Wiki edit footway vs path

Probably both. How many of you knew that highway=path are used for golf cart paths, and also for narrow urban ways in Asia ? And I am sure there exist more useful use-cases we are not aware of as @Minh_Nguyen just showed.

I am all in favor of making ‘highway=path’ an outdoor trail only tag, which seems what a large part of the community is pushing for, but we will need help in finding or creating better tagging scheme for all other highway=path use-cases.

Hiding is a terrible idea in my opinion, at least for this case. When searching for cycling in Austria for example, I find a preset for “Cycle & Foot Path” (tagged as highway=cycleway + bicycle=desiganted + foot=designated), a few meters further across the border in Germany on the other hand I don’t get any preset at all for this kind of way and the closest fit would be “Cycle Path” (highway=cycleway without any further tags), which is arguably worse.
And funny enough “Cycle & Foot Crossing” (tagged as highway=cycleway + bicycle=desiganted + foot=designated + cycleway=crossing) isn’t hidden - as a new user, I would consider the missing preset just as a bug.

Well, and there you have so many e-bikes which are motorized vehicles in the linguistic sense :-D. The motorcycles were suggested by somebody, i do not have stron opinion on them, so feel free to edit :-).

We definitely need a mltipurpose urban built up way - I sugested two_wheel.

1 Like

Good point, the “motorized” distinction may need to be changed, because in the legal sense, e-bikes are not motorcycles, and e-motorcycles are not bicycles.

For me the proposal looks like you are trying to add an implicit informal=yes for highway=path - Which is currently an optional to add.

We had this this discussion about a hundret times in Germany and consensus was that highway=path is explicitly not equivalent for a german “Trampelpfad” e.g. an informal footway.

Flo

2 Likes

Then I think Germany should be added as an exception along with the UK. Some communities, it seems, enforce the distinction I made explicit. Some communities do not. This should be explicit knowledge documented on the wiki (. However, in some countries like the UK or others, this distinction does not hold and highway=footway can be used for these too."). How else is a novice mapper supposed to know?

PS.: If the distinction does not hold at all, pictures in the wiki need to be changed as they all confirm the distinction.

There may be some overlap but I don’t think this is the same thing. My understanding is that a trail with any kind of official status and/or identifiable operator and/or waymarking/trailblazing would not usually be tagged as informal=yes. “Informal=yes” is more for “social trails” in the terminology of the US Trails Project. An official trail might be rough, difficult, and rarely maintained - perhaps it would only be repaired if it is blocked in some way. In that case it would certainly be a path under this proposal, but it wouldn’t be informal=yes.

8 Likes

Looking at the numbers (in km) of ways where both foot and bicycle designation spelt out in attributes, in Austria path and cycleway quite even, ohsome - dashboard - cycleway 2055 km, path 1654 km.

I was surprised: If I allow for designated or yes ohsome - dashboard, track takes the crown: 5281 km. Beware, only fully specced ways counted. Does that mean, in Austria track is the top shared-use infrastructure?

Obviously, not a lot mappers see the use in tagging foot or bicycle access on service, residential, unclassified, etc. roads.

PS: On shared use: most of the streets here – residential, unclassified, tertiary, etc. are shared use. Most only Motorways are not. Why are tracks so prominent in this data: 1) On tracks, a bit like on paths, access is not easy to guess, so it makes sense to record it. 2) Tracks alone make for half of all the “highways” combined here.

UPDATE: Following the example of the second-top-poster I made an edit to the documentation Tag:highway=path: Difference between revisions - OpenStreetMap Wiki - It would be disheartening if all that talk here would be hot air. Improvements welcome!

The wiki page now says: Mostly, it just states, the surrounding area can be passed there, by one or more indeterminate means. All the essence is in the attributes.

IMO the first part is the essence, while the attributes contain the details.

In a Platonic sense, yes, the idea is the essence, the ground layer building block, the how and who is just the worldly stuff.

1 Like

Will I need to be a doctor of philosphy to contribute further to this thread?

:slight_smile:

5 Likes

Having a degree in philosophy, continuing the argument: In the tuple highway=path;foot=designated it is not far fetched to call foot=designated the primary tag.

5 Likes

And in a practical sense, too. The essence of a path is to enable movement. This translates to OSM as a way, showing on maps, and weighed in by routers. Whether or not the path fits in a category or has wings and spits fire, that’s interesting, but the details should not obscure or be considered more important than the essence of a path.

In many discussions, there is endless squabbling over certain details, often in combintion with examples and country or culture specific variants. Instead, look at the essence and then decide how to enable universal (oh well, OSM-wide i.e. world wide) mapping and tagging of these differences and variants.

In short, even if it sounds ridiculously philosophic, I would like our wiki not see the details as the essence. Whch is not to say that details aren’t important; they are just not the essence.

Ah, the eternal question: footway or path—a debate as timeless as footfalls upon the earth. Is it not a reflection of our ceaseless desire to categorize and define, even the most subtle distinctions in our shared map of existence? If foot=designated is the primary tag, does that make the path itself subordinate? Or are we merely assigning labels to the ineffable, forever trying to pin down meaning where perhaps none was meant to be? Such is the journey of both mapper and philosopher.

But, dear mappers, what is a footway if not the shadow of a path? We tread on these digital landscapes, drawing lines between concepts that may not truly be divisible. Can a footway ever be fully disentangled from the broader path? Perhaps not. Perhaps the boundary exists only in our minds. And so, we find ourselves mapping not the Earth beneath our feet, but the labyrinth of our own thoughts. As above, so below—our maps are mirrors of our minds, eternally questing for order in chaos.

Or maybe it’s just semantics.

10 Likes

Hello @Peter_Elderson – Your Wiki edit is fine with me, for me the details are the essence - in a practical sense.

I was expecting different complaints. I found it necessary to get into history, something that is not common on tag description articles. But otherwise, how to explain the vagueness? And otherwise, how to tell the urgency for mappers to add attributes and users to consume attributes?

One thing though I observe when looking at newly mapped bare highway=path. They are mostly ephemeral. Maybe the mappers think, this is just a path with no significance at all, why spend attributes on it?

I think you messing up terms. It is key highway that is making something routable according to openstreetmap documentation. That path gets used by the routers of your choice is not an intrinsic property of path. It is just a choice made by the developers or your router.

I don’t think it needs explanation. Showing the range tells the tale. Also, I don’t think it’s vague, it’s just not very precise. That’s why details are needed, possibly subcategories, and we’re discussion how to record the information consistently.

In the meantime, and long thereafter, - can’t be stressed enough - data users must deal with unspecified highway=path’s. Implementation of any solution should incorporate a quick fix for that, optional but clearly advantageous to maps, routers/navigators and end users, and a virtual path to the point on the horizon where the Holy Grail awaits, filled with Pop Corn.

Hopefully I am not bikeshedding (fellow philosophy graduate here, though I have always preferred Wittgenstein to any metaphysicist :-D):

  • I removed “primarily” from is not designated for 2-track vehicles. I think that if there is an agreement on something, it is that path is not to be used by cars. The German cycleways+footpaths mapped as paths physically allow cars, but it is illegal as far as I understand unless you drive a police car. Therefore I think “primarily” just muddied the definition. Plus without it it is shorter, yay?

  • I reworded the essence thingy. The modality was wrong I think - there is a consensus that only paths actually used should be mapped (something that could be described as highway=pathless should only make it into OSM ifpeople actually use it - in theory people can go pretty much anywhere) .

  • Added some links, fixed some grammar (hopefully :-D).

  • Made it all much more concrete and shorter. Also replaced “attributes” with “additional tag”. I am not sure if “attribute” has a precise meaning when it comes to OSM, “additional tags” has. If you think this more concrete version is not covering some use of path, please build further upon it.

Also, in the examples, the last section is confusing:

  1. For the first photo, there is a footnote “Many mappers consider highway=footway/highway=cycleway (+ access tags as necessary) more appropriate for well constructed ‘urban style’ ways like the one depicted in this example.” I suggest to move it into the main text. Footnotes are difficult to follow (especially if you only see the footnote, the backtracking to what it points to is difficult).

  2. Then there is a text:

Note that adding only bicycle=yes will not tell data users the difference between the paved vs the unpaved paths and adding surface=unpaved without smoothness=* will not tell the difference between the third path that will be acceptable for a much larger portion of cyclists than the second path.

I do not get that at all. The second picture issupposedly this:

The third is this:

The third has no path whatseover. How can it be more acceptable that the second picture, where there is at least some path (and I can imagine riding it on my urban bike very slowly and hating it, unlike the third, where it would be nigh impossible)? Are the references wrong, or are they meant for different pictures? I am confused. Does anybody know?

1 Like

I checked the Wiki page history, the third picture used to be a different one, it showed a relatively smooth unpaved path. It was then changed by @Hungerburg who probably didn’t realise that it was referenced in the text.

https://wiki.openstreetmap.org/wiki/Special:MobileDiff/2615582

1 Like