Today happened over this track:
Seven months ago a user of StreetComplete tagged surface=fine_gravel there. I felt the immediate urge to swap for surface=compacted. Photo shows a prime example, doesn’t it? Somehow I managed to resist. Curious now:
Are those two values just synonyms? Or is SC misleading users? (No SC user myself, do not know how the selection is guided there.)
I consider surface=gravel and surface=fine_gravel to be recently added gravel surfaces of a rather uniform size, and surface=compacted to be a mixture of gravels of various sizes and clay that has been rolled to a smooth (when recently made) surface. Your photo looks like it has quite a large amount of clay, so I would tag it surface=compacted. But it wouldn’t be something I would start an edit war about…
Indeed, better to start a discussion on the changeset, possibly referring to this thread.
With these pictures no wonder the tagging. Makes me suggest:
surface=compacted (where the tires go)
surface:middle=fine_gravel (where ???)
I guess part of the problem: Compacted is not only a surface. It is a recipe for constructing a road cover.
I left the mapping alone. I guess all of those values
might be considered synonyms because – from what I observe tagged in the wild --mappers cannot be relied upon to use them consistently and differentiable.
Apropos pebblestone: I found that on compacted roads, like the one in picture above. bikerouter hates it much more than vehicle=forestry, that is what made me aware of that.
PS: Our administrative GIS only knows “paved (gravel)” and “unpaved” when it comes to tracks.
If mappers can’t reliably distinguish between them, will map users be able to?
I have my doubts about the usefulness of having so many values for unpaved surfaces. For practical purposes, it doesn’t really matter what surface you have below your wheels or feet: knowing the smoothness and maybe surface firmness is much more important. For paved surfaces, knowing the surface can be useful for orientation (“This side road is asphalt-paved, but the one I should take according to the map is paving-stones-paved, so this is the wrong way”) but it’s less so for unpaved surfaces if even mappers can’t distinguish between them.
this photo looks like surface=compacted but maybe it was resurfaced shortly before SC mapper was there as it was in fact covered with thin layer of fine gravel?
either way it looks like it should be retagged
no and no
usage is not ideal (some areas calling and signing surface=compacted as “gravel road” is not helping), but there is some correlation
The actual surface of condition/composition of dirt roads changes drastically due to weather and time since last resurface. On a winter solstice ride, after like two months of rain, ”compacted” roads were like sticky mud. Roads that had been recently resurfaced had a layer of loose pebbles/gravel, with only the very edges somewhat firm.
I don’t think this kind of seasona/lifecycle change can reasonably be tagged. An experienced mapper could be aware and not tag the ”freshly resurfaced” condition, especially on a remote roads that likely won’t be re-surveyed in years.
Locally compacted may be tagged as gravel or even sand (by mappers being naive when translating OSM tag values to Finnish).
In the local cycling scene we’ve been actively trying to instruct mappers that the value gravel should be reserved for surfaces consisting of fist sized rocks, something like this (often without the smaller pebbles which make this particular road slightly less terrible than it could be )
No, in my opinion it does matter, and “smoothness” doesn’t describe it any better. Depending on the tyre width, it makes a big difference to the rideability.
Compacted is already very close to paved and can usually be ridden on very well, even with bicycle tyres. You have to put up with some loose fine gravel, especially at the edges and in the middle.
Fine gravel is pure fine gravel, not compacted. However, because the stones are sharp-edged, they often form a relatively firm and flat surface, which can be quickly destroyed, especially when cornering and braking, and then quickly becomes uneven.
Gravel is coarser rock, say fist-sized. It is quite firm and stable and can usually still be ridden on with large tyres and good suspension. On a bicycle, however, it usually becomes an ordeal.
Pebblestones are round stones, attractive to look at and often found in parks. They are fine for walking or slow driving with wide tyres. With narrow bicycle tyres, you sink in and find no grip in bends. The round, smooth stones simply roll away. Only loose sand is worse.
No, they are not synonyms. And if mappers use it incorrectly, often because the literal translation into the local language has a different meaning, you should explain it to them over and over again.
Sorry Mammi, not something that I’m eager to do. While I share much of what you say those tags represent 
I’d rather have the Wiki documentation and the editors give clear guidance.
