12-Foot Traffic Lanes Are Bad for Safety and Should Be Replaced (2014) | Hacker News Reader

12-Foot Traffic Lanes Are Bad for Safety and Should Be Replaced (2014)

citylab.com

174 pointspanic143 comments

I'm reading a lot of comment from people complaining that this article is wrong and we shouldn't make roads less safe.

Question: what are you talking about? A 10-foot road is wide enough for a car or truck to drive safely at urban speed limits. The maximum vehicle width under Federal law is 102 inches, or 8 feet and a half. And this is for trucks. Cars are around 80 inches. This means over a foot per side for cars and more than half a foot for trucks. If it is not enough for you at city speeds, please give up your driver license, you are not good, and you shouldn't drive.

Yes, I agree, I don't want to drive 50 miles an hour next to a truck on a 10-foot wide lane, but that's not a city or a suburban speed, that's highway speed. On the opposite, no problem at driving at 15 or even 25 MPH on a 10 feet lane. Yes, probably I need to give up texting and driving, and I need to be focused on the road. Guess what: I need to do it anyway.

Honestly speaking, I have enough of our politicians of disregarding scientific studies and expert analysis, but this is somewhat accepted. But a community like Hacker News should have the smartest people, not the dumbest ones that fail to understand reality.

I've found the older cities I've lived in with much tighter roads have much better drivers as a result. People know their distances and are hyper-aware of their surroundings because it is a necessity in order to avoid minor collisions.

In cities that are more 'designed' by traffic engineers there is ample room so people don't know the dimensions of their own vehicle and travel at greater speeds while letting themselves be distracted more frequently. The accidents that result are usually frame-bending as opposed to bumper replacements.

In a lot of older areas the roads were built just big enough for farm to market purposes[1] either officially (as in Texas) or unofficially. This resulted in narrow, twisty & curvy roads that encourage people to drive slower and safer.

The safer a road is perceived to be through visual cues (wider, straighter, less hilly, etc) the faster drivers will go. Hence why in Seattle and other cities dangerous (read wide, "safe" feeling) roads are being put on a diet, and going from 4 lanes to 2 lanes with a center turn lane and bike lanes (sadly usually unprotected). The same thing is being done in Denmark, where they actually add curves & planters to the roads to force cars to curve and feel unsafe, which ultimately results in safer streets for everyone involved.

Another strategy that I've heard has better results is eliminating sidewalks and getting rid of jaywalking laws, so that traffic of all types are forced to intermingle at much lower speeds.

[1] - https://en.wikipedia.org/wiki/Farm-to-market_road

As for mingling different sorts of traffic - I've heard of several spots here in the Netherlands where they've done just that, and it doesn't always result in lower accident rates. At some point you cross the line between "less of a comfort zone everybody is more aware" and "confusing mess that has so many things going on that nobody is aware of everything". A narrow road can still be predictable - you'll have to be alert to prevent minor bumps, but there's no surprises that can result in rear-ends because you have to slam on the brakes.

I don't have any studies numbers to back it up atm.

My town has in the past year eliminated raised sidewalks/pavements. All that happens now in those areas is that cars drive all over the pedestrian bits and park on them. It's called a "shared space" but car drivers, being rather selfish, just drive and park wherever they wish with scant regard for pedestrians.

I think the turn lanes and bike lanes were the motivation, the lane width became the accommodating variable.

I've had several british friends tell me that if they pass cars in America like they do back home (say, within a foot with both cars traveling at 25-30mph), American drivers will go ballistic at them and can't believe at "how dangerous they are driving". The difference is that they have much better training and know that they can pass safely.

Come to Ireland and leave the motorway network for our country roads, we pass within inches of bushes on one side and cars on other at 80 to 100km/h while being blinded by idiots misusing fog lights and reflections of perpetualy wet roads. These "roads" resemble twisty tunnels thru trees :D

You should visit Boston or DC some time. It will be an eye-opener. Both are cities that were designed well before cars. Snowbanks make things even narrower. This should be urban planner paradise, given the much tighter roads.

As for the much better drivers though... well see for yourself.

Frame bending, to say nothing of the squishy bits inside.

Absolutely correct, the way we improve safety is by making the most dangerous people feel less comfortable.

I've read about some places where a rural highway leading into a town will narrow to ten foot lanes, waking drivers out of their daze because the driving environment is about to become much more complex. I wish I could remember where this was.

In my country, high traffic moderate speed (90km/h - 56 mph) sometimes go across rural villages where the speed limit is usually half that. Furthermore, drivers here are notorious for not giving a frack about speed limits. So some of those villages have experienced with various ways to narrow the road. Sometimes it plays only with perception, by adding for instance small plastic poles to separate the lanes. Sometimes it's chicanes or bumps.

I can't give figures on the effectiveness of this, but as a driver you instinctively slow down when the road appears narrower. It seems to me more effective than a sign on the side of the road.

The traffic islands before towns[0] have proven to be extremely effective at actually making people slow down to 50km/h.

[0]: http://www.rediweldtraffic.co.uk/media/9186/RediPaveHeader.j...

It's like the thought experiment of replacing seat belts with a spike sticking out of the steering wheel.

That will guarantee that car accidents and deaths drop dramatically.

The long forgotten irony in this meme is that cars used to be designed like this. There were no seat belts and the steering column was a rigid spear. It was practically designed to kill you.

Historical evidence suggests that there would indeed be fewer accidents but more car drivers would die (because even quite trivial bumps would be fatal), though pedestrians and cyclists would be safer. As someone who chooses not to drive, that sounds pretty good to me.

Earlier this year I was at a day long car control clinic hosted by a BMW CCA chapter (they do a lot to promote driving proficiency and safety).

A comment that one of the instructors made will stay with me forever: "In the states they teach you how to operate a car but they don't teach you how to drive".

> This means over a foot per side for cars and more than half a foot for trucks.

I drove in a place like this (Dominica) for a while ~15 years ago. It was rare to see cars that hadn't been dinged up by minor encounters with other cars... If your mirror struck a pole or another car, well, that's why they're on hinges, right? Right?

> I don't want to drive 50 miles an hour next to a truck on a 10-foot wide lane, but that's not a city or a suburban speed, that's highway speed.

80 MPH is a highway speed. 50 MPH is definitely normal suburban speed; non-cul-de-sac suburban roads are often posted at 45.

https://en.wikipedia.org/wiki/Speed_limits_in_the_United_Sta... suggests that 50mph would be unsually high for a residential street.

As for the rear view mirror hinges: what on earth did you think they were there for, if not for saving your mirrors when they get whacked?!

(A somewhat common optional extra for cars is folding rear view mirrors. Some probably fold away automatically, I'm sure; mine make you press a button first. But my car is old. I bet the automatic folding is more common these days; modern wing mirrors often have integrated indicators, and that's going to increase the cost of replacement.

(Sole purpose of the folding feature? Saving your car's wing mirrors from getting hit when you park on a narrow street.)

> (Sole purpose of the folding feature? Saving your car's wing mirrors from getting hit when you park on a narrow street.)

Sure, because you folded them in after you parked.

