Asking to tag segregated=no on highway=footway bicycle=yes is pointless

[mkoniecz]

What steps will reproduce the problem?

1. Create way with highway=footway bicycle=yes
2. Run validator

What is the expected result?

Nothing, as in this case segregated=no is obvious (and that is why for example StreetComplete is not asking people to tag segregated)

After all, if bicycles would have its own designated space, then it would be bicycle=designated, right?

What happens instead?

JOSM spams "missing tag - Combined foot- and cycleway without segregated. (18)"

Please provide any additional information below. Attach a screenshot if possible.

See #17973

See ​https://www.openstreetmap.org/way/25463963

See also ​https://wiki.openstreetmap.org/wiki/Talk:Key:segregated#What_about_highway.3Dfootway_bicycle.3Dyes

Relative:URL: ^/trunk
Repository:UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b
Last:Changed Date: 2020-12-27 20:15:27 +0100 (Sun, 27 Dec 2020)
Revision:17426
Build-Date:2020-12-28 02:30:52
URL:https://josm.openstreetmap.de/svn/trunk

Identification: JOSM/1.5 (17426 en_GB) Linux Ubuntu 20.04.1 LTS
Memory Usage: 494 MB / 3974 MB (249 MB allocated, but free)
Java version: 11.0.9.1+1-Ubuntu-0ubuntu1.20.04, Ubuntu, OpenJDK 64-Bit Server VM
Look and Feel: javax.swing.plaf.metal.MetalLookAndFeel
Screen: :0.0 1920×1080 (scaling 1.00×1.00)
Maximum Screen Size: 1920×1080
Best cursor sizes: 16×16→16×16, 32×32→32×32
Desktop environment: LXQt
Java package: openjdk-11-jre:amd64-11.0.9.1+1-0ubuntu1~20.04
Java ATK Wrapper package: libatk-wrapper-java:all-0.37.1-1
Environment variable LANG: en_GB.UTF-8
libcommons-logging-java: libcommons-logging-java:all-1.2-2
fonts-noto: fonts-noto:-
Dataset consistency test: No problems found

[gaben]

Why is it obvious?

[mkoniecz]

In case of bicycles have its own separate, designated space, then correct access would be bicycle=designated, right?

I am not 100% sure, access tagging is complex so maybe asking on tagging mailing list would be a good idea.

[GerdP]

I also think that a way with tags

bicycle=yes
foot=yes
highway=path

is not meant by the wiki about segregated. The wiki says "This key is used for designated combined cycle- and footways"
and I understand this as both foot and bicycle are designated, not just allowed.

[Klumbumbus]

Replying to GerdP:

and I understand this as both foot and bicycle are designated, not just allowed.

The questioned rule of this ticket comes from #17973 when the wiki said "This key has no default value and should be tagged on all shared ways!". At this time I understood "all shared ways" as all combinations of bicycle=* with foot=*. Limiting it to designated values only probably makes more sense, as you say.

[gaben]

Replying to Klumbumbus:

At this time I understood "all shared ways" as all combinations of bicycle=* with foot=*.

You understood back then correctly the wiki has been edited since then :) I'm saying the same, it makes more sense this way.

Otherwise, I agree with @GerdP. The ticket's original example is problematic, because "segregated=no is obvious" is an error. Segregated should be null here if I can say that.

[Klumbumbus]

I saw that the wiki was changed. But I think even back then with the old text it was already meant for ways with two designated values only.

[GerdP]

In 17457/josm:

fix #20318: Asking to tag segregated=no on highway=footway bicycle=yes is pointless

• only complain about missing segregated tag if way is designated for bicycle AND foot

[stoecker]

Milestone renamed

[goodidea]

Hello!

I just read that with interest ... is it really so easy that segregated is meant only for ways with two designated values for foot + bicycle?

The logic that the wiki only mentions and shows such examples doesn't mean that footways with foot=designated and bicycle=yes (NOT designated) DO NOT exist with segregated ways. Or is meant in this way, that the segregated tag should not be used or only be optional in such cases (and so it's not a case for a validation warning)?

So, I would clearly contradict to the sentence: "if bicycles would have its own designated space, then it would be bicycle=designated, right?" => NO, it's a wrong conclusion. It also can be bicycle=yes.

