Tag trail_visibility: Proposed Improvements for this Descriptive Tag

[erutan]

@rhhs

I hope that we are all have a common aim: to create the world’s best quality map. One of the aspects of a quality map is its reliability, and for this we need to map consistently.

This seems reasonable to me. As it is the tag is pretty much technical debt. There’s a few ways around this:

1. Keep it as is - this means no work up front, but continual bad data forever.
2. Make some minor changes to help clarify it - this may/will cause some things to “break” in the short term but will cause long term data to be more (if obviously not 100%) consistent. Given how broken usage is now, this seems like less of an issue than in some cases.
3. Make more drastic structural changes if deemed necessary, which causes more short term pain but hopefully even less long term unreliability.

It seems the community is split between 1 and 2.

@rhhs

I will make some edits to my Talk page proposal to reflect what I’m stating here. Hope we have consensus soon, before reaching 200 posts in this thread

Unlikely IMO, I think this is going to have to go to vote vs just signing off informally here. There seems to be too many major directional differences to have minor wording changes sway them.

@rhhs

This thread has been mostly about improving the text of the wiki, because I think it needs improvement. So far we haven’t discussed the pictures much, which I think are OK (ain’t broke). If you think they need improvement, we could start another thread about this.

I feel like fixing the text (or not) needs to come before choosing pictures. The pictures are meant to represent what the text explains, as along as that is in flux then there’s not much point in arguing on the bright lines between values.

@Adamant1

OK cool. That sort of makes sense. But then take the picture, if it’s of a grassy, semi-legible path. I assume we would already know it’s a path because it’s being mapped as a way in the first place and you can tag the existence of grass along the route with surface=grass. So what exactly does adding trail_visibility=good add to the equation that’s not already being mapped with highway=path + surface=grass?

Maybe you could argue there’s grass “paths” that “could” be tagged as trail_visibility=no, but then I’d argue if they aren’t visible they wouldn’t and/or shouldn’t be mapped as a path in the first place. And no I don’t want to get into a meta side discussion about the meaning of the word “path.” The fact is that very rarely, if at all are people mapping invisible (one might even say imaginary) grass paths or if they are other people will delete quickly them.

That’s one example of a path - you can have a dirt path in a forest which is unambiguous, and one which becomes faint at times, they would both have the same surface. Not all paths of the same surface type have the same visibility, how often they are used and how heavily they were constructed (if at all) etc play a large role.

This would be an example of a good visibility (and not on the higher side of it), it’s easily followable but not as unambiguous as a clearly maintained trail and if you weren’t paying attention you could step off of it onto a deer trail etc. It’s definitely not a casual tourist trail, but it’s used enough after being abandoned by the land agency to still be followable without having to routefind.

Unless there’s a surface value which would be “dirt with some duff on top of it and some small plants growing” I’m not sure how you would tag its visibility.

The following, despite being narrow and with vegetation on the side I would consider excellent, the first photo was the least visible it was and someone is not going to confuse the vegetative hillside with it. I could see someone tagging it good.

@Adamant1

What surface a trail is might be one of, if not the main, way a hiker goes about determining how difficult the path will be.

That’s a new take to me.

I’ve never looked at the surface value of a trail and thought it would be the main indication of how difficult it would be.

I guess you could add a whole bunch of keys and break sections of the trail into different segments whenever the surface changes, but that doesn’t seem like a good idea. You’d have to get extremely specific (are there some ankle high rocks solid in dirt, knee high rocks that are loose, waist high rocks with gaps between them, rocks that slide, etc) with trail surfaces in order to show how “difficult” something would be, and that still doesn’t really address how easy it is to follow the path.

Surface=sand could mean that it’s easy to walk on (compacted like in Joshua Tree) or incredibly tiring and difficult (the fine sand along the old jeep road along salt creek in Needles). I guess you could have surface=deep_sand. A sandy trail could have a nice well worn path in it, or be faint and hard to follow because of wind blowing over it and being only lightly used.

SAC_Scale already exists for “difficulty”, and combined with surface it tells you something, but that’s separate from visibility.

@Adamant1

Walking on lava rock is completely different than walking on shale or sediment. Heck, have you ever heard of quick sand? How exactly would the trail_visibility tag help at all in that case?

I don’t think trail_visibility tries to solve for all safety issues, including quicksand. If a trail has pools of quicksand in it that’s separate of how easy is it to follow.

@Adamant1

Like I said, if there’s a log in the middle of the path, cool. Map it. There’s no reason to overcomplicate that with something like trail_visibility though. If anything it just makes things more dangerous because it disincentivizes mapping the actual hazards.

Fallen logs often get cleared by maintenance crew on official maintained trails, mapping every one seems like a somewhat ridiculous task.

@Adamant1

There’s zero evidence that anyone’s life has been saved or made safer by tagging paths with trail_visibility.

I’ve come across “trails” in mapping applications that didn’t exist in any meaningful sense, despite having the same surface as would a path that actually exists. Having some level of trail visibility allows for paths that are harder to follow to be weighted fainter on maps, which is useful IMO. Even if it can’t be proven to save a life, being able to see that a path is likely to be come and go etc is useful vs having everything weighted like it’s a well maintained path. This can’t be done solely by surface type, informal/formal, etc.