Written by someone who does not drive a manual transmission, or live in the mountains. Brakelights, at best, state that a car is undergoing some degree of braking forces. Anything more is a very dangerous assumption. Cars can accellerate with thier brakes on, and can remain stopped with them off. (Also, every motorcycle on a flat surface uses their feet to stay stopped at long lights.)
Cars start moving with brakelights on all the time, especially with automatic transmissions where it can take significant force to keep the transmission in check. And hills. A better light might be one that changes in intensity with brake pedal force, so pedestrians can see when an inattentive driver is about to take thier foot off the brake.
Why green instead of red? Would it be an inverse brake light (green appears when foot lets off the brake?) I don’t see how any color light is easy to retrofit.
> The innovative design of front brake lights adds another layer of practicality. These lights would emit a green glow rather than the traditional red, easily integrating into current vehicle frameworks without extensive modifications.
The problem with green on the front is that cars stopped at a light at night will be presenting a green light to oncoming traffic that should also be stopping, possibly overriding the red traffic light or stop sign.
In most US states blue is for police and red is for fire. Some states let other vehicles, like tow trucks, use blue. Sometimes they'll use blue and red for police - but blue is always restricted to my knowledge.
Yellow/Orange/Amber is for "caution". Some states restrict that and some don't, but it has an understood meaning.
Purple signifies a hearse or funeral procession.
I learned about these laws in the early 2000s, when I was in my late teens and early twenties and wanting to drive a customize minitruck. It turned out that in my state, all colors of strobes and accessory underlighting were restricted... except green.
What's worse is that the law specifically includes a provision where possession is considered to be sufficient evidence of use. Technically it's illegal to buy a battery-powered toy with an LED strobe in it, if you drive home with it within reach of the driver.
A trick I learned years ago from a motorcyclist: pay attention to the front wheels of cars near you in traffic.
Very often, the front wheel turning is the first noticeable visual indication that a car is changing lanes or turning. That gives you a few more fractions of a second in a situation where they could be very important.
After a while, I noticed that I was instinctually able to tell more accurately if a car is accelerating or decelerating. After considering it, I realized that I actually could tell better than before, but it wasn't instinct. Because I was watching cars' front wheel, I was able to see the front suspension load and unload as they accelerated or braked respectively. It was just subconscious.
Direct to this, I'm surprised the light would reduce accidents by so much. That said, I don't see any reason not to explore it more.
If you can't assume vehicles have front brake lights, the lack of an indicator has no direct meaning. If you see on one, great, that car is slowing down -- but without comprehensive deployment, you can't assume that no light means that the car isn't slowing.
thecrumb•1d ago
yCombLinks•1d ago
Ancapistani•23h ago
It's not that front brake lights are necessarily a great idea, but that safety is improved with the communication of states and state changes are entirely automated.
Imagine a future car that knows that you're about to change lanes because you just checked your mirror, glanced over your shoulder, and repositioned your hands on the wheel -- so it turns on the turn signal the moment you begin to move the vehicle toward the other lane. Maybe it even has multiple states - what we have today for "I'm definitely turning/merging", and a dimmer or less saturated pulsing light for "the car thinks the driver is probably about to turn".
That could get too complex to be practical quickly. AI/FSD systems have a lot more "input bandwidth", though, so maybe make those partial/inferred indicators IR so humans don't see them but it communicates to FSD systems.
Actually, scratch that. In addition to signals that don't require manual activation, add a 360º IR emitter on top of the vehicle that constantly streams real-time telemetry. FSD could then integrate the actual, specific state of the vehicles around them into its sensor suite.
If it's bidirectional, you could even have a situation where two FSD systems meeting on a narrow road could silently and nearly instantly negotiate a plan to pass one another with minimal disruption.
Larrikin•1d ago
piva00•1d ago
That's a very culture-dependent thing, it's much more unusual for me to see someone not using turn signals in Sweden than it is in the USA or Brazil. If no one around you uses them I'd blame driving licence requirements, enforcement, and training for it.
JohnFen•1d ago
OTOH, there are other bad driving habits that more common where I am than in many other places, such as running red lights.
jalapeno_a•1d ago
hackeraccount•23h ago
Ancapistani•23h ago
Between that and the implementation of momentary signal activation - when you push the stalk just a bit, but not enough to "click" into place and stay there; it blinks ~3 times and stops - you have a nice carrot/stick pair.
It's annoying when the warning goes off when you change lanes without signaling AND it's easier to signal because you only have to tap the stalk once instead of twice.
Ancapistani•23h ago
Everywhere I've ever lived or driven, the vast majority of people use them consistently - at least when appropriate. The most common place I see people not using them is when changing lanes, and then usually only if there's a vehicle in their target lane close enough that it would matter.
Do bad drivers exist? Obviously.
Actually, now that I really think about it, there are two places I've driven where I can recall seeing people merging and turning frequently without signaling: New York and Los Angeles.
In those cases, my guess is that it's partly cultural but predominantly driven by task saturation. When you're operating a vehicle at 80 MPH surrounded by other cars in rush hour traffic on a six-lane highway, sometimes you have to focus on staying alive.
Aviation has a saying that describes this: "Aviate, navigate, communicate". Basically, your first priority is to ensure the immediate safety of your vehicle. After that, you focus on identifying a clear path that doesn't conflict with the path of other vehicles. Only after those things are done do you worry about telling others what's going on. In the air that's the radio; in a car that's your turn signals and horn.
Here's the Wikipedia article talking about this in the context of aviation: https://en.wikipedia.org/wiki/Airmanship#Principles
If task saturation is actually what's happening here, then I would expect to see turn signal usage decrease as traffic density and speed increase. I bet, given the access to data like live traffic cameras that we have today and the current generation of multimodal generative AI models, I could whip up at least a first pass at measuring this without much time investment.
Great. Now I have yet another project :P