Re: [Tagging] traffic_signals:lanes? (specific signal types for certain lanes)

[Paul Johnson]

On Wed, Dec 27, 2017 at 4:04 AM, Tom Pfeifer  wrote:
>
>
> As for your continuous green light, you might have always observed that it
> is on, maybe because you have never been there at a time when it might be
> off


In the US, where a light is continuous green, I've only ever seen it
implemented one of two ways:

   - The signal that has the continuous green only has one aspect.  The
   only way it's going dark is in a power outage or the bulb blows.
   - The signal is a three-aspect, but only the green is wired up, and
   there is a sign next to it that explicitly says "This light is always
   green".

I've also seen lights that are continuous red, again, only in two different
ways:


   - The signal that is continuous red is two aspect, the top one is steady
   and the bottom one has a blinker bulb; a sign indicates that you may only
   turn right on flashing red after stopping or no turn on steady red.  The
   only signals I knew that did this has been replaced by a stop sign and a
   right turn only sign.
   - The signal is a red globe and has a right turn only sign or a sign
   that reads "This light does not turn green."  This only makes sense where
   right turn on red after stop is a thing that exists.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] traffic_signals:lanes? (specific signal types for certain lanes)

[Tom Pfeifer]

On 27.12.2017 00:04, yo paseopor wrote:
In this case I think it would be more useful and accurate to separate the left lane a couple of 
meters before it really does and put a traffic signal for this new way with one lane that turns 
left. I think it would be unusefull to put a traffic signal on the other way with continuous green.


Please don't separate lanes earlier. The common consensus is to draw separate ways only when they 
are physically separated. For lanes that means to split them only when a physical barrier begins.


For everything else we have the :lanes attribute. Some of them are quite successful and used heavily 
by data consumers (e.g. turn:lanes), others are more experimental.


Proposals that tried to model specific traffic light behaviour were unsuccessful so far, mostly 
because such behaviour is often adjustable by traffic management and/or difficult to observe.


As for your continuous green light, you might have always observed that it is on, maybe because you 
have never been there at a time when it might be off.


tom

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

Re: [Tagging] traffic_signals:lanes? (specific signal types for certain lanes)

[Albert Pundt]

The situation might not be common, but applying the :lanes methodology to
the key traffic_signals is just a logical extension of existing conventions.

I wasn't aware traffic_signals:continuous_green was so uncommon; I just
noticed it documented on the wiki and figured it must be at least somewhat
recognized, though I'm sure there's occasionally dubious tagging methods
that sometimes get added.

On Tue, Dec 26, 2017 at 11:23 PM, Paul Johnson  wrote:

>
>
> On Tue, Dec 26, 2017 at 5:04 PM, yo paseopor  wrote:
>
>> I will not discuss here if continuous_green would be a realistic value
>> with full possibilities in a future. But according to taginfo [1] there are
>> only 6 nodes around the world.
>> Also I am asking myself: if continuous_green is continuous green
>> really...is it usefull for the map? (because there is no action here, it
>> would be a traffic light, a painting, or some big commercial ad pannel in
>> the middle of the highway).
>>
>> In this case I think it would be more useful and accurate to separate the
>> left lane a couple of meters before it really does and put a traffic signal
>> for this new way with one lane that turns left. I think it would be
>> unusefull to put a traffic signal on the other way with continuous green.
>>
>
> Only similar situation I can think of was where the west end of US 30
> BYPASS met US 30 at the northwest ramp of the St. John Bridge in Portland
> until sometime around 2005.  The arrangement was such that all three
> eastbound lanes of US 30 had a traffic light (two general access lanes and
> a bicycle lane.  Westbound, the two left lanes had a full signal (a general
> access left turn lane and a general access through lane), then there was a
> set of no-lane-change stripes to the right of that, then a general access
> through lane with a permanent green through arrow, and a bicycle lane with
> a permanent green through arrow.
>
> US 30 was then (and still is now) a single carriageway with no separation
> at that location; wouldn't be accurate to add a median where there isn't one
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
>


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

Re: [Tagging] traffic_signals:lanes? (specific signal types for certain lanes)

[Paul Johnson]

On Tue, Dec 26, 2017 at 5:04 PM, yo paseopor  wrote:

> I will not discuss here if continuous_green would be a realistic value
> with full possibilities in a future. But according to taginfo [1] there are
> only 6 nodes around the world.
> Also I am asking myself: if continuous_green is continuous green
> really...is it usefull for the map? (because there is no action here, it
> would be a traffic light, a painting, or some big commercial ad pannel in
> the middle of the highway).
>
> In this case I think it would be more useful and accurate to separate the
> left lane a couple of meters before it really does and put a traffic signal
> for this new way with one lane that turns left. I think it would be
> unusefull to put a traffic signal on the other way with continuous green.
>

Only similar situation I can think of was where the west end of US 30
BYPASS met US 30 at the northwest ramp of the St. John Bridge in Portland
until sometime around 2005.  The arrangement was such that all three
eastbound lanes of US 30 had a traffic light (two general access lanes and
a bicycle lane.  Westbound, the two left lanes had a full signal (a general
access left turn lane and a general access through lane), then there was a
set of no-lane-change stripes to the right of that, then a general access
through lane with a permanent green through arrow, and a bicycle lane with
a permanent green through arrow.

US 30 was then (and still is now) a single carriageway with no separation
at that location; wouldn't be accurate to add a median where there isn't one
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] traffic_signals:lanes? (specific signal types for certain lanes)

[yo paseopor]

I will not discuss here if continuous_green would be a realistic value with
full possibilities in a future. But according to taginfo [1] there are only
6 nodes around the world.
Also I am asking myself: if continuous_green is continuous green
really...is it usefull for the map? (because there is no action here, it
would be a traffic light, a painting, or some big commercial ad pannel in
the middle of the highway).

In this case I think it would be more useful and accurate to separate the
left lane a couple of meters before it really does and put a traffic signal
for this new way with one lane that turns left. I think it would be
unusefull to put a traffic signal on the other way with continuous green.

That's my opinion. What do you think?

Salut i semàfors (Health and traffic signals)
yopaseopor

[1] https://taginfo.openstreetmap.org/search?q=continuous-green#values
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging