Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Christian Müller]

Again, please note the context this has
been orginally written in.

It dealt with the suggestion of
cycleway:left:direction=forward

There was no intention of importing
other objects of discussion, like you
and others did.


Greetings 
 

> 03/20/19, 11:57, Paul Johnson wrote:
> 
> > Suggesting forward and backward as tag
> > values is a rather bad thing, as these
> > are values for route relation roles.
> 
> lanes:forward/backward/both_ways and
> maxspeed:forward/backward are pretty
> common tags

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Christian Müller]

I do, for the reasons and in the context given,
but please go ahead and throw that context away.
 
Regards,
cmuelle8
 

> On 03/20/19, 07:27, althio wrote:
> 
> "rather bad thing"?!? Who says? Source?

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Paul Johnson]

On Tue, Mar 19, 2019 at 5:12 PM "Christian Müller"  wrote:

> Suggesting forward and backward as tag
> values is a rather bad thing, as these
> are values for route relation roles.
>

I would say that is true for nodes, though this situation is being handled
for nodes that are part of ways better than I had anticipated.  This isn't
true for ways, though, where lanes:forward/backward/both_ways and
maxspeed:forward/backward are pretty common tags, and the direction is
relative to which way the way points, not necessarily in the direction of
travel (though the reasons you might need oneway=-1 seems to be
disappearing rapidly)
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Volker Schmidt]

Nodes on ways do have an "inherited" direction (unless they are end nodes)
The wiki page
https://wiki.openstreetmap.org/wiki/Key:traffic_sign
says:
" You may use traffic_sign:forward
=* to specify that this
particular sign affects vehicles moving in the same direction as the way,
or traffic_sign:backward 
=* to specify that the opposite direction is affected."
Volker


On Wed, 20 Mar 2019 at 09:39, Martin Koppenhoefer 
wrote:

> on ways the tag is perfectly fine for directional properties, but if you
> look at actual usage, there is also a lot of nonsense, for example here:
> https://taginfo.openstreetmap.org/keys/traffic_sign%3Aforward
> 27000 of 28000 objects with *forward* on nodes (nodes do not have a
> direction).
> Just because people are doing something doesn’t necessarily mean it is the
> best thing to do.
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Martin Koppenhoefer]

sent from a phone

> Am 20.03.2019 um 08:27 schrieb althio :
> 
> "rather bad thing"?!? Who says? Source?


on ways the tag is perfectly fine for directional properties, but if you look 
at actual usage, there is also a lot of nonsense, for example here:
https://taginfo.openstreetmap.org/keys/traffic_sign%3Aforward
27000 of 28000 objects with *forward* on nodes (nodes do not have a direction). 
Just because people are doing something doesn’t necessarily mean it is the best 
thing to do.

cheers,
Martin___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[althio]

On Tue, 19 Mar 2019 at 23:13, "Christian Müller"  wrote:

> Suggesting forward and backward as tag
> values is a rather bad thing, as these
> are values for route relation roles.
>

"rather bad thing"?!? Who says? Source?

This is not what exists in the wiki page:
https://wiki.openstreetmap.org/wiki/Forward_%26_backward,_left_%26_right#Tagging

And actual usage in the database
https://taginfo.openstreetmap.org/search?q=forward#keys
https://taginfo.openstreetmap.org/search?q=forward#values
https://taginfo.openstreetmap.org/search?q=forward#roles
https://taginfo.openstreetmap.org/search?q=backward#keys
https://taginfo.openstreetmap.org/search?q=backward#values
https://taginfo.openstreetmap.org/search?q=backward#roles
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Christian Müller]

Suggesting forward and backward as tag
values is a rather bad thing, as these
are values for route relation roles.

For example, if a physical way as part
of a route relation is travelled along
the direction of the osm way (as given
by its node order) it may be assigned
the role "forward" and this would then
have nothing or little to do with
legal restrictions on the physical
way.

And how exactly would *:direction=forward
be beneficial to *:oneway=1 if both aim
to describe legal precepts on travel
direction for a physical way?

I've seen a lot of people suggesting
:direction over :oneway now, but don't
see why in that context.


Regards,
cmuelle8


> On 03/19/19, 08:23, Markus wrote:
>
> Those exceptions could be tagged
> cycleway:[left|right]:direction=[forward|backward|both].