IIRC there was a recent push from the talking community that even shows in the Wiki documentation. Fine gravel means that touring bike wheels will sink. It is a useful tag for the outdoor seating area of mappers beloved inn. On ways it occurs on footpaths in cemeteries and such.
Curiously, bikerouter.de treats it as a synonym for compacted. How can consumers/mappers be so off of what is documented? Maybe they are following an outdated documentation/editor, that of course may have had influenced dozens of thousands of mappings?
Nice try. I bet not.
In my mind also the distinction between surface=fine_gravel and surface=compacted is sometimes rafter fuzzy precisely because ways that begin their life with a fine_gravel surface eventually turn to compacted if not maintained (if new fine gravel is not periodically added or the surface isn’t raked or roughened up).
This! For a few years, I’ve been meaning to take a good representative picture of a gravel surfaced road. There are a few I know (relatively close by but in a direction I don’t usually frequent), but I’ve never remembered to bring my camera with me. Now that it has snowed, I guess I’ll have to wait until spring again. The large grain size makes these roads difficult to ride for bicycles with narrow tires, whereas they roll fine of a fine_gravel and even compacted surfaces. It’d be great if these large grain-size surfaces could be fairly consistently identified from the surface tag.
Where I live (Denmark) in my experience most unpaved roads are tagged ‘gravel’. However, it is very rare to see roads using a uniform layer of fist sized rocks as pictured in the wiki. Especially farm and forest roads use a combination of materials ranging from golf ball-sized stones all the way down to fine sand. They can be quite uncomfortable to cycle on even though at first glance they don’t look that rough. How would you tag such roads with mainly fine gravel (and sand) but also a few larger rocks?
An analogy: fine_gravel is the icing on the cake, while compacted is the cake itself. What is getting mapped just depends on the time when the state was collected. The different tags just map life-cycle. The next stage would be gravel then, when all the fine material has been washed out, thereby also destroying the compacted layer a.k.a. cake BTW.
This would not be a problem, if mappers would not use fine_gravel to tag other surfaces, where there is no compacted layer involved. The Wiki documentation shows a sample picture.
That makes fine_gravel ambiguous. I’d say, the current Wiki doc tries to heal that. I’d also say, this comes a bit late.
To answer my own question in this topic: I should have changed the tagging of the track in the picture from fine_gravel to compacted. Reasons: It is not of the other kind of fine_gravel surface. It is now in that state of life-cycle. Chances of somebody getting by after fresh fine gravel got added during maintenance are very slim. Frankly, I never observed such work and I see lots of compacted tracks.
Are you talking about something like this?
I would consider dirt / earth . This particular dirt has been excavated somewhere else and the laid here as a road surface, which isn’t exactly what wiki’s ”exposed earth/soil/dirt but it is not sand or gravel or rock” suggests. But still, it’s just dirt.
If something defies classification, there’s always the humble unpaved .
Similar but a little less sand and more small stones. I often consider tagging them ‘compacted’ but not sure if this is ok, when there is still some loose larger rocks.
The often look like this:
tricky
Does not appear to be mechanically compacted (e.g. by rollers or vibratory equipment); rather looks loosely spread and only slightly firmed by vehicle traffic.
in my eyes: surface=ground
in my eyes: compacted
My answer is a double negative: I think this is not an unfair description! I personally prefer to couch this in terms of fuzzyness, though. And that is not uncommon in OSM.
Regarding life-cycleness, it’s turtles all the way down. You know, entropy. So fine_gravel → compacted → dirt → ground → bedrock. Meitner and Sakharov seriously suggested that even protons have a half-life. Since the degradation of road surfaces takes a relatively long time, I’m not sure the OSM lifecycle-concept applies here.
In my neck of the woods, this happens relatively frequently. Most of the time, the road surfaces are raked with a heavy truck. This raking removes large stones that emerge from the subsoil due to ground frost. The rakes push these rocks to the side of the road (and the biggest ones are carried away), and sometimes a new layer of fine gravel is added simultaneously. This is done both for paths and tracks (so roads meant for 2-weheeled and 4-wheeled vehichles).
Seconded. I would add smoothness=bad to describe the riding condition, this looks pretty close to one of the examples we have in the gallery.
It could be helpful to document a track maintenance lifecycle in some wiki page somewhere. Might take a while to get a suitable set of photos though.
