At this point we can have a highway=path with trail_visibility=excellent that has no visible path/trail at all because there are bright tall markers at regular short intervals. I think in that case, pointing out that there is no actual “trail” surface is worthwhile.
I see a few different approaches to this:
A consensus that trailblazed:visibility (the visibility of markers along a path) significantly more important than the visibility of the surface of the trail itself, at which point this is dropped.
The lack of above consensus, at which point this should be adopted given that it passed before, but due to unclear wording has changed over time.
If only the overall visibility of the path matters (which it seems that trail_visibility is now poised to formally become) there should not be a trail_surface_visibility and trailblazed:visibility should be deprecated. The general trailblazed key still provides some useful metadata and seems separate of this discussion.
Don’t create a whole new tag just to extend the current trail_visibility with surface information. I would just add a “surface” postfix. Especially since it is a clarification of the trail_visibility value.
There’s a surface key already which could be confusing, but a ‘trail_visibility:surface=foo’ would work and be somewhat inline with ‘trailblazed:visibility’.
The latter is information relating to ‘trail_visibility’ as well, so it’d seem like ‘trail_visibility:marker=foo’ or ‘trail_visibility:trailblaze=foo’ would be more consistent, but I guess that ship has sailed.
What would you do with the information trailblazed:visibility since it is already included in trail_visibility? Supporting option 3 is a valid stance.
Poor weather can reduce visibility of markers (I’ve gone over a pass in Chile during a blizzard that consisted of somewhat widely spaced markers), though obviously snow can also cover a trail surface. There are numerous examples in the trail_visbility thread.
Marking trails that alternate between markers only and having clearly followable surfaces be clearer would be another potential use case - a trail could have intermediate marker and surface visibility but an overall excellent visibility rating if the lower elevation 80% of it was only surface trail, and the upper 20% was marker only. The trail could also be broken into segments, where one has an excellent surface visibility, and another has a good marker visibility - you can see examples of that in a comment here RfC: Have pathless hiking routes (no trail_visibility) be a type of node_network instead of paths - #10 by erutan
Weather should have little effect on either trail or marker. Even then it should only should be about how it factors into the overall value. Such as regular snow fall hiding a trail or winds that turn it into gully.
In this case, there is no path visible on the surface because it is a path only used in winter when there is snow cover so no surface wear takes place. In summer, without snow cover, hikers use the summer path on the right because it’s easier to walk on. Instead of tagging the winter path with trail_surface_visibility, I’d tag it with a slightly higher hiking_difficulty value (although the difference in difficulty is not so that it can be described as sac_scale=mountain_hiking) because the difference between the summer and winter path, in summer, is essentially one of hiking difficulty.
That we should have trail_surface_visibility because we have trailblazed:visibility is not a good argument: each can exist without the other, and each should be judged by its usefulness separately.
So what’s the usefulness for having trailblazed:visibility in and of itself?
It’s a useful mechanism for thinking about visibility instead of ideal conditions. My example of going over a pass in a blizzard is useful - the markers were spaced out enough that with reduced visibility it was hard to spot them at times, vs on a clear sunny day where they would have been sufficient. It’s probably most useful for designating “excellent”, as when things are intermediate they’re already degraded. Something that’s excellent should be unambiguous even with weather that reduces visbility (fog, snowfall, etc). Perhaps that could be dropped for good.
This is hard to achieve: On a ski-tour I happened to get into bad weather. My companions wanted to go down the uphill way. After we parted and met again, we took that. The poles on an, in summer 3m wide compacted track were spaced so, that one could just not be seen from the other. The only thing that brought us home was, that I had walked this track so often, that I knew every turn by heart.
Winter can change a lot: Where in Summer you walk through a cut-line between 3m high brushes, in Winter on a bright sunny day you walk on a plane of snow, where tiny sticks of heath show.
Of course it has !?? What about a foggy moonless night?
Weather should have little effect on either trail or marker mapping.
If this discussion aim to go somewhere, one have to pay attention to keep on topic and with reasonable assumptions, and not to loose oneself in niche counter examples.
I don’t think we need to write up weather for every state of visibility, but it’s something I took into account for the [overall/combined path/trail visibility here](https://Unambiguous well-constructed path everywhere. Easily followable when visibility is reduced.):
Unambiguous well-constructed path everywhere. Easily followable when visibility is reduced.