I heard in song lyrics “move your ass and your mind will follow” but from talking with people aiming to hike St. Jones Way with an ass, the human has to be slightly ahead but not as much as single-file. I do not think mules any different. On other occassions I learned, the asses (a.k.a. donkeys) do not like paths rightfully tagged sac_scale=demanding_mountain_hiking and rather stay put in place where demands get over board. I also guess mules not different.
I realise that, but I was answering a question about a specific set of maps, where that kind of way is rarely an issue. The surface of tracks and paths is these forests is usually compacted at best, and the only vehicles I’ve ever seen are for forest maintenance.
That’s the definition of a way with smoothness=horrible
2 Likes
When you actaully look at the wiki:
https://wiki.openstreetmap.org/wiki/Key:smoothness
it can be anything from very_bad to impassable. Smoothness is despite what it says in the introduction exclusively geared to cars (just look at the definitions). Also it assumes somebody to go around with a tape, having values different by 3 cm. I do not think that is usable for non-car users. This key is even worse than I assumed. Plus when you look at the pictures in post number 2, how would you apply smoothness? By the wiki, they are impassable. Smoothness does not expect there might be ways that are both smooth and unpassable for cars (and that at the same time require a MTB to enjoy).
More and more I think mapping ways for cars and not for cars should as little in common as possible, these are mutually exclusive worlds.
(expletive deleted - I had to suppress my inner Roger Mellie in this reply)
If you actually read that page, it’ll be clear that the whole point of it is to cover multiple usage modes; to quote the page:
… Many are already offering routing for cars, bicycles, pedestrians, and public transport: wouldn’t it be nice if they could add routing for roller skaters, SUVs, 4WD, etc.?
It’s not the best-described page on the OSM wiki, I’ll grant you that, but it gets the job done. Even someone who has never used one of the example vehicles on the page (from roller blades to tanks in the examples) will at least be aware of the concept and should be able to ask themselves “excluding other factors such as width, height etc., would someone travelling by X be able to get through here?”
Surely no-one with an ounce of common sense would say “I’ve never driven a scooter/car/tank and therefore can’t say what smoothness value should be assigned to this path across this field”.
No matter what I drove, the images
, “usable_by” and description are so contradictory that I personally will not ever use that tag until the wiki starts making sense. Description column in the definition table almost exclusively thinks of cars. BTW: as a daily user of a city bike, what stops being passable is somehwere around very_bad, which for good measure does not even mention bikes. I can understand how ambiguity of smoothness and the way it not very well mixes description and usability can make people sceptical of dividing path. The way forward would however be somebody spending a decent amount of time on the wiki page so that it starts making sense. Even maybe putting it to a vote, like I did with 'sac_scale=strolling. And sac_scale` was in a much better shape to begin with.
Anyway, the text of the wiki (which I read) very much feels like vehicle that are not cares were added as an afterthought. I might be mistaken about this, I do not know the history, but it does feel that way.
And to stress, the way smoothnes is segmented does not map onto mtb_single_track at all, nor does it hint at how smoothness should even be applied on such ways.
For deity’s sake, this gallery: Key:smoothness/Gallery - OpenStreetMap Wiki does not seem to have a single example that is not wide enough for a car. Having this as the title image is highly misleading: File:Jena Track roots.jpg - OpenStreetMap Wiki what smoothness is it? I have no idea.
I should probably slighly correct that smoothness is, as per wiki, exclusive to properties of roads usable by cars. That can be somehow useb by non-car users of such roads. However it does not seem to be usable without wild guesses for pathways.
Oh, and we are compltely offtopic here :-).
And that’s entirely OK - people are free to use whatever tags they want. OSM has always worked that way - different people are interested in different things.
… along with (and I’m quoting here) “… racing bikes, … SUVs … hikers …”.
What it does require is (the thing that I suggested about that anyone with an ounce of common sense would be able to do), to say e.g. “I’ve never driven a tank, but I’d imagine that it would be able to use this OSM way”.
Why should it? Those are different keys, documented elsewhere.
In the case of smoothness and mtb:scale I had a look at ways tagged with both in (mainly) England and there wasn’t a lot of correlation. That doesn’t mean that either is “wrong”; they’re just measuring different things.
For the avoidance of doubt, the wiki page does NOT say that.
SUV are (evil and dangerous) cars last time I checked :-). As I said, “almost esclusively” - maybe 90 % of information in taht column is about cars.
Which seems to suggest that in practice maybe smoothnes is for track and up not suitable for paths.
Not explicitly. But it seems to implicitly assume it. Just look at the pictures. It is kind of like footway - that wiki page also is only about urban paths in practice but somehow the tag is in an undefined way permissible for hiking trails.
Anyway, I would really be interested in how smoothness is supposed to be used on the pictures of mtb_single_track in the second post (and if it could be defended by citations from the wiki, even better). I claim that in its current form, it is not clear at all how it should be used (for carless ways).
I agree the “description” column in the table (which was only added in 2021 by @rhhs) is very car-centric. I don’t think the rest of the page and the “usable by” column (which dates from 2008) are too car-centric.
After discussion on the Wiki talk page, I’ve just edited the description of impassable: the 2021 description said “Ways that used to be passable roads … but have deteriorated so much … that no 4-wheeled vehicle can pass it.”
This may apply to some impassable ways, but over 60% of uses of smoothness=impassable is on highway=path and if you look at example, very often these paths were never roads. The 2021 description just confused people into thinking that impassable is inappropriate for such cases.
If anyone here speaks Hungarian, Portuguese, or Romanian: you can help by translating the new description here.
2 Likes
That is certainly one of the things that has put me off using smoothness in the past.
I also wonder if it is right that “good unpaved roads” are classed as “bad”. It seems to push most non-urban paths down towards the bottom of the scale. Maybe that is right, but it feels odd to tag a perfectly normal hiking path with no obstructions as “very bad”.
2 Likes
I agree that the description of using smoothness on the wiki can be improved. When I added the description column, I was too much focussed on cars. It has been suggested to take the descriptions of what smoothness means for cars in particular out of that column into a separate table that also describes it for other wheeled vehicles, something like Verkehrswende-Meetup/smoothness - OpenStreetMap Wiki (translation to English User:Rhhsmits/Smoothness details - OpenStreetMap Wiki ). There are also inconsistencies between the pictures in the main table and those on Key:smoothness/Gallery - OpenStreetMap Wiki )asp. for smoothness=bad/very_bad); I’ve been planning to work on that issue but found I am using all my available time on following the threads on the issues with highway=path 
Let’s create new threads for the issues that smoothness has, preferably after the highway=path issues have been dealt with.
3 Likes
OK, people, I’ve created a new thread, to narrow-down the focus of the discussions:
Since there hasn’t been many larger qualitative changes recently, I believe we can proceed to polish the proposal.