Hacker News new | past | comments | ask | show | jobs | submit login
New book considers the impact of electronic logging devices on drivers (truckersnews.com)
72 points by edward 9 months ago | hide | past | favorite | 68 comments



The problem with ELD's is the lack of flexibility.

So I'm a former driver and I have a good example of this, with paper logs, I could fudge, I might log that I shut down at X time in Y place, when I actually shut down 20-30 miles further down the road - where I could get safe parking in a rest area - instead of on the side of the highway. I'd still take my 10 hour break from the time that I stopped, and mark my logs up as if I started a half an hour before I did. Was this illegal? yes, absolutely. Was it unsafe? no, I'd argue the opposite.

Thats the rub, the ELD system is meant to eliminate the bad actors, but the flexibility and agency is steals from drivers makes everyone less safe, and make the job less pleasant to do. The road was fun, in the days when Truck Stops still had a decent restaurant, before paper logs went away, before the tightening of the DOT Physical rules - now, its not so fun anymore.


I recall reading in the news recently about a trucker who was upset with the system as his normal route meant he had to take his sleeping break 20 minutes from home. He'd then had to drive those 20 minutes home before he could get off his shift.

Due to the electronic monitoring he couldn't just drive those extra 20 minutes and sleep in his own bed.

I know I'd be pretty mad.


Yep but isn't the problem that his employor demand that he drives right to the limit of his legal limit on corporate time and leaves no buffer for say getting home.

The limits on time is not meant to be "norms" but extreme cases so if they are hit daily thats an indication the the culture in the industry is rotten more then anything else.


From what I recall, it was a fixed route. Like drive to a factory, pick up goods, drive back and deliver at docks.

Sure, an edge case. But inflexible rules lead to more edge cases.


I suspect the issue was a variable amount of detainment while loading or unloading - sometimes the dock was fast - as soon as you bump the dock there is a fork truck in the trailer moving the shit out, sometimes you'd wait for four hours to be loaded or unloaded. All of that is out of the control of the driver and the company. The current structure of the industry has little market pressure to force the shippers to be better behaved.


So what it means is that if he could, he would have exceeded the allowed time every single day just to get home? That's not great either. Also you need to decide what flexibility is ok - is doing extra 20 miles fine? How about 30? How about 60? At which point do you say ok, now you're just taking the piss and this isn't ok.


Indeed, that is the problem. If he drove this route once or twice a week it would probably be fine. Every day, perhaps not? Then again, he argued he sleeps much better at home...

And yeah, adding flexibility is very difficult, simple rules are better as such.

And of course, if they add an exception for drivers to get home, I'm sure the companies would exploit that in their route planning.

Not easy either way.


> But inflexible rules lead to more edge cases.

Or alternately a normal working day is 8 hours, and a 10 hour daily limit allows a supremely flexible 2 hours of overtime :)


It's rare to hit your driving hours while driving, but you'll often run out of on duty hours first. Often those issues, detainment at shippers, whatever - is beyond your employers control.


Many truckers are paid by the mile and have total discretion, but need to push the limits to make a decent wage.

More pay is the answer here, as usual.


This is like the definition of a dystopia. Some obscure "public" needs, being put before someone personal individual needs, enforced by a random computer.


The guy wants to drive an unsafe distance/time every day. How is it dystopian that rules and tech prevent this?

If he wants to get home, he can move closer or find another route. Or park the truck and drive his personal vehicle.


If he's unloaded, bobtailing, and driving back home there's technically an exemption for "personal conveyance"; many trucking companies don't always enable PC mode because drivers will abuse the feature to advance a load (which is illegal under PC rules).


Safe/unsafe aren't binary in this case. It doesn't flip from 0 to 1 just because he drove an extra 20 minutes.

The "unsafety" of driving was non-zero all along, and the longer he goes without a rest break make it climb ever so higher. Saying that this factor was safe when it was 0.9000 but as soon as it rolled over to 0.9001 it was unsafe is asinine.

Likely, enforcing such a rest stop could make it more unsafe... if after the rest he was in a hurry to head home, or groggy from an impromptu nap, or even if driving or lighting conditions had changed.

> If he wants to get home, he can move closer or find another route.

Neither of these are practical, in the "completely unachievable in the real world by real people" meaning of the word.


I don't feel one can categorically call it an "unsafe distance". Less safe, sure.

As mentioned his counterpoint is that he'd sleep better at home, so the next day he'd be more rested than he currently gets when he has to sleep in his cab. So in effect it's trading less safe driving today vs less safe driving tomorrow.

Anyway, I'm not at all against regulations in this field. Drivers have been overworked hard and a tired driver on the road, regardless of occupation, is dangerous.


Sure, but society has deemed that 10 hours is the safe limit. Is it an absolute? Only in the sense that we have to draw a line somewhere, because bad actors will abuse soft limits.