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Markus]

On Tue, 19 Mar 2019 at 09:32, Mateusz Konieczny  wrote:
>
> Mar 19, 2019, 8:29 AM by selfishseaho...@gmail.com:
>>
>> So why not correcting those 850 (6%) incorrect uses of
>> cycleway*=opposite_lane instead of inventing a new tagging system for
>> it? I've already corrected a few dozen, here's the link if you want to
>> help me:
>>
>> http://overpass-turbo.eu/s/H91
>
> I worry about data damage in remote editing so I will not join.

Don't worry, i only made corrections when the cycle lanes are visible
on the aerial imagery. If not, i left a note.

> But I would like JOSM validator to notice such inconsistencies during my 
> edits so I opened
> https://josm.openstreetmap.de/ticket/17488#ticket

Thank you!

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Mateusz Konieczny]

In case that people map separate cycleways not as highway=cycleway but
as a property of road you may in addition have cycleways on each side of the 
road.

Each separate cycleway may be one way or not.



Also, with sidewalks mapped as a property of road, not as a separate way:
there may be in addition legal to cycle on a sidewalk.

Each sidewalk with allowed cycling may be one way for cyclists or not.


Mar 19, 2019, 9:02 AM by marc.ge...@gmail.com:

> Which situations can one theoretically have in a oneway street that
> allows contraflow for bicycles and moped/mofas?
>
> - no separate lanes (rather typical in Belgium)
> - separate lanes, one lane on each side, lanes are 1-way
> - separate lane only in contraflow direction, lane is oneway, (e.g.
> flow direction shares lane with cars, or is track)
> - separate lane only in flow direction, lane is oneway
> - lane(s) on contraflow side, lanes are both way
> - lane(s) on flow side, lanes are both way
>
>
> On Tue, Mar 19, 2019 at 8:55 AM Markus <> selfishseaho...@gmail.com 
> > > wrote:
>
>>
>> By the way: aren't all contraflow cycle lanes located on the left side
>> in countries with right-hand traffic or on the right side in countries
>> with left-hand traffic? If so, cycleway*=opposite_lane could simply be
>> replaced by cycleway:*=lane, as the direction of the cycle lane is
>> already implied by oneway:bicycle=no – thus no need for
>> cycleway:*:oneway=-1.
>>
>> ___
>> Tagging mailing list
>> Tagging@openstreetmap.org 
>> https://lists.openstreetmap.org/listinfo/tagging 
>> 
>>
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org 
> https://lists.openstreetmap.org/listinfo/tagging 
> 
>

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Mateusz Konieczny]

Mar 19, 2019, 8:29 AM by selfishseaho...@gmail.com:

> On Tue, 19 Mar 2019 at 00:30, Warin <> 61sundow...@gmail.com 
> > > wrote:
>
>>
>> On 19/03/19 10:03, Graeme Fitzpatrick wrote:
>>
>>>
>>> On Tue, 19 Mar 2019 at 07:47, Markus <>>> selfishseaho...@gmail.com 
>>>  wrote:
>>>

 Unfortunately, many tags are used wrongly (e.g. name, access tags,
 grass, water park) – do you want to replace all of them?

>>>
>>> Yes!!! :-)
>>>
>>
>> All those used incorrectly should be corrected. If that means deletion .. so 
>> be it. But those used correctly should be left.
>>
>
> So why not correcting those 850 (6%) incorrect uses of
> cycleway*=opposite_lane instead of inventing a new tagging system for
> it? I've already corrected a few dozen, here's the link if you want to
> help me:
>
> http://overpass-turbo.eu/s/H91 
>
I worry about data damage in remote editing so I will not join.

But I would like JOSM validator to notice such inconsistencies during my edits 
so I opened 
https://josm.openstreetmap.de/ticket/17488#ticket

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Markus]

On Tue, 19 Mar 2019 at 09:14, Mateusz Konieczny  wrote:
>
>> Mar 19, 2019, 8:52 AM by selfishseaho...@gmail.com:
>>
>> By the way: aren't all contraflow cycle lanes located on the left side
>> in countries with right-hand traffic or on the right side in countries
>> with left-hand traffic? If so, cycleway*=opposite_lane could simply be
>> replaced by cycleway:*=lane, as the direction of the cycle lane is
>> already implied by oneway:bicycle=no – thus no need for
>> cycleway:*:oneway=-1.
>
> There are some exceptions. Usually belonging in
> "the worst bicycle infrastructure in region X" but
> it sometimes exists.

