Hacker News Reader

Comment by rayiner | Hacker News Reader

> After a 1995 collision of two trains on the Williamsburg Bridge killed a train operator and injured more than 50

Overreaction to safety incidents and inability to do cost-benefit analysis cripple American transit systems. DC's Metro, for example, was designed for trains to be driven automatically ("ATO"). In fact, they operated in automatic mode for decades, since the system was built in the 1970s. But in 2009, after nine people were killed in a collision, the system was disabled, and will remain disabled indefinitely.[1]

The result of disabling ATO is that what was previously a very smoothly running system has become a disaster, with human operators lurching into stations, then lurching some more to place the train correctly relative to the platform. Almost a million riders inconvenienced every day for a decade, just to avert recurrence of an incident that resulted in a handful of deaths in 30 years of operation.

[1] https://ggwash.org/view/62992/metro-kicks-the-automatic-trai....

View on HN

Similarly, I have noticed that a difference between American subways and European subways is that when you stop at a station in Europe, the interlocks allow the automatic doors to open as soon as the train is slower than maybe 1–2 km/h. Whereas in America they don't open until a couple seconds after the train has come to a dead stop.

I keep wondering how many wasted man-hours this adds up to over the course of a day — especially on a line that's running at capacity, so if one train is 4 seconds later then the next train will be 8 seconds later and so on.

(The same seems to be true of elevators.)

Imagine the lawsuits if the train wasn't 100% stopped when the doors were opened.

That and also driven so many European subways (most called METRO) you never seen cars as overcrowded as in USA and barely people leaning over the door! In USA its normal to lean against the door because people are exactly gotten used to the fact the train will stop first before the door gets open.

Ehh, if you changed it I bet people would learn really quickly not to lean on doors. Most people leaning on doors are going to be young people with decent reaction times and balance so I wouldn’t even expect injury more like surprise. It would probally be reasonable to run a publicity campaign to reduce surprise as people don’t like being supprised.

Most people leaning on doors are going to be young people with decent reaction times and balance

Say what?

Also in some places (Berlin S-Bahn comes to mind) leaning against the doors will obscure the button you have to press to open them.

> That and also driven so many European subways (most called METRO) you never seen cars as overcrowded as in USA and barely people leaning over the door! In USA its normal to lean against the door because people are exactly gotten used to the fact the train will stop first before the door gets open.

The only places in the EU that have public transportation on a scale comparable to NYC are London and Paris. I haven't ridden in Paris in a while, but in London, you absolutely do see cars that are overcrowded with people leaning against the doors.

I think the train complains if the doors are being leaned on - you certainly get an announcement from the driver, so I reckon the train won't go if that's happening.

Buses open their doors before they come to a stop all the time. So far the lawsuits (if any) from this, don't seem to have harmed bus service.

difference being that (at least in NYC), youre not standing directly in front of the door of bus, squashed like you are on trains

That definitely doesn't happen in Chicago. Nothing drives me nuts like pushing on the back door of a bus when it's completely stopped only to jam it cause I jumped the gun by like .2 seconds because the light hadn't come on yet.

In Montreal they definitely open the doors on the trains before the train has stopped. People even jump out while it's still moving.

I personally don't understand why it's done... but it is, and it seems common enough that if they were going to get sued they would have gotten sued.

Maybe Canada is a lot different than the US, though.

Montreal is on tires which gives a predictable stop, it’s not metal on metal which varies based on atmospherics

> Maybe Canada is a lot different than the US, though.

Montreal is just different than anywhere on the planet.

>Similarly, I have noticed that a difference between American subways and European subways is that when you stop at a station in Europe, the interlocks allow the automatic doors to open as soon as the train is slower than maybe 1–2 km/h

No, this must only happen with older rolling stock. Newer rolling stock both from European and Asian manufacturers operates like your "American" example

(The same seems to be true of elevators.)

The solution to this is called "Pater noster". Just kidding....

The release of door interlock used to be as you described, but now in Europe train stop is controlled prior to releasing it. The delay you note is generally tied to the reaction time of the ATP computer in charge of authorizing doors opening. Some of these could be slow (sometimes 5s). Today it’s less than 0.5s.

Disclaimer: I design and deliver such systems.

Almost every delay in computerized systems that is greater than the physical/logical sensor debounce interval (ie. on the order of milliseconds) is due to programmer error.

It plagues software literally everywhere. We need to stop!

Every gas pump, payment terminal, elevator door is a testament to poor implementations of event driven state machinery...

</rant>

And every new tech is shinier and slower than the old generation. Nobody pays any attention to latency as a part of UX (except people making pro-photo and music gear, it seems - even gaming these days accepts pretty large latency as a part of the experience).