> but the flexibility and agency is steals from drivers makes everyone less safe

I have a hard time believing this. You gave a great example of fudging logs for the "right" reasons, but I have countless stories from my uncle who was a truck driver before ELDs who would drive well beyond the limit of what anyone would consider safe.

Clearly, not every example of drivers breaking the rules was in the interest of safety.


Do you build rules just for the bad actors or to do you try to build a pleasant and equitable society?

The bad actors in most cases won't follow them anyhow, no matter how draconian they are.


> The bad actors in most cases won't follow them anyhow, no matter how draconian they are.

That's why there's a move to ELDs - it's not a matter of making punishments harsher, it's a matter of automatically enforcing the rules so that it's much more difficult to break them.


an ELD will not stop the truck from driving even if the driver is OOS, it will however make it easy to penalize the driver.


> The bad actors in most cases won't follow them anyhow, no matter how draconian they are.

IIRC, I heard a very liberal biologist (Robert Sapolsky @ Stanford) say capital punishment does stop a certain amount of the population.


(Context: Moist/Spangler is a criminal about to be hanged.)

> 'You're a card, Mr Spangler,' said Mr Wilkinson. 'It won't be the same without you around, and that's the truth.'

> 'Not for me, at any rate,' said Moist. This was, once again, treated like rapier wit. Moist sighed. 'Do you really think all this deters crime, Mr Trooper?' he said.

> 'Well, in the generality of things I'd say it's hard to tell, given that it's hard to find evidence of crimes not committed,' said the hangman, giving the trapdoor a final rattle. 'But in the specificality, sir, I'd say it's very efficacious.'

> 'Meaning what?' said Moist.

> 'Meaning I've never seen someone up here more'n once, sir. Shall we go?'

-- Going Postal by Terry Pratchett


But at what cost?


> Do you build rules just for the bad actors or to do you try to build a pleasant and equitable society?

I'd say it's closer to "do you build rules that treat humans as if they were cogs in a machine"?


>The bad actors in most cases won't follow them anyhow, no matter how draconian they are.

The fallacy about enacting stricter and stricter laws to curb lawlessness is that the lawless never did and never will care about abiding the laws.


This argument is often cited in gun control debate. “Criminals will just buy guns illegally and law abiding citizens will be left defenceless”.

And yet, in countries with gun controls, there is significantly less gun crime.

I daresay this is the case for most laws. Yeah, there will be always be some people skirting the law, but implying laws have no effect is ridiculous and untrue.


Indeed... at least here in Europe, the occurence of sleepy truck trivers plowing into people dropped by quite a bit once logging switched from paper tachographs to GPS remote logging.

That doesn't mean I support this kind of logging for private vehicles (it really feels like the government overstepping the safety/freedom boundary), but it did have very positive effects on heavy road transport safety.


> Do you build rules just for the bad actors

Sure. I mean, who else are the rules for?

A law against murder only applies to murderers.


That's an extremely naive view of the enforcement layer. You're ignoring all the rules surrounding due process, arrest, searches, etc. To use the murder analogy, the equivalent is if a cop says "you're under arrest for murder" and you're jailed for life with no trial.


Someone in my family had both parents killed and was very severely and permanently physically and mentally disabled when a truck driver smashed their car after falling asleep while driving over his time (and was convicted). Truckers as a population couldn't be trusted to self-regulate, and even try to defeat actual regulations, so they get tighter and tighter. Blame those guys for the loss of fun.


No one is saying self regulate here - not me or anyone else - but the rules need to flexible - the spirit of the rules is supposed to require that a driver gets 10 hours of off duty time, with the idea that they'll get 8 hours of sleep in there. If you're requiring drivers to shut down on the side of the highway (lack of parking for trucks is a real issue), then I dont think thats helping safety much.

Being able to shutdown where I can get a parking space at a place with a bathroom, a shower, and a hot meal is better than on the side of the interstate, the next best place is a rest area on the side of the highway.

Particularly in the case as a parallel commenters noted of a driver who runs out of hours 20 min from home.


Leaving aside the question of whether these systems are the right answer: the "flexibility" here should be designed into the system.

https://en.wikipedia.org/wiki/Normalization_of_deviance

If there's meant to be a hard limit of N hours, the point of starting to look for a stopping point should be more like N-1.

If N is meant to be the soft limit of starting to look for a stopping point, the hard limit should be more like N+1.

Either way, the "start to look for a stopping point" soft limit and the "stop immediately" hard limit should be separated by however much time it reasonably takes to find a good stopping point.


There is that kind flexibility, drivers can drive past the limit in order to find a safe spot, but if they do that on a regular basis they’ll get pinged which is fair. The problem again is the real world: there are instances where truck stops are full and they would have to drive many miles until they find a spot. Not to mention sometimes they are stopped and are ordered to move the truck, which breaks their rest time. Now you see trucks driving at 2mph on the parking lots or side streets in order to not trip the gps. But also given the structure and culture of the industry ( which drivers are only a piece ) there’s only a hard limit, which is the one that gets you in trouble, everything else is just a footnote.


> But also given the structure and culture of the industry ( which drivers are only a piece ) there’s only a hard limit, which is the one that gets you in trouble, everything else is just a footnote.

It sounds to me like you're saying that we need hard limits because permissiveness will always be abused.


Everyone "needs" hard limits because permissiveness will always be abused.

That's why police and jails exist.

That doesn't mean the hard limits are align with what I would consider a "healthy" system composed by people.

The current trucking business ( at least in the US ) is insane from an individuals perspective, paper or non-existent logs will not solve anything.

And that's one of the issues, everyone wants to "solve" their own problems and screw the rest ( it's their problem not mine ). Over time that means the drivers were the ones who got screwed the most.


> The problem with ELD's is the lack of flexibility.

Perhaps it is like refueling: it is normally done before the engine stops because there is 0 fuel left.


> The road was fun

I think what you mean is that drivers had some sense of dignity. Let's not confuse that with fun. We need to be clear about what's being lost; it's no laughing matter.


I'm sorry fun doesn't sound dignified enough for you, but I meant fun, yes, - dignity was a part of that package. That said, no one is going to do a job no matter how dignified it is without either great pay or privilege if it's not fun.

I got to be what was effectively a professional tourist for a couple of years. It was fun.


But dignity and ability to enjoy something are connected.


Absolutely, but one of them is easier to argue for.


https://how.complexsystems.fail/#11

> Organizations are ambiguous, often intentionally, about the relationship between production targets, efficient use of resources, economy and costs of operations, and acceptable risks of low and high consequence accidents. All ambiguity is resolved by actions of practitioners at the sharp end of the system. After an accident, practitioner actions may be regarded as ‘errors’ or ‘violations’ but these evaluations are heavily biased by hindsight and ignore the other driving forces, especially production pressure.

(taps the sign)


Sidney Dekker mentions this in "The Field Guide to Understanding 'Human Error.'"

Always remember: people generally show up to work with the intention of doing a good job. If someone causes an accident by doing something that is "obviously" unsafe in hindsight, step back and ask yourself why the person thought their "obviously unsafe" action was reasonable in the moment.


Likewise with smaller scale "accidents".

Did the ups driver fling my package over fence because he hates me, or because my long driveway interferes with his 3 minute per package metric.

Measurements kill sensibility.


What about the people who make these half assed laws and regulations? Either they don't care about their job or their job definition is wrong to begin with


Let's apply Dekker's reasoning. Give me a example of a law/regulation that you think is half-arsed. I bet we can show why it was designed that way with good intent.


I haven't seen this list before, I like it a lot - thanks for posting.


Motor carrier stuff is incredibly complex with a bizarre patchwork of regulatory domains and a network of shitty operators, who employ shitty drivers who work the loopholes.

The most egregious examples are the Chinatown busses, which routinely operate with material safety flaws.

To understand safety data, you have to study the cohorts of carriers that comply with regulations. The Feds and state authorities working with them statistically approach compliance inspection, but the smaller, dirtier carriers tend to be fly by night. If you see a crazy trucker speeding at night in the middle of nowhere, Pennsylvania, you’re probably seeing one of those trucks, schlepping stuff between distribution centers outside of federal enforcement scope.

The skilled drivers are right - it’s bullshit. But you need to infantalize the 30 year Walmart driver who has 20 million miles and no accidents, because there’s a 22 year old driving a truck that’s overweight with bad brakes trying to meet a schedule he can’t keep.


Poor 22 year old driver, just entered the industry and already taking the blame for what a certain percentage of 60yo foxes have been doing.

Also, many of the violations are a consequence on how this god forsaken industry is structured and that includes our Karen who wants that pink outfit NOW! Because they have tickets to see the Barbie movie tomorrow and the pink T-shirt is an essential part.


Every time politicians vote on more privacy invading and controlling stuff — they should be wearing/using similar technology themselves.

In this example with ELDs I’d propose a small wearable device that would monitor time when congressman or congresswoman came to work, time they spent eating and using the bathroom and sleeping. And speed they are driving, certainly. Our goal is safety, right?

Oh, and to not make it too privacy invading, only their constituents should have real time access to this data, they are their employers after all, right?

And make this mandatory legislative testing phase (MLTP) at least one year.

We would be amazed at privacy protecting laws rivaled by none.


Lead by example like the good old days!


Any metric which people feel must be gamed ceases to be a good metric.

https://en.wikipedia.org/wiki/Goodhart's_law

https://en.wikipedia.org/wiki/Hawthorne_effect


Ehh, the research is a little more mixed than that article suggests.

