Roads: Traffic Signals

[dullturtle06]

I would like for most of these crossings to be raised to make it easier for drivers to comply and yield to pedestrians.

Raised crosswalks and intersections are great but not for thru routes. They are used on local or service routes. Each is a tool in the tool box, which are is most effective when used properly. Raised crosswalks on thru routes will created unnecessary congestion thinking roads such as dundas or Jarvis. What toronto transportation should do is identify which roads are for vehicle movement and which are local/services. Then plan road safety and traffic flow to maximize the efficiency of each.

 

[W. K. Lis]

Raised crosswalks and intersections are great but not for thru routes. They are used on local or service routes. Each is a tool in the tool box, which are is most effective when used properly. Raised crosswalks on thru routes will created unnecessary congestion thinking roads such as dundas or Jarvis. What toronto transportation should do is identify which roads are for vehicle movement and which are local/services. Then plan road safety and traffic flow to maximize the efficiency of each.

Having raised sidewalks at entrances to side residential streets would nudge motorists to slow down. The main street would still be level. Consideration however should be looked at rainfall runoff and where the rainwater will flow. Hopefully not collect in ponds caused by the dams at the raised sidewalks or crosswalks.

 

[afransen]

Raised crosswalks and intersections are great but not for thru routes. They are used on local or service routes. Each is a tool in the tool box, which are is most effective when used properly. Raised crosswalks on thru routes will created unnecessary congestion thinking roads such as dundas or Jarvis. What toronto transportation should do is identify which roads are for vehicle movement and which are local/services. Then plan road safety and traffic flow to maximize the efficiency of each.

Raised cross-walks are a good idea on 2 lane roads with low/moderate speeds and volumes (<50 kph). Studies show that driver compliance with yielding at cross-walks drops off dramatically at higher speeds. Encouraging people to use pedestrian cross-overs on wide, higher speed streets is a bad idea if you just have some flashing lights. Full traffic signals like you see on Spadina make more sense on wider streets.

 

[reaperexpress]

Raised crosswalks and intersections are great but not for thru routes. They are used on local or service routes. Each is a tool in the tool box, which are is most effective when used properly.

The profile of raised crosswalks or intersections on thru routes (collector streets and minor arterials) in the Netherlands is much softer than those along local streets, to enable buses to travel over them at up to 30 km/h without getting excessively jostled.

Raised crosswalks are primaily used on through routes because local streets should be traffic calmed to the point that you don't need any marked pedestrian crossings at all, pedestrians can just cross the street wherever, or walk in the street, play hockey in the street etc.

Raised crosswalks on thru routes will created unnecessary congestion thinking roads such as dundas or Jarvis. What toronto transportation should do is identify which roads are for vehicle movement and which are local/services. Then plan road safety and traffic flow to maximize the efficiency of each.

Congestion occurs when demand exceeds capacity.

The capacity of a road is determined by its lowest-capacity intersection. In our case that will almost always be the intersection between two arterial roads. Because at those locations there will be a red light most of the time. You'd need to have an enormous volume of pedestrians (e.g. Yonge & Dundas) before a crosswalk actually limits the road's capacity.

Installing priority crosswalks between major intersections would theoretically reduce congestion by making driving slower, making walking faster (reducing motor traffic demand) without affecting the road's capacity.

 

[dkt]

Not Toronto, sorry, but does anyone here know anything about these new pedestrian crossing signals they installed a few weeks ago at the University Ave/Laurel Trail crossing in Waterloo?

They feel much more responsive from the pedestrian end to me (i.e. often pressing the beg button will immediately turn the vehicle signal yellow when there's no incoming traffic, and the ped light will immediately turn green if the button is pressed while the ION crossing gates are down, which didn't always happen before). Would love to know if there's anything meaningfully different with this new system, and whether these improvements are coming to other pedestrian crossings in Ontario.

When the signalling was rebuilt here, they also reconfigured the crossing to be a much wider separated ped + bike crossing, instead of the narrower shared crosswalk it was before. It looks like they've installed a traffic light (and separate beg button?) for bikes, but it's covered up for now.

I used to cross this intersection several times per day, and I still cross it very regularly. It gets a lot of foot and bike traffic. I'm really glad it was improved, the signal timing used to really bug me.

 

[reaperexpress]

Not Toronto, sorry, but does anyone here know anything about these new pedestrian crossing signals they installed a few weeks ago at the University Ave/Laurel Trail crossing in Waterloo?

