If, however, you are mapping the area of a footway because this provides non-linear routing — for example, large sidewalk areas — then you can map the area as highway=footway and area=yes.
So why does the iD editor flag this tagging as incorrect or discourage it? Is this just a limitation of iD’s validator, or is there a broader consensus that highway=footway areas should be avoided in favor of highway=pedestrian areas, which not only go without warning in iD, but are even listed as the feature type “Pedestrian Area”?
However, since there is a clear distinction between highway=footpath and highway=pedestrian for linear features, the same distinction should apply to areas, shouldn’t it?
At least myself do agree with distinguishing =pedestrian vs =footway for both area=yes and the linear feature for a consistent definition. But this could be about area:highway= linear-shaped areas that can be routed with the highway= lines.
IMHO a footway area makes about the same sense as a primary or motorway area, these are always linear connections and the tag to map their area is area:highway not highway
I think it’s inconsistent to have highway=pedestrian area=yes different from the others.
This encourages contributors to invent fake pedestrian roads in order to map sidewalks, for example, as this is the only case where there is no warning in the editors and it’s render in OSM-Carto and many others.
I think that all highway=* should be linear and surface representations should be with area:highway= for all values concerned, rather than yes for some and no for others. real surface-only pedestrian area are place=, aren’t they?
This difference in treatment means, for example, that it is impossible to route emergency vehicles in several locations, since all sidewalks are mapped as pedestrian even when they are not wide enough for a vehicle to pass
Having tested it, it’s difficult to tell people to abandon a rendered scheme in favour of an unrendered one. I even think that rendering is the reason why this distortion of reality was adopted.
But feel free to try it out. You can find the contact details for this local group on the wiki
You can get questionable tagging for any tag, here is a random example I just stumbled over, footway with area=yes, which is all but a footway: https://www.openstreetmap.org/way/137993284
That looks like the correct usage of tags to me, what’s wrong with it?
Generally, my understanding is that the area of a sidewalk could be area:highway=footway while a square that isn’t signed as a pedestrian road (i.e. no vehicle access at all) should be highway=footway+area=yes.
I suspect that the post that you were replying to was trying to make the distinction between area:highway and highway=; area=yes rather than anything else.
It’s true that highway=pedestrian; area=yes is used more than highway=footway; area=yes for non-linear foot access areas (which was the question at the top of the thread). Whether something deserves a place=square tag is a different question, and “a square is not a …” risks confusing the two.
yes, I did not mean to imply place=square would be appropriate for the paved area in the linked way, (I would think it generally has to be either public or at least “articulate” the square, have a name, …, for place=square), I used the word square in reply to:
