Difference between access=designated and access=yes

[Peter_Elderson]

I have just walked the Hadrian Wall Path. Near Newcastle-Upon-Tyne, along the river Tyne, a long stretch is a (signed) combined foot/cycle path without segregation and without preference for cycling or walking. The width varies from 1 meter to maybe 2 or more. No lineage. Some stretches could be, say, unclassified in a different context. Others are narrower width overgrowth at the sides, typical path.
So I would tag the whole path as highway=path plus bicycle=designated plus foot=designated plus segregated=no. Because it’s neither a footway nor a cycleway, it’s a path for both.

A few stretches have an explicit warning that pedestrians have priority. I think these could be tagged as footway with bicycle=yes.

PS I have photos but I can’t upload them until I get home, day after tomorrow

[SomeoneElse (Andy Townsend)]

@Peter_Elderson

So I would tag the whole path as highway=path plus bicycle=designated plus foot=designated plus segregated=no. Because it’s neither a footway nor a cycleway, it’s a path for both.

The challenge there is how will you record the legal right of access there? Depending on which bit you were on it’s likely to have been a public bridleway or public footpath. It’s important to record that information so that people know where they have a legal right to go (as opposed to just being tolerated by the landowner). Legel access rights would normally go in tags such as foot, bicycle or horse, but it’s not possible to do that if someone wants to use bicycle=designated to say that “this path actually has the appearance of a cycleway”.

[Peter_Elderson]

@SomeoneElse

The challenge there is how will you record the legal right of access there? Depending on which bit you were on it’s likely to have been a public bridleway or public footpath.

The legal right of access on this ongoing shared path section is foot=designated and bicycle=designated. The appearance is: path. Asphalted mostly, but no lineage, no kerbs, no drainage system, no shoulder.
As for PROW, most of the Hadrian wall path uses public footpath, and some short stretches used a public bridleway, but this particular stretch is along the Tyne in the Newcastle urban area. Going fron west to east, it starts with a former railroad track, converted to this shared path for cycling and walking, going on and on through Newcastle. I’m pretty sure it’s not marked as a public anything in Newcastle.

[Minh_Nguyen (Minh Nguyễn)]

I suspect you and @SomeoneElse are debating different understandings of designated. There are some mappers who apparently consider the presence of a pictograph on a sign to be a kind of “more than legally allowed”, yet not compulsory in the sense of arbitrarily compelling someone to go on a bike ride while they’re ill and bedridden.

I’ve heard various vague explanations of this phenomenon. It sounds like some countries decide between yes and designated based on who must give way to whom, as implied by signs. If access keys are about legal access, then why are these keys being overloaded with rules about how to use the way on a vehicle that’s already allowed to access the way? If cyclists are required to announce when they pass pedestrians on a given trail, should we indicate that in the same key as well?

Moreover, we cannot necessarily assume that giving way has so much to do with preeminence on a path. In my experience at least, cyclists and equestrians are always required to yield to pedestrians on shared use paths for safety reasons. For that matter, cars are required to yield to any pedestrian in a roadway, even if they’re illegally jaywalking. But turning highway=primary into highway=path foot=designated motor_vehicle=yes would be considered vandalism.

This priority-based designation definition is also counterintuitive to me as an American, since park signs here often literally say “Designated Bicycle Path” and the like, but it’s functionally no different than yes, unqualified. (The absence of such a sign can also mean yes, depending on context.) Instead, we tend to use designated to mean “not only allowed but also designed for”, or “not only allowed but also recommended”. That recommendation can come in the form of an explicit sign or some other indication, such as a name.

When one of these recommendations applies to a sidewalk, the sidewalk doesn’t automatically become a non-footway and a non-sidewalk. highway=footway pairs nicely with bicycle=designated.

Recommendations don’t have anything to do with the identity or legal access of a way, so why do we use designated in this manner? Because the tag exists, seemingly extraneously; it’s a slot that must be filled, rather arbitrarily.

Which creates a bit of a conflict if the path is anything but public. The trails in this nature preserve are closed to the public but still signposted for use by pedestrians who work at the preserve or study in certain academic programs at the university that owns it. These pedestrians have priority over – well, it’s only for pedestrians. Should it not be foot=designated and foot=private at the same time?

[Nadjita]

Thank you for summing it up so nicely, @Minh_Nguyen. I’ve been preaching for a long time that some people are reading too much into “designated” that should better go into separate tags. The original proposal oh highway=path pretty much said: yes means: not forbidden, maybe allowed by law, designated means: explicitly allowed by a sign or similar.

