Re: [Tagging] [Talk-us] Freeway exit tagging

[Kieron Thwaites]

I can't comment whether this is "proper" usage or not, but I can
illustrate how I've personally tagged lanes at freeway exits:

Take, for example, this freeway interchange close to where I grew up
(note: left-hand drive): https://goo.gl/maps/5eF35SVed452

There's no lane markings to indicate direction itself, but there is a
solid line starting a few hundred metres out and continuing just past
the ramp, indicating no lane changes from lane 2 to lane 1 (where lane
1 is the left-most lane).

There is, however, overhead signage for this particular interchange as follows:

https://goo.gl/maps/ApFqTg7vRQT2 (1 km out)
https://goo.gl/maps/FTtiG5UHaxx (500 m out)
https://goo.gl/maps/VMw9Rcpsua62 (at the ramp)

The sign 500 m out is the key one -- that's the one that has the turn
lane information on this.  As such, were I to add turn lane data in
this area (I wouldn't, as I moved out of that city in 2010 and don't
consider myself to have sufficient local knowledge there any more),
I'd tag as follows:

From 500 m out until the point where the ramp separates:
turn:lanes=slight_left;through|none|none
The section with the painted lane restriction: change:lanes=yes|not_left|yes

Now, this is an urban setting and the overhead signage makes (to me,
anyway!) the tagging rather unambiguous.  In a rural setting however,
all you're likely to get is this:

https://goo.gl/maps/ajR1ccg6g252 (signage 1 km from ramp)
https://goo.gl/maps/NLm9VMNskKK2 (signage at the ramp)
https://goo.gl/maps/6aE1NbsuguG2 (overhead view of the ramp)

There's no explicit lane markings here, but once again, there's a
painted line indicating no lane changes from lane 2 to lane 1 around
the vicinity of the ramp, which implies
turn:lanes=slight_left;through|none (and of course, explicitly
specifies change:lanes=yes|not_left), so even through turn:lanes are
implicit on the ground, I'd explicitly tag it.  I personally see this
as unambiguous as per the urban example.

Now, when it comes to the US example that Jack posted
(http://mapillary.com/map/im/7igAGXSa6EsUYlTIujXchw), it seems like
there's nothing explicit or implicit, either from painted markings
(either arrows or lane change restrictions) or signage (I jumped on
Streetview and the only other sign I noticed was an "exit in 0.5 mi"
sign, similar in nature to my rural "exit in 1 km" sign).  In this
case, I would side with "through;slight_right" -- even though there's
nothing explicitly or implicitly specifying as such, the reality on
the ground appears that one can only exit from the right-most lane
while continuing to continue through, and the remaining lanes are for
through traffic only.

I can, however, see the rationale behind tagging "none;slight_right",
as well as tagging nothing at all, and as such, I think that this is
an issue that we need to find consensus on.  That said, I believe Paul
is quite correct with his statement that machines "need to be told
about these restrictions in order for them to be able to provide
useful feedback from it" -- something that isn't explicitly present
(or maybe not even implicitly so) but appears obvious to a human on
the ground isn't necessary obvious to a machine.

--K

On 26 August 2016 at 08:37, Paul Johnson  wrote:
> On Thu, Aug 25, 2016 at 11:06 PM, David Mease  wrote:
>>
>> Road markings are both beneficial and useful for navigation. Cities and
>> governments have paid a lot of money installing them all over globe
>> precisely for these reasons. OSM would be well served to include them
>> exactly as is. I don't hear a lot of people complaining about how those
>> arrows on the roads led them astray.
>
>
> Arrows on the road, at least in North America, are typically only installed
> to indicate relatively unusual lane restrictions, with the typical lane
> restrictions assumed to be common knowledge.  This is where this trips up
> automation, as machines need to be told about these restrictions in order
> for them to be able to provide useful feedback from it or lane guidance will
> be a NP-complete thing for data consumers to deal with.  I mean, I get it,
> don't tag for the data consumer.  But on the other hand, don't break the
> data consumer with stupid tagging schemes, either.
>
> ___
> 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] [Talk-us] Freeway exit tagging

[yo paseopor]

I think it is so clear. Instead of no painting marks with an arrow, you
have some vertical signs and also no_change lane marks, so for me it is
clear how it works. With all traffic signs, posted I think the junction
will have all the information software needs to process it.
Also I think it is tru there's no turn lane marks...but you can add this
information with turn:lanes so it is not needed to wirte "none" when you
know this lane is for going through.

In Spanish traffic_sign preset menu I have tried some examples of tags with
"Orientation/Confirmation" menú. It will be possible to do it with other
countries.

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

Re: [Tagging] [Talk-us] Freeway exit tagging

[Paul Johnson]

On Fri, Aug 26, 2016 at 7:51 AM, yo paseopor  wrote:

> I think it is so clear. Instead of no painting marks with an arrow, you
> have some vertical signs and also no_change lane marks, so for me it is
> clear how it works. With all traffic signs, posted I think the junction
> will have all the information software needs to process it.
> Also I think it is tru there's no turn lane marks...but you can add this
> information with turn:lanes so it is not needed to wirte "none" when you
> know this lane is for going through.
>

Just to clarify, through would be the correct value if you know the lane is
going through; osmand (and presumably other data consumers) will try to
guess what "none" is, but in my experience, that's a value best reserved
for a lane whose access:lanes value is no...
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] [Talk-us] Freeway exit tagging

[yo paseopor]

I think you can specify the information you know instead this information
is not explicit written on the road. So for me , in my opinion it is better
specify through or whatever will be the direction that none or   |" "
(blank) or something like that.

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

Re: [Tagging] [Talk-us] Freeway exit tagging

[Paul Johnson]

On Fri, Aug 26, 2016 at 8:19 AM, yo paseopor  wrote:

> I think you can specify the information you know instead this information
> is not explicit written on the road. So for me , in my opinion it is better
> specify through or whatever will be the direction that none or   |" "
> (blank) or something like that.
>

Agreed.  If you know (either by regional context/knowledge or by markings
visible) the turns, go ahead and tag it as the explicit direction.  I
suggest the "none" tag be used for lanes that don't go anywhere for some
reason (rare, but consider a permanently closed lane) and a null answer
when it's not clear which way the lane goes (armchair mapping in an
unfamiliar region, for example).
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] [Talk-us] Freeway exit tagging

[Johan C]

It is quite simple, If there is a through arrow, the value is through. If
there is no arrow, the value is none. Groundtruth mapping!

Cheers, Johan

2016-08-26 15:41 GMT+02:00 Paul Johnson :

> On Fri, Aug 26, 2016 at 8:19 AM, yo paseopor  wrote:
>
>> I think you can specify the information you know instead this information
>> is not explicit written on the road. So for me , in my opinion it is better
>> specify through or whatever will be the direction that none or   |" "
>> (blank) or something like that.
>>
>
> Agreed.  If you know (either by regional context/knowledge or by markings
> visible) the turns, go ahead and tag it as the explicit direction.  I
> suggest the "none" tag be used for lanes that don't go anywhere for some
> reason (rare, but consider a permanently closed lane) and a null answer
> when it's not clear which way the lane goes (armchair mapping in an
> unfamiliar region, for example).
>
> ___
> 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] [Talk-us] Freeway exit tagging

[yo paseopor]

If you know which turn:lane it is you can specify it instead there's no
marks. In Spain, with administrations in crisis some of them don't paint
all the lanes (before the crisis they painted all the lanes). So I think we
can explain the destination...if we know it, and if we want to do it (put
this extra information). Truth is not the problem. Ground can change.

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

[Tagging] Question on exit-entrance rendering.

[John Willis]

Shuto Expressway system map, 24mb PDF. 

http://www.shutoko.jp/%7E/media/pdf/customer/use/network/navimap/160701_map_big.pdf

I am wondering about proper motorway entrance-exit tagging when there seems to 
be one tag that handles the intersection (highway=junction) and all the other 
information (ref, name destination) gets put on the junction node or the 
highway=motorway_link way (as I understand it).

The system I have been mapping/ cleaning up in Japan has directionally separate 
entrance-exits with either shared or separated toll plazas, shared or separated 
ETC toll collection, or unmatched entrance-exit pairs.

As a person who drives and sees the mistakes Google and Apple make, I see the 
level of complexity needed to guide a tourist around the system. 

I'm wondering, if motorways are one-way, why there is no specific method of 
tagging that a motorway_link is an "exit" or an "entrance" - it seems to me to 
be a piece of information that could make labeling and rendering a map easier. 

The Shuto expressway in Tokyo is the most complicated system I have ever seen. 
The level of detail provided on the system map is so much greater than a 
Freeway map from California because, besides the complexity of the system and 
the directional restrictions that motorways have because of space constraints, 
the separation and lack of parity of entrance-exit pairs makes figuring out 
where to go to enter or where an exit goes *by looking at a map* quite 
difficult. Even trying to reconcile what the map is showing you and the 
turn-by-turn information is difficult because of the insane density and 
complexity of such a system. 

The system map printed and online has to color the direction of traffic and 
whether an entrance or exit is present at each junction (pink and blue arrows) 
, because of the lack of bi-directional exits or entrances at each junction, 
along with the limited choices at interchanges. 

How how mapping ways are parsed and that information is used to create a 
rendered map, or that additional relational information could be determined to 
make a static rendered map that had enough information to be useful is a 
mystery to me, so maybe everything is fine and I shouldn't worry about this, 
but if I wanted to make a "motorway view" or provide enough information, 
visually, to understand this system with something like -carto renderings, it 
feels like there isn't enough tagging granularity to render it correctly. 

To put another way - as a mapper who is trying to tell others, via tags, what 
this big mess is, not being able to explicitly say "this is an 'up'entrance, 
this is a 'down' entrance,  and this is a 'down' exit, and together they are 
all called "junction C22" seems like a deficiency in tagging. 

(Towards town and away from town are referred to as 'up 上' and 'down 下' 
directions on all transit systems here, they don't use cardinal directions 
[NSEW] for most things) 

This difficulty in labeling is very apparent in Google and Apple maps- where 
labels are sloppily rendered because everything has to have huge official names 
(in Japanese and English) because of the lack of information on how everything 
is connected, so they just cram everything into the name field (sometimes), 
leading to large amounts of redundant rendered data than what is actually 
signed - while not rendering other important things useful to navigation. 

This spills over into service areas and other things that might be more helpful 
rendered as a single important point at Z10 and separate things at Z15, so 
being able to tag everything explicitly (maybe motorway_link=exit) or make them 
some member of a relation (which is abstracted away from me in iD, as it should 
be) is the solution I think of because the parsing and rendering of tagging 
data is a mystery to me (and probably most mappers). 

Is this a problem that could use a solution, or is my lack of understanding the 
parsing leading me to want tags that aren't really needed to correctly render 
such a system? Even if it can be calculated, is it easier and more friendly to 
new taggers to have it explicitly tagged that way?  

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

Re: [Tagging] [Talk-us] Freeway exit tagging

[Georg Feddern]

Am 26.08.2016 um 14:18 schrieb Kieron Thwaites:

I can, however, see the rationale behind tagging "none;slight_right",
as well as tagging nothing at all, and as such, I think that this is
an issue that we need to find consensus on.  That said, I believe Paul
is quite correct with his statement that machines "need to be told
about these restrictions in order for them to be able to provide
useful feedback from it" -- something that isn't explicitly present
(or maybe not even implicitly so) but appears obvious to a human on
the ground isn't necessary obvious to a machine.


Please do not take it personally - I just toke your answer to hook on, 
because you agreed on "machines need to be told about these restrictions".


May be I am a bit on the devils advocates side here ... ;)

1. Where are there any restrictions? There is no solid line between the 
3 lanes. ;)
2. If you want to give the machine any advice, you should take 
"through|through|through;slight_right"

because
3. "none|none|none;slight_right" does not give any advice for the both 
left lines - they still could be considered to take the exit.
Well in reality that is the legal situation here - you just have to take 
care for the traffic on the lines. ;)


But in this common case (standard single lane exit) I still do not see 
any necessarity for any advice to the maschine (or the driver), that if 
the route takes the right road one should use the rightmost lane ...
Same situation with solid lines definitely need case 2 - because even 
you should 'implicitly know' to take the rightmost lane there is another 
point where you already _have_ _to_ be on the rightmost lane - the 
maschine needs this advice to announce it appropriate.


