carriage way, or a some sort of wheelbarrow or cartwheel path
You don’t need to go to Everest to do that, btw. The conditions are already such that, if I tried any of the Autumn scrambles this weekend, the chances are high I’d suffer the same fate. At 8000m even a motorway will be hazardous.
True! That’s exactly why Paths should focus on mode of transport. They’re a communication line.
There. You’ve already eliminated the Track and the Trace from the range of possible options. The margin of error is already extremely miniscule.
I’ll let others argue whether something should be mapped based just on Strava. In the recent thread the majority was against it, from what I can tell.
I added a few more photos. Please don’t call the narrow paths in the new photos trails!
That is then a climbing route as it goes to grade III and above? What is so specific about the mountaineering route that it is neither a Trail/Trace nor a Climbing Route?
Correct. But somehow, the width usually tells you whether it is safe or not. The narrow path distinction is an answer to a simple question - is it safe here to pass another person (whether someone coming in the other direction, or overtake someone slower)? If the answer is No, then it is a narrow path. This implies the danger, or limits, are very close.
The path over a meadow may be narrow visibly but
if you can step to the side onto the grass and pass another person, then it is not a “narrow path”. The potential walking surface is much bigger.
if you step to the side and sink into quicksand or a swamp, then it is a narrow path.
It is much easier explained on mountain photos because there simply is no more path on one side, and the other side is limited by a rock or a side of the mountain. But the walking surface is limited to one person only, and that is already a warning sign. I think it is worth distinguishing. It may not be used often but is distinct enough.
I am not saying it is specific to Mount Everest. Just that people are really interested in Mount Everest for some reason and keep putting it into the map:-). There are no motorways in 8000 m.
Mode of transport is a function :-). I really think the strict distinction between physical look and usage is untenable.
No, I can also not be sure if it is traces or pathless. Or even whatever_we_call_it. Nobody advocates for mapping from just Strava, but with some local knowledge and satellite, it is quite normal according to that thread. And even with the local knowledge, you will not be able to tell the subtag, you will need pictures or going there. Hence unknown is useful. Or lets say you get hold of a GPX track with information that that is some sort of hikable path from a friend. They will tell you it was easy to walk up. Good luck trying to get out of them the width.
New pictures, yay!
Last narrow looks more like traces. I can see the staircase, but what then? The second (especially that) and third are not dangerous for passing somebody). Difference that only manifests itself in the mountains is confusing. People in the lowlands will misuse it. Not such a big deal, but than it will make rendering it awkward. Everything but the last picture in narrow is a trail I think:-). Maybe mountain_trail :-). I really think the diwth distinction is better dealt with in sac_scale. Those being trails tells you: you can probably walk there without using hands and great navigations skills.
My path was promptly deleted when I did that the last time. I think a simple line/way is better. Then add a proper tag once it is known. Adding an unknown path does not help anybody else but the mapper, so its function is the same as a GPX track. But ok, put it as a proposal and let’s see some feedback.
The same about the narrow path. I’m not really hung up on it but I’ve yet to see valid reasons for not including it. Yes, it can/will be mistaken but the damage is pretty low if it is a borderline case, anyway.
With sac_scale this distinction is not that critical, true.
There doesn’t seem much left, if you ask me. Let’s see if there’s any more feedback here and perhaps we can post it as a new suggestion/thread and let people comment over the coming weeks?
I asked AI and it came up with a few suggestions (source order):
Single-Track Path
Multi-Use Trail
Narrow Trail
Bike and Cart Path
Compact Route
Recreational Trail
So, you were close with multi-use but still only second to single-track. :))
Edit: questions
“what would be a convenient name for a path that is suitable for vehicle traffic but is narrower than a 2-track road for cars?” and
“this is not a road but a path narrower than a car, however suitable for bicycles, carts, motorcycles, etc.”
Yeah, I think not many people read it here now. I think we should put what we talked about in the wiki post and then start a new thread.
As for importing GPX tracks, I always simplify them in JOSM. I do sometimes go to places where whole maintained trails are still not in OSM, so getting a GPX file from somebody saying "there is a trail marked with red dots and it is trail all the way is enough for me to upload it. I did it only once or twice but can happen.
Hopefully a native speaker will have some suggestions. Karrenweg does seem specific to Austria but hopefully there is an equivalent in English. Maybe it turns out to be multi_use. I’ll add it to the list of potential names and you can add whatever you feel is missing?
Historically we had “pack horse trails”. Hilly areas not that far from me do have something a bit like this - not wide enough for a Land Rover or other high-clearance 4wd, but wide enough for a quad bike (usually with the farmer’s dog looking expectantly on in a box on the front). I can’t think of a good differentiating name though (in casual conversation “path”, “track” and lots of others might be used).
In a different example, OSM adopted an American rather than English term “sidewalk” in order to avoid confusion - maybe here something like Karrenweg would work here?
Not only can, in the Data users do just that. As far as I can tell, rightfully so, following best practice announcements. I do not see in the documentation why path should be used on the obnoxious outliers only, A path in OSM speak can be nothing on the ground, it can be 8m wide paved. Not to say that I am particularly happy about that, but that’s the current wisdom there is.
I think the same logic should apply to roads/paths added from satellite imagery without local knowledge. A line alone isn’t very useful for mappers since it won’t show up in specialized renderers. Some people prefer using highway=road, but others believe it should not include unknown highway=path. I’ve suggested highway=unknown, but I’m open to other ideas.
I agree that it is better to have something rather than nothing. But what is the point of having an “unknown” road or a path? Doesn’t having a road or a path (without other details) already tell what is known about it?
I draw a line in the editor and save it to a GPX, transfer that to my mobile and then go on a survey – I am in for the fun, and survey is where the fun is.
Of course, if all would map like that, OSM would not nearly be as complete.
What I find is the most hard to do: Learn if a path actually exists – it might be a dozen meters off, it might be trail_visibility=no, it might be a game trail, &c.
PS: Anybody enlighten me on why a (mostly single use) purposely constructed MTB downhill race should be mapped as path instead of cycleway? C.F. Use of highway=cycleway for mtb trails – this concerns picture number three posted by @c0nsumer
For reference, that image does not show a dedicated downhill MTB trail (which I believe you are calling a “race”). It’s slightly downhill in the part photographed – no more than 100m long – then levels out, climbs, makes a turn, and is flat for quite a while. It’s also a hiking trail and designated for both pedestrian and bicycle use.
It’s as much of a footway as it is a cycleway. But mostly it’s a path through the woods where cycling and hiking are specifically allowed, with both having equal access.
So it is in no way different from the other two pictures shown? (I cannot find the quote where I got the impression this being a purpose built MTB trail, while open to pedestrians out of legal considerations, I might have been mistaken, maybe more search needed.)
It was built by a trails organization optimized for mountain biking, but just as open to and commonly used by pedestrians.
Quoting the second paragraph of the highway=path wiki entry:
Current usage of highway=path: Ways used by pedestrians, small vehicles like bicycles, for animal riding or livestock walking. This includes walking and hiking trails, bike paths, horse and stock trails, mountain bike trails, as well as multi-use paths for cyclists and pedestrians or similar combinations. Certain types of paths are tagged more specifically with highway=footway, highway=cycleway, and highway=bridleway, but in many cases none of those quite fit so highway=path is used instead.
I feel this falls directly in line with highway=path, foot=designated, bicycle=designated, and segregated=no, along with the other tags denoting MTB trail difficulty rating, hiking difficulty trail rating, etc.
highway=cycleway states:
Ways that are not marked as cycleway by traffic signs (and are therefore open to non-bicycle traffic (e.g., moped or horse) should typically be tagged as highway=path instead of highway=cycleway, but see duck tagging as well.
This is not a designated cycleway because it is not marked with traffic signs. It also doesn’t match any of the examples which all show much more developed, general-purpose, cycling-first routes.
highway=footway states:
The tag highway=footway is used for mapping minor pathways which are used mainly or exclusively by pedestrians.
Footway is also inappropriate because that is not its primary use either.
The margin of error is already extremely miniscule.