I oppose any change to the documentation, that will permit mappers to mark trail blazed (alpine) routes with trail_visibility=excellent|good|intermediate where on the ground there is no trodden path.
I do not oppose some deliberation: After all, trail blazes sharpen the eye, in that they limit the horizon where to look for the path.
I made some edits to the proposed new text on the wiki talk page. Can we please focus on that?
(Imagine future mappers interested in how the consensus on trail_visibility was established having to read through this discussion…)
Let’s try to keep it simple, both for the wiki text and here. I think we should strive for consistent mapping all over the world as much as possible, so the wiki text should be understood the same all over the world, by Yanks, poms and down under, but also by Germans and other non-native speakers. It should be translatable as well, so using a very precise English word to describe something that has no equivalent word in German is no good either… This will only work well if we use Jip-en-Janneketaal 
I am not at all happy with the new text proposal - This will turn trail_visibility into route_visibility - this is plaguing the key ever since its creation: Any visual clue that helps you stay on track.
So, it may be fine for highway=route, but is it as fine for highway=path?
This thread has now had over 100 posts in it. The wiki page before the process started was excellent - it let the pictures do the talking. What I said a month ago (that adding extra guff to the wiki page would make it less useful) still stands.
Perhaps we can somehow limit contributions to “1 per X miles of new paths surveyed on the ground”? 
(I believe I’m still in credit in this thread based on this last change…)
That’s a badass cred-link you offer, Andy. Respect.
I’m not so sure that counting on “common sense” passes the verifiability principle, though. I would prefer the extra wording (which surely need not be “guff”?) to clarify what is meant. Because it seems that to some it is “common sense” that the e.g. nondescript rocky mountain with few rocks moved together constitutes trail_visibility=excellent and to others trail_visibility=none.
Such discrepancy seems undesirable outcome to me, and is clear indication that the wiki is lacking and should be improved. Sometimes a few words can speak more than a hundred pictures…
If the intended outcome is that only those who map on the ground get voice (which BTW seems will soon be deprecated according to the OSMF Board Task B501 in favor of armchair mapping, but I digress), I’m afraid the proponent of that idea might be in for a rude surprise 
(somewhat offtopic - and neither you nor I are being entirely serious here (hence smileys) but:) I suspect some of those proposals will receive rather more pushback once they have been read in more detail…
You should be aware, that the SAC document you link to and the one that was current at time of the original proposals have some slight differences. First of all, everything has become more difficult, even though the paths remained the same 
Two: Not a native, but the English is lousy at times. Three: Hiking has become more mainstream, so it is a bit an outlier, as it is supposed to be easier today than then. Below image looks a good match on the description to me.
The footpath is constructed and excellently visible. The railing rules out falling. I’d say, the picture shows the gold-standard hiking path.
For your entertainment, a mountain path: There are three trail blazes on display.
The path is not on OSM. I hesitate to map it. So I may loose out in the contest above
Can we please please focus on the text of the proposal on the wiki talk page?
I think there is consensus that the wiki text can be improved by further clarification, so that it is clear for more people.
I think the main changes compared to the version that is on the main wiki page (that hasn’t changed much since last year) are that markers are now not mentioned in the table, but examples of markers are mentioned explicitly as clues that can improve trail visibility. That it is not about route visibility has also been clarified.
Can you please state in your posts if you can accept the new text, and if not, which text changes you propose?
I appreciate your efforts… unfortunately I think the discussion has shown that there are different views on how much markers are allowed to contribute and different interpretations of what “not route visibility” means. This will make it hard to agree on a change.
While I personally agree with your proposed changes (they match how I have understood the page), two people have already made it clear that they are not happy with them. I still think we would all map most paths very similarly but the different views clash in some edge cases. So if we want to make any progress, then we maybe need to focus on what we can all agree on.
How about:
The key trail_visibility=* is used as part of a classification scheme for hiking trails and paths. The key describes how easy it is to see and follow a mapped trail. This includes the visibility of the path on the ground (differences in surface of the trail vs. the surface next to the trail, differences in vegetation, etc.) and in the landscape (levelled path surface, treeless corridor through a forest, etc.).