Those exceptions could be tagged
cycleway:[left|right]:direction=[forward|backward|both].

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Mateusz Konieczny]

Mar 19, 2019, 8:52 AM by selfishseaho...@gmail.com:

> By the way: aren't all contraflow cycle lanes located on the left side
> in countries with right-hand traffic or on the right side in countries
> with left-hand traffic? If so, cycleway*=opposite_lane could simply be
> replaced by cycleway:*=lane, as the direction of the cycle lane is
> already implied by oneway:bicycle=no – thus no need for
> cycleway:*:oneway=-1.
>
There are some exceptions. Usually belonging in 
"the worst bicycle infrastructure in region X" but
it sometimes exists.

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Marc Gemis]

Which situations can one theoretically have in a oneway street that
allows contraflow for bicycles and moped/mofas?

- no separate lanes (rather typical in Belgium)
- separate lanes, one lane on each side, lanes are 1-way
- separate lane only in contraflow direction, lane is oneway, (e.g.
flow direction shares lane with cars, or is track)
- separate lane only in flow direction, lane is oneway
- lane(s) on contraflow side, lanes are both way
- lane(s) on flow side, lanes are both way


On Tue, Mar 19, 2019 at 8:55 AM Markus  wrote:
>
> By the way: aren't all contraflow cycle lanes located on the left side
> in countries with right-hand traffic or on the right side in countries
> with left-hand traffic? If so, cycleway*=opposite_lane could simply be
> replaced by cycleway:*=lane, as the direction of the cycle lane is
> already implied by oneway:bicycle=no – thus no need for
> cycleway:*:oneway=-1.
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Markus]

By the way: aren't all contraflow cycle lanes located on the left side
in countries with right-hand traffic or on the right side in countries
with left-hand traffic? If so, cycleway*=opposite_lane could simply be
replaced by cycleway:*=lane, as the direction of the cycle lane is
already implied by oneway:bicycle=no – thus no need for
cycleway:*:oneway=-1.

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Markus]

On Tue, 19 Mar 2019 at 00:30, Warin <61sundow...@gmail.com> wrote:
>
> On 19/03/19 10:03, Graeme Fitzpatrick wrote:
>>
>> On Tue, 19 Mar 2019 at 07:47, Markus  wrote:
>>>
>>> Unfortunately, many tags are used wrongly (e.g. name, access tags,
>>> grass, water park) – do you want to replace all of them?
>>
>> Yes!!! :-)
>
> All those used incorrectly should be corrected. If that means deletion .. so 
> be it. But those used correctly should be left.

So why not correcting those 850 (6%) incorrect uses of
cycleway*=opposite_lane instead of inventing a new tagging system for
it? I've already corrected a few dozen, here's the link if you want to
help me:

http://overpass-turbo.eu/s/H91

Regards

Markus

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Warin]

On 19/03/19 10:03, Graeme Fitzpatrick wrote:



On Tue, 19 Mar 2019 at 07:47, Markus > wrote:



Unfortunately, many tags are used wrongly (e.g. name, access tags,
grass, water park) – do you want to replace all of them?


Yes!!! :-)



All those used incorrectly should be corrected. If that means deletion 
.. so be it. But those used correctly should be left.


Many name tags are used as description .. just change the key to 
description. I have even changed a note to description as the 
information is usefull to the map user!



___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Graeme Fitzpatrick]

On Tue, 19 Mar 2019 at 07:47, Markus  wrote:

>
> Unfortunately, many tags are used wrongly (e.g. name, access tags,
> grass, water park) – do you want to replace all of them?
>

Yes!!! :-)

Thanks

Graeme
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Markus]

On Mon, 18 Mar 2019 at 19:54, "Christian Müller"  wrote:
>
> What is the "normal traffic flow" of a two-way road?
>
> After all, opposite_lane made it to be used in combination
> with these as well - despite the fact that [1] documents
> combination with oneways exclusively.  (And because of that
> [1] presumably spares details on right/left-handedness of
> traffic systems.)

I've looked at several cycleway*=opposite_lane w/o oneway=yes: the
majority were former one-way roads where people forgot to adapt the
cycleway tags.

> The problem is that the value has moved out of that solid
> definition and -in the wild- is used in different context.

Unfortunately, many tags are used wrongly (e.g. name, access tags,
grass, water park) – do you want to replace all of them?

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Volker Schmidt]

These tags need a revision of definition to eliminate ambiguous or unclear
definitions (selecting as far as possible the mainstream practises) and
then a careful plan on how to retag the remaining cases. Main problem will
be that armchair mapping will in many cases not work for lack of street
level photos.

On Mon, 18 Mar 2019, 19:54 "Christian Müller",  wrote:

> What is the "normal traffic flow" of a two-way road?
>
> After all, opposite_lane made it to be used in combination
> with these as well - despite the fact that [1] documents
> combination with oneways exclusively.  (And because of that
> [1] presumably spares details on right/left-handedness of
> traffic systems.)
>
> The problem is that the value has moved out of that solid
> definition and -in the wild- is used in different context.
>
>
> Greetings
>
> > On 03/18/19, 13:34, Markus wrote:
> >
> > Maybe backward_lane would have been a little clearer, but the meaning
> > of cycleway=opposite_lane ("a cycling lane going in the opposite
> > direction to normal traffic flow") is clearly defined on its tag page
> > [1].
> >
> > [1]: https://wiki.openstreetmap.org/wiki/Key:cycleway
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Christian Müller]

What is the "normal traffic flow" of a two-way road?

After all, opposite_lane made it to be used in combination
with these as well - despite the fact that [1] documents
combination with oneways exclusively.  (And because of that
[1] presumably spares details on right/left-handedness of
traffic systems.)

The problem is that the value has moved out of that solid
definition and -in the wild- is used in different context.


Greetings

> On 03/18/19, 13:34, Markus wrote:
> 
> Maybe backward_lane would have been a little clearer, but the meaning
> of cycleway=opposite_lane ("a cycling lane going in the opposite
> direction to normal traffic flow") is clearly defined on its tag page
> [1].
> 
> [1]: https://wiki.openstreetmap.org/wiki/Key:cycleway

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

[Tagging] Do we still need cycleway=opposite_lane? (Was: Do we still need cycleway=opposite?)

[Markus]

On Mon, 18 Mar 2019 at 11:41, Martin Koppenhoefer
 wrote:
>
> I disagree, cycleway:left=opposite_lane is clearly about a bicycle lane in an 
> "opposite" direction, but it remains unclear to what "opposite" refers 
> (direction of OSM way or direction you would expect from the jurisdiction), 
> and it doesn't make it clear whether this has oneway implications (the wiki 
> says so, but the data doesn't support it, see below).

Maybe backward_lane would have been a little clearer, but the meaning
of cycleway=opposite_lane ("a cycling lane going in the opposite
direction to normal traffic flow") is clearly defined on its tag page
[1].

[1]: https://wiki.openstreetmap.org/wiki/Key:cycleway

> There seems to be more confusion about this tag in the wiki, e.g. here 
> https://wiki.openstreetmap.org/wiki/Tag:cycleway:right%3Dlane I read that 
> cycleway=opposite_lane is used for locations with 2 bike lanes (trail is to 
> mean "lane" on that page?)?

That was obviously wrong and i've corrected it. [2]

[2]: 
https://wiki.openstreetmap.org/w/index.php?title=Tag%3Acycleway%3Aright%3Dlane&type=revision&diff=1822549&oldid=1608868

> Looking at actual data it seems more than 10% of the roads with this tag also 
> have an explicit oneway=no and roughly one third (!) do not have any oneway 
> tag, which directly contradicts the definition.

I've just done some overpass turbo queries and came to completely
different numbers:

cycleway=opposite_lane: 9,210
cycleway=opposite_lane + oneway=yes: 8,645
--> 6% of cycleway=opposite_lane are missing oneway=yes

cycleway:right=opposite_lane: 3,626
cycleway:right=opposite_lane + oneway=yes: 3,414
--> 6% of cycleway:right=opposite_lane are missing oneway=yes

cycleway:left=opposite_lane: 956
cycleway:left=opposite_lane + oneway=yes: 883
--> 8% of cycleway:left=opposite_lane are missing oneway=yes

All cycleway*=opposite_lane: 13,792
All cycleway*=opposite_lane + oneway=yes: 12,942
--> 6% of all cycleway*=opposite_lane are missing oneway=yes

This isn't very good, but i doubt that it gets better with an
additional and cryptic cycleway:left:oneway=-1 tag.

Regards

Markus

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging