Clashing access tags

classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

Clashing access tags

Richard Fairhurst
Hi all,

Occasionally I encounter tag combinations like this:

        bicycle=designated
        highway=proposed

        (from https://www.openstreetmap.org/way/335831004)

where the "bikes can ride along here" of the first tag is contradicted
by the "this hasn't even been built yet" of the second.

Similarly, on occasion I've found ways which are tagged access=no
("nothing is allowed along here") but are part of a bike route relation
("bikes can ride along here").

To some degree they're similar to "trolltags"
(https://wiki.openstreetmap.org/wiki/Trolltag) - where the meaning of
one tag is "radically changed" by another.

Two questions:

1. Is there any precedent for how to parse these contradictory tags? At
present cycle.travel will assume the most optimistic outcome, which is
good for a cycle route which goes over a private road and the mapper has
forgotten to add bicycle=permissive, but not good for a new cycleway
which hasn't yet been constructed.

2. Can we get warnings about this into validators etc.? I note iD
doesn't warn about it. (No idea what JOSM does.)

cheers
Richard

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Clashing access tags

dieterdreist


sent from a phone

> On 14. Jul 2019, at 14:07, Richard Fairhurst <[hidden email]> wrote:
>
> 1. Is there any precedent for how to parse these contradictory tags?


not sure I follow your analysis that these are contradictory. One could read bicycle=designated as a property (refines a feature), so in this case highway=proposed defines a planned highway and bicycle=designated states it will be explicitly for bicycles. No contradiction.


Cheers Martin
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Clashing access tags

Leif Rasmussen
I'd map the way as highway=cycleway + proposed:highway=*, since until the proposed highway is finished, it's effectivity a cycleway.
Adding foot=yes could also be good if relevant.

On Sun, Jul 14, 2019, 3:38 PM Martin Koppenhoefer <[hidden email]> wrote:


sent from a phone

> On 14. Jul 2019, at 14:07, Richard Fairhurst <[hidden email]> wrote:
>
> 1. Is there any precedent for how to parse these contradictory tags?


not sure I follow your analysis that these are contradictory. One could read bicycle=designated as a property (refines a feature), so in this case highway=proposed defines a planned highway and bicycle=designated states it will be explicitly for bicycles. No contradiction.


Cheers Martin
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Clashing access tags

Mateusz Konieczny-3
In reply to this post by Richard Fairhurst

14 lip 2019, 14:07 od [hidden email]:
Occasionally I encounter tag combinations like this:

bicycle=designated
highway=proposed
Based on tagging is it a not yet existing
road that will have some part designated
for use by cyclists (lane/cycleway?).
Maybe it will be a cycleway.

Other major issue is that source tag
is missing what is not ok for 
proposed ways (assuming that mapping highway=proposed
is desirable)
Similarly, on occasion I've found ways which are tagged access=no ("nothing is allowed along here") but are part of a bike route relation ("bikes can ride along here").
It may be worth reporting as error by validator. Even with really badly designed
bicycle routes where bicycle=no
happens still allow foot passage.
1. Is there any precedent for how to parse these contradictory tags? At present cycle.travel will assume the most optimistic outcome, which is good for a cycle route which goes over a private road and the mapper has forgotten to add bicycle=permissive, but not good for a new cycleway which hasn't yet been constructed.
Personally I would assume that all 
highway=proposed are not passable
(including ones with cycleway=lane etc).

I would treat ways with supposedly
no access but in cycleway route as 
passable. But maybe add some penalty 
and assume that dismounting and walking
is necessary?

Not sure about ways in route, with 
explicit bicycle=no, foot=no.
2. Can we get warnings about this into validators etc.? I note iD doesn't warn about it. (No idea what JOSM does.)
As usual, opening issue in their bug trackers
is usual a good first step.

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Clashing access tags

Mateusz Konieczny-3
In reply to this post by Richard Fairhurst

14 lip 2019, 14:07 od [hidden email]:
Occasionally I encounter tag combinations like this:

bicycle=designated
highway=proposed
Based on tagging is it a not yet existing
road that will have some part designated
for use by cyclists (lane/cycleway?).
Maybe it will be a cycleway.

