Opened 5 years ago
Closed 5 years ago
#18471 closed enhancement (fixed)
warn about segregated=no and sidewalk=left/right/both
Reported by: | Famlam | Owned by: | Klumbumbus |
---|---|---|---|
Priority: | normal | Milestone: | 20.01 |
Component: | Core validator | Version: | tested |
Keywords: | sidewalk segregated | Cc: |
Description
Please see https://overpass-turbo.eu/s/PgT
There are 435 cycleways where a sidewalk (either left
/right
/both
, optionally also the already-warned-for yes
) is set, but on which segregated=no
. A cycleway with a sidewalk, but no segregation (i.e., separation for bicycles and pedestrians), is hard to imagine.
What is the expected result?
A sidewalk on a cycleway implies that there is a separate area for pedestrians. Hence, the combination with segregated=no is quite likely a mistake. Therefore, a warning about a suspicious tag combination should be given.
What happens instead?
No warning.
Please provide any additional information below. Attach a screenshot if possible.
URL:https://josm.openstreetmap.de/svn/trunk Repository:UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b Last:Changed Date: 2019-12-01 23:10:15 +0100 (Sun, 01 Dec 2019) Build-Date:2019-12-02 02:30:57 Revision:15553 Relative:URL: ^/trunk Identification: JOSM/1.5 (15553 nl) Windows 10 64-Bit OS Build number: Windows 10 Home 1903 (18362) Memory Usage: 751 MB / 1820 MB (501 MB allocated, but free) Java version: 1.8.0_231-b11, Oracle Corporation, Java HotSpot(TM) 64-Bit Server VM Screen: \Display0 1920x1080 Maximum Screen Size: 1920x1080 Plugins: + OpeningHoursEditor (35242) + changeset-viewer (22) + imagery_offset_db (34908) + tageditor (34977) + turnlanes-tagging (281) + undelete (34977) + utilsplugin2 (35238) Map paint styles: - https://josm.openstreetmap.de/josmfile?page=Styles/Lane_and_Road_Attributes&zip=1 - https://josm.openstreetmap.de/josmfile?page=Styles/PublicTransport&zip=1 Last errors/warnings: - W: No configuration settings found. Using hardcoded default values for all pools.
Attachments (0)
Change History (13)
comment:1 by , 5 years ago
Component: | Core → Core validator |
---|---|
Summary: | Validator: warn about segregated=no and sidewalk=left/right/both → warn about segregated=no and sidewalk=left/right/both |
comment:2 by , 5 years ago
Milestone: | → 20.01 |
---|---|
Owner: | changed from | to
comment:3 by , 5 years ago
way[highway=~/^(cycleway|footway|path)$/][segregated][sidewalk =~/^(left|right|both)$/] { throwWarning: tr("{0} together with {1} and {2}", "{0.tag}", "{1.key}", "{2.key}"); group: tr("suspicious tag combination"); }
follow-up: 6 comment:5 by , 5 years ago
So as summary if that text was too confusing ;)
I think I understood it, thanks :)! I always interpret segregated yes
as "a single road dedicated to pedestrians and bicycles (/mopeds/mofas), where they do not share each others travel area", so without restriction to whether that separation of travel areas is due to a line on the road or a kerb like a sidewalk. But then this was always duplicate tagging when a sidewalk was present, so it was confusing for me why I had to add both (and, clearly, for many others too)
Indeed, I wouldn't warn about residential (or other car-allowed-roads), even if it were only because the wiki states that there are sidewalks where cyclist may go on the sidewalk. (Although I would personally probably call that cycleway=track
+ sidewalk=no
, when used as property of the road; I'm not aware of such a case though).
So to summarize what I wrote, I agree with your proposal :)
This however also requires that https://josm.openstreetmap.de/browser/josm/trunk/data/validator/combinations.mapcss#L629 is modified to exclude ways with the sidewalk=left/right/both tag, or it will warn to re-add the segregated tag.
comment:6 by , 5 years ago
Replying to anonym:
This however also requires that https://josm.openstreetmap.de/browser/josm/trunk/data/validator/combinations.mapcss#L629 is modified to exclude ways with the sidewalk=left/right/both tag, or it will warn to re-add the segregated tag.
I don't think that rule should be changed. What would be a valid tag combination for which we would see a false positive warning? Can you please give an example?
follow-up: 9 comment:7 by , 5 years ago
The rule from comment 3 would trigger on a cycleway with:
highway=cycleway
+ foot=yes
+ segregated=yes
+ sidewalk=left
Assuming a sidewalk was present, I would then (based upon the suspicious tag combination warning) change that to:
highway=cycleway
+ foot=yes
+ sidewalk=left
The rule that I mentioned (sorry, wasn't logged in, but anonym was me), specifically way[highway=cycleway][foot][foot!=no ][foot!=use_sidepath ][!segregated]!.unpaved_surface
would then trigger the warning Combined foot- and cycleway without segregated
, because it (highway=cycleway
+ foot=yes
+ sidewalk=left
) has no segregated.
Or do I miss something here?
follow-up: 10 comment:9 by , 5 years ago
Replying to Famlam:
The rule from comment 3 would trigger on a cycleway with:
highway=cycleway
+foot=yes
+segregated=yes
+sidewalk=left
Assuming a sidewalk was present, I would then (based upon the suspicious tag combination warning) change that to:
highway=cycleway
+foot=yes
+sidewalk=left
I was about to say that in this case foot=yes
and segregated=yes
should be removed instead as sidewalk=left
is the tag that allows pedestrians to use the sidewalk of this highway. But then I read osmwiki:Sidewalks again and it says: "The above tagging provides information about the physical infrastructure rather than legal access details which can be included using foot=yes/no." So by removing foot=yes
the access of the highway would change. Thats not the right way, so highway=cycleway
+ foot=yes
+ segregated=yes
+ sidewalk=left
seems to be a full valid tagging example and in the end we can only warn for cases with segregated=no
like you suggested at the beginning :)
way[highway=~/^(cycleway|footway|path)$/][segregated=no][sidewalk =~/^(left|right|both)$/] { throwWarning: tr("{0} together with {1} and {2}", "{0.tag}", "{1.tag}", "{2.key}"); group: tr("suspicious tag combination"); }
follow-up: 11 comment:10 by , 5 years ago
Replying to Klumbumbus:
only warn for cases with
segregated=no
In the end false positives are still possible: ways where pedestrians are allowed to walk on the lane together with the cyclists and where additional a sidewalk is present. However as already said this seems very unlikely and I think we can live with very few false positives.
New overpass query: https://overpass-turbo.eu/s/PQo
segregated
is independent fromsidewalk
assegregated
refers to the lane itself whilesidewalk
not. However also roads withsegegated=yes
are suspicious as well. There are cases like this but then it ishighway=residential
(nothighway=cycleway
) +bicycle_road=yes
+sidewalk=both
(and nosegregated
). While it could theoretically besegregated=yes|no
on the lane it is still wide enough to behighway=residential
. If the road is small (highway=cycleway
) then it is unlikely that if a sidewalk is present, pedestrians are allowed to walk on the lane too and interfer the cyclists.So as summary if that text was too confusing ;) I would warn for all:
highway=cycleway|footway|path
+segregated=*
+sidewalk=left|right|both