a bit weird cases, but yes it seems to more or less fit?
and I do not believe we will get non-weird cases
though this smoothness=horrible rather qualifies as some barrier= value with this steeping stones… This is quite moot case as it is not open to general traffic anyway.
or actually highway=crossing+crossing=uncontrolled+ crossing:markings=stepping_stones and traffic_calming=cushion 
I agree that any traffic_calming=* should be mapped separately and not taken into account when evaluating smoothness
It is open for pedestrians, after paying an entry fee, and parts have been made accessible for wheelchairs. Visitors with baby strollers could be interested in the smoothness of these roads. They are now mapped as paths and pedestrian streets, with street names, wikidata IDs, etc. They are some of the oldest streets with sidewalk=both
In my opinion not very_bad but horrible, this mud.
by tire tracks we see it seems that "heavy-duty off road vehicles " are not actually needed here
if anything, I would consider bad as viable
Why? I don’t see any obstacles that would require high clearance, so it could be smoothness=bad. You can’t see how deep the pools are, though…
It’s my opinion that is more bicycle than car centric. And no, bad is much too good for such ways - in my opinion.
smoothness should not consider only cars. Else we need
smoothness:car
smoothness:bicycle
...
type of tire marks left there indicates that they were passed without requiring heavy-duty vehicle
Key:smoothness - OpenStreetMap Wiki does not distinguish using bicycles between very_bad and horrible
and probably it does not count as comfortably passable on trekking bicycle pushing it below bad
There are some surfaces that are particularly hard to pass on bicycle, and soft mud is one of them (loose sand and loose gravel are other examples). But I think these are difficult because of the mechanical properties of the surface and this should not be taken into account when evaluating smoothness, which is about the physical shape of the surface (roughness, bumpiness, …). If this road was perfectly smooth mud, it would still be hard to cycle on, but that doesn’t make it smoothness=horrible.
I think very_bad is fine in this case and our data tells everything we need to know to evaluate how passable this way is for bikes. Because it is combined with surface=mud, which is hard to pass anyway (-> yes, regarding both tags, it’s “horrible” to drive there by bike 
).
The need to differentiate smoothness by vehicle type (e.g. using smoothness:bicycle) has long since been discarded in the long discussions of recent years (fortunately in my opinion). We have created a relatively objective classification that can be interpreted independently of the vehicle, and only argue about the nuances of the delimitation of some categories.
Again and again: very_bad is not requiring high clearance (when driving there by car), it’s just that a very bad road is much more comfortable to use with high clearance by car / it’s very uncomfortable to use a very bad road by car without high clearance.
Sorry about the delay (again). File:Wooden bridge in poor condition.jpg - Wikimedia Commons
Maybe one of these for intermediate (the gap between lengthwise planks would be nerve-racking on narrow tires)?
I propose to use it as surface=wood smoothness=very_bad example with extra note “it looks like it would collapse under a car, or heavy duty vehicle and probably deserves extra tags”
it is a bit weird case, but I cannot imagine surface=wood going below
or maybe surface=wood should be added to top part of table with surface=very_bad marked as probably impossible to happen on a passable road?
The way to this bridge is tagged as highway=track, which seems correct. A normal car could physically and possibly legally drive there. Google Maps will route a car there. I’m not sure if the road on the far side ever connected anywhere except the fields on that side of the river.
The surface is pretty much gone, but it was possible to cross by carefully walking over the supporting beams, which seemed surprisingly sound.
I think past this level of deterioration one has to ask what the surface value should be. If a road still asphalt if the entire surface layer is gone? Or for wood, up to which point are sinking and rotting duckboards still a wood surface…