I am with you in this general rant. However, in this particular case, the delay was a consequence of the introduction of computers and asynchronous telecommunications. I was not around at that time but it seems nobody anticipated that putting all the logic in a computer would result in worst performances as opposed to older designs were computations were essentially carried out by relays wired together. Relay logics worked in parallel, when CPU have to do things sequentially.

Fun fact: some operators around the world refuse to use computer-based systems for some applications because of that, and they are getting a strong support now because of cyber security threats. A relay is hard to hack :D

Further to your point, missing a few seconds due to this will reduce capacity but will not create delays, since this time will be factored in the timetables with a margin.

Disclaimer: I realize such studies :O)

Delays have pushed more people onto the roads (bikes, taxis, etc.), where there is a much higher chance of injury or fatality. So their decision has probably killed more people than if they had changed nothing.

Yes, and see the TSA for another example.

We've easily lost another 3000+ people on the roads since 9/11, simply because the professional pearl-clutchers have made it such a pain in the ass to fly anywhere.

"pearl-clutcher" is a sexist term, and the perpetrators of the TSA are for the major part by far not pearl-wearers.

That’s so interesting! I see now what you mean (hi google), but my initial mental image was a rich old man with a death-grip on his riches.

I hate the driver who pens the L line going to Brooklyn from Manhattan. He speeds way too much and then breaks way too hard. I have seen passenger fall because of it time and again. If we can't get rid of the union workers, I would rather have MTA pay these guys wages to sit at home and replace them with computer-driven subway.

TFA mentions train operators will preemptively slow down to keep their job.

Nothing to do with a union, but keep trying to shoe horn that in.

> TFA mentions train operators will preemptively slow down to keep their job. Nothing to do with a union, but keep trying to shoe horn that in.

As alluded to in the article, the TWU has been fighting very aggressively against the reforms for most of the issues documented in this article.

Who need allusion? TFA explicitly states TWU wants to keep safety improvements. That's literally the job I would expect for a union.

Safety is already worse in NYC than other areas.

It all hinges on if the safety concerns are valid.

> TFA explicitly states TWU wants to keep safety improvements. That's literally the job I would expect for a union. Safety is already worse in NYC than other areas.

You're actually going to accept the reasoning they give uncritically and at face value?

Of course they'll frame nearly everything in rhetoric of safety, because politically that's the equivalent "but think of the kids".

Of course, when you actually look at it, it's obvious that the TWU has spent the last several decades fighting actual safety improvements (including those that are SOP at the other Big Four systems), because their only interest is in protecting their members' jobs.

As linked elsewhere in this article, if the MTA tries to introduce newer and safer machinery, the TWU actually extracts a "technological advancement" fee to compensate them for the missed job opportunities, in an attempt to disincentive them from adopting new technology.

The TWU is literally putting passengers' lives at risk because their only goal is to extract as much money as possible from the MTA, and they don't actually care about the end state of the subway system.

> You're actually going to accept the reasoning they give uncritically and at face value?

That question should be directed at you. It should be obvious this is a nuanced and complex issue.

Why would anyone trust an internet-rando and a single article?

You accept what appears to be a one-sided view. Do you believe that should increase or decrease your trustworthiness?

I'd like to know what's actually wrong in NYC so I can prevent the same in my own city.

> That question should be directed at you. It should be obvious this is a nuanced and complex issue. You accept what appears to be a one-sided view. Do you believe that should increase or decrease your trustworthiness?

I don't know where you get "one-sided" out of any of this. I responded to a claim that the problems in this article have nothing to do with unions, pointing out that the union for MTA workers has been the main driving force against solving the problems described in the article. Between the information in the article, the other information linked elsewhere in the comments here, and some basic Googling, that's all pretty easy to verify.

If anything is "one-sided", it's the original claim that I refuted.

> I'd like to know what's actually wrong in NYC so I can prevent the same in my own city.

Rampant corruption. The TWU isn't the only source of that corruption, but it's foolish to pretend they're not a significant part of it.

Corruption isn't the only logical conclusion. It's foolish to pretend otherwise.

> Corruption isn't the only logical conclusion. It's foolish to pretend otherwise.

Instead of blindly speculating on what could be true, when you've already said that you're not familiar with the topic, how about doing your research on the facts and learning why this is what's actually true.

You started off with a claim that was directly debunked by actually reading the article. Then you pivoted to speculating (but not substantiating) the existence of information that could hypothetically make your original statement a little less incorrect.

At this point, the only foolish thing is pretending that you're contributing to the discussion in any meaningful way.

It has everything to do with unions, specifically public-sector unions that shouldn't be allowed to exist anyway.

You can't expect the development and deployment of efficient automation in an environment where it's impossible to fire the existing human operators. Or where firing them would be just as expensive as keeping them around doing nothing useful.

> It has everything to do with unions, specifically public-sector unions that shouldn't be allowed to exist anyway.

OP complained about a driver slowing down and speeding up. TFA explained it. There's no magic.

Your viewpoint is staunch, one-sided, and brought no new information. That point of view sounds questionable.

Why don't public sector employees deserve rights?

Because those "rights" come from extorting the taxpayer, who has little or no say in the matter.

First, public sector employees don’t deserve special rights everyone else doesn’t have. Second, public sector unions present unique problems privat sector unions do not. If a private sector union pushes too hard, the company goes out of business and everyone is out of work. If a public sector union pushes too hard, nothing happens—it can hold critical government services (often monopoly services) hostage until it gets whatever terms it wants.

If MTA were a private company, it would have gone bankrupt long ago. But we accept that public services should sometimes be subsidized at the public expense. That subsidy should go toward making the service cheaper and more available to the public. But public sector unions capture some of that subsidy to get higher pay and benefits for public sector workers than those workers would get in a private company.

The L train already uses CBTC (i.e automatic train operation); fast speeds are issued by the computer.

There's also at least one conductor who starts closing the doors during the morning rush our while passengers are still existing at the 1 AV stop.

One of those times the doors smashed into a person walking out on crutches.

The L is entirely computerized already. The two people working on each train are there to press a "keep the union happy" button every 30 seconds.

> The L is entirely computerized already. The two people working on each train are there to press a "keep the union happy" button every 30 seconds.

In case anyone thinks this is a joke, it's not. The MTA pays the Transit Workers Union massive sums of money when they use technology they use which does a job otherwise performed by a human, even when it's been standard practice worldwide to automate that job for decades.

One example, from tunnel boring:

> The critics pointed to several unusual provisions in the labor agreements. One part of Local 147’s deal entitles the union to $450,000 for each tunnel-boring machine used. That is to make up for job losses from “technological advancement,” even though the equipment has been standard for decades.

https://www.nytimes.com/2017/12/28/nyregion/new-york-subway-...

I don't understand why New York has two employees per train. Nearly everywhere else in the world has 1.

The person in the middle's job seems to be operating the doors and making sure the train is berthed correctly, but all you have to do is put a line where the front of the train has to line up and the driver can do it. As for the doors, it's not like the driver has much else to do when in the station.

Nominally, the trains are long, and sometimes the stations are curved. A person in the middle of the train can see more of the doors to make sure they're clear.

This is easily solved by cameras.

Now, yes, but when the union started it wasn't, and they've been working ever since to never allow cuts in the number of personnel.

> I don't understand why New York has two employees per train. Nearly everywhere else in the world has 1.

Because the TWU penalizes the MTA financially for technological advancement. They are incredibly powerful, mostly because neither the MTA nor the state government which oversees the MTA bother to keep costs in check.

There's no reason the L couldn't be completely automated, with zero employees per train. As described elsewhere in the thread, the person operating the train doesn't actually have to do anything at the moment; the entire process is already automated.

Kinda sounds like he's doing his part to keep the trains actually on time in a totally broken signalling environment!

The L train is actually the one train with fully upgraded signals, and no track sharing with other trains. Probably for that reason, it also has the highest on-time percentage of all trains.

I've never been bothered by that (and as others point out, the L is automated, anyway). What I'd like to know is why, when the L reaches the 8th Ave stop in Manhattan, the operator waits 10 seconds after coming to a full stop before opening the doors. The train has already stopped, so this seems unnecessary.

There's also this weird thing where, while the train is sitting there waiting to return the opposite way back to Brooklyn, they close all the subway doors except one in each carriage. It doesn't prevent people from getting on.

They do this at every terminal. I have no idea why.

I found an answer here [1], via a Reddit thread [2] (where the answers are apparently gone):

    I won't go into details, but if the conductor wants to exit
    the train and leave the doors open at the terminal, he needs 
    to walk to the next car at the last stop in order to open
    the doors, which is why there's a short delay before they
    open. If the doors open immediately at the last stop, then
    it means the train is either going to go to the yard, or the
    crew is going to manually key open one door per car (which
    is only done if the train is going to sit there for at least
    10 minutes, and helps the air comfort system).

[1] http://www.ibtimes.com/mta-worker-goes-underground-share-sub...

[2] https://www.reddit.com/r/IAmA/comments/kx5fa/iama_new_york_c...

Very interesting. I definitely always hear the conductor moving between cars at the last stop but I never connected that to being why the doors can't be opened.

I think it's funny that the train's computer knows it's the last stop and announces that fact to the passengers... but the conductor has to manually override some sort of system before he or she can exit the train.

