I did vote for “It’s complicated”, may I still deliberate: tracktype=* is not a good name for a tag that better be called firmness=*. Has this already been mentioned: Radically changing the definition and from what I have read above, completely swapping out both the text and the images thereunder, of what has been the base for millions of mappings not a good idea? @rhhs Who is going to verify the new definition meets the mappings?
1 Like
You mix two different things: The proposal process, that starts with an RfC, goes thru sometimes lengthy discussions and ends with a vote that establishes a newly approved tag; And a poll in the forum that allows you to the change definition of a tag approved by mere usage. You are right, both times, talking heads lead your way. Reminder: Mappers vote with their feet, neither on the Wiki, nor on the Forum, but by mapping.
Perhaps you witnessed the recent change in surface=fine_gravel and now want to usurp tracktype just the same? Let met tell you: Most of the mappings of surface=fine_gravel I am aware of do not meet the new specification but rather the old. IMO, this tag has been hosed by Wiki Fiddling, bike routers well advised to not listen to what the Wiki says! You really want to repeat that?
That may be true where you live, but definitely not true worldwide
3 Likes
Ah, the beauty of selective quoting..,
I was trying to solicit comments here Poll: How do you grade surface firmness but that failed. That looks to me a grade5 but quite firm, at least at the time of photo. BTW: Not sure this even shows a highway=track in OSM terms at all.
- Would you mind linking/sharing street level imagery of some tracks in your area?
- What would you gain by having a tag like
firmness=excellent|good|intermediate|bad|…?
& this is the thing.
As others have mentioned, tracks can change dramatically over the year, or due to weather!
I’ve seen paths that are a footway in summer, but a ski run in winter - what surface / firmness are they?
Again, we have vehicle tracks that are hard, solid, dirt when it’s dry, but one shower of rain & they turn into bottomless mud. They’re usually tagged as firm or mainly firm, but should they be?
1 Like
@willkmis @Carnildo @arctic-rocinante - Would you do jump in?
After all, the pictures on the tracktype=* documentation article not applicable all over the world, so a future article on firmness=* greatly would be helped by that, I’d suppose?
People have already covered most of my objections to tracktype being a synonym of firmness, my take on it is well-summarized by Tracktype as a measure of surface firmness - #5 by ezekielf . But I can provide a specific counterfactual I have in mind: I’m personally quite familiar with the mountains and deserts of Southern California. These are fairly-to-extremely arid ecosystems with little topsoil, and bedrock is often either exposed or not far down. Such an ecosystem is not especially unique, portions of e.g. the Mediterranean, Africa, South America, and Australia have similar climates.
A prototypical highway=track in this area is a fire road, built and maintained such that emergency vehicles can access an area during a wildfire emergency. They’re usually off-limits to motor vehicles otherwise but open for foot, bicycles, and/or horses for recreation. A very well-maintained one will be fairly smooth (probably smoothness=intermediate or bad), unpaved, and wide enough for a truck (>3 m). Here’s a photo I took of this road, which I tagged as tracktype=grade2:
As you can see, the surface is quite firm, well-graded, and cleared of obstacles. These are the highest quality tracks in the region and would be quite easy to drive on (if you had permission), walk, or bicycle. However, you can probably see from this picture that the material the road is made of doesn’t differ all that much from the ground on either side, it’s just what it looks like if you rip out all the plants and grade it. I can’t say I’m totally sure, but I wouldn’t be surprised if no outside material has been added at all when making this road. It’s tagged as surface=dirt.
The problem with equating tracktype to firmness is that as tracks get worse in this area, firmness isn’t the first thing to go. A grade3 track in this area would be similar to this, but with more divots and gullies from erosion and/or rocks sticking out. But it would probably have the same “firmness”, just be less even. Toward grade4 or grade5, I’d expect the road to be poorly cleared, perhaps with plants between the tire treads and lots of rocks jutting out. But it could be just as “firm” as a grade2, nothing stays naturally muddy around here (unless it’s rained, in which case everything is equally muddy). However, in some cases lower-tier tracks are actually sandier; this would be a good case for a separate firmness axis. Here’s a photo I found online of a track I would probably call grade3, although it depends what the rest of it looks like. It’s not less firm per se than the grade2, as the surface is still mostly hard dirt, it’s just lower quality: less smooth, more erosion, etc.
I hope this illustrates my perspective on why redefining tracktype to mean firmness doesn’t really make sense. I think tracktype is useful as a broad, qualitative judgment of a road’s quality, even if it’s not 100% objective. Of course, further subtags like surface, smoothness, and maybe now firmness can only help to clarify the situation.
10 Likes
well, not everything, those roads with a draining or sealed surface will not become muddy, and this is interesting to know as a data consumer when the weather is wet.
2 Likes
Thank you for the reply. It all makes sense to me. When I did ping you, I thought you held the same opinion as @bradrh , that tracktype should concern itself with firmness only.
PS: In my area, tracks mostly look like shown in the 20 years old Wiki photo. But there are also regions with little top soil.
From the photo I’d call that a grade3, in the data it is a grade4 even, perhaps because it is skunked (the track, not the tag) in other places. I do not remember. I added access permissions but left the grade. The reddish hill BTW houses a prehistoric copper mining site. That’s why the photo.
What information would tracktype convey that smoothness=very_bad, and surface=unpaved (or ground) doesn’t handle?
3 Likes
My I start from personal experience of opentstreet-mapping? It is not uncommon here, that tracktype is the only attribute, if any at all.
So if these two tracks in your pictures had a single attribute grade5 I’d infer: There was little to no effort spent into improving the ground. I’d not hold high expectations into using that for anything but walking, while for this, I’d even might prefer it over a grade2.
If they were of grade2 I’d infer: Quite some efforts were spent into improving the ground and likely also into maintaining that state over time. I’d expect them to not fall all too much behind an unpaved public road.
To your question: I’d say, together with these two attributes, it would convey just the same to me. But I would have a hard time imagining what to expect in case of grade2 while in case of grade5 it would fit not so badly.
Two tracks photographed less than a hundred meters apart. The surface firmness is identical (a mix of fine gravel and sand), but one was created with a bulldozer and the other was created by repeatedly driving over the same route. Many tracks in the area are similar regardless of construction technique: the ground is a thin layer of soil over deposits from the Missoula Floods.
The firmness of this little-used track isn’t much different, despite being heavily overgrown.
1 Like
Cross-linking some potentially useful context on the different interpretations and usage of tracktype:
https://community.openstreetmap.org/t/improving-the-wiki-documentation-of-tracktype/106918/5
I’m still a bit confused about why surface=concrete:lanes is mapped with tracktype=grade1. Based on the wiki samples, if the lanes’ surface were compacted instead, it wouldn’t be tracktype=grade2 but rather tracktype=grade4.
My view is that surface already conveys information about ground firmness (e.g. mud vs dirt vs clay vs rock). In my region, tracktype is mainly useful to indicate how well a double-track is used or maintained.
In my view, this depicts a grade5 type track. What would you gain, if tracktype would rate firmness only? What would you loose, if tracktype was not about firmness only?
I could not vote on the other poll. Looking at results here, I’d say there is a way to sort this out in a manner that does not make millions of mappings dangle from the heavens without support, a.k.a. turning trackype=* from a rough type with warts into a tag that maps a minute detail of its surface with scientific rigor.
Obviously, a change in the wording of the documentation still necessary; Should be the least intrusive, so to invalidate the least number of millions of mappings: I propose to move the mentioning of “particularly firmness” to a separate paragraph that says perhaps, preliminary wording:
In regions with deep soil -- where the tag was coined and first used -- the amount of developmental efforts into improving tracks to accomodate them for a wider number of users positively corresponds with the firmness of its surface. This though might not apply all over the world.
@bradrh @Fizzie-DWG @ezekielf @rhhs and others, would you be fine with that? (I do not think the addition of that interpretation of the pictures as being particularly about firmness went through such scrutiny, BTW)
2 Likes
I still don’t know what info it adds that isn’t already in smoothness and surface. The user wants to know what vehicle is appropriate; passenger car, high clearance, or jeep; dirt bike or adventure bike.
1 Like
See here:
It doesn’t provide info that smoothness and surface lack, it provides info when those tags aren’t present at all.
3 Likes
That’s fine then. Looking the pics on the tracktype wiki page, I could drive a passenger car on any of those. Is that the way its intended, & the way it’s used?
Not owning a car nor a drivers licence, looking at the new grade5 picture: I guess, you can drive a passenger car right through the meadow and the field too, at least at the time the photo was taken. What makes this a track is only the fact, that a certain part of the area is actually used for repeated driving. Nothing whatsoever has been done to improve that area, and this is why I see a grade5 track there. Your 2020-07-08 photo is similar, the most obvious difference, there the track was cut clear from trees. I am not certain if that this is sufficient for an upgrade. BTW I think this shows something were key:smoothness for cars and key:smoothness for bicycles do not align to the same value. I certainly can ride that with my touring bike.