My car has crumple zones and airbags, too, but that doesn't mean I want to use them. If you've struck something, then while some mitigation of the damage is great, well.. you struck something.

Where I currently live, in the US, if you hit something with your car, you've had a car accident. Avoiding accidents is a very high priority for most people, even if they're low-damage, survivable accidents. If I tap someone's bumper or mirror, even if there's no visible mark at all, I have to be prepared to furnish my insurance information and get theirs, and even if there are no monetary costs to my insurance company, my rates are probably going to rise.

In places where lane widths are very low, hitting things or people with your car can turn into something that's a regular occurrence, such that if a year went by without any bump or scrape you'd find it remarkable.

In any case, my original point was that an environment in which cars, trucks, and vans are passing with an average foot clearance is going to be an environment in which collisions happen frequently. A related point made elsewhere in this thread is that a follow-on problem to additional collisions will be that narrower lanes will no longer allow traffic to proceed around the scene of some minor collision where the parties are arguing over whose fault it was and waiting for the police to sort it out.

Not residential street - suburban thoroughfare - as in 4-6 lanes wide.

In Houston 2 lane (1 each way) thoroughfares are often 45 mph.

Those are de-jure numbers. Add 50% to get de-facto numbers. So yeah, 50 MPH is normal for residential. The highways marked 65 MPH are actually 100 MPH.

Speed limits suffer from a NIMBY problem. Everybody wants to the limit to be 5 MPH around their own home, but 125 MPH when traveling elsewhere. The locals control the limits. The same goes for speed bumps, often loved by the locals but hated by everybody else.

I thought the hinges are there to be able to adjust the mirror to the driver's eye position. If you move the seat forward or your upper body is longer the angle at which you look at the mirror changes.

This is what they're there for:

https://youtu.be/JIO7lZuokpQ?t=7m25s

Also, I'm not sure why the OP referred to them as rear view mirrors. I've only heard them called side view mirrors.

For automatic-type mirrors, no. That sort of mirror has a hinge to let the mirror rotate forwards and backwards on one axis. This is there to fold the mirror in, and to help absorb whacks. The mirror angle itself is adjusted by a separate, finer mechanism inside the unit.

For manual-type mirrors the same joint - some kind of ball-and-socket affair, I imagine? - performs both functions. The joint has a much wider range of movement than you'd expect if it were just for catering for the position of the driver's head.

Out of curiosity, have you ever driven a box van on roads with 10ft lanes?

I just looked this up: an ambulance is 9' wide. A pumper truck is 9'9". A ladder truck is 10'.

Which is why European cities with narrow roads have special fire trucks that accomdate them.

Today we build cities to fit fire trucks instead of building fire trucks to fit cities.

Here's an example of a Japanese fire truck for narrow residential roads: http://www.synapse.ne.jp/hozumi/P1000996.jpg

It fits into the "Kei car" tax category, which means max 3.4m (11.2ft) long, 1.48m (4.9ft) wide

And Europeans and Scandinavians generally, especially here in Norway, get out of the way when an emergency vehicle is heard or seen. Often this happens on roads that are only six metres wide yet there is still room for the ambulance or fire engine to storm down the centre with room to spare.

Of course not all Europeans are so public spirited, I've seen video from the UK and Germany where people made no attempt to get out of the way.

I know which fire truck I want if my place is burning. The bigger the better. I'll take a 20-foot wide truck please, with extra water. Little trucks are not the same.

Looks like we've got an expert here. Please sir, cure us of our collective ignorance. What is the ideal street width, in millimeters?

Edit: I'm going to leave my comment as-is, and just point out that this is a question that you should feel comfortable answering, even if indirectly, to weigh in on this matter. HN is getting way out of hand with people prescribing solutions to problems that the prescriber knows jack shit about.

Done that without problems, also because I know how to drive.

Out of curiosity, where you found 9' wide ambulance when the US GSA (General Service Administration) standard says that they should be 96 inches or less?

Amherst, MA.

https://www.amherstma.gov/DocumentCenter/View/24390

The Fire Apparatus Manufacturers Assn. thinks 26 states limit their equipment to 102" and another 17 are "exempt."

Whoa whoa whoa. You are getting it wrong. The size reported by your Amherst document is the maximum width, not the body width.

I have an idea of the importance of this distinction, and I could go at great length explaining it. But the linked document reference the AASHTO's Green book as a reference. And if that manual says that 10-foot road are ok in one chapter, and acknowledge the vehicles dimension in another, chance are that is ok. Obviously, if you think it is not, you have to demonstrate they are wrong.

Full disclosure, I know nothing about traffic engineering and I'm an average driver. This is the reason why I check expert sources and not the first document found on the internet...

The "10' wide" fire engine there supposedly conforms to the AASHTO BUS-14 standard... which means 2.6m wide. (This is the widest type of vehicle in the list.) 2.6m = 102"; 102" = 8.5'.

Go figure, as they say.

Please allow me, not that you have the option or anything, to link to a previous comment of mine, which is probably appropriate: https://news.ycombinator.com/item?id=12472990

(Maybe 10' includes the wing mirrors? 1.5' = 45cm. 22.5cm per side. Not unreasonable for lorry-size wing mirrors.)

That is a "medium duty" ambulance, and those aren't especially common (some areas use them quite heavily, but most don't).

Ambulances are generally built on pickup truck or van chassis (the Ford F350/E350 lines are popular).

Here is the model lineup of a popular ambulance manufacturer:

http://www.braunambulances.com/models/

My issue with the article is that it doesn't really scope the problem. A six lane boulevard or highway in a city is a different animal than a typical street, and the examples seem to be those large through streets.

I live in a state where the the state only maintains state designated roads outside of cities. So maybe I'm weird. But generally, I think it's safe to say that boulevards, regardless of lane width tend to be unfriendly to non-automotive traffic. Side streets with marked lanes are rarely wide.

So other than stating that engineers are Uber conservative, the problem really isn't defined.

It implicitly scopes by arguing for pedestrian safety. This means it goes for all roads were pedestrian crossings are done.

> Honestly speaking, I have enough of our politicians of disregarding scientific studies and expert analysis, but this is somewhat accepted.

The problem is, the linked article is not scientific. Yes, it links some studies and cherry picks a few sentences from them, but exactly in the way that media do in pieces like "Scientists have discovered 3 more ways make your road safer!".

For example, the article cites a study that says that higher speed is expected on wider straight streets between traffic lights. Then it says that there is no reason to not apply to other types of road (like not being straight? or maybe 20 other even unworded reasons, which is why scientists tests hypotheses and don't publish papers 'there is no reason this is not true' all day). Then it says that pedestrian hit at speed X have better chances than when being hit at speed Y. But does each car hit a pedestrian? By this same method you can say that blindfolded drivers will drive more slowly. Will it be safer?