It’s not really part of the narrative of the book but fatigue was only a contributing factor for ~13% of crashes historically so the maximum positive impact couldn’t be that large. Meanwhile there’s been a long term trend of significantly increased fatalities. https://www.fmcsa.dot.gov/safety/data-and-statistics/trends-...

The comments on studies looking back 1 year while the transition happened 5 years ago looks like cherry picking data to fit a narrative. Yes the year after it was introduced was a 30 year high, so was the year after that, and the year after that…


Right, but the ELDs didn't create the metric. The metric existed long before that, it was just gamed to an extreme degree because there was no accountability.

ELDs brought accountability and made it harder to game.

They didn't create the metric, they just reduced the gaming.


Aren’t electronic logbooks for truckers used in Europe since 2006? And no one whines about here?


Indeed. I hitchhiked into the tens of thousands of km in Europe in the 2000s, much of it by the kindness of truck drivers, and I never heard a single complaint about the tachograph. What I did hear complaints about were GPS tracking systems, but in that case it was the employer and their quirks keeping a tight leash on the driver, not the state and its road-safety aims.

Fun fact: it was helpful for a hitchhiker to ask the driver for one of the blank paper discs for the tachograph. Later, when standing beside the road and waiting for another lift, you could hold that up whenever a truck approached, and truck drivers would take it as a sign that you were relatively trustworthy.


Truck drivers insurers don't allow them to take hitchhikers (or anyone, really) nowadays.


The insurer simply does not cover costs for certain events or entities. For example, if the truck driver picked someone up, got into an accident, then the passenger’s injuries would not be covered by the insurance.

In which case, the passenger might sue the driver, and the driver might not want that, so the driver chooses to not pick up hitchhikers. One might say this is a consequence of not having taxpayer funded healthcare for everyone, since exorbitant healthcare costs + needing to find someone to milk mean everyone ends up looking out for their own ass.

Another scenario is the passenger steals the truck or its contents , in which case the insurer would say not our problem. You were covered for random acts of weather/other drivers/machine failure/thieves breaking and entering. But not if you willingly let them into the truck. Seems like a reasonable caveat.


Depends where. In the USA (and perhaps Canada) this has long been a concern -- dating at least back to Kerouac's time -- and a hitchhiker there relies e.g. on drivers who are owner-operators. In the rest of the world, including the EU, hitchhikers still easily move via trucks.


Riding as passenger in a French long haul truck should be almost impossible. Although most trucks accross France aren't immatriculated in France, so it might still be possible.


My direct experience is a few years old now, because I moved on to slightly more bourgeois means of travel (but I remain active in travel communities where the younger, less bourgeois members hitchhike), but I don't remember seeing many French trucks whenever I was jumping from one motorway service station to another. The trucking industry in Europe very quickly moved eastward, and already Polish drivers were the stereotypical long-distance lift across Western Europe, with Hungarians and (after 2006) Romanians also being common.


speed and driving times have been recorded mechanically for drivers in Europe since at least the 90ies so going electronic was maybe not as radical a change for European drivers when they went electronic in 2006.

It's also worth noting that a lot of the culture around the labor market is different as the European regulators tend to be a bit more of an aggressive stance on companies creating pressures on employees and then try to shift all blame when those pressure's lead to rule breaking and accidents.


ELDs have been a predictably miserable failure - as the article points out, their very rigidity in draconian enforcement of "safety" rules actually creates perverse incentives to do unsafe things (like ignoring needed maintenance, etc.)

This does not just affect truckers - as "driver monitoring" becomes an anti-feature in many new cars (especially EVs or "self-driving" cars), there will be more and more cases of people (quite reasonably) bristling at Big Brother looking over their shoulder and into their eyeballs every moment they are in their own car.

As for me, I'm leaning heavily towards a life goal to never again buy a car with a screen in the dashboard.


Yep, eyeball monitoring is over the line for me. Not planning to drive stock cars in the '30s.

Looking ahead to electric crate motors and open source infotainment:

https://electricvehicleforums.com/articles/electric-crate-mo...

https://i-carus.com


Imagine a society that never had cars, but did have things like eyeball monitoring technology, speed limiters, auto driving functionality, etc.

If cars were then invented, would the populace entertain the notion of giving everyone complete control to a 2,000lb+ machine capable of moving at 100mph+?

Is our acceptance of the risks of giving everyone complete control of these machines based on the cost:benefit analysis at the time of their advent, but not necessarily a cost:benefit analysis in the current?


Yeah I used to make that argument about humans driving automobiles too.

Before the pandemic.


How is the pandemic relevant?


Repudiation of most global organizations.


The high priests of AI are trying to make computers more like people and in the process making people more like computers.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: