157 comments

[ 4.8 ms ] story [ 230 ms ] thread
Imagine if we had the tech to apply speed governors in places with high crash rates! What if departments of transportation could lower speeds dynamically to make sure that risky road infrastructure is made safer.

I can't help but feel a little bit salty about these baindaid solutions to our ever increasing rate of road deaths in the US. I'm not sure what the right solutions are to keeping people safe while allowing them to get where they want to go but the solution of adding yet another notification on an electronic device seems disheartening to me.

An article about increased pedestrian deaths: https://www.npr.org/2023/06/26/1184034017/us-pedestrian-deat...

>>I'm not sure what the right solutions are

Giving the federal government the ability to dynamically control my car based on crash data is 1000000% NOT the right solution....

The black box is bad enough... giving them control HELLLL no

Curious to hear your objections to the black box.

Black box data (speed, steering+pedal input logs) were invaluable in determining fault in an accident involving one of my family members recently. As long as the data doesn't leave the car under normal circumstances, what's the harm?

There are a lot of people that don't want to be held accountable for their negligence and blatant disregard for the safety of others. Driving like you're the main character is so normalized that it will take generations to fix the problem.
I dont think giving the government the power to force by law people record their movement, operating conditions, and other factors is inline with spirit of constitution.

Sure it makes it easy to "determining fault in an accident" but same could be said if everyone was required to wear a body cam, or install surveillance camera's in their home that the police monitor 24x7.

I am opposed to any and all government mandated surveillance, I do not want to live in the surveillance state we do, I sure as hell do not want to see it expanded

I also reject the narrative that another commenter basically claims, one of "if you have nothing to hide you have nothing to fear", that old trope has been proven through out history to be moronic statement

I care very much about privacy. I use only free and open source software on my personal and work computers, as well as my phone and router. I do not use social media, except for this site. I do not connect anything to my network that I do not control. For this reason, I don't have a smart TV, smart speakers, or smart appliances. I drive a car from 2009 with no cellular connection or GPS unit. I agree that the "nothing to hide" argument is weak.

I am opposed to indiscriminate government surveillance. This is not that. I have seen what it looks like when this data is used as evidence in court. The EDR provides information about the driver's last inputs to the car, and it has a very small amount of storage. The data is only accessible through physical access to the unit.

In the event of a fatal accident, there are very few sources of information:

- Survivors (biased)

- Witnesses (not always present)

- Forensics (squishy)

- EDR data

The EDR is a crucial source of truth for determining the cause of an accident. When you drive on busy roads, you are taking others' lives into your hands. In the event of a serious accident, it's critical that the details of the seconds before the accident are available to inform legal decisions. If we remove the EDR, we are left only with less reliable sources of information.

I do not support making black boxes mandatory because it feels like it violates the 5th Amendment. I will also never purchase a car without one, because I think it is more likely to prevent harm than cause it.

> federal government

I had assumed it would be state government since they normally control traffic law?

What other solutions would you propose for enforcing speed/safety regulations on state owned infrastructure? Personally, given my experience interacting with police, hate our current system that relies on being pulled over and interacting with them. I would much prefer an automated solution (e.g. speed governors or automated speeding cameras).

>>I had assumed it would be state government

I dont want them either, but these types of things are really controlled by Federal Highway regulations in practice.

> I would much prefer an automated solution (e.g. speed governors or automated speeding cameras).

Well I have the right to face my accuser, and speed camera's have been ruled a few times to be unconstitutional for that reason. I for one DO NOT want automated speed camera's and much like Red Light Camera's you will find they are not deployed to increase safety but solely for Revenue generation, and often have the impact of REDUCING safety not increasing it.

red Light camera's caused more crashes then they prevented, and speed camera's would do the same

>What other solutions would you propose for enforcing speed/safety regulations on state owned infrastructure?

Well I would start with a reduction of said regulations, and a more rational objective way to not only set the regulations but also appeal them.

LOTS of speed regulations are less about safety and more about setting up revenue streams for towns and cities.

You seem to believe that all of these regulations and government actions are purely out some altruistic motive to protect you the driver. Nothing could be further from the truth. Most of the regulations are designed to

1. Create Revenue stream 2. Give the police a pretense to look for other criminal activity

Safety is maybe 3rd on the list, maybe

in short, i do not trust government, and I do not look to government to protect me.

"Imagine if we had the tech to apply speed governors in places with high crash rates! What if departments of transportation could lower speeds dynamically to make sure that risky road infrastructure is made safer."

This, of course, could be done, but what if people driving fast are rushing to flee a rapidly advancing forest fire, tidal wave, active shooter, etc?

As a check on the ever-growing power of government and surveillance tech, anytime a system is capable of taking away someone's liberty, there should be a human in the loop on the other side to make sure those decisions are reasonable.

Edit: Or just don't take away their liberty at all

Sure, reasonable. But not a this-or-that kind of thing.

Hit a button - override! Be your own boss. There was a fire, everybody will understand.

But you just press that button because you're a type-A dick, well pay the ticket then.

Liberty!

Sounds nice, but eventually, the button is always taken away.
No, the button stays but the price increases so much so only rich assholes can use it.
So... a speed camera?
Curiously, different in a particular way. See, your car is reporting your speed. Self-reported speeding is a different niche.
Maybe you could have a speed governor that can be overridden if the gas petal is fully pressed to the floor. (Fine for gas cars but probably not a good experience in EVs).

I live near a blind turn into an intersection where the speed limit is 50. When there’s heavy fog, it’s deadly. I’ve seen 2 fatal accidents, one with a dump truck and school bus, multiple people died. Another one I witnessed the next year was someone head on hitting a telephone pole. Same intersection. Same conditions (heavy fog).

Right now the only other solution is to reduce the speed limit or to add a red light. Or do nothing and let the accidents continue

How do we balance safety regulations with these emergency situations?

We're experiencing some of the highest traffic fatality rates in decades and I think the liberty of those dying and being injured on our streets need to be taken into account as well.

I prefer the technological solution of a speed governor to automated traffic cameras because I think automated traffic cameras can easily become surveilance devices while speed governors tend to be implemented locally (i.e. within the car).

Are there other solutions that you prefer that can protect liberty law abiding drivers and pedestrians from the recklessness of speeding and dangerous driving?

Ticket drivers who drive dangerously. Address the problem (the person's decisions), not the tool they're using.
Eh your right to drive like an aggressive maniac is not enshrined in the constitution.

People have consistently shown that they’re incapable of driving responsibly. If we’re not going to fix the actual problems, I’d prefer some limits on car speeds when inside cities and more populated areas.

> "People" have consistently shown that they’re incapable of driving responsibly. Therefore "You" should have less freedom and if it results in You and Your family dying because a device You paid for refused to do what You told it to, that's an acceptable loss to Me

That may not be the best line of reasoning. There's probably a lot of things you enjoy doing that aren't "enshrined in the constitution" that I might want to take from you in order to make me feel more comfortable, and the same can likely be said for a lot of other people as well. If we all played your game there's no telling how much you could end up being prevented from doing.

I like speed limits too, I'm glad that we already have them. Draconian enforcement of rules by machines is supposed to be a cautionary tale, not something to aim for.

You are driving on publicly funded roads in a licensed machine where the police have the legal authority to demand you submit to inspection at any time, and you legally must purchase insurance from a private company to operate it.

I would say local speed governors would be less authoritarian than many interventions that already exist for driving, which is an act that is already a leading cause of death in North America.

Just make it an IR blaster on the street lamps over the road that the municipality can remote command to send out the "low speed limit next N meters" signal. Let it be legal for drivers to override the governor, but doing so is effectively an admission of guilt that you're speeding.

> rushing to flee a rapidly advancing forest fire, tidal wave, active shooter

You design for the normal, not the "so unlikely that 99% of drivers will never be in this situation in a lifetime of driving."

So much of what engineers design for is the abnormal. 100 year flood, hurricane strength wind, web traffic surge, etc etc.

If you can't anticipate contingency in your design, it's not a very good design.

I am not actually in favor of speed governors, but it makes for a bit of an interesting thought experiment: I wonder if it might actually increase throughput in a widespread emergency–many of the worst gridlock situations I see in the NYC area are caused by aggressive drivers who crash and wind up blocking the road, so I could imagine it being the case that, at least in dense areas, restricting speeds could mitigate the stampede effect and keep traffic flowing.
Competent authorities don't ask for technology to be added to the car to deal with very dangerous spots: They redesign them. The dangerous spot was designed by an engineer, and approved by an official or eight. If it's that dangerous, it was designed putting safety too far down the priority list: It happens. So why add a speed governor, with its bonus risks, instead of just redesign that section of road correctly?

It's the same as with the pedestrian deaths: It's dangerous vehicles that hit in the chest, and roads that make pedestrians cross a surface of cars going way too fast to react. We can avoid the situation altogether by fixing the road.

I agree with the idea that I would like to see roads redesigned to make them safer.

I'm not sure why it's an either or though. Redesigning roads can certainly slow people down and make them safer but that often comes at the cost of a large planning process and a lot of money spent. Dyanmic speed limits and speed governors could be much simpler to manage.