Sorry, this is no science, this is pushing an agenda. It uses nice subtitle "The Studies: Rare but Conclusive" right before citing a paper, where even cherry-picked quote says it was statistically insignificant. It uses grand wordings to play as having all in hard data, but it stretches conclusions by untested implications. It doesn't compare costs and outcomes to other alternatives (optical speed "bumps", lenghtwise wave markings, etc.). It doesn't mention effects on driver fatigue (you have your nice slower speeds on several streets, but people drive half an hour through that and pay less attention elsewhere), it doesn't consider long term impact - many traffic solutions shows changes, because, well, they ARE changes. Some of them regress very quickly - e.g. make a few roads narrower? It is a change in both the temporal aspect (a week ago, the street was wider) and the citywise movement (you drive and suddenly the next road is narrower). Change several roads throughout the city and this mind trick will be gone, people will get accustomed and speed up, now on narrower street. What is the worst of this, is that five years from such a change, articles like this one playing it "scientifically" could compare the 'before' and 'after' and still claim decrease in accidents and deaths regardless of the true effect of the narrower road. All you need to do is:

- not account for changing vehicle park, where old cars full of sharp edges are scraped and newer cars are shaped all around accidents, there must be clearance between engine and bonnet, some bonnets even pop up, etc.

- not account for common driver assists (ESP, blind spot indicators, bird's eye parking view, rear passing traffic alarm, etc.),

- improving systems like autonomous braking (talking not only about Tesla, it has been out there for quite some time, it only needs to come down to cheaper cars),

- healthcare improvements, new methods, new apparatus.

All of these could make previously deadly accidents survivable or could prevent them altogether cooperating with the driver or even autonomously, regardless of whether you make your roads narrower or not. Accounting for all these effects is hard (not impossible) and it is beyond the level of "scientificness" often found in articles like these.

The article is simply the same politics as from politicians, only targeted to a little more sophisticated reader than the typical "experts found out X. end of article." articles from media, yet the reader must be one who doesn't orient themselves in science and politics of agenda pushing transportation-proposing-articles.

Urban design, and transport in particular are intrinsically political subjects. You can optimise things in one way or the other in a rigourous way. But there is no way to objectively determine how to balance different issues. That has to be decided politically, and then a solution is "engineered". Using science to draw political conclusions is destined to bias however well informed the person. But there is nothing wrong with pushing an agenda.

Exactly. The author came across as smug and not credible, since he was clearly pushing an agenda. The first red flag was his straw man:

> Think about your behavior when you enter a highway. If you are like me, you take note of the posted speed limit, set your cruise control for 5 m.p.h. above that limit, and you're good to go.

I can't remember the last time I used cruise control, but it was definitely on a 200+ mile trip. If the author is such an un-engaged driver that he relies on cruise control every time he gets on a highway, I'd hate to think how lackadaisical he is when he's "only" going 40 MPH around town. Certainly drivers like this—to the extent they actually exist—would need encouragement to stay attentive. But pretending that everyone is this checked-out while driving isn't realistic (I've ridden with many drivers from all over the US in the last 20 years, and no one reaches for the cruise control right after getting on a highway).

It may be the case that narrower lanes is the way to go, but the tone of this article undermined the persuasiveness of the evidence cited.

What correlation does using cruise control have with not paying attention? I think that's a strawman. I certainly feel much safer on the highway with people driving at a relatively constant speed, rather than gunning up to 80 mph, then falling back to 60 because they are fiddling with their radio or taking a call, then accelerating back up.

> Think about your behavior when you enter a highway. If you are like me, you take note of the posted speed limit, set your cruise control for 5 m.p.h. above that limit, and you're good to go.'

This is exactly what I do, every time. Except I set it at the speed limit, and trundle along in the right-hand lane. You get there just as fast, and don't run the risk of attracting any attention from the five-o.

As a US driver living in France, I can share just the anecdote that French driving far less pleasant, more stressful, and feels more dangerous due to its narrow traffic lanes. I can also report seeing high numbers of highway accidents in Paris, and I would be surprised if per capita traffic fatalities were lower in France. I dream of driving in the luxurious, relaxing wide lanes of the US.

According to wikipedia numbers, France has half the traffic fatalities per capita, and and a bit less than the US per distance driven.

https://en.wikipedia.org/wiki/List_of_countries_by_traffic-r...

Making driving more dangerous forces drivers to focus harder, and focusing harder makes drivers less likely to screw up. The thing the article fails to address is the stress this adds to everyone's commute. The stress of a daily commute can have a serious impact on a person's health. Deliberately increasing the stress associated with driving without considering the ramifications is irresponsible imo.

http://time.com/9912/10-things-your-commute-does-to-your-bod...

"Making driving more dangerous forces drivers to focus harder, and focusing harder makes drivers less likely to screw up."

The inescapable conclusion is that we should get rid of seat belts, mount the driver in a plexiglass bubble in front of the front bumper, and put a bloody great spike in the middle of the steering wheel.

Any takers?

In a related question, there is some evidence that bike helmets may lead to greater risk taking and thus increase the likelihood of a crash (though bike helmets pretty definitively reduce the risk of head injury should a rider be in a crash).

http://bigthink.com/neurobonkers/the-bike-helmet-paradox

Your "inescapable conclusion" would probably depend on whether the driver actually perceived the increase in risk. The spike and bubble might, but removing airbags might not.

That is a pretty far off base conclusion, the goal is to make roads feel unsafe for drivers so they go slower, and reduce or eliminate commute distance, that is how roads are made safer.

For example, a common strategy is to narrow a road from 4 lanes to 2 with a center median, being either a turn lane, planter with half height jersey barrier (directs cars wheels back onto the roadway), or barrels filled with water (which absorbs impacts and makes roads much safer). Tactics like these make roads feel more claustrophobic and thus "dangerous", but the road is ultimately much safer in the event of an accident, and accidents are less likely due to drivers being more alert because of the narrow roadway.

TL;DR: Wide roads lull you into a false sense of safety, narrow them!

I had an other smart acquaintance who believed replacing airbags with spikes would make people safer.

Ridiculous.

That's like saying that decreasing did production will make people use food smarter and starve less. It'll improve the efficiency of food usage, but fewer people won't starve.

Steering wheel spikes will improve driver habits, but people won't be safer overall.

Typo: decreasing food production

Some things make us feel much safer, but don't actually increase safety much (wide streets).

Some things make us much more safe, but don't actually increase safety much (seatbelts).

The latter increase safety (for drivers, at least--they could conceivably decrease safety for pedestrians). The former do not.

I agree with you up to the stress bit. Until we have reliable autonomous driving we should be fully alert while piloting heavy objects through shared space at high velocities. It adds to your cognitive load sure, but so does killing an equally distracted pedestrian.

Hey, there's a great way to significantly reduce commuter stress: Efficient and comfortable public transport.

Can you name a city where it's normal for public transit commuters to get seats at all (unless they live an hour+ ride out of the city, where the train starts), let alone comfortable ones? I'm skeptical that there is such a thing as efficient and comfortable public transport. In my experience only underutilized public transport is comfortable, and public transport is only underutilized if it's slower than driving.

The trend in most systems is towards inward-facing seats along the sides for the very lucky few, a huge mass of standees in the middle (the denser the better until additional passengers physically can't make it on), and width of seats/separation between seated passengers makes basic economy on a budget airline feel positively luxurious.

Off-peak errands and entertainment on public transport can be fine, but not commuting. How anyone can find physical contact with strangers in a cramped, enclosed space less stressful than driving is way beyond me.

First of all, I didn't say you get a seat, I said comfortable. I can stand for quite a while, read a book on my phone and be comfortable. Secondly, there is a huge margin between having to stand but getting your own little space and being cramped. Former is expected, the latter is something you can plan around when choosing workplace and home. Also, how annoying cramped spaces are is heavily dependant on who else is there. That ranges from drunk football (soccer) fans at one extreme to Japanese. With the latter I don't care at all being cramped, everyone stays relaxed, gives way when needed and is perfectly behaved in queues. I'll take that any day over standing in traffic - because I can completely switch off, no responsibilities apart from getting off at my stop.

Caltrain in the SFBA whenever there isn't a Giants game. There are bike cars where I can park my bike and sit down.

I believe it's slower than driving, on average, (especially with the non-express trains) but the variance is lower (fewer traffic accidents).

> I'm skeptical that there is such a thing as efficient and comfortable public transport.

Have you tried living in a country made for people, as opposed to cars?

Like, just about anywhere in Europe.

Yes, narrowing lanes will increase drivers' stress (whether by a significant amount, I don't know). But it will also decrease the stress of cyclists, pedestrians, and other people using the same roads. So this is not a straightforward consideration against narrowing lanes, or any other measure that will increase safety but increase drivers' stress as a side effect. To me, it sounds like a good trade: when the stress of driving goes up and the stress of cycling or walking goes down, more people will switch away from driving, and there are many reasons why that would be a good thing.

Actually will narrowing lanes increase driver stress?

Presumably, the increased stress is why they drive slower in the first place, which should then have a canceling effect lowering their stress back to the original level.

If you get too stressed, maybe you should drive a bit slower.

I know I am not controlling for road conditions, car conditions, traffic signals conditions and culture of respect to traffic laws, but... in Brazil, IIRC, the lanes are 10 feet (3 meters) and we are world champions for road fatalities per vehicle-km (https://en.wikipedia.org/wiki/List_of_countries_by_traffic-r...)

It does not disproof the OP, but it is good evidence that lanes width is not the most important factor. But I honestly don't know if all other more important factors are fairly good in the USA.

edit: I was checking brazilian regulations and it lane width is defined on a rage from ~8 to 12 feet, depending on the "class" of the road (from small, unused roads in small towns, passing through regular traffic heavy urban roads to interestual highways with heavy trucks).

Even 10 feet sounds ridiculously generous by UK standards! Only on motorways would you normally find 12 foot lanes (and even then, they're not always so wide).

I suspect many unfamiliar American drivers would be horrified by UK roads and wouldn't consider them "safe" - yet accidents and road deaths, per mile driven, are significantly lower.

As an American who got to drive over central Ireland on a vacation and my wife spent half of the time covering her eyes as big farm trucks passed by with just an inch or two of space on my side and an ancient looking stone wall just inches from her side. We weren't even in a big vehicle, it was just a tiny Ford.

It certainly didn't feel any safer, and I couldn't help but to notice all of the sideswipe marks on various vehicles when we drove through towns.

Sounds like the west coast of Norway, just replace the farm truck with a semi trailer and the stone wall with a mountain or a 200 meter straight drop.

Sounds like parts of my journey to and from work: one side is a drainage ditch backed by a rough granite cliff rising vertically, the other side a short drop into the fjord (short drop because I live on the east coast of Norway not the west).

I'm in Norway - and this honestly makes me hesitant to drive, whenever I do wind up with a valid license here.

The author was on a roll, and then he decided to bring Florida DOT and Florida drivers into the article. For anyone that's driven in or through Florida you've seen first hand some of the worst traffic engineering in the US. FDOT excels at creating weird intersections, dangerous on and off ramps, etc.

Likewise Florida drivers along the east coast are some of the worst I've seen. Even with 12' lanes they seldom get up to the speed limit. Driving in Jacksonville during rush hour is not for the faint of heart. Miami traffic is own special version rapid lane switching without signals that is the hallmark of Boston. Except in Miami, lane changes can take 1/4 of a mile and blinker can stay on for miles afterwards.

Should have offered up another place to try the idea out, I'd suggest Dallas. Those drivers would be happy with 8 and a half foot lanes, I've seen them shove giant Dodge and Ford pickups in to smaller spaces at 60MPH.

The article agrees with you.

"It is striking to hear this news from FDOT, the agency that may preside over the greatest pedestrian massacre in U.S. history. Four out of the five deadliest American cities for walking are currently in Florida. This is by design: in no other state has the DOT had such a powerful influence on the design of urban streets."

"Those drivers would be happy with 8 and a half foot lanes, I've seen them shove giant Dodge and Ford pickups in to smaller spaces at 60MPH."

Why wouldn't they fit there? I've watched many Texans "happily" park their F-350 dual wheels in compact spots (http://cdn1-www.craveonline.com/assets/mandatory/legacy/2014... and https://ardent-camper-production.s3.amazonaws.com/uploads/im... just to name two)

I've lived in Boston, Jacksonville, and Miami and i think you're right on point. Miami has a special intersection (pun!) of retirees, newer immigrants, tourists, and young people (not saying any of these groups are bad drivers as a rule, but they certainly have a wide array of driving styles); combined with poor infrastructure design it's a deadly mix. I have friends who've moved from Miami to LA, and they all agree the traffic in South Florida is worse.

The evidence here doesn't seem that strong -- just manuals that hint that "10-foot lanes are as okay as 12-foot lanes" and the odd (non-randomized or large scale) study here and there.

I'm also curious whether a good part of the author's conclusion is driven by his like of bicycles and pedestrians. I live in a city, and almost exclusively walk and bike, so I'd be on his side. But a key part of his argument seems to rest on it potentially being able to free up more bike lanes, about it being faster for pedestrians to cross the street.

If part of the motivation is that towns need to be more pedestrian and cyclist friendly (potentially at the cost of cars) that's fine, though I would appreciate it being explicitly pointed out. Right now it's a bit confounded.

Over there in Europe, we have mixed approach which IMO is great

- Highways (including city bypasses etc highway-like roads, not only posted highways) are plenty wide. Like, dual-carriage can easily fit 3 or even 4 trucks next to each other.

- Major fast traffic city streets (no pedestrians, good visibility, preferably crossings are 2-level or controlled) Are wide-ish to drive comfortably, 3 trucks could barely fit on dual-carriage, but 3 cars - easily

- City streets (pedestrians highly possible, cars coming in/out of drives, no speed-up/slow-down lanes, uncontrolled crossings) are just wide enough to drive comfortable at 50km/h. May have speed control features.

- Residential streets. Narrow streets. Parked cars on the sides. Driving at > 30km/h is not comfortable. 2 trucks could hardly pass. Lots of speed control features.

In addition to that, we usually have traffic features that warn of upcoming speed limit. For example, small towns frequently got an artificial S turn or shallow speed bump-. You can easily pass it at 50-ish km/h. Sometimes it's possible at 30 or 70. Which is usually the speed limit for upcoming stretch of the road.

I love this approach. The road itself should tell it's limit. Safe-looking roads that are actually not safe suck. Point in case, we recently had some 2-wide-lanes-one-way streets converted to 2-narrow-lanes + bus lane. I was sceptical at first, but in general traffic calmed down, less assholes, less speeding.

This approach sucks if there're no enough highway(-ish) roads and bypasses though. It's a torture to do a long drive through small cities. On the other hand, it's safer to residents of those cities if people passing through don't speed 2x the limit..

In summary: Make lanes narrower to annoy drivers into driving slower. (Although 12 foot does seem excessive to me.)

I think narrowing lanes/roadways to slow down drivers is probably one of the least annoying ways to do it, while remaining effective.

Posting low limits isn't very effective without enforcement which is annoying. Speed bumps can be effective but are annoying. Poor road conditions or shifting lanes are effective but annoying, but effectiveness depends on traffic: When there's no traffic, I may straddle lanes if they're going back and forth all the time.

As a driver it's also annoying to be driving on a road where the safe speed is much lower than the apparent safe speed, because I have to keep slowing down. I don't really intend to go 40 mph on streets with houses, but when they have wide, well paved lanes, sometimes it happens.

Absolutely. The road's design language should communicate these things to drivers. Knowing 65 vs 70 may require signs, but you should be clearly informed by overhanging trees, narrow lanes, and activity when you are in a busy area.

It also builds aversion to law enforcement, who have to stand in for proper design by charging people large fines. I don't think it's a stretch to assume some link between police enforcing seemingly malicious and arbitrary rules, and growing public mistrust of police.

Annoy? Maybe scare-the-sh*t-out-of is more like it.

The interstate 74 suspension bridge over the Mississippi river where I live was built in 1935 and upgraded in 1970.

It's 2 lanes, no shoulders, and speed limit 50. Getting next to a semi truck on that thing is never fun.

They recently finished replacing the box girder bridge over the Tennessee near where I live.

2 lanes (one way, for the last couple decades), no shoulder, 55mph limit, 14' clearance. Log trucks got both lanes, if you were fond of your life.

I'll miss it.

"If we make driving less safe people will drive safer"

Yes. This is absolutely true. People don't speed around mountain passes on snowy nights. So I'm not clear what your point is.

A better way to frame it is do we want to optimize everything for automobile travel speed and convenience? That's what the US has done for the last 100 years, and the result is poor public transit, unwalkable neighborhoods, urban cycling as extreme sport, and an epidemic of road rage.

Rather than making roads visibly more dangerous, there are better options.

One I heard long ago on a motorcycle forum is: get rid of seat belts, mount the driver in a plexiglass bubble in front of the front bumper, and put a bloody great spike in the middle of the steering wheel.

I wonder why that idea hasn't caught on?

> People don't speed around mountain passes on snowy nights.

I don't follow. Driving on a snowy mountain road is still incredibly dangerous though, even after I've adapted my behavior.

You're implying that slippery, curvy roads are safer?

No. The implication is that driver behavior adapts to road conditions, and that slippery curvy roads demand higher attention and care to navigate. In many instances, improved driver attention compensates for the increased risks of dangerous road conditions: more "dangerous" roads may not have proportionally increased accident rates or fatalities.

In the extreme case where a road appears life-threatening, a driver with that knowledge chooses not to drive.

The most dangerous conditions occur if the road appears safe but has unforeseen or transient hazards.

That's actually one of the most basic principles of traffic design.

Roads built in the Netherlands and Germany are designed so that you absolutely can't drive safely at speeds higher than the supposed speeds - it feels unsafe, and you just don't.

For example, in areas before schools, you usually limit the road to one lane, and switch which side that is every 50 meters, requiring cars to drive slalom, which in turn requires cars to slow down (or crash into concrete).

You don't make it actually unsafe (there's enough space that it won't ever cause any fatal accident), it just has to feel unsafe.

EDIT: More examples:

Another example are the emergency lanes and outer lane markings on the Autobahn: they are a lot louder to drive on intentionally, so if you accidentally steer onto them, you'll hear a very loud noise, and your car will vibrate just slightly - if you fell asleep and steered onto them, you'll instantly wake up again. Another example would be speed bumps.

they are a lot louder to drive on intentionally, so if you accidentally steer onto them, you'll hear a very loud noise, and your car will vibrate just slightly - if you fell asleep and steered onto them, you'll instantly wake up again.

These are present in the US too:

https://en.wikipedia.org/wiki/Rumble_strip#Continuous_should...

I've seen a similar principle in the Alps. Every town is preceded by a really sharp turn. Works much better than any speed limit sign.

Yes, this is part of what is not just good, but great design. It just works, thanks to subconscious influence.

If we'd spend more of our resources on designing such systems that any human instantly gets, I'd assume humanity would be a lot better off.

Yes.

It will. And it does.

Massachusetts roads are infamously lined with trees, rocks, and all sorts of other barriers that make it a very bad idea to drive carelessly. Drive like an idiot and you'll send a body shop owner's kid to college.

However, that steep price for careless drivers makes life a lot safer for those of us biking or walking in MA.

I don't know where this is coming from; I've been in MA for a long time and never any road obstacles like the above. Now having said that driving backroads is not easy to do fast simply because they are windy and narrow.

I guess if you compare to someplace like Arizona where if you spin out you just end up in sand instead of hitting a tree, sure, but why single out MA instead of just anyplace with trees? :)

I am not kidding about this: there are places in the Midwest where people have been required to remove trees from their front yards in order to make things safer for drivers who might drift off roads.

My grandmother lived o a very wide but otherwise normal street in South Bend, IN. She had a row of GIANT oaks in front of her house. There were at least 3 crashes I heard about that killed drivers. People would drive down the street way too fast, hit a slight bump about 40-50 feet down the street from her house and lose control. The city offered to cut down the trees after 4 teenagers died a while back. She declined though, because they would end up in her living room if they didn't hit the tree.

That being said, there's only one tree left, cause that last accident killed the tree they hit.

I'm from Minnesota and have put in 100,000+ miles of driving there, much of it in the winter. I'm not sure how I feel about removing trees and other side-of-the-road obstacles, but just wanted to point out that it's possible in very bad conditions to drift off the road, spin out on the highway, rear-end the car ahead, etc. even if you're being extremely cautious. I hit black ice at 30 mph on the highway and my car rotated a full 540 degrees, but fortunately didn't end up in a ditch. Sometimes conditions and visibility can be so bad in Minnesota, that I don't think any outsider can understand how difficult it is just to stay on the road. I've taken drives where you see cars in the ditch every quarter mile. I rear-ended a guy once, but he wasn't even angry because he saw my car slide slowly toward him from at least 50 feet away.

I once hit black ice at 10 MPH, just as an old man was crossing. Luckily for me, the black ice ended 12 feet before I'd have hit him.

At that speed, I'd rather have hit a tree.

Point is, when the roads are narrow, and you know that any mistake will cost you money at the body shop, you don't travel at dangerous speeds.

I understand the need for life to go on in places with hostile weather, and also that unexpected situations come up.

But at some point, a reasonable standard must be set, and if you can't drive safely you shouldn't be driving. Clearing obstructions on the side of the road won't really help if you're spinning out of control on a busy highway.

That's to make the safety zone larger. You know, so drivers can safely drive off the road. https://youtu.be/P9BUyWVg1xI

In greater Boston, the DCR (Department of Conservation and Recreation, formerly the MDC) maintains many arterial roads such as Storrow Drive, Memorial Drive, VFW Parkway, the Jamaicaway/Arborway, etc.

These roads are considered "pleasure roads" and date back to the day when a drive in the outdoors was primarily recreational. The roads connect and traverse major parklands.

DCR is not part of MassDOT (Department of Transportation). These roads are not necessarily built to traditional highway standards.

For example, trees along Memorial Drive and the Jamaicaway are closer to the roadway than would generally be recommended. These characteristics are linked to the historic character of these roads and their surrounds. Periodically, there will be plans to change the roads in some way, but they are usually met with significant pushback (neighbors, parks people, preservationists). On the other hand, the DCR for years resisted standard pedestrian crossings as well.

Currently, I think there is a slow trend to make these roads more standard where possible but also slow traffic speeds and improve pedestrian access.

I'm assuming you haven't had much chance to compare, so I'll give you my impression (having driven rural roads in CA, NV, NM,UT,CO, AZ, KS, MO, IL,TN,PA, and every East coast state except FL in the past six months.) RI, CT, and MA are very uncomfortable because it's common to have narrow roads with rocks and trees no more than 18 inches off the roadway. In VA, for a contrast, trees grow everywhere, but along rural highways there is only relatively short grass for the first 15 or so feet off each side. And everyone who drives VA enough hates the state for its police (which I've heard frequently from people living in every state from new York to south Carolina.) There are also places in VA where that dynamic doesn't exist (west of the Appalachian trail) because, like you said, the winding roads, rather than police, do the job of slowing cars down.

And admittedly AZ is easy. Passing in the oncoming lane while going 90 is a pretty comfortable thing there.

Yep. Traffic fatalities increased after clearing the sides of intersections to remove obstacles.

If your field of vision is covered with trees, you drive much slower. If you feel like you're driving on an open road, you drive much faster.

Obviously there are some cases where the increase in safety outweighs the increase in recklessness, but not all cases.

This relation is often true, but not always. For instance, if a driver can't see past an obstruction and needs to pull into a crosswalk to look for oncoming traffic, he or she may neglect to look for a crossing pedestrian because of a prioritization of risk.

Progressive traffic engineering is required to balance these competing interests and safety needs on a case-by-case basis.

"A crowded society is a polite society."

I expect a lot of resistance to this idea, no matter it's actual impact on safety, from drivers. I know that driving in narrow lanes makes me nervous, and I feel better driving in wider lanes. If narrower lanes will make driving slightly more unpleasant and stressful for most drivers, they're likely to be opposed to it.

People bought big SUVs to feel safer, but it's an illusion. 10 foot lanes would do much to contain this trend. They would also stop texting and pay attention.

Completely anecdotal, but I agree with this article 100%

I spent the first 6 years of my (oldest) kid's lives in the immediate bordering towns of Madison, WI. Nice, straight, flat, wide lanes. The people tore through those streets; I did as well, before I had kids.

We moved a few years ago and now live outside of Boston. Narrow, winding streets that are easily 1/2 as wide as WI.

I feel much safer with my kids in this neighborhood than in WI. And it's for every reason listed in this article.

In 50 years, when autonomous vehicles have become a reality, all the 4x 12' lane roads can be converted to 6x 8' lane roads. Everyone wins!

So.. New Zealand has 3.25m wide lanes typically. If you are interested in studies about how variables such as this and speed affect yearly fatality outcomes, NZ Transport may be a good place to start. With ACC and public health in general safety and reducing healthcare costs(read: traffic accidents) is a huge concern.

That being said, anyone driving from back to Auckland from the snow at 120kph+ knows that you don't drive slower due to the narrow lanes and crazy windy roads as much as you drive to the tolerances ;) Just like in US mountain areas you may be timid at first and get smoked by and annoy the locals, but you will soon learn to push the envelope.

I've said it before. Putting sidewalks right next to vehicle traffic is stupid. I suggested to a friend that putting walkways through the back yards in a suburban neighborhood would be a better idea. You'd also have half as many because they'd run between two properties. He said oh yeah, they do it that way in... I think it was Texas?

That planning decision articulates to the drivers and pedestrians that the cars will encounter few or no hazards, and that the pedestrians are not welcome near roadways. It also supports things like large safety areas (25 ft) on either side of the road that allow drivers to be minimally injured if their cars leave the road. It tells everyone, "cars have margin of error to wander out of their space" and, following from that, "pedestrians shouldn't feel overly confident about not encountering cars in pedestrian space." That's a pretty good way to do highways.

The alternative is the reverse, where pedestrians wander, and drivers know not to get too comfortable even in their own spaces. This arrangement supports commerce and community and vitality.

You can't have it both ways. You're either moving cars, or you're building places. Both have their place, but they are separate places.

They've started doing exactly that in a couple of the newer developments here in Gilroy (California). The roads within the development are significantly narrower, but there are now bike lanes separated from the traffic by actual curbs instead of just a painted line, and the sidewalks are a bit wider, too. The biggest walkability boost is that all cul-de-sacs now have walkways at the top of the 'bubble', so that foot and bicycle traffic can pass straight on through. Also, roundabouts have replaced stoplights.

On the more negative side, the other feature of these development is the size of the lots vs. the size of the homes on them. The houses are close enough that you could probably reach out a window and hand something to your neighbor without much effort, and while each house does have a (small) garage, there is very limited on-street parking.

I have mixed feelings about whether it will succeed, simply because the locations chosen for these developments are many miles away from the nearest stores, so anyone who lives there is going to need to drive most of the time anyway. Some of this is just the chicken/egg problem, some of it unchangeable without completely flattening and rebuilding the city. It will be interesting to see how it goes. I personally am glad not to live in such a place, as it as cramped as any city dwelling, but without the positives that dense cities provide.

I love cul-de-sacs with open ends for pedestrians. I'm glad to hear they're starting to get more love.

Yeah, that is probably my favorite aspect of the entire system. The slow road speeds and open space that allow kids to safely play outside, but close to the walkability of a grid.

Neighborhood A: You stroll down the sidewalk under a canopy of trees and wave to your neighbors on their porch.

Neighborhood B: You walk behind people's houses completely isolated from everything.

Human drivers are bad for safety[1] and should be replaced. Good news is that we're already on track to solve the real problem.

[1] 1.25 million deaths per year: http://www.who.int/gho/road_safety/mortality/en/

In a neighborhood I drive through every day the city cut off with a painted divider roughly 4 feet on each side off the road for bikes. I think as a result the speed I drive has decreased somewhat.

Drivers adjust speed according to visual clues. The speed limit is only one factor and probably not even the most important one.

I'll keep this bit:

we, as a society, are more than willing to sacrifice lives for automobility

Perhaps that's the point we need to reconsider, the cause-not-symptom behind the effect that the OP associates with a simple metric (which, to me, sounds a bit superstitious: why precisely 10 foot lanes, and not 9.5 or 10.5 ones?)

All kinds of things can go wrong when the lanes are too wide: https://www.youtube.com/watch?v=CwinnODU0yo

I had a difficult time agreeing with very much of the author's arguments made in this article. In particular, the opening argument "What's the number one most important thing that we have to fight for? ... Well that's easy: 10-foot lanes instead of 12-foot lanes.", which caused me to read the rest of the article. This argument was, at best, poorly supported and at worst, an attempt to inflate the argument and keep me reading the remainder of the article.

To back up this statement, he cited two studies and states that there's few good studies. The lack of studies is, in itself, a big problem. Making large changes on little data invites unintended consequences -- much like those he's stating were caused by the original change to 12-foot lanes[0]. This would be OK if the studies had a very strong supporting conclusion about current circumstances facing drivers, today, but the studies' conclusions are weak. One states that "10 feet or more [rather than 12 feet] resulted in accident rates that were either reduced or unchanged" and the other indicated "analyses conducted were generally either not statistically significant or indicated that narrower lanes were associated with lower rather than higher crash frequencies."

He later draws a few conclusions about drivers "feeling more safe" in 12-foot lanes, causing more risky driving (supported only by anecdote). He stretches quite a bit with the anecdote about how when we merge onto a freeway with their really wide lanes, we set the cruise control 5MPH higher than the speed limit and relax more because the 12-foot lanes give us the illusion of additional safety. This is presented as a "Common Sense" anecdote, but I'd be willing to bet there's more data that people feel safer on highways because traffic all moves around the same speed, there are no traditional intersections, and no sudden stops except in cases of high traffic volumes (where it's anticipated) and accidents. I've never even noticed that highway lanes are wider, but I can picture it now. He then goes on to compare things that are not comparable -- residential roads that are 12-foot wide with traffic going in both directions -- those streets are safer because there's almost no traffic to compete with and people are usually driving far slower than they are on county roads.

I support the last part of the article, about adding bike lanes, but that argument suffers from the same weaknesses. Though I suspect adding marked bike lanes improves roadway safety for cyclists, I can't point to data showing that and the author did not provide it. In addition, his original argument that reducing lanes to 10-foot is "The #1 most important thing" isn't covered by this unless road engineers -- in designing roadways at 10 feet -- automatically include bike lanes as a result. We have a strong drivers lobby in the state that I live -- Michigan. In areas where shrinking the lanes would leave a half or more of another lane, you're more likely to see the road widened that extra half-lane to increase capacity for vehicles, not cyclists. That would make intersections like 12-mile and Telegraph[1] more dangerous with yet another lane of traffic that might not notice a pedestrian.

As much as I'm sure this is something the author feels strongly about, the facts provided don't add up to the conclusion he drew and if I'm being particularly uncharitable, after reading this article, I felt like I'd been taken in click-bait fashion. Before reading the article, I thought interesting, here's an issue I didn't know about and apparently I might want to care about it and after reading, unfortunately, I concluded almost the opposite -- that we shouldn't change anything until we have solid data proving that 12-foot lanes are a problem.

[0] One that comes to mind is how does this change affect the safety of self-driving vehicles that we're likely to see in a few years?

[1] About a decade ago, this intersection was in the top 20 or so of most dangerous in the United States. It's still high on the list in Michigan, so either a bunch of new intersections became more dangerous or when they reconstructed the intersection, they improved safety. I'm not sure, but I worked near there and the reconstruction, which moved the sidewalk closer to the road, made it a lot easier to notice pedestrians preparing to cross the street when turning.

But with 12 foot lanes, you can adjust the lanes after the fact to 10 foot lanes and put bike lanes everywhere!

2 feet is inadequate for a bike lane. I just measured the width of my bicycle and it's 23.5 inches, so even if I rode perfectly central I'd have only a quarter inch clearance each side. And my bike is not unusually wide.

But if it's a 2 lane road (4 total), then you'll have 4 feet for the bike lane.

Sounds like bro-science to me

I'm confused, why are we designing streets for human drivers? :)

Because while AI driving will no doubt be popular over the next 50 years, its not right now and won't be ubiquitous for another 20 or so.

Humans still drive, if you haven't noticed.

If you care about pedestrians, add more pedestrian controlled pedestrian crossings. Train pedestrians to use them correctly.

Are you talking about beg buttons?

I am talking about pedestrian crossing traffic lights, that can be called by pedestrians to turn green. Appears that these are indeed called beg buttons.

They are called like that because they don't actually do anything. They are there just to calm down impatient pederastians by giving them illusion of control ;)

Not universally true. I have seem some crossings in Texas and elsewhere where the light is solely for pedestrian crossing and will change (after a short delay) on command. On intersections that are sensor-controlled they will also change the light if there is no cross traffic to change it for you.

That said, on intersections where the lights are purely timed or where there are always cars waiting in both directions, it pretty much is a "beg button". At that point, there is no way to allow on-demand pedestrian crossing without making the roads useless to cars. In fact, some places like NYC don't even have buttons because they are pointless and possibly counterproductive.

That's largely a myth, at least in the US. Everywhere I've lived really needs a button or the "Walk" cycle is skipped to give longer turning times for cars. But because of the myth, you get infrequent pedestrians who believe in the myth and just wait until somebody on either side actually pushes the button.

Traffic "engineers" don't deserve the name. I'm certain not one of them has ever taken a bike and tried to travel on their own design.

Their allegiance is to parking spaces and car throughput, in that order.

Meet Seattle'a traffic engineer:

https://twitter.com/dongho_chang

Scroll through his Twitter a bit and you'll probably find him taking pictures of interesting things and possible safety concerns he encounters as he bikes through his own city.

the exception, not the rule.

I highly disagree. This article assumes perfect driving conditions it also suggests that people drive faster with wider lanes -- then the problem is a function of speed limit and not lane width and also assumes that pedestrian safety is somehow affected solely because of lane width. If pedestrians are getting hit by cars one of two things has happened: the pedestrian jaywalked or a driver didn't stop when required -- both cases which have nothing to do with lane width.

I live in rural France with narrow roads and it's incredibly stressful to be driving with nearly no margin of error preventing a head on collision.

What about a stalled vehicle on the road? Debris?

I get it, the author doesn't like cars, but making lanes narrower to improve safety is counterintuitive. if the problem is excessive speed, then traffic enforcement is the problem.

This is like reducing the size of your front door to make it harder for thieves to steal your couch.

If you build a road through downtown the same way you build a superhighway, people are going to drive on it at superhighway speeds, which is unsafe for pedestrians, cyclists, and drivers because downtown has more intersections and interactions between modes of travel and directions of travel.

You can try to fix that by posting a low limit and ticketing. However, that only works while enforcement is high. It's much simpler to build the road to encourage appropriate speed, which means narrowing the lanes. You can still have a wide shoulder for avoidance of debris or stalled vehicles: It helps to have it well marked so drivers don't feel it's part of their lane.

It's not jaywalking to cross the street at unmarked intersections, but it can be dangerous when drivers are traveling faster than speeds where humans are good at estimating, or where there is poor visibility due to turns in the road or vegetation or whatever.

Also, since the author is writing about Florida, age and humidity are factors.

For highways where high speed is appropriate, wide lanes are too.

> If you build a road through downtown the same way you build a superhighway, people are going to drive on it at superhighway speeds, which is unsafe for pedestrians, cyclists, and drivers because downtown has more intersections and interactions between modes of travel and directions of travel.

Sorry, but that contradicts itself. If you build a road through downtown with the properties of superhighway (side barriers, on/off ramps, only motorized vehicles allowed, separated oncoming lanes), there would be none of said intersections and interactions between other modes of transport and the only concern would be noise pollution and more frequent merging ramps.

People don't drive at superhighway speeds only because they have wider lane markings and there is a sign posted, but because there are no oncoming cars with potential for head-on collision, because merging cars speed up almost to the same speed and you suddenly don't get a car appear sideways in front you from a side street and other similar factors (e.g. empty peripheral vision on highway vs. fast moving tall buildings downtown). If you put a sidewalk with people or 90-degree intersection on a highway, you wouldn't get highway speeds neither. Speed is not a simple function of lane width.

There is however another issue - why is the problem of safety reduced only to speed, especially when the speed is obtained by placing more cognitive load onto the driver? When the driver must constantly check the oncoming vehicle whether it doesn't swerve too far into his lane, eyeball the vehicles next to him whether they won't start changing the lanes without indicators (being closer, there is less time to use the horn or react quickly, and having less space to the another lane/oncoming lane, there is less elbowroom to avoid them) - then, yes, drivers would drive slower, but also pay way less attention to pedestrians on sidewalks, cars pulling away from intersections, parallel parked cars about to pull away and other things. Even the argument with the crossing pedestrian - yes, when you have narrower lanes, there is less distance to cover. At the same time, the pedestrian appears right in front of the cars in their lane. Another problem related to the cognitive load and perception - you may just be shifting your problem elsewhere. If you have a city and suddenly, four blocks have narrower streets, people will slow down, because it is suddenly different and driving through takes some of their brainpower. If you have the entire city like that, then: a) it will become the norm and the speed will go up again after some acclimation, b) after a drive to the other end of the city, the drivers will be much more exhausted and likely to be involved in an accident.

Anecdata: 150-200 mile drive, one break. If you drive that on highway, in the end of the journey back in your city, you are somehow tired physically from all that sitting and paying attention, but otherwise still alert and still would spot suspicious behavior (car about to pull in front of you, pedestrian in dark clothing in the night, etc.). If you ride that on country roads (think central Europe), you eyeball every tree and bush while driving to forest (a boar or deer jumping right in front of you, common occurrence), you drive through sharp corners with who-knows-what behind them, you drive through many small cities and villages with no sidewalks and pedestrians in dark clothing. Yes, you drive way, way, way slower than on highway, but in the end, you are tired not only physically, but also mentally and you just drive home following the central lane marking ignoring surroundings to a certain level. Is that more safe?

This approach just tightens the safety of drivers to the limit, so instead of using other tricks working on the same human brain principles, you make it really more dangerous. Yes, it will make a few people catching pokemons while driving put their phone down and pay attention, but at everyone else's expense. And this is road to hell - just like the "phase out schools for talented children" discussion we had in Europe some time before - surely it will help other children in normal (slow) classes, where these children will move.

And it is the same stupid move as installing tall metal speed bump - it works, it lowers the speed, eureka! But:

- it is PITA to drive through,

- it slows down emergency services (here, we have two speed bumps right on a road from ambulance standby post, genius),

- it increases dust emissions (from brake pads, really nasty - try washing your rims from the inside), because the cars have to brake almost to stop,

- it increases fuel emissions, because the cars has to pull away again,

- drivers are trying to hypnotize the speed bump finding the best path through and in the end completely ignore the poorly marked, lit and signed pedestrian crossing behind it, not mentioning the approaching pedestrians.

Yet, you can install optical speed retarder markings, which doesn't impede ambulances, cars instinctively slow down, but doesn't have to almost stop. You can install speed-detecting traffic lights. You can use retroreflective marking built into the crossing. Replace streetlamp to better lit not only the crossing, but also the "approaching" area. But bolting a piece of metal to the road is the easiest, just like narrowing the lanes and - hey, let's ding yourselves, run over pedestrians, drive more fatigued, but you will do that more slowly!

> If pedestrians are getting hit by cars one of two things has happened: the pedestrian jaywalked or a driver didn't stop when required -- both cases which have nothing to do with lane width.

The driver will be able to stop when required when they're driving slower, which they certainly will do when the streets are designed to make them drive slower.

>if the problem is excessive speed, then traffic enforcement is the problem.

This is the solution that we've already tried for decades that doesn't work. When you make it more effective with solutions such as speed cameras, people decry it as a cash grab and they elect politicians that promise to remove them.

Bullshit. If your cars are unable to stop from 50 or 60 mph with a red light then there are 3 possibilities:

1)People driving them should not be allowed to drive them because they are either fucking blind, unable to pay attention, drunk or too stupid

2)Those cars should not have passed the yearly technical evaluation and their registration is not valid

3)The lights themselves are not set up properly or are obscured by something

If you fix all those 3 issues there is no reason not to have higher speed limits on main roads that have only intersections/crossings with lights on them

> 2)Those cars should not have passed the yearly technical evaluation and their registration is not valid

There are places in the US, like the very state mentioned in this article, that do not have annual technical/safety inspections.

Even at 40 mph if you hit a pedestrian there's over an 80% chance you'll kill them so we really shouldn't have people going 50-60 mph in areas where there are pedestrians at all.

Pro tip:

stop cuddling everybody

If you have poor infrastructure, no crosswalks with lights, no separated bike lanes, cars that run red lights then sure, go ahead, go fix that

If some idiot jaywalks on a main road with proper safety infrastructure and gets himself killed, it's his problem

"I live in rural France with narrow roads and it's incredibly stressful to be driving with nearly no margin of error preventing a head on collision."

This is the crux -- I think the author is in fact saying that roads are safer while you are in this stressed state.

"This is like reducing the size of your front door to make it harder for thieves to steal your couch."

The problem is fundamentally different in that the agent on the road at not trying to cause accidents.

Reducing the size of your front door CAN indeed make it harder for thieves to steal your couch. For one thing, it will not stop them from entering, but it can make it difficult enough for them to get the couch out that they select a different target

These are all valid and factual observations, so I have no idea why you're being downvoted.

When Waze alerts me about roadkill I'm usually annoyed, because I only want to be alerted about actual obstructions in the road, but that's because it's trivial to veer within my lane to avoid a small critter.

There are many possible solutions to make people slow down (enforcement, speed limiters) but intentionally decreasing safety should not be one of them.

Reply on news.ycombinator.com