But may be I am a bit too old and have driven too long with my own eyes 
and head - and without a navigation assistant. ;)


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

Re: [Tagging] Question on exit-entrance rendering.

[yo paseopor]

Instead of this "messy map"... I don't think it is so difficult. It is hard
because of all the amount of information you need to describe correctly but
it is "easy". OSM can work with a  3D system, with in layers. Also you have
the incline tag,ele, exit_to, ref, ref:colour, destination: , turn:lanes
... forward, backward, sides (relative to the way is drawn) Are you sure
OSM is missing some tag? Can you create other one?

For example, in Spain we can use:

destination:symbol
distance
turn:destination
destination:ref
destination:ref_int
colour:ref
colour:ref_int
distance
ref
lanes
arrow
colour:arrow
colour:text
name
motorway_junction
exit_to
direction
side
forward
backward

... (other value variables, other key variables...)

I'm sure I'm missing some value or interesting key. Do you need some more?

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

Re: [Tagging] [Talk-us] Freeway exit tagging

[Paul Johnson]

On Fri, Aug 26, 2016 at 11:31 AM, Georg Feddern 
wrote:

> Am 26.08.2016 um 14:18 schrieb Kieron Thwaites:
>
>> I can, however, see the rationale behind tagging "none;slight_right",
>> as well as tagging nothing at all, and as such, I think that this is
>> an issue that we need to find consensus on.  That said, I believe Paul
>> is quite correct with his statement that machines "need to be told
>> about these restrictions in order for them to be able to provide
>> useful feedback from it" -- something that isn't explicitly present
>> (or maybe not even implicitly so) but appears obvious to a human on
>> the ground isn't necessary obvious to a machine.
>>
>
> Please do not take it personally - I just toke your answer to hook on,
> because you agreed on "machines need to be told about these restrictions".
>

No problem, I'm willing to take it at face value for clarifying my position.


> May be I am a bit on the devils advocates side here ... ;)
>
> 1. Where are there any restrictions? There is no solid line between the 3
> lanes. ;)
>

Right, but we're not talking about change:lanes=* here.


> 2. If you want to give the machine any advice, you should take
> "through|through|through;slight_right"
> because
> 3. "none|none|none;slight_right" does not give any advice for the both
> left lines - they still could be considered to take the exit.
>

In the physical sense and a "no cop no foul" sense, sure, under ideal
conditions.  In practice, depending on where you're at, it's going to
garner some variation or combination of improprer lane change, or not
turning from the nearest lane, or cause a collision.  We shouldn't be
helping Mayhem play GPS in https://www.youtube.com/watch?v=4h0Qvc6_MfQ if
we can avoid it, even if it still is the driver's responsibility to not be
an idiot (or trust a driverless car to the data without supervision).  It's
still not helpful or going to gain many new end users or contributing
mappers to make it easy for data consumers to give them incorrect
information due to incomplete information just because the DOT left out a
sign or a pavement arrow.


> Well in reality that is the legal situation here - you just have to take
> care for the traffic on the lines.
>

At the risk of being moderately offensive with the stereotypes, this does
remind me of a Family Guy bit about driving...
https://www.youtube.com/watch?v=LLuaPZWkvZ0

But in this common case (standard single lane exit) I still do not see any
> necessarity for any advice to the maschine (or the driver), that if the
> route takes the right road one should use the rightmost lane ...
> Same situation with solid lines definitely need case 2 - because even you
> should 'implicitly know' to take the rightmost lane there is another point
> where you already _have_ _to_ be on the rightmost lane - the maschine needs
> this advice to announce it appropriate.
>
> But may be I am a bit too old and have driven too long with my own eyes
> and head - and without a navigation assistant. ;)


This isn't about humans driving in familiar territory, it's about
facilitating automated driving or providing cues for drivers unfamiliar
with the area how to best navigate a new region.  Oddly enough, on
particularly well lane-mapped areas, I'm actually quite surprised how close
Osmand is to lane choices a local with extensive experience with an area
would pick.  Omitting things or leaving "none" in there generally throws
off some oddities (like suggesting a spot where there's two marked left
turn lanes, three lanes with no markings and a right turn lane with an
arrow that you can turn left from the five leftmost lanes or right from the
four rightmost).  Rather than expecting data consumers have to have
programmatic assumptions for every jurisdiction, imparting this information
even if it's not explicitly signed or posted yet only debatable in the most
farcical of situations, should be considered ground truth for the purposes
of mapping and not just killing potential developers interested in
providing lane guidance.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] [Talk-us] Freeway exit tagging

[Paul Johnson]

On Fri, Aug 26, 2016 at 9:44 AM, yo paseopor  wrote:

> If you know which turn:lane it is you can specify it instead there's no
> marks. In Spain, with administrations in crisis some of them don't paint
> all the lanes (before the crisis they painted all the lanes). So I think we
> can explain the destination...if we know it, and if we want to do it (put
> this extra information). Truth is not the problem. Ground can change.
>

Indeed I think it's fairly safe to say, explicitly painting more than
channelization lines isn't exactly a high priority for a lot of highway
authorities unless it's not clear by context; I'm a firm believer that the
context should be captured.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] [Talk-us] Freeway exit tagging

[Tod Fitch]

> On Aug 26, 2016, at 10:48 AM, Paul Johnson  wrote:
> 
> On Fri, Aug 26, 2016 at 9:44 AM, yo paseopor  > wrote:
> If you know which turn:lane it is you can specify it instead there's no 
> marks. In Spain, with administrations in crisis some of them don't paint all 
> the lanes (before the crisis they painted all the lanes). So I think we can 
> explain the destination...if we know it, and if we want to do it (put this 
> extra information). Truth is not the problem. Ground can change. 
> 
> Indeed I think it's fairly safe to say, explicitly painting more than 
> channelization lines isn't exactly a high priority for a lot of highway 
> authorities unless it's not clear by context; I'm a firm believer that the 
> context should be captured.

Perhaps we are dealing with a difference in semantics with one group leaning 
toward show what is painted and the other group leaning toward what is meant. 
The wiki is definitely in the “what is painted” camp but the only data consumer 
I am aware of treats it as “what is meant”.

While I’ve been following the wiki in my turn lane tagging, this discussion has 
led me to think I have been wrong: The biggest purpose for putting lane tagging 
in is for driving assistance while following navigation instructions. So the 
tagging should keep this in mind.

Consider a divided roadway with three continuous lanes in each direction, some 
intersections having purpose built turn lanes and others not.

In the parts of the world I drive, in the absence of marked turn lanes or 
specific turn restrictions, you are allowed to turn right from the right most 
lane and left from the left most lane. So implicitly there is an unmarked 
turn:lanes=left;through|through|through;right tagging leading up to each 
intersection that the data consumer should assume even though per the wiki the 
tagging, if it exists at all, will be turn:lanes=none|none|none

And if turn lanes exist, only the actual lane will be marked with an arrow on 
the pavement: Most through lanes are not marked even though turns from them are 
illegal. So the implicit marking needed by the data consumer is 
“left|through|through|through|right” even though the wiki says to tag it as 
“left|none|none|none|right”.

So the questions are:

1. How can the data consumer determine the difference between “none” meaning 
“left;through”, “through” or “through;right”?

2. If the determination of the meaning of “none” is by ad hoc heuristics, do 
those vary by jurisdiction?

While the system used where I drive seems natural and obvious, I’ve come to 
learn that there are significant differences around the world about what is 
considered natural and obvious, so I can’t assume the ad hoc rules that would 
decode “none” into “left;through”, “through” or “through;right” which apply 
where I drive would apply elsewhere. This, I think makes a strong case for 
tagging the intent rather than the paint on the pavement.

So based on this discussion, I will probably start using “through” instead of 
“none” when I am adding or correcting turn lane tagging.




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