But as already noted, obstacle as a top level/feature tag (which it would have to be for “obstacle” nodes) isn’t even discussed on the relevant wiki page, and it is totally unclear what it would even imply (from a practical pov to be used in routing for example it would need to be on a way node).
To me, this makes sense. Or from a different angle:
An obstacle is something that you may overcome or remove (if you have the means to do so).
Passing a barrier in the forest may be illegal. Removing or circumventing obstacles is okay. (At least that would be my linguistic interpretation of the tag names.)
And I guess the editor settings should be changed too in that case.
Here in the forest, there are some unmaintained paths where it is very unclear whether and when fallen trees or growing vegetation gets removed. From a user perspective, I had the issue that I planned a tour where I ended up on a totally unusable way with little options for alternate routes. This annoyed me and has actually been one of the reasons why I signed up on OpenStreetMap: I like usable maps that are suitable for offroad traveling with the bicycle.
My specific case was this path, which is apparently used by hunters (it leads to some hunting stands) but it’s only accessible from the south (with smoothness=horrible, I would say). The northern part of the path doesn’t seem to be maintained at all for the moment, and almost not noticeable anymore. In addition, there is at least one fallen tree covering the path. Yet, I think I remember to have seen some marker indicating that there is still meant to be a path, even though the path doesn’t practically exist.
Deleting the path doesn’t seem to be right. It might be cleared from obstacles at some point in the future (and maybe even is part of some official maps from the administration). But can I say whether this tree will be removed within a year? I don’t really know. I wouldn’t expect it to be cleared soon.
Having an obstacle=fallen_tree and/or obstacle=fallen_tree tags on the map, which eventually (hopefully) would get rendered properly, then other bicyclists would benefit from that.
As an editor, I can only guess whether and within which time frame such obstacles would be removed. From local knowledge, I know that some paths (e.g. official trails) will be cleared again eventually, and I wouldn’t place obstacle tags in those cases.
But if obstacle gets formalized, couldn’t these viewers easily (or at least eventually) be updated to show the obstacles?
To my understanding the following three criteria are orthogonal (i.e. to some extent independent of each other):
- whether it is a single tree or a couple of fallen trees
- whether the tag is used on a node or on a way
- whether the obstruction has been placed intentionally or not
Getting back to my specific example which led me to start this thread, only the northern end was impassable due to one or more fallen trees and overgrowing vegetation. Even if it’s multiple trees, I could still imagine it may be okay to use a tag on a node? Otherwise I would have to split the way into two objects.
Like I said
Pretty much all existing use of barrier=log and obstacle=fallen_tree, what little there is, is with the wrong image in place. This period has also seen massive overall interest in mapping paths more accurately, demonstrated here with sac_scale and mtb:scale.
By which I’m simply trying to say, what you’ve concluded as incorrect documentation has been in place while a lot of mapping in this sphere has taken place. Let’s at least make sure this discussion happens before changing the documentation.
Nobody is asking for definite answers in such situations and OSM will defer to the mappers judgement in any case. Lets look at it from the other side: would the obstacle tags seriously imped the use of the path if they remained longer than necessary (aka till they have been cleared)?
Okay, I would then try to judge whether I expect the obstacle to be removed within a year or not. Maybe if I’m not sure and when it’s a corner case, I would also consider the severity of the problem. A totally impassable way with bad options to circumvent might be worth tagging, while a small trunk on a path for hikers and bicyclists (where there are no cars anyway, and where you can simply walk around it) might not require to be mapped.
Not sure how well this matches actual usage, but FWIW I really like this distinction and I would generally support efforts to move things further in that direction.
Doing this without permission would be illegal in my jurisdiction, as would cutting down live trees forming anobstacle=vegetation. On the other hand, a barrier, gate or fence may be completely legal to pass. A land owner is free to build barriers after all, but whether they have any legal meaning depends on the land use behind the barrier.
Done by NP Rangers inside a National Park, probably having a bit of fun!, to allow renewed access along a popular walking trail.
Well, I think the legal issues of removing (unintended) obstacles goes beyong the scope of OpenStreetMap? With respect to intended barriers, there is also the access tag.
I still think seeing obstacles as something that is meant to be removed (of course by whom and in which way may be another question), is a good rule of thumb. (I.e. of course, laws apply.)
If a path is meant to be overgrown and use is prohibited, it could additionally be marked as access=no.
Paths and tracks are overgrown from simple disuse. That doesn’t imply it’s forbidden to use them.
Similarly, a landowner doesn’t need to care about moving fallen trees that happen to block a path somewhere in a forest they happen to own. This has no implications on the legal use of said path. A new path may then form that goes around the fallen tree, or people just go over it.
Now that is extremely jurisdiction dependent and is not any kind of general rule.
Yes. So is every other stated legal implication in this thread. The point is to stop associating tags that describe trees or parts of dead trees with this other stuff.
I am not sure, but I think there are some situations where this is not true. In nature protection areas (German: Naturschutzgebiet), it may be forbidden to leave paths. I have seen former paths in those areas (which used to be official paths but have been abandoned), which were overgrown by be vegetation. Destroying that vegetation or using those paths despite the growing vegetation may be illegal in those protected areas. I am not sure if the rules are well-defined in all cases and each protected area may have distinctive rules. There are also Special Areas of Conservation on the level of the European Union.
Do you suggest to not distinguish between deliberately placed barriers and unintended obstacles at all? Or do you only want to avoid tieing legal questions to barrier vs obstacle?
As I noted above, there is also the access tag.
Even if neither barrier nor obstacle have direct legal implications, it may still be of interest whether a trunk has been placed intentionally or not.
Generally in Germany you may walk everywhere in the open (fields and forests, no paths required), unless they are fenced or it is a nature reserve, so yes, protected areas are the big exception (and each can have individual rules, but leaving official paths is generally forbidden).
Specifically about barrier=log: when I suggested the tag the idea was to use it both, for cases where they intentionally block a way, and for cases where they delimit an area (e.g. often seen to delimit parking or drivable areas in natural context, on picnic sites, etc.), just like barrier=fence. The implication is you can climb over these on foot but it blocks vehicles (might eventually be possible to lift your bike). May provide seating. There are even more assumptions you might associate, for example these are likely more permanent/not (easily) removable, compared to a chain (which has similar effects but different connotations).