Other major issue is that source tag
is missing what is not ok for 
proposed ways (assuming that mapping highway=proposed
is desirable)
Similarly, on occasion I've found ways which are tagged access=no ("nothing is allowed along here") but are part of a bike route relation ("bikes can ride along here").
It may be worth reporting as error by validator. Even with really badly designed
bicycle routes where bicycle=no
happens still allow foot passage.
1. Is there any precedent for how to parse these contradictory tags? At present cycle.travel will assume the most optimistic outcome, which is good for a cycle route which goes over a private road and the mapper has forgotten to add bicycle=permissive, but not good for a new cycleway which hasn't yet been constructed.
Personally I would assume that all 
highway=proposed are not passable
(including ones with cycleway=lane etc).

I would treat ways with supposedly
no access but in cycleway route as 
passable. But maybe add some penalty 
and assume that dismounting and walking
is necessary?

Not sure about ways in route, with 
explicit bicycle=no, foot=no.
2. Can we get warnings about this into validators etc.? I note iD doesn't warn about it. (No idea what JOSM does.)
As usual, opening issue in their bug trackers
is usual a good first step.

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Clashing access tags

Tagging mailing list
In reply to this post by Richard Fairhurst
On 14/07/2019 13:07, Richard Fairhurst wrote:

> Hi all,
>
> Occasionally I encounter tag combinations like this:
>
>     bicycle=designated
>     highway=proposed
>
>     (from https://www.openstreetmap.org/way/335831004)
>
> where the "bikes can ride along here" of the first tag is contradicted
> by the "this hasn't even been built yet" of the second.

No. it doesn't. Highway is the primary tag. It usurps sub-tags which are
merely adjectives of the primary.
When it has been completed it will be designated for use by bicycles.

>
> Similarly, on occasion I've found ways which are tagged access=no
> ("nothing is allowed along here") but are part of a bike route
> relation ("bikes can ride along here").

Route relations should be aware of tags on ways. access=no can be used
in part to indicate road works.
Just because a way may be inaccessible to bikes doesn't mean it's not
still officially regarded as a cycle route (the NCN ref won't be removed)

DaveF

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Clashing access tags

Andrew Harvey-3
+1

On Mon., 15 Jul. 2019, 3:02 am Dave F via Tagging, <[hidden email]> wrote:
On 14/07/2019 13:07, Richard Fairhurst wrote:
> Hi all,
>
> Occasionally I encounter tag combinations like this:
>
>     bicycle=designated
>     highway=proposed
>
>     (from https://www.openstreetmap.org/way/335831004)
>
> where the "bikes can ride along here" of the first tag is contradicted
> by the "this hasn't even been built yet" of the second.

No. it doesn't. Highway is the primary tag. It usurps sub-tags which are
merely adjectives of the primary.
When it has been completed it will be designated for use by bicycles.

>
> Similarly, on occasion I've found ways which are tagged access=no
> ("nothing is allowed along here") but are part of a bike route
> relation ("bikes can ride along here").

Route relations should be aware of tags on ways. access=no can be used
in part to indicate road works.
Just because a way may be inaccessible to bikes doesn't mean it's not
still officially regarded as a cycle route (the NCN ref won't be removed)

DaveF

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Clashing access tags

Mateusz Konieczny-3
In reply to this post by Tagging mailing list
14 Jul 2019, 21:03 by [hidden email]:
Route relations should be aware of tags on ways. access=no can be used in part to indicate road works.
Just because a way may be inaccessible to bikes doesn't mean it's not still officially regarded as a cycle route (the NCN ref won't be removed)
Probably highway=construction is a better tagging for such case.


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Clashing access tags

Tagging mailing list
Even if 'construction' was to be used, it would still cause the same confusion to Richard F

On 15/07/2019 20:33, Mateusz Konieczny wrote:
14 Jul 2019, 21:03 by [hidden email]:

Route relations should be aware of tags on ways. access=no can be used in part to indicate road works.
Just because a way may be inaccessible to bikes doesn't mean it's not still officially regarded as a cycle route (the NCN ref won't be removed)

Probably highway=construction is a better tagging for such case.




_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging