Proposal talk:One-way for pedestrians

Consistency

I do not believe this would be consistent because oneway is defined as applying to vehicles only, a subtag that inverts this would be considered troll tagging. “formalizing” it would not prevent the contradiction. —Dieterdreist (talk) 09:24, 1 September 2024 (UTC)

As mentioned in the forum discussion, that is de-facto not the case and there are many instances where oneway=* is clearly meant to apply to pedestrians, regardless of what the wiki might say. That is the whole point of this proposal.
The 'troll tagging' argument also doesn't make a lot of sense because by that logic `access=no` itself could be considered a troll tag because it 'inverts' the default access restrictions. (Alternatively, if `oneway:foot=*` doesn't change anything about the interpretation of `oneway=*` for vehicles which that tag is supposedly only concerned with, how can it be a troll tag?) Woazboat (talk) 11:59, 1 September 2024 (UTC)

On a purely logical level, I think that Dieterdreist is correct. But as I don't see a better alternative offered (some alternatives are mentioned in the rationale of the proposal, and they each come with they own problems) and since there is clearly an problem to be solved here, I support the proposal. I think this little semantic inconsistency is better than the current abhorrent (and, in some cases, life threatening) ambiguity. Thank you Osmuser63783 for addressing the issue! -- Martianfreeloader (talk) 06:46, 3 September 2024 (UTC)

oneway=reversible / oneway=alternating

The proposal should also consider how oneway=reversible / oneway=alternating can be represented on footways. (e.g. by specifying/clarifying that `oneway:<access_mode>=yes/no/alternating/reversible` is also fine instead of just `yes/no` values)
https://wiki.openstreetmap.org/wiki/Key:oneway#List_of_values
https://community.openstreetmap.org/t/effect-of-oneway-on-pedestrians/7631/81 Woazboat (talk) 11:51, 1 September 2024 (UTC)

Sounds good, I'm happy to add that! Do you have any real-world examples? Osmuser63783 (talk) 17:51, 3 September 2024 (UTC)

Unfortunately not at the moment, no. I mostly mentioned this because the ability to express this would be lost otherwise. Woazboat (talk) 20:18, 3 September 2024 (UTC)

Thanks. I've added this to the proposal. Osmuser63783 (talk) 19:24, 27 October 2024 (UTC)

Why the path exception?

Paths are, by default, usable by bicyclists, so oneway=yes should be a perfectly valid tag on a highway=path that applies to the bicyclists there. It should be discouraged only on "highways" that don't allow vehicles by default:

• highway=footway
• highway=pedestrian
• highway=steps
• highway=corridor
• highway=via_ferrata (but since it's explicitly mentioned there, we can exclude it)

Nadjita (talk) 06:53, 2 September 2024 (UTC)

+1. --Langläufer (talk) 09:33, 2 September 2024 (UTC)

+1. --Martianfreeloader (talk) 06:48, 3 September 2024 (UTC)

I’ve spent some time looking at examples on Overpass. It’s not hard to find a highway=path where the mapper clearly meant pedestrians when they used the oneway tag. These are usually hiking paths (which tend to be mapped highway=path instead of highway=footway, regardless of bicycle access). Examples 1, 2, 3, 4, 5. But these examples are clearly in the minority. They are far outnumbered by things like a sidewalk where the mapper clearly only meant bicycles. There are also a lot of mtb trails mapped as highway=path with oneway=yes where the mapper clearly meant bicycles (and I am unsure if pedestrians are allowed at all). So it might be better to say that on highway=path, oneway=yes is only for vehicles, and presets and validator warnings should warn people to use oneway:foot=* instead when they mean pedestrians. In other words, in this proposal, treat path like cycleway, not like footway. Does that address your concern? Osmuser63783 (talk) 17:04, 3 September 2024 (UTC)

I believe this was implied in Nadjita's statement. --Martianfreeloader (talk) 17:18, 3 September 2024 (UTC)

Exactly. In the proposal, it says, "Mappers will be discouraged from using oneway=yes on highway=footway and highway=path." The "and highway=path" is what I would replace with "highway=pedestrian,
highway=steps, and highway=corridor". Nadjita (talk) 19:28, 3 September 2024 (UTC)

@Nadjita: What exactly did you mean by "it" in "It should be discouraged"? oneway=yes mentioned earlier, or a topic of this proposal (i.e. oneway:foot=yes)? The whole point of the proposal is to make currently potentially ambiguous uses non-abmbigous. If that highway=path in your area is (implied or explicit) bicycle=yes and e.g. bicycles must use only one direction, but pedestrians may use either, than you should always tag at least two oneway-related tags on it to be clear. (E.g. oneway:bicycle=yes + oneway:foot=yes would make it non-ambiguous) --mnalis (talk) 14:23, 6 September 2024 (UTC)

What I mean is: the use of oneway=yes should be discouraged only on highways that allow purely pedestrian traffic by default (footway, pedestrian, steps, corridor). highway=path allows pedestrians and bicycles everywhere, that was the whole point of introducing this highway type. The proposal wants to discourage the use of highway=path + oneway=yes, which I am rejecting, because the meaning of oneway=yes is well-defined, and only causes irritation on pedestrian-only ways, because it would usually not apply there.
On purely pedestrian highways however, it is unclear what people mean with highway=yes, because some used it for marking the way oneway for pedestrians as well. To clarify this, adding oneway:foot=yes or oneway:foot=no should be used on these ways. But only there. Nadjita (talk) 16:19, 6 September 2024 (UTC)

@Nadjita: Firstly, your assumptions on highway=path usages ("allows bicycles everywhere") seem to be wrong. Unless there is explicit bicycle=* on that highway=path, its exact meaning (and whether bicycles or anything other then pedestrians are allowed/possible there) will depend heavily on the region. Even its wiki introduction sections links to this discussion, showing that most such assumptions are incorrect and lead to ambiguity. And the whole purpose of this proposal is making things explicit exactly for the reason of avoiding ambiguity. You claim that "oneway=yes is well-defined" which is obviously untrue, or we wouldn't have that gargantuan discussion on community forum. Just because something is written on the wiki, does not mean people actually use it that way, nor has it ability to change reality, no matter how much we would want such wiki-pixie dust to work! In fact, that forum discussion has shown time and again (with many examples) that the wiki claim is (unfortunately, I agree) provably incorrect. And when reality and wiki claims clash, guess who wins every time? Thus, we need clarification of oneway:foot=yes on those highway=path ways currently tagged with oneway=yes! --mnalis (talk) 00:23, 8 September 2024 (UTC)

> Firstly, your assumptions on highway=path usages ("allows bicycles everywhere") seem to be wrong.
Well, that's how it's defined:
> The default access restriction of highway=path is "open to all non-motorized vehicles, but emergency vehicles are allowed".
And the definition of oneway=yes is that it only applies to vehicles. That's what is well-defined. The fact that people are mis-using oneway=yes on pedestrian-only ways is the problem we're facing – not its usage on highway=path. And to prevent this from happening in the future, the usage of a plain oneway=yes on these highway-types should be discouraged, and editors should warn about this.
But: if in any part of the world, the default access restrictions of a highway=path do not include any vehicles, then in these parts, the usage of a plain oneway=yes should be discouraged as well. Nadjita (talk) 08:11, 8 September 2024 (UTC)

After this discussion, and the discussion in the community forum, I changed the treatment of "path" in the proposal: it now says that on highway=path, oneway=yes applies to bicycles and any others vehicles that may be allowed there, but not to pedestrians. It has the same meaning on highway=pedestrian: it's quite common to have a pedestrian street that's one-way for those vehicles that are allowed to use it (e.g. buses) without being one-way for pedestrians, and mappers often tag this as oneway=yes. As for steps and corridor: personally I would use oneway:foot=yes for consistency. But where plain oneway=* has been used, it's generally been used to mean pedestrians, so it doesn't suffer from the same ambiguity as on footways. Osmuser63783 (talk) 17:53, 27 October 2024 (UTC)

A few theses

• Avoid redefining tags - o.k.
• The meaning of oneway must not change if the highway type or the access tags are changed. (-> also for highway=steps).
• oneway should apply only to vehicles (-> oneway:foot on steps and footways if pedestrians are effected).
• existing correct tagging for vehicles should continue to work. (no need to change oneway to oneway:bicycle or even more strange oneway:vehicle)
• Resolve unclear situations for pedestrians by explicitly tagging oneway:foot=yes/no/*

--Langläufer (talk) 13:17, 4 September 2024 (UTC)

This proposal doesn't deprecate foot:backward=*.

(statement in the proposal). Then in it's the case, foot:backward=no would remain valid. So, the purpose of the proposal is... ? --Nospam2005 (talk) 19:48, 15 September 2024 (UTC)

The proposal is about us deciding as a community how pedestrian one-ways should be tagged, and not using oneway=* any more on highway=footway because we can't agree what it means. As for oneway:foot=* vs. foot:backward=*: as it stands, if the developer of an editor was to include oneway:foot=* in a preset, proponents of foot:backward=* would be annoyed, and vice versa. We should make a decision which tag we prefer, to provide clarity to developers. This doesn't require deprecating the other tag. Same as how vehicle:backward=* is a valid tag (not deprecated), but when we're talking about something like highway=residential, the community consensus is that tagging oneway=* is better. Osmuser63783 (talk) 22:06, 16 September 2024 (UTC)

Rendering and routing

Anyone who wants to know if a way is one-way for pedestrians, only needs to look at the oneway:foot=* tag. They can ignore the oneway=* tag.

This sentence seems to be wrong in the current form of the proposal (it's not true on highway=footway and highway=pedestrian. --Martianfreeloader (talk) 13:02, 20 October 2024 (UTC)

Thanks for commenting. Can you explain what you mean please? The proposal holds that highway=pedestrian is a street, so just like with highway=residential or highway=unclassified, oneway=* doesn't affect pedestrians. If there is a pedestrian street somewhere that is one-way for pedestrians, it should be tagged oneway:foot=yes. As for highway=footway, it should always have oneway:foot=yes if it's one-way for pedestrians. Osmuser63783 (talk) 21:45, 24 October 2024 (UTC)

It appears I hadn't read the proposal carefully enough when I wrote that comment. Strictly speaking, this sentence would be correct after the proposal is approved. However I think it might still be good to amend a reminder that even if the proposal gets approved, it might still be dangerous to ignore oneway=* on a highway=footway as not all mappers might adhere to the proposal -- plus there will be a considerable amount of highway=footway with legacy tagging remaining in the database for quite some time. --Martianfreeloader (talk) 14:31, 27 October 2024 (UTC)

That's right. If the proposal is accepted, I'm planning to suggest validator warnings, that will lead to clearer tagging, but it will be a slow process. I'll mention this in the proposal. Osmuser63783 (talk) 17:55, 27 October 2024 (UTC)