Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

classic Classic list List threaded Threaded
60 messages Options
123
Reply | Threaded
Open this post in threaded view
|

Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Nick Bolten
Hi everyone!

I have two proposals out regarding the crossing tag and how it is not orthogonal, leading to all kinds of issues in mapping crossings and later interpreting that data. As currently written, if both proposals were accepted, crossing=traffic_signals/uncontrolled/unmarked would become two tags: crossing=marked/unmarked and crossing:signals=yes/no.

Both Tobias and Masteuz have made an interesting suggestions about crossing=marked/unmarked, which is that it still has the problem of declaring that a crossing has a "type" (marked or unmarked) whereas it could be considered another attribute, just like having traffic signals.

To give background, I initially chose crossing=marked/unmarked because (1) both are in use in the wild, (2) the schema is equally non-ambiguous, and (3) if I had to decide on the "type" of a crossing, I'd separate those with no indication of their presence aside from regionally-varying conventions (which is currently mapped as crossing=unmarked) from all the rest. But point 3 isn't completely true: a crossing that has only signals but no clear ground markings is less abstract/"fictitious" than a crossing established solely by convention, with no infrastructure saying where to cross.

In contrast, crossing:markings=yes/no would let us avoid making decisions about the "type" of crossing entirely. If it were swapped out for the crossing=marked/unmarked proposal, it would result in this schema for crossings:

crossing=no (for crossings that should be specifically called out as not doable/allowed)
crossing:markings=yes/no
crossing:signals=yes/no
crossing_ref=* (unchanged)

There has also been the suggestion that crossing=* could be left unchanged, and these two new tags added as alternatives. I like that this potentially avoids conflict and therefore makes it easier to start mapping this data separately, but think it would result in competing schemas and redundant data.

So, what are you thoughts? Is crossing:markings=yes/no better than crossing=marked/unmarked? Are there any downsides/upsides I've missed? If crossing:markings were preferable, what should happen to the crossing=* tag?

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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

osm.tagging

This is not in line with hat others have suggested, and invalidating 2.5 million existing crossing=* tags (everything with a value different from yes/no) is a complete no go.

 

As you said, what others suggested, and what would be a welcome addition, is to leave the existing tag untouched (it seems to work fine for most people except you), and tag the special exception where a crossing=traffic_signals doesn’t have road markings with crossing:markings=no

 

What can be done here is to basically define that the different crossing=* values imply default values for various other tags (the same way as the wiki currently already documents what e.g. crossing=zebra or crossing=pelican implies).

 

 

From: Nick Bolten <[hidden email]>
Sent: Saturday, 25 May 2019 03:55
To: Tag discussion, strategy and related tools <[hidden email]>
Subject: [Tagging] Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

 

In contrast, crossing:markings=yes/no would let us avoid making decisions about the "type" of crossing entirely. If it were swapped out for the crossing=marked/unmarked proposal, it would result in this schema for crossings:

 

crossing=no (for crossings that should be specifically called out as not doable/allowed)

crossing:markings=yes/no

crossing:signals=yes/no

crossing_ref=* (unchanged)

 

There has also been the suggestion that crossing=* could be left unchanged, and these two new tags added as alternatives. I like that this potentially avoids conflict and therefore makes it easier to start mapping this data separately, but think it would result in competing schemas and redundant data.

 

So, what are you thoughts? Is crossing:markings=yes/no better than crossing=marked/unmarked? Are there any downsides/upsides I've missed? If crossing:markings were preferable, what should happen to the crossing=* tag?


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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Paul Allen

On Fri, 24 May 2019 at 20:06, <[hidden email]> wrote:
 

As you said, what others suggested, and what would be a welcome addition, is to leave the existing tag untouched (it seems to work fine for most people except you), and tag the special exception where a crossing=traffic_signals doesn’t have road markings with crossing:markings=no


I think this is the nub of the issue: what is meant by crossing markings.  I think Nick's interpretation
is different from that of some on this list.  However, your paragraph seems to conform to Nick's
interpretation.  What do you mean by a crossing with traffic signals AND with road markings?

Hint: crossing=unmarked is defined as being a crossing without road markings or traffic
lights.  Have you ever seen a crossing with lights AND zebra stripes?  Which of the two takes
precedence?  Motorists have right of way if their signal is green; pedestrians have absolute
right of way just by stepping on the crossing irrespective of the lights.  Does not compute.

However, if you include the zig-zag lines before and after the crossing that do NOT define
the interaction of pedestrian and motorist but impose conditions on the motorist alone (cannot
park, cannot wait, cannot load or unload, etc) as being crossing_markings=yes then you have
the dangerous situation that the map leads people to think that a light-controlled crossing
(pedestrians and motorists are controlled by the lights) is a marked crossing (like a zebra)
where pedestrians have priority.  See the problem?  But I suspect this is Nick;s interpretation
of what a marked crossing is - there are some marks on the road (I can't make sense of his
proposals without that interpretation).

I don't consider the zig-zag markings before or after the crossing to be relevant to tagging the
crossing.  Any more than I consider a white line down the centre of the road to mean that it's
a marked crossing.  Those markings do not define pedestrian/motorist interaction.

I agree with Nick (that will surprise him) that these things matter.  Somebody with macular
degeneration may have lost all of their central vision.  It may be far easier to spot a zebra
stripe than to see the lights on crossing signals because of relative sizes.  In fact, you don't
even have to see the stripes, just know that they are there, because pedestrians have priority.
That's why it's a bad idea to tag in a way that could lead somebody to conclude that a crossing
with signals is a marked crossing.  Instead of hunting for the button and listening for the signal,
they'll just step into the road knowing (incorrectly) that traffic will stop for them.

Could we make the tagging more explicit?  For sure.  Could we improve the documentation?  Yep.
Should we say that light-controlled crossings are marked?  Nope.  traffic_signals and marking
are NOT orthogonal, they are mutually exclusive alternatives.  Well, in the UK they are - it's possible
there's some country where you can have  zebra-light-controlled crossings.

--
Paul


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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Mateusz Konieczny-3

24 May 2019, 21:52 by [hidden email]:
Have you ever seen a crossing with lights AND zebra stripes? 
This is a very popular situation in Poland.
Motorists have right of way if their signal is green; pedestrians have absolute
right of way just by stepping on the crossing irrespective of the lights.  Does not compute.
Note that legal implications of zebra stripes differ vastly across the world.
traffic_signals and marking are NOT orthogonal, they are mutually exclusive alternatives. 
Well, in the UK they are - it's possible there's some country where you can have
zebra-light-controlled crossings.
In Poland it is pretty hard to find light-controlled crossings without zebra markings.


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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

osm.tagging
In reply to this post by Paul Allen

The way I see it:

 

crossing=no – crossing here is not legal/possible

 

crossing=unmarked – there are no road markings (or traffic signals) that indicate this is a designated crossing, but based on other factors, it’s a location where pedestrians common cross, e.g. because of lowered kerbs, or because the sidewalk on one side of the road ended

 

crossing=uncontrolled – there are road markings indicating this is a designated pedestrian crossing, but no traffic signals that explicitly tell pedestrians when they have to stop

 

crossing=traffic_signals – there are explicit traffic signals that tell pedestrians when to stop. There are very likely road markings, but even if not, the absence of road markings, in the presence of actual traffic signals, is irrelevant for how this crossing operates.

 

All other crossing=* values that are currently in use are either simply undefined in meaning, or, like the ones listed in the wiki (zebra, pelican, toucan, …) are shorthand for one of the 4 values above + implicit values for additional tags.

 

 

 

From: Paul Allen <[hidden email]>
Sent: Saturday, 25 May 2019 05:53
To: Tag discussion, strategy and related tools <[hidden email]>
Subject: Re: [Tagging] Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

 

 

On Fri, 24 May 2019 at 20:06, <[hidden email]> wrote:

 

As you said, what others suggested, and what would be a welcome addition, is to leave the existing tag untouched (it seems to work fine for most people except you), and tag the special exception where a crossing=traffic_signals doesn’t have road markings with crossing:markings=no

 

I think this is the nub of the issue: what is meant by crossing markings.  I think Nick's interpretation

is different from that of some on this list.  However, your paragraph seems to conform to Nick's

interpretation.  What do you mean by a crossing with traffic signals AND with road markings?

 

Hint: crossing=unmarked is defined as being a crossing without road markings or traffic

lights.  Have you ever seen a crossing with lights AND zebra stripes?  Which of the two takes

precedence?  Motorists have right of way if their signal is green; pedestrians have absolute

right of way just by stepping on the crossing irrespective of the lights.  Does not compute.

 

However, if you include the zig-zag lines before and after the crossing that do NOT define

the interaction of pedestrian and motorist but impose conditions on the motorist alone (cannot

park, cannot wait, cannot load or unload, etc) as being crossing_markings=yes then you have

the dangerous situation that the map leads people to think that a light-controlled crossing

(pedestrians and motorists are controlled by the lights) is a marked crossing (like a zebra)

where pedestrians have priority.  See the problem?  But I suspect this is Nick;s interpretation

of what a marked crossing is - there are some marks on the road (I can't make sense of his

proposals without that interpretation).

 

I don't consider the zig-zag markings before or after the crossing to be relevant to tagging the

crossing.  Any more than I consider a white line down the centre of the road to mean that it's

a marked crossing.  Those markings do not define pedestrian/motorist interaction.

 

I agree with Nick (that will surprise him) that these things matter.  Somebody with macular

degeneration may have lost all of their central vision.  It may be far easier to spot a zebra

stripe than to see the lights on crossing signals because of relative sizes.  In fact, you don't

even have to see the stripes, just know that they are there, because pedestrians have priority.

That's why it's a bad idea to tag in a way that could lead somebody to conclude that a crossing

with signals is a marked crossing.  Instead of hunting for the button and listening for the signal,

they'll just step into the road knowing (incorrectly) that traffic will stop for them.

 

Could we make the tagging more explicit?  For sure.  Could we improve the documentation?  Yep.

Should we say that light-controlled crossings are marked?  Nope.  traffic_signals and marking

are NOT orthogonal, they are mutually exclusive alternatives.  Well, in the UK they are - it's possible

there's some country where you can have  zebra-light-controlled crossings.

 

--

Paul

 


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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Paul Allen
In reply to this post by Mateusz Konieczny-3
On Fri, 24 May 2019 at 21:00, Mateusz Konieczny <[hidden email]> wrote:

24 May 2019, 21:52 by [hidden email]:
Have you ever seen a crossing with lights AND zebra stripes? 
This is a very popular situation in Poland.

I knew there'd be at least one.  :)

Motorists have right of way if their signal is green; pedestrians have absolute
right of way just by stepping on the crossing irrespective of the lights.  Does not compute.
Note that legal implications of zebra stripes differ vastly across the world.

OK, so let me ask this.  Do zebra stripes on their own have any legal significance?  Can
you have zebra stripes without lights or are they only ever present with lights?

If you can have zebra stripes without lights that mean something different to zebra stripes
with lights, that could be a problem for the blind.  But if you can that complicates matters a
lot.  And still means that making markings and lights orthogonal is a bad idea if the markings
have two different meanings depending on whether or not lights are present.  Because then
we need to ensure we distinguish between "these zebra stripes mean pedestrians have
priority" and "these zebra stripes do not mean pedestrians have priority" to avoid potential
mistagging.  If zebras only occur in Poland with lights then it's just a crossing=traffic_lights.
And yes, that would raise problems for people mapping from aerial imagery if the stripes can
have different meanings.

--
Paul



--
Paul



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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Kevin Kenny-3
In reply to this post by osm.tagging
On Fri, May 24, 2019 at 4:09 PM <[hidden email]> wrote:
> The way I see it:

> crossing=no – crossing here is not legal/possible

> crossing=unmarked – there are no road markings (or traffic signals) that indicate this is a designated crossing, but based on other factors, it’s a location where pedestrians common cross, e.g. because of lowered kerbs, or because the sidewalk on one side of the road ended

> crossing=uncontrolled – there are road markings indicating this is a designated pedestrian crossing, but no traffic signals that explicitly tell pedestrians when they have to stop

> crossing=traffic_signals – there are explicit traffic signals that tell pedestrians when to stop. There are very likely road markings, but even if not, the absence of road markings, in the presence of actual traffic signals, is irrelevant for how this crossing operates.

Does any of this change in a jurisdiction where there is an implied
crossing at every intersection unless posted otherwise?

What sort of feature gets tagged crossing=no? Does one draw a line or
node to represent the footway that isn't there?

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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Paul Allen
In reply to this post by osm.tagging
On Fri, 24 May 2019 at 21:09, <[hidden email]> wrote:

The way I see it:

 

crossing=no – crossing here is not legal/possible


Yep.

 

crossing=unmarked – there are no road markings (or traffic signals) that indicate this is a designated crossing, but based on other factors, it’s a location where pedestrians common cross, e.g. because of lowered kerbs, or because the sidewalk on one side of the road ended


Yep.

 

crossing=uncontrolled – there are road markings indicating this is a designated pedestrian crossing, but no traffic signals that explicitly tell pedestrians when they have to stop


Yes, but.  At least in the UK those road markings not only indicate a designated crossing but also
give the pedestiran right of way.  Once the pedestrian places a foot (or a wheel of a buggy or
wheelchair) on the crossing the motorist MUST stop.  If the pedestrian is not on the crossing
the motorist can blithely proceed.

 

crossing=traffic_signals – there are explicit traffic signals that tell pedestrians when to stop. There are very likely road markings, but even if not, the absence of road markings, in the presence of actual traffic signals, is irrelevant for how this crossing operates.


Yep.

That was how I interpreted it all until the Polish contingent threw a spanner in the works.  I'm
waiting for a response to see if it's a big spanner or a little spanner.

 

All other crossing=* values that are currently in use are either simply undefined in meaning, or, like the ones listed in the wiki (zebra, pelican, toucan, …) are shorthand for one of the 4 values above + implicit values for additional tags.


Depending on the answer from Poland, we may have to drastically revise that and explicitly
tag crossing types.  It depends if zebra stripes in Poland are only in conjunction with traffic signals
(cosmetic road markings) or if they can be independent of signals and have different meanings.

--
Paul


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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Mateusz Konieczny-3
In reply to this post by Paul Allen



24 May 2019, 22:10 by [hidden email]:
On Fri, 24 May 2019 at 21:00, Mateusz Konieczny <[hidden email]> wrote:

24 May 2019, 21:52 by [hidden email]:
Motorists have right of way if their signal is green; pedestrians have absolute
right of way just by stepping on the crossing irrespective of the lights.  Does not compute.
Note that legal implications of zebra stripes differ vastly across the world.

OK, so let me ask this.  Do zebra stripes on their own have any legal significance?
Yes - it marks place as a pedestrian crossing with some legal implications
for pedestrians and drivers.
Can you have zebra stripes without lights or are they only ever present with lights?
You can have zebra stripes with lights and without lights.
If you can have zebra stripes without lights that mean something different to zebra stripes
with lights, that could be a problem for the blind. 
Main difference is that without traffic lights who can move depends on situation
(neither drivers nor pedestrians have absolute priority), with lights who can legally enter
is marked by traffic lights.
If zebras only occur in Poland with lights then it's just a crossing=traffic_lights.
Unfortunately - not only.

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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Paul Allen
In reply to this post by Kevin Kenny-3
On Fri, 24 May 2019 at 21:15, Kevin Kenny <[hidden email]> wrote:

Does any of this change in a jurisdiction where there is an implied
crossing at every intersection unless posted otherwise?

In the UK you can legally cross just about anywhere it's physically possible (with the
exception of motorways).  However, there are legal ramifications if you cross without
due care and attention and you are ADVISED to use designated crossings where
feasible.

--
Paul


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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Mateusz Konieczny-3
In reply to this post by Paul Allen



24 May 2019, 22:16 by [hidden email]:
On Fri, 24 May 2019 at 21:09, <[hidden email]> wrote:

crossing=traffic_signals – there are explicit traffic signals that tell pedestrians when to stop. There are very likely road markings, but even if not, the absence of road markings, in the presence of actual traffic signals, is irrelevant for how this crossing operates.


Yep.

That was how I interpreted it all until the Polish contingent threw a spanner in the works.  I'm
waiting for a response to see if it's a big spanner or a little spanner.
AFAIK once traffic lights are present markings are not changing anything (and crossing
with traffic lights without markings are really rare, I suspect that almost always result of worn-out
painting or recent surface reconstruction).


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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Kevin Kenny-3
In reply to this post by Paul Allen
On Fri, May 24, 2019 at 4:11 PM Paul Allen <[hidden email]> wrote:
> On Fri, 24 May 2019 at 21:00, Mateusz Konieczny <[hidden email]> wrote:
>> 24 May 2019, 21:52 by [hidden email]:
>> Have you ever seen a crossing with lights AND zebra stripes?
>> This is a very popular situation in Poland.
> I knew there'd be at least one.  :)

Common in the US as well.

If you scroll south along Route 146/Balltown Road from
https://orthos.dhses.ny.gov/?Extent=-8225903.698222782,5284508.54974362,-8225376.103578491,5284728.604782014&Layers=2018_cache,2017_cache,2016_cache,2015_cache,2014_cache,2013_cache
you'll see a bunch of zebra-striped crossings. They all have traffic
lights with pedestrian-activated signals.

At those, the pedestrian has the right of way but only when the lights
are in the pedestrian's favour. This is still significant, since a
right turn at a red signal is lawful here if it is safe to do so and
the right-turning vehicle has come to a complete stop.

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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

osm.tagging
In reply to this post by Kevin Kenny-3
> Does any of this change in a jurisdiction where there is an implied
> crossing at every intersection unless posted otherwise?

Such purely implied crossings would be crossing=unmarked, and under the "do not map local legislation" rule, I would only map them if they have a physical presence (e.g. lowered kerbs).

> What sort of feature gets tagged crossing=no? Does one draw a line
> or node to represent the footway that isn't there?
It's a tag that should be rarely used, and it's primary purpose is if there is a context in which people may think that there should be crossing here, to indicated that there really isn't one. Mainly to keep armchair mappers from later coming along and thinking "hey, someone forgot to tag the crossing" and add some other crossing=* value.

It's important to mention that crossing=no does NOT get a highway=crossing tag, to prevent data consumer that only look for highway=crossing and not interpret the crossing=* value from wrongly thinking there is a crossing here.



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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Jmapb
In reply to this post by Paul Allen

On 5/24/2019 4:10 PM, Paul Allen wrote:

Have you ever seen a crossing with lights AND zebra stripes? 
This is a very popular situation in Poland.

I knew there'd be at least one.  :)

It's common in the USA too.

OK, so let me ask this.  Do zebra stripes on their own have any legal significance?  Can
you have zebra stripes without lights or are they only ever present with lights?

In *my* experience in the USA, stripes are basically there to give drivers a visual clue to look out for pedestrians and not to block the crosswalk, and thus to inform crossing pedestrians where on the pavement is safest. Of course these marking and the relevant laws are decided on a local level, so officially there may be many differing legal meanings to the stripes.

J


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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

osm.tagging
In reply to this post by Paul Allen

 

 

From: Paul Allen <[hidden email]>
Sent: Saturday, 25 May 2019 06:17
To: Tag discussion, strategy and related tools <[hidden email]>
Subject: Re: [Tagging] Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

 

 

crossing=uncontrolled – there are road markings indicating this is a designated pedestrian crossing, but no traffic signals that explicitly tell pedestrians when they have to stop

 

Yes, but.  At least in the UK those road markings not only indicate a designated crossing but also

give the pedestiran right of way.  Once the pedestrian places a foot (or a wheel of a buggy or

wheelchair) on the crossing the motorist MUST stop.  If the pedestrian is not on the crossing

the motorist can blithely proceed.

 

Yes, I would assume that’s the same in most jurisdictions, designated pedestrian crossings give pedestrians priority.

 

In cases where the exact type of marking is important, that’s what the crossing_ref tag is for, which has to be interpreted under consideration of local legislation.

 

 

Yep.

 

That was how I interpreted it all until the Polish contingent threw a spanner in the works.  I'm

waiting for a response to see if it's a big spanner or a little spanner.

 

All other crossing=* values that are currently in use are either simply undefined in meaning, or, like the ones listed in the wiki (zebra, pelican, toucan, …) are shorthand for one of the 4 values above + implicit values for additional tags.

 

Depending on the answer from Poland, we may have to drastically revise that and explicitly

tag crossing types.  It depends if zebra stripes in Poland are only in conjunction with traffic signals

(cosmetic road markings) or if they can be independent of signals and have different meanings.

 

I think there is a miscommunication there. It’s common in many countries that you have signal controlled crossing, and zebra strips as road markings. That does NOT imply that pedestrians can just walk across at any time and have priority. As long as the traffic lights are present, they take priority, and red means red.

 

I’m pretty sure it’s the same in Poland.


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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Nick Bolten
In reply to this post by osm.tagging
This is not in line with hat others have suggested (...)

I think it's in line with what Mateusz suggested, but sorry if I mischaracterized your ideas.

Also, apologies to you both because I somehow managed to screw up both names.

and invalidating 2.5 million existing crossing=* tags (everything with a value different from yes/no) is a complete no go.

Here's my attempt at restating this as a downside: using crossing:markings=yes/no combined with deprecating crossing=* tags is even worse, as it deprecates even more tags.

As you said, what others suggested, and what would be a welcome addition, is to leave the existing tag untouched (it seems to work fine for most people except you), and tag the special exception where a crossing=traffic_signals doesn’t have road markings with crossing:markings=no

I wouldn't call this a special exception, as traffic_signals does not currently imply markings. Some people on this list kind of sort of say it does, but the wiki doesn't. I don't think any editors do either, but I guess I haven't checked every single one. There's multiple things wrong with the current tagging schema - the others still apply if you leave crossing=* unchanged.

What can be done here is to basically define that the different crossing=* values imply default values for various other tags (the same way as the wiki currently already documents what e.g. crossing=zebra or crossing=pelican implies).

I'm interested in this, in theory, but doesn't it actually imply redefining those 2.5 million tags? Previous mappers were never told these meanings, nor do I think they had them in mind. Redefining those tags post-hoc is actually a harder problem to address via editors / QA / data consumption than deprecation.


On Fri, May 24, 2019 at 12:06 PM <[hidden email]> wrote:

This is not in line with hat others have suggested, and invalidating 2.5 million existing crossing=* tags (everything with a value different from yes/no) is a complete no go.

 

As you said, what others suggested, and what would be a welcome addition, is to leave the existing tag untouched (it seems to work fine for most people except you), and tag the special exception where a crossing=traffic_signals doesn’t have road markings with crossing:markings=no

 

What can be done here is to basically define that the different crossing=* values imply default values for various other tags (the same way as the wiki currently already documents what e.g. crossing=zebra or crossing=pelican implies).