They feel much more responsive from the pedestrian end to me (i.e. often pressing the beg button will immediately turn the vehicle signal yellow when there's no incoming traffic, and the ped light will immediately turn green if the button is pressed while the ION crossing gates are down, which didn't always happen before). Would love to know if there's anything meaningfully different with this new system, and whether these improvements are coming to other pedestrian crossings in Ontario.

When the signalling was rebuilt here, they also reconfigured the crossing to be a much wider separated ped + bike crossing, instead of the narrower shared crosswalk it was before. It looks like they've installed a traffic light (and separate beg button?) for bikes, but it's covered up for now.

I used to cross this intersection several times per day, and I still cross it very regularly. It gets a lot of foot and bike traffic. I'm really glad it was improved, the signal timing used to really bug me.

View attachment 501305View attachment 501306

The operation you describe is known as uncoordinated operation. So the signal sits in the main street for some minimum amount of time, after which it responds immediately when someone presses the button.

This is possible with pretty much every traffic signal. It's not an equipment limitation it's just a question of priorities. In most places they don't like it because it can interrupt waves of cars, but it's far better for pedestrians, cyclists and transit priority.

There are some uncoordinated bicycle/ped signals in Toronto (e.g. Finch Hydro trail at Norfinch, Grantbrook and at Talbot), but they programmed a main street pedestrian signal in the timings even though there is no pedestrian signal across the path. So when it detects a pedestrian or bike, it waits 5 seconds for the pedestrian countdown on the non-existant pedestrian signal.

All it would take to resolve this issue is to change the main street recall mode from "pedestrian recall" to "vehicle recall", which is a matter of pressing a single button. But the city doesn't do this because it's not in their procedures. So at all uncoordinated pedestrian/bicycle signals in Toronto, pedestrians and cyclists always need to wait 5 seconds for no reason whatsoever.

 

[W. K. Lis]

If the automobiles getting an automatic reading by cameras or radar, why can't the same be done for pedestrians? Or are they afraid raccoons, deer, dogs, coyotoes, etc. may trigger them?

 

[junctionist]

There are plenty of intersections in Toronto where you have to wait at least a minute if you press the "beg" button.

I've never seen an implementation of the system locally where if the traffic light for cars is green but the pedestrian signal is "don't walk" (because no one pressed the button before the drivers got the green), it will automatically turn on the "walk" signal during that same light cycle. The pedestrian always has to wait an additional two light cycles to legally cross in that situation, which is an unacceptable delay that drivers don't have to deal with.

These kinds of traffic lights are set up in a way that marginalizes pedestrians in most Ontario municipalities.

 

[reaperexpress]

There are plenty of intersections in Toronto where you have to wait at least a minute if you press the "beg" button.

I've never seen an implementation of the system locally where if the traffic light for cars is green but the pedestrian signal is "don't walk" (because no one pressed the button before the drivers got the green), it will automatically turn on the "walk" signal during that same light cycle. The pedestrian always has to wait an additional two light cycles to legally cross in that situation, which is an unacceptable delay that drivers don't have to deal with.

These kinds of traffic lights are set up in a way that marginalizes pedestrians in most Ontario municipalities.

It is possible with our existing signal equipment to allow the Walk signal to come on after the green has already started using a feature called a "permissive period". Basically the cutoff time for pedestrian calls is kept as late as possible, so if there's enough time left within the maximum green time to fit the Walk + Flashing Don't Walk, a pedestrian call can be served right away.

Here's an example presented by Jerry Schippa from the City of Madison, WI. He calls it "pedestrian reservice".

For minor streets, the maximum green duration is typically equal to the pedestrian crossing time, since that's already far more green time than required to clear the typical queue of vehicles. In which case regardless of how the signal is programmed, there will no longer be enough time to serve the pedestrian phase after the light has turned green. To make pedestrian reservice work, you need to increase the maximum green time by the equivalent of the desired permissive period. Jerry increases it by a couple seconds to account for the time it takes to walk to the button. After these couple seconds, the signal won't immediately respond to pedestrian calls, because there isn't enough time anymore.