Trail blazing (way marking) can enhance trail visibility, but even excellent trailblazing is unlikely to turn a path that is otherwise barely visible into one of the top grades. See trailblazed:visibility=* for attributes regarding trail marking quality.
The key is not about the visibility of any hiking routes that the trail might be part of, e.g. how easy it is to choose the right trail at a junction.
To me, a path with excellent trail visibility should be as easy to follow as a road even in poor weather, or in the dark, with only a torch. Most paths that aren’t visible on the ground are not, no matter how well they are marked.
@SomeoneElse don’t you think this is better than the current version? If the current page was excellent, then surely we wouldn’t be having this long discussion?
Also: even if no changes are made to the page in the end, I have learned a lot from this discussion about how people around the world think about this tag, whether they have far more experience in OSM than me or less
The more text you add in a definition, the more it is prone to bad interpretation.
In particular it does not help saying “It is not about …” for concision.
How about:
I’m not even sure we should add the last part. To me, those ‘Controversy’ sections should probably be deleted.
Sure, the stuff about route visibility can go. (Then it should also be removed from the infobox)
I intentionally didn’t include markers in the first paragraph because of @Hungerburg’s objection that it could encourage people to map an otherwise invisible path as “excellent” just because it’s well marked.
On two changes in wording in value section:
Both make sense, see my hiking path picture, but will the illustrating pictures still match the words? How about all the previous mappings on the old terms?
For me, all it would take is make the page no longer mention markers. If the markers work, people will see the path/trail See my second picture, the blazes make a horrible out of a no.
I second @yvecai - more words make matters worse: Why mention mapped paths? We are mapping! Why mention which way to follow at a crossing? I hope nobody ever understood the key like that!
Edit to add: The combination of no trail_visibility and marking other than no should continue to make sense. It doesnt do that, if trail visibility is already a mix of both.
I think given the massive differences in opinion on what the original wiki page meant, including the fact that the current consensus is the opposite of it’s original intent is an indication that clarification is needed.
Yup.
I agree on principle, but since it (according to consensus here as well as data analysis) is being used as route/path_visibility it is what it is. I think bowing to what this is, and then proposing a trail_surface_visbility might be the most practical path forward.
The existing documentation already allows it for excellent and good, so that ship has sailed. Not having it below there is a strange inconsistency.
This seems reasonable to me, and the “mapped trail” I think helps clarify the semantic disconnect people (like me) have between a trail and formal OSM highway=path.
That unfortunately would make breaking changes with how it is currently written / being used.
“mapped paths” is to clarify that trail_visibility is not actually trail_visibility, but the visibility of the overall path shown on OSM.
The description of excellent currently is “None, orientation unproblematic, no map required”. I like the following point and would change it a bit. “None, orientation unproblematic, no map required. It would still be easy to follow in poor weather or at night with a personal light source.” (gets around torch, headlamp, flashlight).
Good makes less sense to me. If a trail is a “Continuous established path, always visible” then why is a basic sense of direction required? “Path always visible, but sometimes has to be searched for” makes more sense to me and is less likely to break current usage. Perhaps “A continuous established path which is always visible but occasionally has to be searched for. Could be difficult but possible to follow at night or in poor weather.”
Intermediate is a sentence fragment that doesn’t make sense:
“Path mostly but not always obvious/visible” mostly what?
Intermediate should probably be fleshed out to "The path is often not obviously visible, but can be found after searching. In poor weather or at night the path might not be followable and orientation/routefinding would be required.
Bad could be "The path is often not obviously visible and occasional stretches of the mapped tail may be completely pathless.
I am curious as to what the different between being “mostly visible” and “sometimes invisible” in intermediate and bad. It seems like they’re describing the same thing - something which is mostly visible is sometimes not visible. Adding in some sense of frequency seems to keep with the original intent and wording while making things clearer.
Skills in orientation vs orientation skills should also just be normalized to one or the other.
[...] but also human-added clues such as trail blazes, poles and cairns.
[...]
The key is not about route visibility (how easy it is to choose the right trail at a trail junction) and quality of trail marking. See [trailblazed:visibility](https://wiki.openstreetmap.org/wiki/Key:trailblazed:visibility)=* for attributes regarding trail marking quality.
Huh?? the key isn’t about markers, like trail blazes, poles, and cairns, but is about clues like trail blazes, poles and cairns?
The usage “route visibility” is still weird to me. Even in Europe “route” is used to refer to low/no visibility paths, see the Austrian system, or the OMS wiki page describing climbing routes.
www.austria. info/en/active-outdoors/summer/walking-hiking/signposting-difficulty-levels-hiking-climbing
wiki.openstreetmap. org/wiki/Climbing#Climbing_Routes
(I can’t post more than 3 links, and can’t post another reply, so remove spaces above)
“Alpine routes:
These are not hiking trails – they lead into alpine or high alpine terrain, and can contain sections that jut out, pose a risk of falling, or are unsecured.”
Using the word route doesn’t automatically refer to the relationship of paths.
Routing, or route relationships, or navigation between paths would be clearer IMO.
The different between ground and landscape for path visibility also seem a little unclear to me. A level or slightly sunken surface I would just think of as the ground.
Something like:
The key trail_visibility=* is used as part of a classification scheme for hiking trails and paths. The key describes how easy it is to follow a mapped trail using all available visual aspects of a path. These aspects include the visibility of the path on the ground (differences in surface of the trail vs. the surface next to it, differences in vegetation, a leveled or concave path surface, a treeless corridor through a forest, etc.) as well as human-added markers such as trail blazes, poles and cairns.
It is based on the classification of the Swiss Alpine Club (SAC).
The key is not about routing visibility (how easy it is to choose the right trail at a trail junction), as that is covered by guideposts.
If a path does have markers they should be indicated in via trailblazed. See trailblazed:visibility=* for specific attributes regarding trail marking quality.
Base on your description, there are only markers. The fact that there are no trails means that including a trail related tags is just confusing. This is more a matter of orienting between marked way points.
I’m not sure what you’re referring to, but trail_visibility (not as originally intended) is being used for paths that consist solely of markers with no trail surface. This has been explained a few times in this thread.
I personally wouldn’t consider routes that don’t have any trail surface and consist only trailblazed=horrible trails or paths, but there are many formally maintained and recognized trails that consist only of markers that are well spaced and easily visible.
Is it worth having a cut-off for a “path/trail” that consists of only sporadic markers (no and horrible, possibly bad)? I feel like an abandoned trail with some come and go surface trail is more legitimately a path than a cairn every 30-100 meters, or ones that are informally and inconsistently made. This might actually have been the original intent as markers were only mentioned on good and excellent trail_visibility. Anyone can stack up a few rocks on a pathless route and just call it a trail.
I think is reasonable to reduce trail or marker quality based on a combination of quanity and quality. As for a lower threshold, that may be a slippery slope. Pun aside, it is hard to set a clear cut-off where a trail becomes useless.
trail_visibility=no indicates a mostly pathless trail.
There’s no guidance on what that means - if there’s three informal cairns over a kilometer, does that constitute a trail? If it does then anyone can create a “trail”.
In my mind there should at least be some significant stretches of a visible path - coming over Red Pass in the Sierra Nevada there’s an old trail from a historical Sierra Club pack group + recreational users over the years that is clearly visible in places and can be followed for tens of meters before it fades then appears again, there’s some visible rock work on switchbacks. That’s not currently marked as a trail in OSM (and I don’t think there’s a lot of utility in doing so as it’s easily navigable terrain for anyone that got that far) but I could see just mapping those areas where the path is visible, then having it start, then stop again as matching ground truth. Creating an arbitrary path that leads all the way to the pass down to Marion Lake seems less palatable.
note: this isn’t visible by satellite, and I wasn’t recording a GPX track and starting / stopping it (and I’m not sure that I would have found all existing paths) so can’t add it in.