 

 

From: Nick Bolten <[hidden email]>
Sent: Saturday, 25 May 2019 03:55
To: Tag discussion, strategy and related tools <[hidden email]>
Subject: [Tagging] Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

 

In contrast, crossing:markings=yes/no would let us avoid making decisions about the "type" of crossing entirely. If it were swapped out for the crossing=marked/unmarked proposal, it would result in this schema for crossings:

 

crossing=no (for crossings that should be specifically called out as not doable/allowed)

crossing:markings=yes/no

crossing:signals=yes/no

crossing_ref=* (unchanged)

 

There has also been the suggestion that crossing=* could be left unchanged, and these two new tags added as alternatives. I like that this potentially avoids conflict and therefore makes it easier to start mapping this data separately, but think it would result in competing schemas and redundant data.

 

So, what are you thoughts? Is crossing:markings=yes/no better than crossing=marked/unmarked? Are there any downsides/upsides I've missed? If crossing:markings were preferable, what should happen to the crossing=* tag?

_______________________________________________
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: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

osm.tagging

 

> What can be done here is to basically define that the different crossing=* values imply default values for various other tags (the same way as the wiki currently already documents what e.g. crossing=zebra or crossing=pelican implies).

 

I'm interested in this, in theory, but doesn't it actually imply redefining those 2.5 million tags? Previous mappers were never told these meanings, nor do I think they had them in mind. Redefining those tags post-hoc is actually a harder problem to address via editors / QA / data consumption than deprecation.

 

Nothing I said changes the meaning of any existing tags. You seem to be one of very few people that is incapable of understanding the existing tags, and you shouldn’t be projecting your seeming inability to understand them onto all mappers.


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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Kevin Kenny-3
In reply to this post by Kevin Kenny-3
On Fri, May 24, 2019 at 4:22 PM Kevin Kenny <[hidden email]> wrote:

>
> On Fri, May 24, 2019 at 4:11 PM Paul Allen <[hidden email]> wrote:
> > On Fri, 24 May 2019 at 21:00, Mateusz Konieczny <[hidden email]> wrote:
> >> 24 May 2019, 21:52 by [hidden email]:
> >> Have you ever seen a crossing with lights AND zebra stripes?
> >> This is a very popular situation in Poland.
> > I knew there'd be at least one.  :)
>
> Common in the US as well.
>
> If you scroll south along Route 146/Balltown Road from
> https://orthos.dhses.ny.gov/?Extent=-8225903.698222782,5284508.54974362,-8225376.103578491,5284728.604782014&Layers=2018_cache,2017_cache,2016_cache,2015_cache,2014_cache,2013_cache
> you'll see a bunch of zebra-striped crossings. They all have traffic
> lights with pedestrian-activated signals.
>
> At those, the pedestrian has the right of way but only when the lights
> are in the pedestrian's favour. This is still significant, since a
> right turn at a red signal is lawful here if it is safe to do so and
> the right-turning vehicle has come to a complete stop.

Oops, hit 'send' prematurely.

At https://orthos.dhses.ny.gov/?Extent=-8225929.286534256,5288785.407348416,-8225401.691889965,5289005.46238681&Layers=2018_cache,2017_cache,2016_cache,2015_cache,2014_cache,2013_cache,
there are pedestrian- and cyclist-activated flashing amber lights and
signs warning that the crossing is there. There is no other traffic
signal, and motor traffic gives way to the foot and bicycle traffic.
(Well, it's supposed to. With the local drivers....:shrug:)

The zebra stripes are pretty much the same as at the other crossings.

The low-traffic cycleway-unclassified intersection at
https://orthos.dhses.ny.gov/?Extent=-8220641.091134535,5281227.32128579,-8220113.496490244,5281447.376324184&Layers=2018_cache,2017_cache,2016_cache,2015_cache,2014_cache,2013_cache
is marked with signs but no pavement markings.  I'm not sure I'm
comfortable calling it 'unmarked' when there is the signage.

The one at https://orthos.dhses.ny.gov/?Extent=-8214441.788886785,5278866.292470276,-8213914.194242493,5279086.347508671&Layers=2018_cache,2017_cache,2016_cache,2015_cache,2014_cache,2013_cache
has both pavement markings and signs. I can't tell whether it never
had zebra-stripes or whether they've just worn away. I suspect the
former, since https://orthos.dhses.ny.gov/?Extent=-8214441.788886785,5278866.292470276,-8213914.194242493,5279086.347508671&Layers=2018_cache,2017_cache,2016_cache,2015_cache,2014_cache,2013_cache
 https://orthos.dhses.ny.gov/?Extent=-8210714.566462569,5279836.451841815,-8210186.971818278,5280056.50688021&Layers=2018_cache,2017_cache,2016_cache,2015_cache,2014_cache,2013_cache
are  also stripe-less. These, too, are very low traffic. Cars yield to
pedestrians and cyclists at all of them.

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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Paul Allen
In reply to this post by Mateusz Konieczny-3
On Fri, 24 May 2019 at 21:19, Mateusz Konieczny <[hidden email]> wrote:

Can you have zebra stripes without lights or are they only ever present with lights?
You can have zebra stripes with lights and without lights.
If you can have zebra stripes without lights that mean something different to zebra stripes
with lights, that could be a problem for the blind. 
Main difference is that without traffic lights who can move depends on situation
(neither drivers nor pedestrians have absolute priority), with lights who can legally enter
is marked by traffic lights.
If zebras only occur in Poland with lights then it's just a crossing=traffic_lights.
Unfortunately - not only.

OK then, you've said you can have lights without stripes.  Is there any legal difference in
the way pedestrians/motorists interact with and without stripes?  If not, then the stripes are
cosmetic, not functional.

However, they still pose a problem for the blind.  With macular degeneration you might be
able to make out stripes but not see the signals.  Which would mean that without OSM
making a distinction they wouldn't know which type of crossing it was.  But right now they
have that problem anyway, and if it was a serious concern I'd hope Poland would have
done something about it by now.  If I understand correctly, they don't have right of way
in stripes alone and only have right of way at lights if the light changes and/or audible or
other alert, so they're not going to walk into traffic whichever type of crossing they encounter.

--
Paul


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

Re: Non-orthogonal crossing=* tag proposals: crossing=marked/unmarked vs crossing:markings=yes/no

Jmapb
In reply to this post by Jmapb
On 5/24/2019 4:28 PM, Jmapb wrote:

On 5/24/2019 4:10 PM, Paul Allen wrote:

Have you ever seen a crossing with lights AND zebra stripes? 
This is a very popular situation in Poland.

I knew there'd be at least one.  :)

It's common in the USA too.

OK, so let me ask this.  Do zebra stripes on their own have any legal significance?  Can
you have zebra stripes without lights or are they only ever present with lights?

In *my* experience in the USA, stripes are basically there to give drivers a visual clue to look out for pedestrians and not to block the crosswalk, and thus to inform crossing pedestrians where on the pavement is safest. Of course these marking and the relevant laws are decided on a local level, so officially there may be many differing legal meanings to the stripes.

Just to be clear -- zebra stripes occur at both with stop signs and with traffic lights. At a stop sign, pedestrians always have right of way. At traffic lights, pedestrians only have the right of way when obeying the lights.

In some localities zebra stripes may also be used for pedestrian crossings that are specifically signed on the roadway to let drivers know that pedestrians have the right of way at all times.

But unstriped crossings are also used in all of these very same scenarios! So the stripes themselves have no universal legal meaning on their own.

(I'm not aware of anywhere in the USA where there are stripes without traffic signs/signals. I'm sure this exists somewhere but if I saw it I'd think that a sign was missing.)

J


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
123