For major streets, the policy in Toronto is that the signal should rest in Walk until there's actually some specific reason for the light to end (either a timed end-of-green moment or someone detected on the side street). In which case there would only be Green + Don't Walk if there is currently a TTC bus/streetcar extending the green light (and the signal therefore doesn't know exactly when the green will end). But in some other places, including York Region, signals can also rest in Green/Don't Walk, which works really well at intersections with few pedestrians. In that situation, when someone is detected on the side street, the main street signal can immediately change to yellow, whereas in Toronto the person on the side street would first need to wait for the main street pedestrian countdown. If a pedestrian is detected on the main street while the signal is resting in Green / Don't Walk, the pedestrian signal can change immediately to Walk.

There's an example of this right at the start of this video of mine (filmed at YMCA Blvd & University Blvd near Unionville GO Station).

There is some debate in the traffic engineering community about whether the Walk signal should be allowed to start after the green has already started for a movement with a permissive conflict with the crosswalk (e.g. right turns on green are permitted). The concern is that someone turning right could see a pedestrian standing still or even see the "Don't Walk" signal and assume that they can turn without yielding, but then the pedestrian gets a Walk light and steps out in front of the turning vehicle which wasn't expecting to stop. In the Netherlands, it is fairly common for signals with permissive right turns to include the constraint that the Walk or Bike signals can only start simultaneously or earlier than the parallel vehicle signal. If the signal is resting in Green/Don't Walk and a pedestrian/bike requests the main street green, the signal will first go to yellow, then red along the main street, before the ped/bike signals can start and the main street signal goes back to green.

While there is some logical justification for this constraint, I'm not convinced that it actually improves safety overall. There is a signal in a Dutch city which I programmed where there are no right turn signals along the main street, so the City required the bike and ped signals to start simultaneously or earlier than the parallel vehicle signals. No problem I thought, we'll just make the bike signal turn green every time the parallel vehicle signal turns green (like it does virtually everywhere in Toronto). But no, the City insisted that the bike signal should only turn green if there is a bike detected, because the red clearance time for bikes is (marginally) longer than the clearance time for cars. As a result, most of the time cyclists arrive and get stopped at a red light for no apparent reason. And after seeing that the parallel traffic has a green, they just ride through the red light.

Traffic signals only work when people obey them, and people only obey them when the signals seem reasonable. Making pedestrians or cyclists wait for no apparent reason will reduce signal compliance, which I suspect degrades safety more than allowing the Walk light to begin while a permissive conflict already has a green.

 

[mistersg1]

What you are proposing as I understand is for pedestrians in a semi-actuated signal is for the green/don’t walk “rest” on the main road to instantly change to amber on the push of the pedestrian call button?

I always thought that “rest”, which you often see in suburbia, was for a dedicated time when pedestrians (in theory) shouldn’t be in the intersection to allow left or right turns to happen more easily.


Also, if the main road light is programmed to end instantly at the push of the button, I assume this means that the loops on the side street will also change instantly? The problem with this scenario is that you can run into situations where the side street light will cycle to green when there’s no reason to, such as when a right on red vehicle from a side street makes a right turn. In the loop on the side street, the vehicle will cause an amber on the main road. While in the current situation, the vehicle that makes the right turn on red will start the countdown, if the vehicle can get out of the loop by the end of the countdown, the main road signal will revert back to green/walk.

Another thing to consider, here in Brampton, if we’re talking about a fast road, like say Airport Road which mostly has a speed limit of 80km/h, the flashing don’t walk signal acts as a kind of pre-amber signal to drivers warning that the light could change to yellow and thus to prepare to stop. It is somewhat jarring to have to be going at a quick pace, say 80km/h, and thus have to jam hard on the brakes when the amber seemingly appears out of nowhere

I know I may have trouble articulating what I’m trying to say and it’s my two cents on the matter.

EDiT: There’s one other problematic scenario that I missed with holding on green/don’t walk. Presuming the light stays green like a semi actuated signal, how does the pedestrian travelling on the main road crossing the side road make it through the sidewalk safely. Because any push of the call button or a car activating a loop on the side road would trigger an amber on the main road and the pedestrian trapped.

The only safe option the pedestrian would have in this situation would be to push the button to cross the main road (travel on side road) and wait until the pedestrian fades a walk signal on the main road.

 

[W. K. Lis]

What you are proposing as I understand is for pedestrians in a semi-actuated signal is for the green/don’t walk “rest” on the main road to instantly change to amber on the push of the pedestrian call button?

I always thought that “rest”, which you often see in suburbia, was for a dedicated time when pedestrians (in theory) shouldn’t be in the intersection to allow left or right turns to happen more easily.


Also, if the main road light is programmed to end instantly at the push of the button, I assume this means that the loops on the side street will also change instantly? The problem with this scenario is that you can run into situations where the side street light will cycle to green when there’s no reason to, such as when a right on red vehicle from a side street makes a right turn. In the loop on the side street, the vehicle will cause an amber on the main road. While in the current situation, the vehicle that makes the right turn on red will start the countdown, if the vehicle can get out of the loop by the end of the countdown, the main road signal will revert back to green/walk.

Another thing to consider, here in Brampton, if we’re talking about a fast road, like say Airport Road which mostly has a speed limit of 80km/h, the flashing don’t walk signal acts as a kind of pre-amber signal to drivers warning that the light could change to yellow and thus to prepare to stop. It is somewhat jarring to have to be going at a quick pace, say 80km/h, and thus have to jam hard on the brakes when the amber seemingly appears out of nowhere

I know I may have trouble articulating what I’m trying to say and it’s my two cents on the matter.

EDiT: There’s one other problematic scenario that I missed with holding on green/don’t walk. Presuming the light stays green like a semi actuated signal, how does the pedestrian travelling on the main road crossing the side road make it through the sidewalk safely. Because any push of the call button or a car activating a loop on the side road would trigger an amber on the main road and the pedestrian trapped.

The only safe option the pedestrian would have in this situation would be to push the button to cross the main road (travel on side road) and wait until the pedestrian fades a walk signal on the main road.

Brampton is a "suburban city" of stroads, built for the automobile, SUVs, and private pickup trucks. Not for pedestrians. Not for cyclists. The stroads are designed for the "safety" of speeders, not for the safety of pedestrians. They may change the speed limit signs downward, but the stroads remain designed for high speeds, so they do.

The stroads are very wide, without pedestrian refugee islands in the middle. Many of the so called "safety" islands don't even have beg buttons to activate the pedestrian signals if the pedestrians get caught half way.

Brampton just copied what "worked" in America. We should be copying what works in Europe for all traffic signals; pedestrian, motor vehicles, and cyclists. Starting with locating f traffic signal on the nearside of intersections, instead of the farside of intersections like we have here. See the video at this link.

 

[mistersg1]

Brampton is a "suburban city" of stroads, built for the automobile, SUVs, and private pickup trucks. Not for pedestrians. Not for cyclists. The stroads are designed for the "safety" of speeders, not for the safety of pedestrians. They may change the speed limit signs downward, but the stroads remain designed for high speeds, so they do.

The stroads are very wide, without pedestrian refugee islands in the middle. Many of the so called "safety" islands don't even have beg buttons to activate the pedestrian signals if the pedestrians get caught half way.

Brampton just copied what "worked" in America. We should be copying what works in Europe for all traffic signals; pedestrian, motor vehicles, and cyclists. Starting with locating f traffic signal on the nearside of intersections, instead of the farside of intersections like we have here. See the video at this link.

It would be nice if you could actually respond to what I was saying rather than the condescending hostile reply you just gave.

I was responding to what the actual traffic engineer (which I presume he is with the terminology he uses) said in regards to this.

You said absolutely nothing to help the discussion.

 

[reaperexpress]

What you are proposing as I understand is for pedestrians in a semi-actuated signal is for the green/don’t walk “rest” on the main road to instantly change to amber on the push of the pedestrian call button?

Yes.

I always thought that “rest”, which you often see in suburbia, was for a dedicated time when pedestrians (in theory) shouldn’t be in the intersection to allow left or right turns to happen more easily.

No, that's not what it's for. If that were the intention they would provide a green right turn arrow. Green/Solid Don't Walk typically occurs because the signal is extending in real time based on vehicle detection. It can't extend the pedestrian signal in real time based on approaching vehicles because there would be pedestrian countdown between the last vehicle being detected and the light changing to yellow, making the vehicle detection virtually useless.

Also, if the main road light is programmed to end instantly at the push of the button, I assume this means that the loops on the side street will also change instantly? The problem with this scenario is that you can run into situations where the side street light will cycle to green when there’s no reason to, such as when a right on red vehicle from a side street makes a right turn. In the loop on the side street, the vehicle will cause an amber on the main road. While in the current situation, the vehicle that makes the right turn on red will start the countdown, if the vehicle can get out of the loop by the end of the countdown, the main road signal will revert back to green/walk.