The space for the bicycle may be separated or segregated on footways with bicycle=yes, but it must NOT be a way with "bicycle=designated". Because this has another meaning (under the current agreements and documentation in the wiki as far as I know) – it means, that you HAVE to use this way, and may not use a parallel street for example. That's also the difference of signs with bicycle="yes" vs. bicycle="designated".

I know some examples in my city of such ways with foot=designated, bicycle=yes and segregated=yes. And following the logic above, wouldn't ONE counter example be enough to falsify this "rule"? Here is one:

• ​https://www.openstreetmap.org/way/914823660 (and it even has different surfaces for the footway and the bicycle side)

• The sign with foot=designated, bicycle=yes for this way is just here (DE:239,1022-10): ​https://www.openstreetmap.org/node/7781939900

• And the following pedestrian/bicycle crossing also has segregated ways: ​https://www.openstreetmap.org/way/833631510 and it is correctly tagged with "bicycle=yes" like the way before and not "bicycle=designated" – you can also use the street by bicycle if you want, so it is NOT designated for bicycles, even if it is a segregated foot/bicycle way. The sign is also very clear about that.

• And it goes on ... The next way here is a segregated=no way (also with foot=designated/bicyle=yes): ​https://www.openstreetmap.org/way/931896322, then follows a crossing WITH segregated ways: ​https://www.openstreetmap.org/way/931896321, again a sign (DE:239,1022-10): ​https://www.openstreetmap.org/node/7781939907, and after the crossing a small way segment with segregated=no followed by a segment with segregated=yes.

I only want to say and show, that it can be quite a litte mess with such ways in reality ... and that is not a very rare case.

So is segregated (and its check) not also "valid" and reasonable for designated footways with bicycle=yes, and perhaps only quite unnecessary with highway=path or highway=track (and foot=null or yes and bicycle=null or yes)?

[skyper]

Yes, at least in Germany, I find highway=path, foot=designated, bicycle=yes and segregated=yes. As footway and cycleway with foot=[yes|designated] and bicycle=[yes|designated] are invalid in Germany, anyway, this rule might fit in the Rules/GermanySpecific.

[GerdP]

How do you detect that the way is segregated when there is no corresponding sign?

[skyper]

It is the lane_marking and the symbols on the ground. The usual case are former combined + segregated ways where the "blue sign" was changed but the lane markings and symbols still exist. As a cyclist you are only allowed to use the former designated part but pedestrian are allowed on "both sides" and for sure cyclist have to respect that these ways are designated for "foot", now, and "bicycle" is only allowed.

[GerdP]

As a cyclist you are only allowed to use the former designated part

That's new to me. Doesn't the sign (DE:239,1022-10) say that a cyclist can chose freely?

[goodidea]

Replying to GerdP:

How do you detect that the way is segregated when there is no corresponding sign?

In my case (see comment 12) there are no old markings (symbols) on the way, but there are 2 different surfaces – at least on some parts of the way (asphalt left and paving stones right). And there are street crossings on the way where the side for bicycles has the typical red colour – and this part continues after the crossing with the asphalt surface ...

I made some photos in April and put it in Mapillary. You can see it quite good here:

​https://www.mapillary.com/app/?pKey=1487129391650794

What you see in Mapillary is this part of the way (plus the crossings):

• Part 1: ​https://www.openstreetmap.org/way/914823667 (segregated=yes; surfaces both paving_stones but with a clear segregation)
• Part 2: ​https://www.openstreetmap.org/way/914823660 (segregated=yes; surfaces asphalt/paving_stones = a clear segregation)
• Part 3: ​https://www.openstreetmap.org/way/931896322 (segregated=no; surface only paving_stones without segregation – the way is also more narrow here)

It's clear that the sign says that pedestrians may walk everywhere and cyclists, too - with due respect to the pedestrians. Officially/legally there is no segregation. But in reality, the cyclists are ALWAYS on the left side there and pedestrians (mostly) walk on the right side ... the segregation by the 2 surfaces is "stronger" for the people than the sign (and healthier) ... That's why I would also tag these parts of the way with "segregated=yes", it corresponds exactly with the situation there.