> There's also this weird thing where, while the train is sitting there waiting to return the opposite way back to Brooklyn, they close all the subway doors except one in each carriage. It doesn't prevent people from getting on.

My understanding is that it's to keep the subway cars relatively climate-controlled. Especially in the summer, it's better not to let all the cool air from the AC leak out the open doors.

Oh yeah, that's it. Because non-union-workers would never drive like that. :-P

I agree with your point, but strongly disagree with your example. You think DC should be running the risk of accidents which kill riders because of trains that lurch into stations? Loss of life is a massive cost, even if only economic factors are accounted for. Please show me a cost-benefit analysis where that is outweighed by lurching trains (even many millions of lurching trains).

Also, it's incorrect to say that we could have used decades of problem-free operation to extrapolate and predict a low risk of similar accidents in the coming decades. The 2009 crash was caused by an aging system, and unless your solution involved replacing all the switches/sensors, we would have seen new failures at a more frequent rate.

Finally, the 2009 accident highlighted an inherent problem with the old system: putting a human operator in a train who is only responsible for acting in an emergency will not work. Just like the recent self-driving car fatality. It wasn't a case where we knew and should have accepted the risk, it was a case where the crash caused us to understand that the prior risk assessment was wrong (because it assumed that a human operator could prevent these kinds of accidents, when that's just not going to happen). So it was appropriate to change operations.

> You think DC should be running the risk of accidents which kill riders because of trains that lurch into stations?

DOT estimates the cost of a human life at about $6 million. Times 9 is $54 million. Let's say such an event happens once every 10 years, so $5.4 million per year. The D.C. metro has 180 million rides per year. That works out to 3 cents per ride. Do you think people would pay 3 cents for a smoother, ATO-controlled Metro ride? I bet they would. (That's not even counting all the delays caused be forgoing automated control.)

As to the switches and sensors--Metro already replaced the track circuits (per the NTSB's recommendation): https://wamu.org/story/17/04/06/metro-shelves-plans-bring-ba....

It's worth elaborating on a central point; the idea that 'preventable deaths are unacceptable' only makes sense at ordinary human scales. In someone's day-to-day life, they aren't reasonably going to be in a situation where letting someone die will be the best alternative.

The situation changes completely at the level of the decision makers of countries and large cities work at. /Everything/ is a variant of the trolley problem and all resources could in theory be used to save a life. Not accepting a risk of death can do more harm than good and realistically infrastructure; it just obscures the cause and effect.

Eg, we'll never know if someone died because their doctor was late because a train operator made a mistake operating a train. The cause-effect chain is too convoluted. But statistically speaking, something like that will happen every so often. Deaths from inefficiency are just as bad as deaths from direct accident; and if the scale is large enough that they can reasonably be detected then they need to be considered too.

Obviously most of the time a high standard of safety is going to be best, but trains are a high-risk environment moving 100s people at speeds they were never meant to move. True perfection over 30 years is probably not a required standard. Great goal though.

This has been interesting to chew on. Let me say that if there are actually delays, obviously those impose other costs (as a sister post explains here) and so the cost-benefit calculation may come out differently. But your original post was only about whether it's worth risking lives to avoid lurching trains, which is why I thought it was interesting.

I think the problem with your cost-benefit analysis is that you are using different methodologies to value the cost and the benefit. If you value the benefit of non-lurching trains by asking what people would pay, then you need to use a value for human life that asks what people would agree to receive to be killed--which I believe is much more than $6 million.

Conversely, if you are going to value a human life based on lost earnings/support to family/etc., or some other figure which might produce results in the ballpark of $6 million, then on the other side you need to consider only the lost earnings/hard economic costs of an annoying train ride, which are $0 -- there is no economic loss caused by lurching into a station.

I think the devil's advocate argument here is: would deciding not the run trains on auto cause an increase of car riding which would lead to greater than the number of deaths generated by the aging automatic driving system?

The problem is that ATO relies on the underlying technology to be working and maintained. Maintenance requires money, which WMATA doesn't have. That's the issue. Humans don't require maintenance the same way something like ATO does.

Politicians continue to underfund WMATA, which means they can't assure safety using an automated system, so that have to use humans.

Got to wonder why - while they were paying people to begin 'installing and modifying hundreds of signals to prevent trains from going too fast' - they didn't modify the signals so they could be remotely re-configured.

Something in that system must be configured to be resistant to superior technology. Cuz it couldn't be the humans, right?

The purpose of this change probably isn't so much to make the trains more safe as to facilitate transferring blame for a fatal accident from the transit system to a human operator.

CYA is rule #1 in bureaucracies. If something happens and you "do nothing," you're responsible, so you have to "do something."

This example is exactly why I'm dubious of the idea of mass adoption self driving car technology.

Edit: I should add: in the United States.

Reply on news.ycombinator.com