But giving that so many people read “priority” and “compulsory”, or “especially made for” into designated, we should really come up with tags for these properties.

[Msiipola]

@SomeoneElse

so that people know where they have a legal right to go (as opposed to just being tolerated by the landowner

As already said several times in this thread, such rules depends much which country we talk about. In the Nordic countries (at least in Sweden and Finland), we can walk almost everywhere. In most cases landowners can’t stop this. Private signs with “pedestrians/cyclist not allowed” can be ignored. (Of course fenced industrial and military areas, and signs set up by the authorities are an exception to this).

[ezekielf (Zeke Farwell)]

@Minh_Nguyen

The trails in this nature preserve are closed to the public but still signposted for use by pedestrians who work at the preserve or study in certain academic programs at the university that owns it. These pedestrians have priority over – well, it’s only for pedestrians. Should it not be foot=designated and foot=private at the same time?

I explored this issue a couple years ago but people mostly told me there was no problem or that I could just ignore *=designated . I see a need for new tags indicating what a way is designed or intended for separately from what the access rules are.

Access tagging ambiguity

I’ve been trying to gain a better understanding of access tagging and I’ve realized there are some tag combinations of the designated, private, customers, and permit values where the meaning is not clear to me. How do you interpret the meaning of the following three tag combinations? Example 1: A designated snowmobile trail where a permit is required. Since snowmobile=designated + snowmobile=permit is not a possible tag combination, access=permit + snowmobile=designated seems reasonable. How…

[Nadjita]

@Minh_Nguyen

Shared spaces defy neat categorization into primary feature types

Maybe it’s finally time for highway=footway;cycleway Oh wait, maybe even highway:footway=yes + highway:cycleway=yes? I’ll see myself out …

[aighes (Henning)]

A long time ago someone came up with highway=path

[Hungerburg (Asa Hundert)]

@Minh_Nguyen

I hope one of the proposals for iterative refinement with path=* will help to close that gap.

I repeat myself: path=* will inevitably end up a semicolon separated list.

[Minh_Nguyen (Minh Nguyễn)]

If so, then truly we don’t know what the feature “is” more precisely than a generic path.

[Peter_Elderson]

@SomeoneElse

I’m guessing they mean that you saw “there is some sort of sign that says that it is currently OK to walk or cycle there”

Ok to cycle there is bicycle=yes. If the way is explicitly meant for cycling (assigned to cycling) we tag bicycle=designated. If that is explicitly the main use, we tag highway=cycleway, which has the implicit tag bicycle=designated.
If walking is allowed on the cycleway (pedestrians are guests on the cycleway), we add foot=yes (which I feel should be the world wide default, but currently the default is foot=no).
This combination can indeed be interpreted as preference for cyclists.

Ok to walk there is foot=yes. If the way is explicitly meant for walking (assigned to walking) we tag foot=designated. If that is explicitly the main use, we tag highway=footway, which has the implicit tag foot=designated.
If cycling is allowed on the footway (bicycle is a guest on the footway), we add bicycle=yes (world wide default currently is bicycle=no, which I feel is correct).
This combination can indeed be interpreted as preference for pedestrians.

If a way is explicitly meant for both walking and cycling, without preference, the highway tag is neither footway nor cycleway. Some communities then simply assign a preference, tagging it as a footway with bicycle=yes, or a cycleway with foot=yes, usually based on the customs or legal rules in their country. I think that is a correct way of translating the effect of the local rules to OSM-wide tags.

But it is also correct NOT to assign a preference if there is none, by tagging highway=path|trail and then explicit foot=designated and bicycle=designated, as found on the legal traffic signs for combined foot/bicycle path.

Data users can handle the tags in a universal way, because the local specifics are expressed in OSM-wide access tags, combined with world-wide access defaults. (I know world wide access defaults are not exactly applied, but general purposerenderers and routers will usually come to the same conclusions).

One thing, some people require that there is a legal sign, in order to set the tag …=designated. I prefer to require that the designation is explicit, visible on the spot, so that walkers and cyclists can see what they are meant to do).
Logo’s on the road, kerbs, special pavings, and on private land signs put up by the owner, are al just as explicit and can amount to …=designated. The tag should let the user know what is going on.

Last thought, for now: the famous English public right of way. Where PROW materializes in public footpaths, I think highway=footway is appropriarte and sufficient on most sections (which implies foot=designated). Where the public footpath uses a different kind of highway (e.g. a public bridleway or track or cycleway or service road), simply add the tag foot=designated to this way.
Where a public footpath crosses a field, to my experience it almost always is very clear where you are supposed to walk, from each gate/stile/steps/ladder to the next. On the Hadrian Wall walk, we even were admonished time and again to NOT walk single file, but find a different trail, so as to prevent erosion of the grassy hills. Still, usually a clear trail was visible. Probably, the first walkers in the early spring will create a fresh trail from ladder to ladder along the field or within the corridor, but in all practicality these are clearly visible paths most of the year, and when not, still clear where the path will be once the first say 3-5 walkers have passed

[alan_gr (Alan)]

@Peter_Elderson

Logo’s on the road, kerbs, special pavings, and on private land signs put up by the owner, are al just as explicit and can amount to …=designated. The tag should let the user know what is going on.

So how does the tag let the user know what is going on if a path on private land is intended for pedestrians? Consider a hotel with large grounds with a mixture of paths signed for different uses. Access might be “permissive” if they generally let anyone roam the grounds, or “customer” if only guests can do do. How do you express that if the foot= tag has been set to designated?

@Peter_Elderson

Where PROW materializes in public footpaths, I think highway=footway is appropriarte and sufficient

The how can users tell a PROW apart from any other footway? A path inside a theme park, for example?

[Peter_Elderson]

@Richard

bikes are allowed on public bridleways, but bike racing isn’t!)

The ones I have encountered were not fit for racing at all.
I would tag horse=designated, bicycle=yes, foot=yes, if it’s applied to the paths, and add a route relation if it’s a route in the sense of a string of ways.

If one route (e.g. to the next village) is for three different modes of travel, we have another as yet unsolved issue: routes with multiple designations. As it stands, a string of values for route=… is not supported by data users, so separate route relations per mode of travel is currently the only working solution.

[Richard (Richard Fairhurst)]

@Peter_Elderson

As it stands, a string of values for route=… is not supported by data users

Well, some data users

local route  = relation:get_value_by_key("route")
-- [...do some stuff here...]
elseif not route:find('bicycle') then return end

[Peter_Elderson]

@Nadjita

Not forbidden → yes. Allowed → designated.

What is the difference between “Not forbidden” and “Allowed”? In my book, that is the same. Designated adds that the path is especially meant for/assigned to/committed to something.
Foot=yes means walking is allowed on the way; foot=designated means that the way is visibly/verifiably meant for pedestrians.

[Nadjita]

@Peter_Elderson

What is the difference between “Not forbidden” and “Allowed”? In my book, that is the same.

We’re reading different books it seems. Not forbidden means: there is no sign/icon allowing or forbidding the use of a transportation mode. The examples given in the discussion of highway=path include “if signed cycleways are always usable by pedestrians, but there is no sign saying “pedestrians allowed”, then it would be bicycle=designated + foot=yes”. This tells data consumers that there will be no sign on the ground that allows pedestrians, but rather a law granting them the right to walk there, or simply the absence of a law forbidding them to go there.

But I know that a lot of people read much more into designated than “there is a sign or symbol somewhere that explicitly allows it”

[Peter_Elderson]

@Nadjita

bicycle=designated + foot=yes”. This tells data consumers that there will be no sign on the ground that allows pedestrians, but rather a law granting them the right to walk there, or simply the absence of a law forbidding them to go there.

I don’t agree. With access tags, we do not map laws or signs, but what they mean for the people on this road or this path. Not forbidden is allowed, whether or not there is a sign. If local or national rules are in place, we tag what they mean for the people on the road.
If no rule applies and no sign forbids access, it is allowed: foot=yes. If a rule applies so walking is allowed where in other countries it would not be allowed: foot=yes. If a sign authorizes walkers to go there: foot=yes. If a traffic sign (or other visible signs) says it’s a footway: highway=footway implying foot=designated: allowed for AND specifically assigned to foot traffic.

[Nadjita]

@Peter_Elderson

I don’t agree

It’s not about agreeing or not. I was quoting from the proposal of highway=path, which is also where access=designated comes from.

@Peter_Elderson

specifically assigned to foot traffic

Now what does “specifically assigned” mean?

[Peter_Elderson]

@Nadjita

Now what does “specifically assigned” mean?

The powers that be have decided the way is specifically meant for walkers, and you can see that on the road, usually because there is a traffic sign saying so, or other visible and verifiable characteristics such as a painted logo, kerb and paving. If there is no sign saying that it’s a footway, and no special characteristics, then it’s not a footway, but pedestrians can still be allowed to use the way. Then it’s foot=yes, not designated.