Request to render highway=residential & area=yes

[Drahflow]

Expected behavior

The same arguments for highway=service & area=yes objects also apply for highway=residential & area=yes. Thus ways with highway=residential & area=yes should be rendered as residential-street-textured areas.

I checked a few instances of this tag combination around the lat/lon of the one linked below, in all cases they were not mis-mapped areas of essential linear ways, but square-like areas across which vehicles could drive omnidirectionally.

Actual behavior

Such objects are invisible.

Screenshots with links illustrating the problem

https://overpass-turbo.eu/s/1usa

vs.

https://www.bing.com/maps?q=hannover+map&FORM=HDRSC6&cp=52.282895%7E10.612627&lvl=19.0&style=h

[imagico]

Would revert #4096 - see #3995 for a lengthy discussion of this. See also #3985, #4637.

Please bring up any new arguments on the matter here.

[Drahflow]

In somer other issues (e.g. #3995 (comment) , and thus a few years back), it is claimed that no guidance exists on the wiki. I derived my understanding from this section:
https://wiki.openstreetmap.org/wiki/Key:area:highway

area=yes on a highway=* describes a routable highway area on which the dedicated traffic can route omnidirectionally; i.e. from any point along the edge to any other. This is typical for pedestrian areas, or service areas where vehicles drive without prescribed lanes. The rationale is that routing engines should consider those omnidirectional objects in addition to nodes and edges in classical graph theory; thus no linear highway lines would be required within. However, as of 2019, OpenTripPlanner is the only OSM-based router that supports routing over areas. It does so by computing a visibility graph over the area. [1]

[Drahflow]

Another argument: What would otherwise be the correct way to map https://en.wikipedia.org/wiki/Shared_space ? (And I think some of the problem examples I found in Germany so far were attempts at implementing it.)

[Drahflow]

@imagico Is there value in doing some statistics on actual usage? I found some more examples, most seem to be "parking space at end of road"-situations, where rendering an area would likely be more accurate; a bunch are truly drivable area features (i.e. neither parking, nor pedestrian, just a wide open space available (also) for cars); and I also found a few I'd classify as wrongly mapped (e.g. an entire plaza tagged highway=residential & area = yes, even though it obviously has non-car subareas).

If that would constitute an argument one way or another, I could do a more methodical checking of the map.

[imagico]

Changes in actual use of highway=residential on polygons or in mapper consensus on what this means since #4096 would be the main reason for revisiting the decision. But i don't see any indications for that so far. I also would not assume that the shown example is by mapper consensus considered a valid use of highway=residential on polygons. If it is it would be important to have documentation of that.

[matkoniecz]

Note that highway=service area=yes is heavily abused in tagging for renderer, with deliberately wrong tagging for renderer more common than valid uses. highway=residential is very likely to be abused in the same way if rendered and actual correct use is extremely rare.

Also, this specific case looks like amenity=parking to me, though I have not surveyed it.

[pnorman]

I would be against rendering it, for the reasons matkoniecz mentions, and were previously established in t he linked issues. I don't see that anything has changed since then.

[imagico]

Closing as the general impression seems to be that the situation is mostly unchanged compared to previous discussion.