one we’ll need to figure out, as some of the information belongs to users or websites and not really belongs to Mapeak.
One of your original points was that only if the route has a responsible operator it should be in OSM. That’s where I have a problem. BTW I never saw a strict OSM guideline like that.
In the cycling community we have some “pirated” routes that are very useful. They are outside your criteria and would rather they are mapped. This is in addition to KKL singletracks for example.
That Mishmar loop could be a nice hiking trip. I’ll refine my previous answer: First let it’s creator upload it in the new Mapeak mechanism. He would appreciate it. Then delete in OSM.
A suggested hiking loop is a different scenario from useful known singletracks that are on the ground.
Once a definition is vague, like what you just described, then it’s hard to know if one should put the route in OSM or not, and if to remove a route or not.
If you have a better suggestion for a criteria that we can maybe layer on update our wiki with it would be helpful.
“Truth on the ground” have stood the test of time, I’m not sure I’m familiar with a different criteria that did. Feel free to suggest one.
In addition, I believe OSM should not promote external sites or apps, especially for-profit companies or organizations. A trail should either exist in the OSM database or not; referencing or ‘moving’ a trail to a different site is strictly out of scope for the OSM community.
@peter_hrb Can you share your relation to the Israel OSM community? I’m asking mainly because Mapeak, the successor of Israel Hiking Map, is probably the main reason people in Israel are contributing to OSM. We, at Mapeak, are using the OSM user base as our authentication mechanism, the app is built to contribute to OSM: titles, places, description and it uploads recordings.
The need to find a home for routes that should not be in OSM was brought up by this community in our telegram group.
So yes, the scope of this discussion is relevant to the OSM community in Israel. Even if Mapeak is a commercial app.
If you are not aware of how Israel Hiking Map and Mapeak had transformed the Israeli OSM community I suggest to take a step back and learn this topic before posting here…
@Harel_M With all due respect to the Israeli OSM users, and to your need to promote your commercial app, your comment is very rude. What do you mean by “taking a step back”?
OSM is a community project, and does not promote commercial apps. Period. You should not promote your app or use the OSM community in and way for doing it. Being rude to other users by asking them to “take a step back” is over the line and you should apologize.
There are many commercial apps that add data to OSM, and people all over the world, including in Israel, are contributing to OSM and using our data.
I apologze if you found the message rude. This was not my intention.
Having said that, this is still the Israeli OSM community forum. I don’t know how to put it nicely, but it seems like you are missing some basic information about our community, our problems and how we operate. Also I find your messages rude too and patronizing, so that’s not helpful to the discussion here. I’m trying to solve a problem for the community while you are trying to lecture me about OSM.
I’d like to pull the discussion back the OSM framework and get everyone on the same page, starting with the discussion that started in the OSM Israel Telegram channel.
Over the years, Israeli mountain bikers have been using type=route relations to map single-tracks names. Many such names were given by the MTB community without any “truth on the ground” basis. Lately, members of the hiking community asked why they are requested to avoid using type=route relations to map their favorite/personal hiking routes.
The concern is that without clear criteria, OSM type=route relations will become a way for individuals to add their personal favorite routes to OSM. The understanding is that many MTB routes do not comply with the “truth on the ground” and “verifiability” requirements of OSM and should not be part of the OSM data set.
The conclusion of the discussion was, as written above, that OSM type=route relations can only be used for MTB routes if (1) there are trail marks on the ground and (2) the route has an operator that maintains it. A similar approach has been used for hiking routes in Israel for years.
Note that this conclusion refers only to the route relations. The member highway=* elements have “truth on the ground” presence and are not subject to any proposed change.
Since the MTB community is a major contributor to OSM in Israel, the other Israeli OSM contributors did not want to delete the information right away. Instead, there was a request to find an alternative host for relations that do not comply with the OSM guidelines. In response to the request, Mapeak is now offering to become such a host.
Zeev Stadler
OSM Israel community moderator
Co-founder of Israel Hiking Map and Mapeak
If a route is marked on the ground it can be added, regardless of if there is an operator maintaining it.
If there is no on-the-ground evidence of a route then it won’t generally belong in OSM. If a route is only a set of suggestions that a user thinks would make a good hike with no additional documentation then it’s not a route in the OSM sense of the term. There’s no need to wait for some third-party site to add those route relations, they can be safely removed.
On the specific matter of single tracks with informal names, I see them as fit for OSM under mtb:name. Although the name is not on the ground it is kind of a local name. The wiki agrees on that: “Sometimes trails have names that are only used by mountainbikers and not official. Use mtb:name to specify them. This is not Key:loc_name”.
On the matter of longer unmarked multi-track (sort of like a trip recommendation) routes I am ambivalent and not sure I have a well educated opinion but it does seem out of OSM’s scope on a first glance. (e.g. the one shared by Harel Relation: Mishmar HaCarmel Loop (18245859) | OpenStreetMap )
Further elaboration on the distinction: In my mind, there is a clear difference between the two types. The first is a physical entity with some sort of on-the ground presence despite the lack of a sign or formality. It’s a short track on the ground, quite often created and/or maintained informally by the mtb community, with a name that is well known locally. They are rarely (if ever?) relations because they are short single tracks “סינגלים”. An example is Way: 280522491 | OpenStreetMap - I am very much against touching these. They are just fine in OSM. They also usually have plenty of physical non-sign evidence like the arrangement of the rocks and the compacted dirt and the clearance of plants and more.