Effect of oneway on pedestrians?

[easbar]

@Peter_Elderson

I also recognise the likelihood that this will continue until the end of time, no matter what is said in this thread.
It’s just so easy and common sense to add oneway=yes to a footway signed as oneway.

Yes… but for me this is rather a reason to be very clear about this tricky situation wherever possible, rather than accept the ambiguity.

[osmuser63783]

For what it’s worth, I’ve looked at examples of highway=pedestrian oneway=yes in various countries around the world. As far as it’s possible to judge this from afar, in the vast majority of cases, what mappers meant was that the street is one-way for vehicles. So I would put pedestrian in the same category as living_street or residential and say the oneway tag in unambiguous on pedestrian.

Has anyone looked at highway=path yet?

[easbar]

@osmuser63783

Has anyone looked at highway=path yet?

For example in Frankfurt, Germany (but not only) there are quite a few matches for way[highway=path][foot=designated][oneway=yes]. These cases are not meant to affect pedestrians. The reason being that there are separate lanes for pedestrians and bicycles where bicycles need to stay on the lane on the right side of the road (in the right direction), while pedestrians can choose the direction and side freely. Based on just these examples alone I would say it is not safe to assume the oneway tag is meant to affect pedestrians on paths. Btw maybe that is part of the reason why German mappers are against oneway applying to pedestrians.

[Nadjita]

@osmuser63783

Do German mappers think of the oneway tag in OSM as primarily a “translation” into OSM tagging of the above one-way street sign? Is this the reason why especially German mappers in this thread argue against not only oneway applying to pedestrians, but also against oneway:foot?

I cannot speak for the others, but the reason I’m arguing for this is, because that has always been the documented behaviour in the wiki. Actually, it applied to vehicles and boats initially, not sure what became of this.

But as for literally translating: this wouldn’t work, because in Germany, oneway-streets apply to horses as well. It’s probably just that everyone forgets the poor riders all the time.

@Peter_Elderson

I retagged all the fake village_greens in Nederland, but they still keep on appearing, and they still flourish in other countries.

I feel with you. I’ve experienced the same And don’t get me started in the 20cm² parks surrounding single trees along the street…

[osmuser63783]

@easbar

For example in Frankfurt, Germany (but not only) there are quite a few matches for way[highway=path][foot=designated][oneway=yes].

Thanks - but I meant internationally. When I looked at highway=footway oneway=yes, in most examples in Germany the mapper had meant bicycles when they set the oneway tag (e.g. shared sidewalks). In most examples outside Germany, the mapper had meant pedestrians.

I don’t know if that’s because German mappers tag differently from mappers in other countries (e.g. UK) or because footways where cycling is merely permitted but not equal (hence footway bicycle=yes) are more common in Germany than elsewhere.

[Nadjita]

@osmuser63783

In most examples outside Germany, the mapper had meant pedestrians.

I just checked for Chile, because I’m there very often, and most highway=path that are oneway=yes are used on hiking paths where mountainbikes are allowed downhill only. Some are even highway=path + access=no + bicycle=designated

There are some exceptions in some national parks where the way is so small that people are only allowed to pass in one direction, but that shouldn’t be a path anyway, because neither bicycles, nor horses are allowed there.

[easbar]

@osmuser63783

I don’t know if that’s because German mappers tag differently from mappers

Part of this question is where mappers are mapping separate bicycle lanes at all. There is rather more such ‘micro-mapping’ in Germany than in other countries, isn’t it? But I also looked at Amsterdam for example where separate cycleways are mapped also as oneways (and I doubt pedestrians are only allowed to walk in one direction there). Also please note I was refering to path not footway when I mentioned the cases in Frankfurt, Germany.

[osmuser63783]

OK I agree with you both for path. Lots of sidewalks, lots of mountain bike trails. Some exceptions where oneway was probably meant for pedestrians, but a very small proportion.

Looks like most of the ambiguity is for highway=footway.

[Mateusz_Konieczny]

@easbar

For example in Frankfurt, Germany (but not only) there are quite a few matches for way[highway=path][foot=designated][oneway=yes]. These cases are not meant to affect pedestrians.

have you tried skipping bicycle=designated and bicycle=yes? Germans love having one-way cycling infrastructure.

[Mateusz_Konieczny]

@easbar

“oneway only applies to vehicles on footways” is closer to the ‘truth’ in current OSM than “oneway applies also to non-vehicles on footways”

I would go with “Oneway applies also to pedestrians in cases where oneway=yes tag is present and only pedestrians use such path or footway. And in cases where oneway:foot=yes is present.”

[dieterdreist]

Unfortunately we don’t have a more descriptive name for the top level value, so oneway:access=* sounds quite weird but it’s technically correct and consistent.

access:forward=yes
access:backward=no
is more consistent and it has some few hundred uses https://taginfo.openstreetmap.org/keys/access%3Abackward#values

oneway:access does not: https://taginfo.openstreetmap.org/search?q=oneway%3Aaccess#keys

[easbar]

@Mateusz_Konieczny

I would go with “Oneway applies also to pedestrians in cases where oneway=yes tag is present and only pedestrians use such path or footway. And in cases where oneway:foot=yes is present.”

These cannot reliably be identified. See also this post. Nevertheless, I agree that probably most mappers mapping a highway=residential add the oneway=yes tag (rightfully) without the slightest concern that this will break a pedestrian router for example. And they also add oneway=yes to a highway=footway that is a waiting queue in an amusement park without even considering that this might not apply to people standing in line. Yet, highway=footway+oneway=yes is too ambiguous in other cases that not always can be distinguished from these allegedly obvious ones. And to avoid this ambiguity we need to make these people aware about the issue with very good documentation and an alternative tag that ideally seems just as fitting as the ambiguous oneway=yes.

[easbar]

@Mateusz_Konieczny

have you tried skipping bicycle=designated and bicycle=yes?

Have I tried skipping these tags to achieve what?

[dieterdreist]

are you arguing for foot:forward=only?
IMHO what you wrote regarding foot:backward does not make sense, because nothing was “removed”, the backward does not refer to foot backward direction but to the direction of the way in OpenStreetMap

[Nadjita]

@dieterdreist

access:backward=no

That is actually the best suggestion for a generic oneway path/footway so far. And you can even allow foot:backward=yes/designated if you wanted exceptions. Nice.

[ManuelB701 (Manuel B.)]

@easbar

Part of this question is where mappers are mapping separate bicycle lanes at all. There is rather more such ‘micro-mapping’ in Germany than in other countries, isn’t it? But I also looked at Amsterdam for example where separate cycleways are mapped also as oneways (and I doubt pedestrians are only allowed to walk in one direction there). Also please note I was refering to path not footway when I mentioned the cases in Frankfurt, Germany.

It’s more of a general thing on OSM than specific to Germany (e.g. it’s definitively the case in Poland), though I ultimately disagree with the idea to map cycleways separately all the time (including some of the most minor streets) to the point where I consider it to be a bad mapping practice.

[dieterdreist]

note that this is relatively rare situation so there will be no thousands/millions of cases - but some were linked already

reviewing my region:

do you know how these are signed? Are these really legal restrictions? Is it true once you began walking any of these you may not turn back and have to walk to the end? Maybe some are just recommendations?

[Mateusz_Konieczny]

@easbar

Have I tried skipping these tags to achieve what?

to skip such ways from your list of ways, as for them it relatively clearly refers to cyclist being limited to one direction

[Mateusz_Konieczny]

@dieterdreist

do you know how these are signed?

similar to Effect of oneway on pedestrians? - #136 by Mateusz_Konieczny

@dieterdreist

is it true once you began walking any of these you may not turn back and have to walk to the end?

yes (with some exceptions, oneway for vehicles also has some exceptions)

@dieterdreist

Are these really legal restrictions?

at some point you would need a lawyer, but these are at very least enforced (if you run zoo and put “one way only” sign and security enforces it - is it legal restriction or not?)

[dieterdreist]

it is about cases like

are these legal restrictions or recommendations? Once you set a foot behind the sign you may not step back and have to walk to the end, or could get fined, just like a oneway for vehicles, right?