No. All detector inputs have a setting called "delay", which delays the call to the controller for a certain duration. In York Region they put a 10 second delay on detectors in lanes where a right turn on red is permitted so that the light doesn't change for someone who turning right immediately after arriving.

Another thing to consider, here in Brampton, if we’re talking about a fast road, like say Airport Road which mostly has a speed limit of 80km/h, the flashing don’t walk signal acts as a kind of pre-amber signal to drivers warning that the light could change to yellow and thus to prepare to stop. It is somewhat jarring to have to be going at a quick pace, say 80km/h, and thus have to jam hard on the brakes when the amber seemingly appears out of nowhere

The pedestrian signal is not a signal for drivers. It is a signal for pedestrians. The amber interval for vehicle signals is determined such that it is sufficient for drivers to react to the start of the amber and reach the stop line if they are too close to stop considering the speed limit. Drivers do not need to jam on the brakes - the amber duration is based on a typical routine deceleration rate. If additional warning is desired, a flashing "prepare to stop when flashing" sign can be installed upstream which flashes just prior to the start of yellow.

Expecting the pedestrian signal to act as a signal for drivers is very problematic, because the two signals don't necessarily need to correspond with each other. For example, at high speed intersection approach, there will often be a detector about 4.5 seconds in advance of the intersection which holds the green for 3 seconds to prevent the light from changing to yellow while there is a vehicle in the 'dillemma zone' (where it's ambiguous whether to stop or not). This will produce Green + Don't Walk after the end of the countdown. Trying to guess when the light will change based on the pedestrian countdown will likely cause drivers to speed up or slow down, undermining the effectiveness of the dilemma zone protection.

Please stop expecting the pedestrian signal to be a signal for drivers. That is not what it is.

EDiT: There’s one other problematic scenario that I missed with holding on green/don’t walk. Presuming the light stays green like a semi actuated signal, how does the pedestrian travelling on the main road crossing the side road make it through the sidewalk safely. Because any push of the call button or a car activating a loop on the side road would trigger an amber on the main road and the pedestrian trapped.

No. The pedestrian on the main road presses the pedestrian button and immediately receives a Walk indication, followed by the required Flashing Don't Walk clearance time. See the example from York Region I showed at the start of the video I linked in my earlier post. This is literally the reason we provide pedestrian signals in the first place: it takes longer for a pedestrian cross than the amber and red clearance time available for vehicles, so we provide them with a dedicated signal which guarantees them enough time to cross the street.

The only safe option the pedestrian would have in this situation would be to push the button to cross the main road (travel on side road) and wait until the pedestrian fades a walk signal on the main road.

Why would someone travelling along the main road suddenly choose to cross perpendicular to their intended direction? Even if you assume that the Walk light can't come on after the parallel vehicle phase has turned green (which is not what I'm suggesting), surely someone travelling along the main street would just press the main street pedestrian button and wait for the main street signal to change to Walk.

 

[mistersg1]

I think I was somewhat misunderstood in the point I was trying to make in my post. Sorry for taking so long to reply.

The purpose of my original post was in response to one specific portion of your earlier post.

For major streets, the policy in Toronto is that the signal should rest in Walk until there's actually some specific reason for the light to end (either a timed end-of-green moment or someone detected on the side street). In which case there would only be Green + Don't Walk if there is currently a TTC bus/streetcar extending the green light (and the signal therefore doesn't know exactly when the green will end). But in some other places, including York Region, signals can also rest in Green/Don't Walk, which works really well at intersections with few pedestrians. In that situation, when someone is detected on the side street, the main street signal can immediately change to yellow, whereas in Toronto the person on the side street would first need to wait for the main street pedestrian countdown. If a pedestrian is detected on the main street while the signal is resting in Green / Don't Walk, the pedestrian signal can change immediately to Walk.

Just so we don't get confused which I think we did last time, let's say that the main street is Islington, and the side street is Jutland. Thus, a pedestrian who wishes to CROSS Islington is travelling on Jutland, while the pedestrian who wants to CROSS Jutland is travelling on Islington. The Islington/Jutland intersection is simple in that it's a simple RYG intersection with three signal heads, no protected turns or anything.

Therefore, Islington's phase, if we are going to have a rest on a steady "Don't Walk" would have three separate pedestrian "phases", which are of course

1. Green + Solid WALK

2. Green + Flashing DONT WALK (Countdown optional)

3. Green + Steady DONT WALK

Normally of course, signals rest in Phase 1 indefinitely unless the pedestrian call button (on the side street (Jutland) to cross the main street (Islington)) is pushed OR a vehicle is in the loop on the side street.

The point I was trying to make, if signals naturally rest in Phase 3, how does a pedestrian on Islington cross Jutland? As Islington wouldn't have a call button, (it doesn't need one as it's set up for semi-actuated) therefore my solution I presented was that the pedestrian on Islington would need to push Jutland's call button to get a walk signal again on Islington? (as that's the only way to return to Phase 1 as it would return there after Jutland gets its phase)

Unless you are implying that a call button should be installed on Islington? So if you push that call button, that the signal would return to Phase 1 (presuming Jutland's button isn't pushed and no vehicles are in Jutland's loops?)

On this site especially, people have a serious problem with call buttons and posters almost always refer to them as "beg buttons" as it's seen as having to ask a favour to cross the street. Ideally, wouldn't having less call buttons be more what the pedestrian ideally wants?

The pedestrian signal is not a signal for drivers. It is a signal for pedestrians. The amber interval for vehicle signals is determined such that it is sufficient for drivers to react to the start of the amber and reach the stop line if they are too close to stop considering the speed limit. Drivers do not need to jam on the brakes - the amber duration is based on a typical routine deceleration rate. If additional warning is desired, a flashing "prepare to stop when flashing" sign can be installed upstream which flashes just prior to the start of yellow.

Expecting the pedestrian signal to act as a signal for drivers is very problematic, because the two signals don't necessarily need to correspond with each other. For example, at high speed intersection approach, there will often be a detector about 4.5 seconds in advance of the intersection which holds the green for 3 seconds to prevent the light from changing to yellow while there is a vehicle in the 'dillemma zone' (where it's ambiguous whether to stop or not). This will produce Green + Don't Walk after the end of the countdown. Trying to guess when the light will change based on the pedestrian countdown will likely cause drivers to speed up or slow down, undermining the effectiveness of the dilemma zone protection.

Please stop expecting the pedestrian signal to be a signal for drivers. That is not what it is.

Drivers shouldn't use the pedestrian signal as a signal, sure. But there's many things that can be deduced from the current state of the pedestrian signal.

Heck, even when I did driver's ed, the instructor told me to be careful during a flashing don't walk, as the light could change soon. (This was in the era before countdown was mainstream)


There are a few universal truths (or rather axioms) that one can easily deduce by the state of the pedestrian signal, and I'm sure most drivers do pay attention to it somewhat, especially in suburbia.

1. A circular green along with a white walk signal being displayed means that the green phase will NOT end soon

2. When a countdown is shown, the green phase will NOT end on any number greater than zero.

While it is true, that the zero on the countdown will not necessarily immediately correspond to a yellow signal, the other two situations are always true.



I know that higher speed limits and line of sight lead to longer amber times, as I believe this is laid out in OTM Book 12, one of my main concerns is the rather seriousness from a legal standpoint the amber light holds in Ontario, and thus why it is important when to anticipate when the amber will occur.

Section 144 (15) plainly states that drivers must stop when they see an amber light unless they are unable to do so safely. There is a sort of dilemma zone in the mind of a driver as to when this point of safety exists. At what point is this safety point, does it mean if I can stop, even if I jam on the brakes, that I should do it. I could "safely" stop in that situation, and therefore I must do it.

(15) Every driver approaching a traffic control signal showing a circular amber indication and facing the indication shall stop his or her vehicle if he or she can do so safely, otherwise he or she may proceed with caution. R.S.O. 1990, c. H.8, s. 144 (15).

Advanced warning flashers can help, I understand that, but seeing a white light adjacent to the traffic light, let's me know that I won't have to worry about making a split second decision if I should stop or not.

 

[reaperexpress]

Just so we don't get confused which I think we did last time, let's say that the main street is Islington, and the side street is Jutland. Thus, a pedestrian who wishes to CROSS Islington is travelling on Jutland, while the pedestrian who wants to CROSS Jutland is travelling on Islington. The Islington/Jutland intersection is simple in that it's a simple RYG intersection with three signal heads, no protected turns or anything.

