If a way is tagged with sidewalk=*, is it reasonable to assume implicitly that foot=yes unless tagged otherwise?
Does anyone know of data consumers that make no assumption at all faced with sidewalk=*, and would treat the way as not accessible to pedestrians unless other rules provide access?
Yes, I would assume sidewalk=* except for sidewalk=separate is an additional indication for foot=yes.
Additional because if you scroll through OSM tags for routing/Access restrictions you see that for many countries most frequent occurring ways/paths give foot=yes.
See taginfo, sidewalk=* is used ~4.000.000 times but highway=residential ~67.000.000 times so only a very small fraction of all ways have sidewalk=* mapped.
which tags a data consumer considers and which assumptions he makes is his decision.
With sidewalk:left/right/both and no/separate it becomes a bit complicated.
For Germany I would guess that the following means foot=use_sidepath
sidewalk=separate
sidewalk:both=separate
sidewalk:left=separate + sidewalk:right=no
sidewalk:right=separate + sidewalk:left=no
sidewalk(:left/right/both)=yes/left/right/both/shoulder would be a hint for foot=yes
sidewalk=no/none or no sidewalk tag â default access
As a mapper, you cannot assume that sidewalk is evaluated for access and must therefore set all necessary access values ââthat differ from the default for the highway type.
Sure, but I wanted to know if anyone positively knows of data consumers that make no such assumption.
Key:segregated - OpenStreetMap Wiki says:
The presence of a sidewalk implies segregated=yes (because sidewalks are segregated by definition) and foot=yes. segregated=yes and foot=yes can be omitted in this case.
The edit in the wiki
It comes from this discussion in the NL community forum
@Famlam Can we put it under the headline sidewalk at cycleways?
@balchen you missed this part
On a ââdesignated cyclewayââ, pedestrians could be required to walk on a sidewalk, if present. This situation is especially common in The Netherlands.
I think itâs about the Dutch specialty sidewalk on cycleways.
highway=cycleway
sidewalk=*
But Iâm not sure if it is a good idea I think it is a bad idea to omit the âfoot=yesâ in this case.
Why are you not sure?
to omit foot=yes if sidewalk=* on a highway=cycleway is present
But in the Netherlands foot=yes is the default for cyclway.
But why are you not sure?
Yes, Iâm sure that it is a bad Idea.
Aber warum bist du nicht sicher, oder warum bist du sicher dass das eine schlechte Idee ist?
Ich bin der Meinung, dass man nicht davon ausgehen darf, das der Datennutzer sidewalk berĂźcksichtigt um access zu bestimmen. Und schon garnicht, wenn die einzige Erklärung dazu im Wiki unter âsegregatedâ zu finden ist. Ich halte das generell auch nicht fĂźr eine gute Idee das sidewalk oder cycleway die access-defaults Ăźberschreiben sollte. Die Auswertung von access wird dadurch noch komplizierter (insbesondere auch durch :left/right/both).
OK, so there were two questions, and I will rephrase them to make them clearer:
Is is reasonable â as a data consumer â to assume foot=yes when sidewalk=left|right|yes (removed no and separate and so on)?
Does anyone know of a data consumer that positively does not make this assumption?
I assume now you mean cycleway=* as opposed to highway=cycleway?
And youâre saying that for example cycleway=lane|track should not imply that bicycle=yes on this road?
I say sidewalk=* and cycleway=* on a street say something about physical infrastructure. This does not change the access of the way. You have to set it explicite if default access of the way is not allowing pedestrians or cyclists.
When does it ever make sense to have a sidewalk or a cycle lane/track and not be allowed to walk or cycle on it?
Also, cycle track normally implies something that is not a part of the carriageway, so any hypothetical restrictions on the carriageway would not apply on the cycle track⌠But because of the one-line crowd, the cycle track is mapped on the carriageway line.
You can not expect every user to consider all possible tags. We have access-tags for access. We have other tags for other things. Of cause all the other tags may contain information from what one can guess access.
I suggest you read my rephreased questions, or read them a little more thorouhghly.
Also, in terms of assessing foot access, sidewalk=* is most certainly not âall possible tagsâ. Itâs a highly relevant tag that you should be considering anyway if youâre doing anything foot related.
I wasnât asking if it was reasonable to expect every user to consider all possible tags. This is unnecessary hyperbole in a discussion.