I think most proposals for speed governors would make it difficult to go over 10 mph above the speed limit under non emergency operation. I don't think anyone would argue that it's safe to drive 10 mph over the speed limit on a redesigned street so why wouldn't a speed govenor be another tool in the toolkit?

Arguably your plan would take longer to implement as you would need standardization across vehicles, implementation into actual cars, extensive testing, adoption by drivers, codification into the law with some enforcement mechanism and implementation on selected locations. It's a cool creative idea but redesigning a junction seems like something more within reach and that can actually prevent crashes in the meantime.

If you want to tackle speeding on highways, income-based fines from speed cameras seem like the way to go. It is more feasible, can be implemented more quickly and easily and is cheaper overall.

So really, it seems like a new expensive tool with questionable relative benefits that would take really long to deliver (not to mention the privacy / freedom concerns its existence would raise) while the existing tools haven't even been used yet.

I agree that we would need testing and trials before it could be rolled out. Many trucks in the US are already implementing these for various reasons. Other countries are already mandating this.

The European Union is introducing mandatory “Intelligent Speed Assist” in 2024 which requires a lot of the technology that you’re referring to.

I think the real obstacle to this is no longer technological but political

Fair, I didn't know about this. They are farther ahead than I knew.

But Intelligent Speed Assistance is way more lenient and it is up to the car maker how stringent they are. It can be as little as a beep which should be very easy to wire into a dashboard that already shows the speed. Indeed, this removes a lot of plausible deniability for speeding and arguably really is helpful to the driver.

Standardizing and implementing a limit, however, would take more time and testing. Mistakes could result in dangerous situations.

> I don't think anyone would argue that it's safe to drive 10 mph over the speed limit on a redesigned street

Speed limits are set for many reasons. In good conditions, such as very light traffic, good visibility, dry road, limited access, reasonable maintenance of vehicle and road, it's not unreasonable for the safe speed to be 20 mph or more over the posted limit, depending on what went into setting the posted limit.

we have the technology to engineer safer roads
I love the idea of a surface layer without cars at all (other than emergency vehicles, and some rare permitted / licensed construction stuff).

All of the motor vehicles on a dedicated layer beneath, an industrial transit zone.

The right solution is to drastically raise the bar for obtaining a driver's license, and make it far more easy to lose one. Periodic testing and stringent enforcement of basic rules. Don't use your blinker when turning? No license for a year. Bobbing and weaving in and out of traffic like a maniac? Electric chair. (/s)
Meanwhile we can't even agree whether automated traffic enforcement using cameras should be legal
DOWNSIDES

- Your car can get hacked to extort you to pay the ransom to unlock normal/any ability to move

- The government can get hacked for the same reason crippling an entire region

- A terrorist or enemy state can strike a major blow against our economy

- The government now has the capability to use this to shut down protests. Nice protest you had there shame you can't get to it. Remember also that we are on the cusp of a dice roll to decide whether we slip into fascism

- The government now has a record of where you drive at all times. What if someday psychotic municipalities think they can punish you for driving through their county to get an abortion... wait that already happened.

- Your individual car could be hijacked on the road to rob you

- Your car could simply malfunction and not work right

- Your car could apply the wrong rules based on inaccurate GPS reading and cause you either an inconvenience or an accident

- Your car could make it hard to willfully exceed the speed limit in an emergency.

BUTTON OVERRIDE

- Governments often speed limits too low or impractically low given how much traffic has to traverse this space that if it was automatically applied would destroy throughtput to the point that traffic would utterly break down.

- Use of the button even though it might enable emergency traffic would be one more thing you have to do in an emergency that might well not get done. For example you got hit because you couldn't maneuver or you freaked out and mashed the pedal but didn't disable the misfeature in a timely fashion.

- There isn't enough courts to have time to hear why people did or didn't use the button it would end up being an automatic ticket unless you affirmatively hired a lawyer and harassed them and as a revenue center it would be optimized for sucking your money out of your wallet. Think about how red light cams lead to shorter yellows to extract rent not increase safety

COST

- It would cost 150B to outfit all existing cars

- Outfitting only new cars would still add substantially to the costs born by Americans, depress sales of new cars drastically, hurt the deployment of electric cars in turn, and probably be complete by the time it no longer matters.

CONCLUSION

A truly awful idea

How about instead of adding authoritarian restrictive technology to private vehicles, we just take cars out of the cities?

With proper public transit, this would be a no brainer.

Imagine if this country was not so car centric, then this problem would not exist at all. Highways that run through dense urban cores. Suburban sprawl (more expensive highway infrastructure to support). Bulldozing once walkable communities and replacing with parking garages, parking lots, and of course street parking.

Cars. Cars. More cars. Endless amounts of time spent in cars. Now we want aUtoMatEd cArS because of perceived safety gains.

I want walkable and other alternate forms of transportation to be viable for a majority of cities with efficient city planning.

We are digging ourselves into another massive financial meltdown. We won’t see corporations or individual families fail. This time it will be cities filing for bankruptcy because the private muni bond payments become exorbitant.

Nah. What happens when the anti car urbanists take over and then your car gets disabled if it enters city limits?
Vision zero. In Europe it reduced road fatalities almost tenfold without any extra devices added to cars.
What about better driver training and much stricter requirements to get a license.
The digital equivalent of a caution road sign!
And what a great excuse to take the eyes off the road where they matter most!
Waze is really good. No comment on the accuracy of the directions, but I loved the crowdsourcing, and the warnings were the best of any of Apple, Google, & Waze. Loved the very accurate "police ahead" warnings.

Like most of you I'm sure (beating a dead horse here) I really wish they hadn't been bought out by Google.

For all we know they would have been out of business now if it weren't for the acquisition.
Yeah, besides some adverts (when you stop, and I haven't seen them in a while), there doesn't seem to be any monetization going on.
Do you still appreciate the police warnings knowing you only get them because a group of people saw one and started fiddling with their phones, probably while driving, to let the app know?
I don't fiddle with my phone. I touch 3 times my onboard touchpad via carplay without looking.
Or they pressed the report button on their screen, not too dissimilar to changing the AC. Or do you pull over before operating any of the controls in the car? :)
Can someone build a little Waze keyboard that maybe Bluetooth links to do the phone and does a one-touch accident report?

Should be doable with an auto-hot key type of script.

More likely possible on android.

> More likely possible on android.

Just as easy on iOS. You can have an NFC button associated with a Shortcut, so as long as Waze exposes "send accident report at current location" to Shortcuts, it would be very simple.

Just checked and Waze isn’t exposed to Shortcuts at all :(
Uhh okay?
It's two button presses on carplay. Not exactly fiddling.
You can do it with your voice, something like "hey google, report police" when you're in Waze
Typically I'm only worried about police during highway driving, so I don't actually mind the phone usage that much. You make a point: it's not the greatest & safest, but Waze has a much better UI than Google Maps for this, so it's very quick.
Maybe the police reported themselves. Either way they could just not be there, causing accidents and unsafe road behavior.
The alternative was Meta if I recall correctly, but they couldn’t make the deal fast enough.

The founders wanted to sell, they were ready to exit, it was a matter of when not if

I used it recently on a roadtrip with some friends, it seemed quite nice, but I really don't want advertisements in my navigation app, I'm guessing there is some premium subscription to avoid these?

I'm also not sure about those gamified prompts that come up all the time, I feel like if I was driving alone I could get distracted by the strong call to action to always tap something on the phone to earn points.

The gamification made it feel like an app targeted at children.
Haven’t seen a price to remove ads. yet. They’re pretty annoying at night when you have dark mode on but bright white ads pop up whenever you stop.
I wish Waze would tell me less things. It is getting to the point that it's issuing warnings every 1/4 of a mile. Car stopped, object in road, railroad crossing, etc. I like using Waze for navigation, but I know keep the sound off to avoid it's constant chatter and interruption of my music stream.
You can deactivate railroad crossing warnings, which is useful if there are many on your route.
Yea they need to allow people to disable notifications by type.
You can: Settings > Alerts and Reports > Reports. Enable or disable "Alert while driving" for each type.
Except cars on the side of the road. That never turns off.
Why not just idk fix the issue?

Place like that need government intervention not digital warnings for a small percentage of drivers

It's in a development approval process please check back in a few years when it is rejected for budgetary concerns.
How would Google or Waze 'fix the issue'? Do you want them to start a road construction company?
Free wifi the whole way, but 100x the usual number of billboards? Hmm, might be worth it
> Why not just

'just' do it yourself then if it is so easy.

Another interesting addition would be to warn drivers if a close by car/driver also had a bad driving history. May be also track if the close by car started of from a bar/restaurant and weight the probably of the driver ending up with a crash with other cars in the vicinity

/s

Given the lack of data security in cars, I bet we could see an open source project doing just this
Waze is good, but it does have some very weird over-optimizations.

Anecdote: A couple years ago, while driving into Toronto on the 401, we hit some traffic as always. Waze told me to get off at the next exit which I was fortunately right beside, so I assumed it was going to route me around the accident and back onto the highway after it cleared up. What it really did was take me up the off ramp, do a U-turn, and get back onto the on ramp to gain about 700m of distance from where I was before. It theoretically saved me 5 minutes of travel time, but it only really moved me a few dozen car lengths ahead.

Their wealth of real-time and historical data is certainly beneficial, but it also causes some truly bizarre navigation now and then.

I use extensively when in LATAM. It sometimes feels as if the route chosen was for the algo to "probe" an alternative.
It is a technique commonly used in gaming, StarCraft for example. It is called “send a drone” or “send a probe”.
But do drivers earn points for taking these side missions? Like if you accept multiple probing missions, do you earn a badge that the algo uses to find drivers known to be willing to accept these new assignments which should increase the likelihood of the missing being accepted?
It’s like in StartCraft, no? Do drones earn points ;) ?
Waze used to do the same thing to me on my morning commute. The worst thing was that the backup I was "avoiding" was itself caused by the merging traffic from that onramp, so Waze was perpetuating the vicious cycle.
I know some human drives do that without Waze. It is a faster and legal route, why wouldn't the algorithm send you down that path?
It seems like it would mess up traffic more? I have heard that Waze sends drivers through residential areas as shortcuts, which, you know, kills kids.
My two cents: weaving through dense traffic is stressful and probably puts you at a higher risk of an accident than remaining in the same lane for an extra 5 minutes. I prefer to ignore those kind of shortcuts.
It's arguably an asshole move, even if legal. You really just gain the same time as the other people collectively lose between the off ramp and on ramp. In addition you cause additional traffic on the ramps, which might hold up other road users using these ramps normally.

I imagine the Nash-equilibrium of this pattern if more people does this is that there is blocked traffic on the highway between the ramps and also on the ramps.

Not arguably. It's objectively selfish. Why is your time worth more than those that you're passing?
This seems like over-optimizing for time. For the individual there are trade offs in wear and tear, fuel consumption, stress, accident risk, and as a sibling comment indicates, a social component.

For traffic in general this increases the number of lane changes and merges, increasing risk and slowing traffic further. Depending on location it may benefit by using more of the road surface (and alleviate problems further back) or may cause problems off the freeway at the local exit either due to greater traffic at the intersection or the freeway queue backing up into surface streets. (Or both.)

It should also be noted that this is an illegal maneuver in some US states. I presume Waze doesn’t suggest it in those states.

I'm not totally averse to taking shortcuts on routes I know well. But excessive shortcuts through suburban developments to save a few minutes are sort of obnoxious even if they're legal.
(comment deleted)
My experience has been that map apps are overly reliant on average times when traversing city streets. A shortcut thought LA might save you 10 minutes if the traffic light gods are in your favor or might add 30 minutes if you hit a red light at every intersection.

There are two parallel roads between my home and office and google likes to tell me that they will take the same time. One goes though about 10 stoplights in busy retail areas, one goes though about 5 stoplights in quiet industrial areas.

On average both roads might get me there at the same time, but on a unlucky day the busier road may cost me 20 minutes. I always take the other path

Traffic flows better if played as a collective game, where the objective is to help the others get through it along with helping yourself.

This strategy, if followed by enough people, massively reduces traffic stress. It may also reduce travel time, but honestly the difference is likely just a few minutes each way.

Yes, I've driven in cities that play by the "we are in it together, let's work together to get us all out", and cities that play by "any advantage I can get regardless of the cost to others".

Problem is, once enough people start playing the selfish game, the equilibrium breaks and everyone has to go selfish.

that's how you kill a society. Kill the social contract, the care for your fellows, one form of interaction at a time.

Waze seemed to not understand the express/collector system in Toronto for the longest time, and would have you merge back and forth between the two unnecessarily. It seems to have become better over the last year or two, however.

https://en.wikipedia.org/wiki/Local–express_lanes

Every GPS I've ever used seems confused by the express/collector system on the 401. Google Maps, the one built into my Toyota, and the old Garmin all had trouble with it.
There should be a way to adjust how aggressively it optimizes.
I'd like the option to add small positive or negative weights to intersections / roads I know to be problematic.

Generally I also prefer to stick to more major roads unless there's a major time savings (E.G. >= 5 min, this is part of 'how aggressive' you propose) or accident.

The other aspect of aggressive would be the relative cost for 'unusual movement', such as (legal) U turns, cutting across on small side street connections rather than a major intersection, etc.

Yeah that happens to me sometimes, to the point where I just ignore it and stay in traffic instead of going down exits and B-roads, which are more effort / more tiring than just creeping forward.

It has sent me down some interesting roads though.

I sometimes feel that Google Maps goes into what I call its "I feel like a country drive today" mode. I'm pretty sure it's usually slower because half the time I miss one of the 20 turns it's telling me to make--especially at night or if there's some multi-road intersection where the roads aren't at right angles. It's particularly annoying if there has been recent snow because the back roads tend to have snowdrifts or are otherwise just slow.
Once upon a time, I was trying to cross from VA into MD via the Woodrow Wilson Bridge. The bridge was closed due to an accident. Waze happily directed off the highway and onto side streets, where it had me do circles in and out of parking lots and strip malls until I gave up and went home. It was really weird - literally couldn't go anywhere, so instead of telling me to stay put, it just ran me around in bumper-bumper traffic for 90 minutes.
Did it really take you 90 minutes to realize you were going in circles?
No, that was the time it took me to give up and go home (vs waiting for the bridge to re-open). It wasn't literal circles, but looping in and out of feeder roads, neighborhood streets, etc. Completely silly - nobody with a destination in MD was going to get there, why would Waze take us off the interstate at all?
Clarification for those not familiar with the area: if that bridge is out, your options are a 20-mile detour that involves driving through downtown DC, or a detour of over 100 miles to the next crossing downstream.
This actually works in some cases. There is a section of I95 near me with an exit "shunt" that's divided from the main highway. The main highway has a merge where another interstate joins, and that "shunt", where you can enter/exit is protected from the merge traffic. So taking the shunt can save you 5-10 minutes when the merge point is at a standstill.
What it really did was take me up the off ramp, do a U-turn, and get back onto the on ramp to gain about 700m of distance from where I was before.

I'm pretty sure that isn't even legal, at least in the state of Washington. I rarely go that way, so help me out here Seattle-area Eastsiders, but a lot of folks used to do this on eastbound WA520 at 108th in Bellevue, to the point that there at least used to be a sign at the bottom of the exit ramp saying (in much fewer words) no, you can't get back on the freeway by going up the entrance ramp. (Sorry, street view doesn't show the sign that said, IIRC, "no reentry", so I could be imagining it.)

There is a similar sign next exit 17 on 405N prohibiting existing and re-entering back.
So you saved 5 mins? Seems like an overall ok experience?
Any time I try to use waze it gives the most insane route recommendations. Would they shave 1 or 2 minutes off my commute? Sure! But will I be tasked with crossing several 5 or 6 lane stroads in doing so? You bet! It's like what do you value more, saving like 4 or 5 minutes of driving a day or your life?
For me it’s depended on where I am at the time. In LA it would do exactly that, having me cross many lanes in heavy traffic, but elsewhere it was fine.

My theory is that Waze must model what maneuvers are possible/easy based on other drivers in the area. So if you live in an area with more aggressive drivers it assumes those maneuvers are easier to do.

Yeah Waze needs a "UPS" setting where it doesn't take you straight across stroads and heavily favors turning right, while not routing you straight through everyone's residential neighborhoods.
Google does this to me too, sends me on some really annoying route because it's slightly faster.

I also wonder if they kind of "round robin" routes so not to cause traffic?

I can imagine a future where map routes are optimized to give subscribers to Google services priority while allocating free users to alternate ones to free up traffic.
I think it pays to learn how to drive and ride without Google maps, it's amazing how many people I know who are completely dependent on Google maps for navigation.
I have a very similar experience. Truth be told, it is what I have on my car's navigation whenever I know exactly where I'm going: the warnings about traffic jams, stationary radars, police and all that is all I care about. But whenever I need directions, I quickly switch back to the big G.
Waze is great but I have no idea how they make money and I fear now I’m the product.
It costs them money to develop and operate, they were bought by Google, and you don’t pay anything for it. All I know is my gut says, maybe.
There is ads. You often get a pop up saying: A trip to Taco Bell add 5 minutes to your drive. Would you like to go?

It seems less invasive then most other apps out there.

Waze is a Google product, so yes, you are the product. Also, they sell ad spots. Published a few for a restaurant, worked OK actually as we got some customers who got there because of the ad.
> One feature of Waze that was unique for a long time was its ability to crowdsource traffic information. Users add live traffic information to the app as they're driving, like a car stopped by the side of the road or a crash.

I wish apps wouldn’t encourage their users to interact with it while driving.

I would love Waze to sell a set of bluetooth buttons that I can mount on the back of my steering wheel. Press a button to report speed trap, accident, or other hazard.
Yes! And/or a hot button bar in the UI for frequent reports.

Object in road, stalled vehicle, police are the top things I report.

Programmable Bluetooth buttons already exist, it seems like you would just need to connect them to the Waze API (which I presume exists)
I was hoping Waze would have a deep link scheme to report speed traps (https://developers.google.com/waze/deeplinks) but they don't. If they did, I could potentially pull this off with iOS shortcuts.

They do not appear to offer it as part of any other API.

I hope they'll start warning about bad drivers using historical data (one day)
I’m very hit and miss with Waze as an early user.

On the one hand, more users = more data to share back with me.

On the other hand, more users = some of the amazing shortcuts it’s shown me aren’t so amazing anymore.

Waze is the tragedy of the commons in action.

It’s great for the individual, but the more people use it the worse off everyone becomes, including the users and the surrounding community.

I mean, I’m sure they’ve implemented some logic to avoid “over-suggesting” a 1-lane shortcut from a 4 lane highway, but I have taken shortcuts where I’m clearly in a line of Wazers.

I feel bad when it sends me toward a difficult unprotected left (in the LHD world) and there’s a bazillion cars behind me waiting to turn right but can’t until I can move.

I do think Waze and the like have effectively “added” 10% more roads out of thin air. Departments of Transportation should be paying them.

Long overdue. I use both Waze and Google for long trips from MA to ME and despite Boston being a clusterf*ck every. single. time. These apps will routinely behave like they've never seen rush hour (or any hour) of Boston traffic.

It always baffles my mind that these apps consume mountains of data, each day about traffic flow, yet can't predict slowdowns. I have one, human speed brain and I could probably estimate traffic jams better than Google/Waze.

> I use both Waze and Google for long trips…

You use both simultaneously? Or if you alternate, based on what?

I do this.

I use Google Maps for trips with multiple destinations or extended routes, and Waze for spontaneous, on-the-fly directions when I need to make unplanned stops.

I tend to use multiple apps simultaneously (Waze, and either Apple Maps or Google Maps, depending on how I'm feeling or the area). Especially for longer trips or trips outside of my home area, I trust non-Waze apps to keep me more on the "safe route" and not make detours into unsafe neighborhoods or winding back-roads in areas I'm unfamiliar with.
I alternate at random, although I prefer Waze for the hazard warnings. I will use both when my wife is driving, and I want to plan ahead or deal with the aforementioned traffic snarl. BTW Apple Car play is terrible for navigators. Once a phone is "locked in" it's useless for the navigator. You need another device to "look ahead".
Not sure I understand here... you can tap on the CarPlay map and then drag to scroll around. It's better on capacitive screens, but still works well. Then, tap the compass indicator to go back to current location.
> It's better on capacitive screens, but still works well

From my experience? It's wonky. Tap map, zoom out, drag map and then randomly it will snap back (Waze is worse) - it works (for this one task), it's just clunky. Clearly all the focus in car navigation is on the driver, who shouldn't be diddling the screen but it would be nice to have a few features for "navigators" riding shotgun.

> These apps will routinely behave like they've never seen rush hour (or any hour) of Boston traffic.

Interesting - when I pick a route it typically says something among the lines of "increased traffic at this hour".

It does indeed seem to assume that the traffic situation won't change for the duration of the trip, but I got around that by checking different departure times.

Hey, does anybody know why Waze still exists as a separate app?

It was acquired a decade ago. Is there a legitimate reason, or is it just a reflection of Google leadership's lack of a holistic app strategy? (see: Google's history of messaging apps)

> Google kept Waze independent and as a separate brand to preserve the spirit of innovation of the team behind it and because there was really no reason to kill it, but there was also no reason to promote it.

but sounds like its days are numbered https://www.calcalistech.com/ctechnews/article/go92isunj

If Google are planning to merge the two, why haven't they added the ability to provide real-time feedback to their app? A few times I have met a closed road and have no options. I can't reroute because Google doesn't know the road is closed, it just tries to turn me around to go on the closed road. I can't report it closed etc. other than a long-winded Google Maps thing.

I don't know how they are supposed to know about temporary road closures but I do sometimes see them marked on maps.

At least here signs do that. Safety is the reason they are put in but drivers use apps to avoid police. The best way to avoid fines is to follow rules, but I hope users will follow apps more than signs.
What Waze needs is a proper zoom function, especially on Apple Carplay.

Waze is almost unusable in that its auto-zoom doesn't work properly, and I'm forced to switch to manual zoom, which also doesn't work when I want to see more precise directions to places that I've never been before.

Apple Maps gets this right, in that it properly zooms between long distances when I'm on the highway, and short distances when I'm on the streets. I don't know why Waze doesn't get this right and I've moved to Apple Maps entirely for new locations.

This doesn't account for homeowners that regularly report fake accidents on their streets to prevent Waze from directing traffic through their neighbourhood.
Is this happening often? (Wondering, I don't know.)
I did quite like Waze for most reasons but once the ETA started to drift inexplicably, I started to trust it less. I would leave for a 2 hour journey and it would estimate 1hr30, which would creep up to 2 hours by the time I arrived at home. No traffic, no delays so why didn't it start at 2 hours ETA? Some kind of gaming algorithm?

However, Google nav is far worse. It doesn't seem to log traffic on more minor roads and where I work seems to keep me in the traffic instead of taking me a slightly longer route which is generally quicker. As you sit in the traffic, the ETA just creeps up and up proving it is not measuring real-time positions which would tell it how slowly we are moving and therefore how long it is likely to take over that path.

We also had a very bad experience with Google where it sent us on a detour to avoid a serious accident on the motorway and pointed us, as well as everyone else, up a very narrow road despite a slightly longer route on a much wider "normal" road. It took an hour to get about 1 mile, when we finally rejoined the empty main road. Again, very concerning that Google doesn't seem to understand that sending everyone up a slightly more direct route doesn't work when it blocks everything up. It took us a number of other interesting routes and took us about 3 hours longer than the 3 hours it would normally take. Sometimes I wonder if it shouldn't just say, "Everything's rammed, just wait it out here".

> [...] Google nav is far worse [...]

I have Google Maps in my car and I can't understand how that thing got into production. It keeps suggesting "alternative routes", which is basically a good thing. Except when the shortest route is like 1.5 hours and the alternate route usually adds 45min or more to the drive. The worst thing being, there's no setting to turn suggestions off.

I've consistently observed the same creeping ETA on Apple Maps. Sometimes it's blatantly clear when planning a route when the estimated travel time and distance results in needing my average speed to be higher than any speed limit along the route. More often it's just that I'm traveling at a reasonable speed yet the ETA just keeps creeping out and out.

Additionally, maybe it is some settings I have on my iPhone, but Apple Maps seems to not make recommendations but just execute them, with no input from me. Recently we tried were navigating out of Burlington, VT, USA and specifically chose a route to avoid the ferry only to have Apple Maps decide we really should take the ferry without ever asking if we wanted to accept the route change.

I expect your idea of "reasonable speed" does not align well with the actual average speed of travelers along the route. The only time I see this ETA creeping behavior is when I'm on long a long trip, not in a hurry, and I intentionally drive at the speed limit instead of at the average speed of traffic to conserve fuel.

A navigation app that estimated based solely off of the posted speed limits would be utterly useless for 99% of all drivers. Ideally the app would learn from your actual driving habits and adapt based on them, but I don't know if any of the big 3 do that.

I think it's totally unreasonable for a mapping app to ever assume you can travel at a higher speed than the posted speed limit, but they all seem to do it.
They all do it because within a very small margin of error, that's how everyone drives.
It should be illegal to estimate times lower than you can achieve within the speed limit - it means people leave later and have to speed to show up on time.
I feel that's what missing from Google Maps is an analysis of the driving difficulty. Last month I was driving across Europe, where we were driving all day, doing close to 1000km per day. Where there wasn't a direct highway connection, it would quite often suggest going through local roads, which more often than not, had just one lane each way.

If there is an option that involves more highway driving for only 5 or 10 minutes more, I'd rather take that, as it's going to require a lot less mental energy to drive along that, than a narrow windey road I've never been along before. Sometimes the alternative routes suggest that, but most of the time I had to just look at the map and figure it out myself, then create a route from A to B via C to see it's only going to add 10 minutes to this 3 hour leg.

Toll roads are another fun thing. When driving through France from Switzerland to Normandy, the most direct route would have resulted in close to €50 of toll fees. We ended up driving half an hour more via Belgium (where fuel is cheaper too), on a highway that had no toll fees.

Glad to see this! Waze, while still an excellent routing and traffic utility, has felt stagnant and neglected for years. There’s so much room for improvement over Google or Apple Maps with features like this. I’d say I’m surprised by the lack of innovation, but acquisitions left to languish is well known to be Google’s playbook.
I mostly like Waze and use it a lot. But its inability to determine which way to turn when exiting a parking lot at the start of a trip is consistent and baffling. Often as not it'll show the wrong direction, followed by a literal U-turn or more complex sequence amounting to the same thing. I resort to using the compass and/or taking the extra time to zoom out on the map and see if it's sane.
I wish waze would use computer vision to auto-report roadside police
Hey Waze, how about telling me exactly which lane the 'Object on road' is? Surely users can be asked the lane when reporting.
Good idea!

There used to be constant crashes outside my house as it was a crossroads but not very visually obvious and with poor side visibility so they wouldn't notice the car coming at right angles till too late. An app warning probably would have improved things although eventually they re-engineered the junction with mini roundabouts which mostly fixed things. That was many decades into the junctions existence though.

That said rather than users adding information you'd probably get better data from motor insurance data as they probably log most accidents.

I used Waze and loved the idea before I became so cynical -- around 2013-ish -- about the "surveillance economy" that I started deleting apps from the likes of GOOG and FB. This included Waze post-GOOG.

Maybe there could have been a revenue stream for Waze pre-GOOG from something other than ads and tracking? We'll never know.