Therefore, Islington's phase, if we are going to have a rest on a steady "Don't Walk" would have three separate pedestrian "phases", which are of course

1. Green + Solid WALK

2. Green + Flashing DONT WALK (Countdown optional)

3. Green + Steady DONT WALK

Normally of course, signals rest in Phase 1 indefinitely unless the pedestrian call button (on the side street (Jutland) to cross the main street (Islington)) is pushed OR a vehicle is in the loop on the side street.

The point I was trying to make, if signals naturally rest in Phase 3, how does a pedestrian on Islington cross Jutland? As Islington wouldn't have a call button, (it doesn't need one as it's set up for semi-actuated) therefore my solution I presented was that the pedestrian on Islington would need to push Jutland's call button to get a walk signal again on Islington? (as that's the only way to return to Phase 1 as it would return there after Jutland gets its phase)

Unless you are implying that a call button should be installed on Islington? So if you push that call button, that the signal would return to Phase 1 (presuming Jutland's button isn't pushed and no vehicles are in Jutland's loops?)

On this site especially, people have a serious problem with call buttons and posters almost always refer to them as "beg buttons" as it's seen as having to ask a favour to cross the street. Ideally, wouldn't having less call buttons be more what the pedestrian ideally wants?

I didn't get confused. What you describe here is the same as what you described in your previous post, and the same as what I described in my response.

For the third time, I ask that you watch this video, since it answers all of your questions..

Drivers shouldn't use the pedestrian signal as a signal, sure. But there's many things that can be deduced from the current state of the pedestrian signal.

Heck, even when I did driver's ed, the instructor told me to be careful during a flashing don't walk, as the light could change soon. (This was in the era before countdown was mainstream)

There are a few universal truths (or rather axioms) that one can easily deduce by the state of the pedestrian signal, and I'm sure most drivers do pay attention to it somewhat, especially in suburbia.

1. A circular green along with a white walk signal being displayed means that the green phase will NOT end soon

2. When a countdown is shown, the green phase will NOT end on any number greater than zero.

While it is true, that the zero on the countdown will not necessarily immediately correspond to a yellow signal, the other two situations are always true.

Your driver's ed instructor was wrong and these are not universal truths.

The light can change to yellow while the Walk signal is on on the left side, if there is a lagging left turn phase in the oncoming direction.

The light can change to yellow while the Flashing Don't Walk countdown has not finished. According to OTM Book 12, section 3.5: "... it is permissible to continue the FDW through the amber or all-red clearance intervals as this may provide additional information or reassurance to crossing pedestrians". Toronto does not currently do this, but it would be a good idea to do so where the side street always has a leading pedestrian interval, to avoid the massively excessive Don't Walk time which currently exists between the end of FDW and the start of the next green across that crosswalk. Massively unrealistic clearance intervals result in non-compliance, which undermines the basic function of a traffic signal.

I know that higher speed limits and line of sight lead to longer amber times, as I believe this is laid out in OTM Book 12, one of my main concerns is the rather seriousness from a legal standpoint the amber light holds in Ontario, and thus why it is important when to anticipate when the amber will occur.

Section 144 (15) plainly states that drivers must stop when they see an amber light unless they are unable to do so safely. There is a sort of dilemma zone in the mind of a driver as to when this point of safety exists. At what point is this safety point, does it mean if I can stop, even if I jam on the brakes, that I should do it. I could "safely" stop in that situation, and therefore I must do it.

(15) Every driver approaching a traffic control signal showing a circular amber indication and facing the indication shall stop his or her vehicle if he or she can do so safely, otherwise he or she may proceed with caution. R.S.O. 1990, c. H.8, s. 144 (15).

Advanced warning flashers can help, I understand that, but seeing a white light adjacent to the traffic light, let's me know that I won't have to worry about making a split second decision if I should stop or not.

Your descriptions of the HTA and the dilemma zone are correct, but that in no way justifies using pedestrian signals when you are a driver. The duration of the amber interval calculated as per OTM Book 12 is based on a routine deceleration rate from the speed limit. You do not need to anticipate when the amber will occur. And if you would need to jam on the brakes to stop by the stop line, then clearly you are beyond the dilemma zone. Trying to anticipate the start of amber using a signal which is not even intended for drivers, let alone predicting their amber intervals, is dangerous and counterproductive. You even mentioned in your post that seeing the white pedestrian signal causes you to ignore the signal which actually applies to you, on the assumption that it won't change to amber.

You should only anticipate amber signals based on signs which specifically warn you of an imminent amber (i.e. the flashing "Prepare to stop" signs).