Hacker News new | past | comments | ask | show | jobs | submit login
Much of the technology in the NY subway hasn't been updated in over 100 years (businessinsider.com)
102 points by dankohn1 on Oct 19, 2015 | hide | past | favorite | 90 comments



Business Insider copied that from another article in Gothamist [1] several months ago. It was even discussed on HN back then.

Most of the NYC subway signaling technology is relay-based from the 1950s, so it's not quite a century old. Those lever machines at West 4th St shown by Business Insider are rare; almost everything is electrical.[2] The 1950s technology from General Railway Signal is quite good; it's just high-maintenance. Here's a 1950s era panel, running automatically and unattended during off peak hours.[3]

That technology really is quite good. All vital relays are open in the stop condition. All broken wires result in red signals. This is a basic design principle of classic railroad signaling. The mechanism for stopping trains[2] is brutally simple; a steel train stop comes up at red signals which, if a train tries to pass it, will be hit by a valve lever on each subway car which will directly trip the air brakes. The position of those levers is monitored by circuits which insure that there are always train stops in the up position between trains. If a train stop lever won't go up (ice, mechanical breakage, somebody or something holding it down) the previous train stop won't go down, maintaining protection, and signals will be forced red until the problem is fixed.

One reason it hasn't been replaced is that the newer technologies don't have as good a track record. The old stuff is very rugged, and is known to survive dirt, ice, snow, vandalism, traction power, vibration, rocks, water, and idiots.

[1] http://gothamist.com/2015/07/28/subway_steampunk_video.php [2] http://nycsubway.org/wiki/Subway_Signals:_Train_Stops [3] https://www.youtube.com/watch?v=RStJ621auaU


How can you say they copied this from Gothamist, especially considering Gothamist hardly ever creates their own content? Business Insider summarized the exact same video Gothamist did in your link, but they added more pictures, relevant quotes and helpful diagrams. Gothamist barely did any work on this article, they just posted a quick summary of the video. Oh, but they added this insightful gem: "This shit is OLD, like grizzled dude who won't stop stabbing at the back of your plane seat because he can't figure out the TV touchscreen old." Oh and if you want information, you can click on one of the equally vague gothamist links.

It's like, people complain about how shitty reporting is these days, but when a publication puts some time and effort into its stories, people complain about how they ripped off a rag that covered the story earlier even though they did a shitty job at it.


A few years ago my Dad did some consulting (HAVC power) for the Tube in the UK and he was amused to find that the 1930's vintage equipment was lasting much better than the new 1970's gear.


On the contrary, that should be expected.

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

...now, we should figure out why the older stuff lasts longer. But it's inevitable that the older stuff is more durable. If it weren't, it wouldn't exist.


In general, the things which survive are over-engineered. The wires are several guages larger than they need to be, there are backups for the backups, and the underlying technology (magnetism, switching) is hundreds of years old. In contrast, many of our electronics use components which are barely larger than they need to be (if at all), and rely on technology which is younger than a single generation.

Now then, we do still create technology which lasts - it's just not as frequently found in consumer tech. Take stoplight timing switches as an example. They use electronics, and work in some incredibly extreme environments, and are reliable more often than not.


Exactly it was only 15 or so years ago that one electrical firm local to me stopped producing knife switches - thats the big switches (steam punk style) you see in Frankensteins lab film sets.


Its the fact that it tended to be built like the proverbial brick outhouse.

And survivorship bias cant really be used for a tube line as opposed to a widget off of a production line.

A tube line is a one off custom system.


The tube line is a custom one-off system. The components aren't though.


1970s electronics wasn't very good. Until the 1970s, complex electronic devices mostly stayed in labs, computer rooms, and military facilities, where they were cared for by technicians. In the 1970s, consumer electronics beyond the TV set appeared. Transistors and early ICs were reliable enough to make this possible.

But now a host of secondary problems had to be solved. Soldering of PC boards wasn't a reliable mass production process yet. Plastic cased ICs weren't perfectly airtight. Connectors were not yet reliable. IC line quality wasn't consistent. IC wire bonding was done by hand. "Purple plague", caused by contamination from the wire bonding process, corroded ICs internally. Phenolic PC boards were subject to hairline cracks. ICs lacked electrostatic discharge protection on pins and static electricity could easily damage them. Power supples could not maintain voltage during power line sags and spikes. RF shielding was poor; devices both emitted too much RF and were sensitive to RF in the area. Many signals between boards and units were relative to ground, not differential, and ground loops were a problem. On top of all this, it took a lot of components to do anything, so everything had a high part count and many interconnects, all points of failure.

Every one of those problems has been solved. But that's why 1970s electronic devices were troublesome, and didn't age well.

Now, though, we have electromigration. Slowly, atoms creep across barriers just from statistical diffusion helped by electric fields. This is a very slow process, but with IC features so small, not that many atoms have to move before something fails. It's possible to make high-reliability electronics at a larger line size with current quality fab technology, and this is done for aerospace and some automotive applications.[1] Still, few consumer parts today have a design life longer than 20 years.

[1] http://www.freescale.com/products/ultra-reliable-microcontro...


If it's really a rip of the Gothamist content, without much value added, I hope the link can be changed by admins. Such behavior should not be acceptable.


This isn't 100 years old, but a few years ago, during the 25th birthday of OS/2, it was said that OS/2 is still the software for the Metrocard swiping machines:

http://techland.time.com/2012/04/02/25-years-of-ibms-os2-the...

> Try though it might, IBM couldn’t force an unwilling world to use OS/2 as its primary operating system. But it also couldn’t extinguish demand for the system simply by declaring that it wasn’t going to sell or support it anymore, an announcement it made in 2005. In 2012, OS/2 is invisible to the naked eye, but it’s still out there, in more places than you’d think.

> In New York City’s subway system, for instance, the travelers who gain entrance by swiping their MetroCards over 5 million times each weekday do so with the assistance of IBM’s theoretically defunct software. “While OS/2 is not running any visible part of the system, it does serve an essential purpose, and there are hundreds of OS/2 computers in service,” says Neil Waldhauer, a consultant who helps New York City Transit and other clients keep their OS/2 applications running.

> “OS/2 is not a superior solution in the places where it is still in use,” he explains. “Rather, it is a vital part of a larger system. Many enterprises have big investments in OS/2 programs that have no equivalent on other operating system … For many users, it would be expensive or inconvenient to move years or decades of their data and programs to a new platform. Replacing OS/2 would mean replacing their entire system.”


Well, one positive thing about this is that it's probably less exposed to cyber warfare. So when the big attack comes, the subway will still run.


OS/2 emulators exist, so if there is an actual need to switch to modern hardware (e.g. old hardware is failing), switching to emulators would be cheap.


It's still sold and supported on new hardware too; just look for eComStation - definitely not exciting, but I think they have a decent business.


Do they ?

It's a shame the OSFree project has never got much support, it would be great if it got to the point where it could run gui apps.


I understand Parallels was formed by a bunch of Russian hackers who were originally contracted to make a huge bank's OS/2-based software system run on new computers without OS/2 drivers.

The big issue was that OS/2's kernel operated in Ring 0, and no existing virtualization/emulation systems supported the relevant CPU instructions.

I wonder if that bank is still running on OS/2... :P


I heard that Parallels got a head start by, ahem, inheriting a substantial amount of code from VMware. It's something I read online a while ago but haven't been able to google any confirmation to.

Anybody know anything they're willing to share publicly?


Lots of ATMS use XP, recently dropped support by MSFT (saved perhaps for some military contracts). It would be less expensive to buy new systems than update the old. Hackers know and exploit the holes.


XP is still supported via POSReady 2009. It's still got 5 years of life left in it for updates.


This article promulgates the myth that the MTA is underfunded. In fact the MTA has lots of money which is spent incredibly wastefully. For example, it costs the MTA $1B/km to build underground, compared to $100-200M/km in the rest of the world.

I discuss in more detail here: https://news.ycombinator.com/item?id=10247109

This article even hints at it: Installing the system on the L train took over six years, with multiple delays and cost overruns reaching $288 million. The so-called "robot trains" require far fewer operators, too, which drew ire from the Local 100 of the Transport Workers Union of America.


Both are true. The MTA is terribly inefficient in its capital and operating budgets. It is also underfunded on a per capita and per $ GDP basis. Very similar things can be said about the NYC area roads, bridges, tunnels, and airports.

The second avenue subway is way over-budget, will take far longer than it should have to build, and is poorly designed. But when it is finally done it will bring a lot of value to a lot of people's lives. The same cannot be said of the F35.


I'd be interested in seeing a better comparison - it seems odd to compare a metro commuter rail vs. one stop through one of the most congested, densely populated infrastructures in the world. Presumably there are economies of scale at building a long metro commuter train that extends aboveground outside of the city.

It's hard to compare construction and maintenance costs between two different cities where one has 30% more track and 10x the square mileage AND is 24hours in operation.


Building underground in the most congested, densely populated infrastructures in the world? You mean like the undergrounds of Hong Kong ($500M/km), Fukuoka ($300M/km), Singapore ($500M/km), Cairo ($300M/km), Sao Paolo ($250M/km) and Seoul ($100M/km)?

https://pedestrianobservations.wordpress.com/2013/06/03/comp...

(Admittedly, one of the Sao Paolo lines is 16% above ground.)


I feel like you took only one part of my comment to disagree with out of context with the rest- I'm not even disagreeing that the MTA is inefficient, I'm just saying that by looking at only one data point (cost/mile), you're disregarding many others.

I think NYC is a bit of a perfect storm situation:

- 24 hour operations

- High labor costs, coupled with strong unions

- Legacy infrastructure to maintain (vs. new subways)

- extremely dense infrastructure to manage (NYC under-street infrastructure is famously messy/complex)

- Huge distance of track to maintain

- Large square mileage of city area to contend with

Surely this has to play a big part in the cost of building subways in NYC vs. simply, "The MTA wastes money"?


While size and square mileage would make things worse per km (or per km^2), they shouldn't affect unit costs.

See also these tweets (also from the linked thread) and various others by MarketUrbanism: https://twitter.com/MarketUrbanism/status/639265131885686784 https://twitter.com/MarketUrbanism/status/639445006084886529

Apparently turf wars ("Your job is to keep the LIRR out of Grand Central") and union contracts do contribute a lot to this perfect storm. I don't understand why you disagree with my use of the word "waste" to describe this.


But the cost of labor in those countries is much lower than in the US, no? (It actually makes sound $1B/km in the US sound way better than $500M/km in HK.)


On average, no. If you adjust for PPP (as my source did), Hong Kong and Singapore are wealthier than the US.

https://www.google.com/publicdata/explore?ds=d5bncppjof8f9_&...

However, due to exceedingly powerful transit unions, labor costs are higher in the specific case of infrastructure. As I said, waste.


GDP (and thus wages) in Singapore and Hong Kong are pretty high.

If they can get cheaper labour for construction, than only because they make it easier to import cheap labour, because they have saner visa policies.


I think you may be promulgating a myth yourself - that building underground is the same, irrespective of location. The fact that London's Crossrail costs similar amounts per km may have something to do with the peculiar challenges of building under cities like London and New York.


Spend some time with union relations and you'll find that you are more right than you know.


"Every signal on each line is mapped on this board, which looks more like a an old board game"

It looks a lot like a physical version of a new signalling centre in the UK [1], except that uses monitors, which presumably make it easier to update the "board". Many of the diagrams are visible online, since there's an open data feed [2] (this one is about half the London Overground in London, orange on the familiar map).

[1] http://www.railengineer.uk/wp-content/uploads/Sig04-online-9...

[2] https://www.opentraintimes.com/maps/signalling/NLL


Here's a link to the video that this article pretty much summarized [0]

[0] https://youtu.be/Mjx3S3UjmnA


Why do the floors of the subway have to be much so worse than the floors of Grand Central Station, or for that matter, the ground in Bryant Park? All are open to the public and incredibly heavily trafficked.

The only difference is funding. NYC subways remain woefully underfunded. There's no big secret of what it would take to fully fund a clean subway system. The answer is $100 M a year [1]. Now, should the capital infrastructure cost less? Of course. But there is no reason for the economic arterial system of the whole region to remain so filthy in the meantime.

The only potential revenue source that could credibly provide the MTA with the money it needs to bring all stations to a "State of Good Repair" [2], install valuable upgrades like train locator screens and sliding platform doors, and keep the stations clean is a congestion tax like Move NY [3].

A congestion tax would also have the benefit of making Manhattan a much more walkable and ridable city and would even make rush hour taxis move faster. But until there's political will to implement a congestion tax, I don't expect any meaningful improvements in the subways.

[1] http://secondavenuesagas.com/2008/09/30/cleaning-the-system-... [2] http://secondavenuesagas.com/2015/09/10/cbc-mtas-state-of-go... [3] http://iheartmoveny.org/


One of the reasons is that they have giant trains running next to them, in a way your other two examples don't. Trains create a lot of dirt and dust from wheels and brakes among other things. If you compare the actual platforms at Grand Central to a subway station they're pretty similar actually. Also both of your other examples close at night, the subway does not.

The subways are generally not that different in cleanliness from the corresponding streets and sidewalks above them. You could ask a different question, which is why NYC as a whole is not very clean, but that's a much more involved and complicated question than this facile complaint.


Better example then is PATH, which is 24 hours and has markedly cleaner stations. However, its costs are closer to a commuter railroad at $10 per rider [1].

[1] http://www.nj.com/news/index.ssf/2015/01/whats_the_real_reas...


A congestion tax is a fine idea, but I disagree that it is the only potential revenue source that could credibly provide the MTA with the money it needs.

Other options include bringing back the commuter tax which was foolishly eliminated for short-term political gain, and getting more of our money back from Albany and/or Washington.


Money is fungible, so all sorts of new taxes could work. But Move NY actually might be able to pass, as it is intelligently designed to actually subsidize outer borough bridge users who were most penalized by past congestion tax proposals. It also represents a Pigovian tax on congestion.


It might well reduce the cost for the occasional driver who wants to leave the city or those few that commute from one outer boro to another, but the people that most opposed the former proposal were those that commuted from Queens or Brooklyn to Manhattan. This proposal looks to lead to the same unhappiness from that group -- indeed any realistic proposal would, since a basic goal is to get them to stop doing that.


New Yorker and a straphanger for decades here. Commenting on the "{City} subway system is clean. Why not NYC's?". The subway is cheap, and runs 24 hours a day. This limits how much one can clean it - Washington DC subway system washes all its platforms with hot water and bleach every night, when the subway system is closed. I believe it is a combination of non-stop, utilitarian and ancient system that brings bad rap to NYC subway. I personally favor it over cabs for a reliable, cheap ride.


They also wash the platforms with water and bleach in NYC, while people are waiting for the train.

The times I've watched it, they do an incredibly lazy job, so the net result is not much cleaning. If the goal was "clean the platform" instead of "look like you're cleaning the platform", the system would probably be a lot cleaner.

The MTA has a large number of labor relations problems they need to fix. That's what's really holding them back.

(I also watched an interesting exchange recently. Someone jumps the turnstile. Someone reports that to the station agent, who just watched it happen. She replies "are you sure, I didn't see that", and went back to playing with her phone. This attitude of neglect completely kills the "if you see something, say something" culture the MTA tries to instill. If I see something, I'll keep it to myself to avoid the ridicule I'd receive from the employees or police.)


NYC is also by far the largest and most complex subway system. The larger stations are built as underground, multilayered labyrinths. Times Square/Port Authority Bus Terminal is probably the worst offender -- it's two stations linked by an underground tunnel that services most of the subway lines in Manhattan (1/2/3, A/C/E, N/Q/R, 7, and the crosstown shuttle). Walking from the shuttle platform to the bus terminal feels like you're walking half a mile, and you go up and down several staircases and make several turns in the process.


Ppl in software are familiar with tech debt.

There is also organizational debt: http://steveblank.com/2015/05/19/organizational-debt-is-like...

This article seems to be an example infrastructure debt.


Actually I'd not call this "debt" but "assets". It's a proofed system with an incredible track record. Please go ahead and show me any freely programmable system that has been in continuous operation for that long.

Think about it that way: It's a system where over time all the "bugs" have been identified, resolved and failure modes mitigated. If you were to replace the system with something new, you'd likely introduce the same old bugs, and new ones, without zero operational gain. Train tracks are rather fixed in their operation; in fact their design is dictated by the train movements. Changes in possible train movements come with changes to the tracks, and those changes are easily reflected by hardwired circuitry.


There would be operational gain. Modern signalling systems can reduce the time between trains to 30s or less, and computer-controlled acceleration and braking is more efficient and smoother for passengers. Also, at some point trains do wear out — i.e significant parts are worn out, like the chassis, that it makes sense to replace the whole train.

I'm just repeating the press statements of London Underground, who have been investing in new trains and signalling systems for the past 10-15 years. (It follows a long period of under-investment.)

New equipment does bring new bugs, which is why they introduce new trains on 'spare' track at night (without passengers), then with passengers on Sunday nights, then Sunday mornings, etc.


We're talking about the track system here, not the trains. For trains it makes of course a lot of sense to use modern locomotion control technology.

However when it comes to the tracks and signaling, there's only very little to be gained to have the actual signalling being computerized on a standard two track railway with the occassional switch track.

The distance between trains, both in time and space, is determined by the blocking of the tracks. There's a hard constraint on the minimum distance between blocks: For any given train running a track, it must be able to come to a full stop within the length of a block. And for safety reasons there must be at least two free blocks between trains (exercise to the reader: why two blocks? Hint: Emergency braking).

Now there is of course the possibility to subdivide a track into microblocks, where a fixed number of microblocks make up a "block" and the "block" borders are dynamically moving along with the trains on a track, so that there's always two blocks of distance between them. But again, this does not require a programmable computer to implement. It can be done using hardwired circuitry (and I'd feel much safer using such a system); just throw a few 74HCT… ICs onto a PCB and mass manufacture that.


Visited NYC for the first time last week and god, I have been to a lot of subways and NYC's metro system was one of the creepiest.

Relative to how much it is used I wonder why they just don't throw money at the problem to make it a more attractive way of transportation instead of being only a necessity.


These comments happen all the time, and represent a fundamental misunderstanding of what the NYC subway system is for. Namely, getting truly massive numbers of people (typically 5-6 million per day) to places they need to go quickly, completely around the clock.

You can compare apples to staplers by comparing it to other systems that don't do that, but that misses the point. When one actually moves here you start to realize just how the whole thing works, and realize that systems that close at night, or don't have express trains, are just hopelessly inferior and not a valid basis for comparison.


The real question to ask is, why can Tokyo Metro make a profit with those ridership numbers, but not the MTA?


I believe Tokyo Metro owns much of the retail space and some shopping developments connected to their stations. Based on casual observation it seems that they're better at making that real estate desirable compared to the NY Subway. I wonder how much of their profits comes from that?


Meanwhile, as someone from DC, I love the NYC subway. It is faster, the trains run more frequently, the stations are numerous and convenient, and it's cheap. I'd rather have a dirty, dimly lit system that works than the crap we have in DC.


Over the years, many of us New Yorkers grow to love the vibe of the subway, and over the years the cleaner smaller newer systems in places like DC just start to seem antiseptic and unnecessarily cute.


I wouldn't mind someone using a little antiseptic in many parts of the New York subway.

But underneath the grime there's some really gorgeous stuff down there. The original tile signs from the days of the IRT and BMT (and even some of the IND) are lovely, and have much more personality than any other subway I've taken.

These sorts of things: https://www.pinterest.com/fennecfoxen/old-subway-signs/


And this is still happening. Someone among the powers that be, budget crunches and old equipment be damned, is still prioritizing the creation of beautiful things underground, with no less personality than what came before.

http://nyclovesnyc.blogspot.de/2012/12/new-years-eve-themed-...

I'm not currently living in NYC, but a big perk of my yearly visit is being surprised and delighted with the new things that have popped up.


You could try London [1], the style is very similar.

The London Underground seems reasonably clean to me, but I used to live there. It's closed overnight, but it's also cleaned during the day — you will see a cleaner on many trips you make in the evening, and trains are often (usually?) cleaned quickly when they reach the end of the line.

Stockholm Metro also has some wonderful design, although it's modern.

[1] https://www.google.dk/search?q=leslie+london+underground+til...


The Guardian ran a photo-gallery "article" today on the Stockholm Metro's art:

http://www.theguardian.com/cities/gallery/2015/oct/20/art-st...


The MTA is treading water -- even factoring in inflation, a subway ride has gone from $1.50 in 2000 ($2.07 in 2015 dollars, according to the BLS inflation calculator [0]) to $2.75 today. On top of that, the MTA is stingier with "bonus" fare costs -- the threshold for the bonus is higher ($10 to $15) and the amount is lower (15% to 10%) -- and you now pay a $1.00 fee for a new card.

[0] http://www.bls.gov/data/inflation_calculator.htm


It's a way to get from point a to point b, I don't think creepiness is a big factor for me. Good lighting is a different story.


As a very infrequent visitor (typically during the summer), I'd prefer a bit more ventilation.


When there are homeless people taking a dump on the platform and hundreds of positively disgusting surfaces.. Creepy doesn't even begin to describe it. Dirty, disgusting and dungeon like isn't a virtue. Combine that with the horrid lack of accessibility and you have the NYC subway. Even PATH looks like luxury compared to riding the metro. Funny how NYC claims to be 'progressive' yet someone in a wheelchair or a mom with small children can't safely use much of the system. Seoul has double the population, yet their system is twice as clean.


Talk about sour grapes... It's anecdote against anecdote, but have you ever actually seen someone taking a crap on a platform? I've ridden the subway several thousands of times in the past decade and I've never seen someone take a dump on the platform or train. That's one of those stories that everyone talks about, imagines, and hits the front page of Gothamist, but is in all likelihood farther and fewer between in reality, than in the stories we tell.

A "progressive" city can still have legacy parts, and those legacy parts are supplemented by custom features. All regular NYC Buses are handicap accessible. There are 24/7, door-to-door, paratransit services for handicapped people that cost the same as a subway fare [0]. Sure, the NYC Subway isn't Seoul or Singapore's, but it runs 24/7 (Seoul's only runs 5:30-Midnight) and the stations on average are probably ~20-30 years older than the oldest parts of the Seoul Metro (early 70s). The unofficial handicapped accessible subway map is depressingly sparse [1], but it seems on par with London's [2] and better than Paris' [3]. (Neither of which are 24 hour systems, BTW.)

I'll admit that the surfaces are gross and the subway collects the most unfortunate, down on their luck members of society, but your post reads as a total condemnation of the best mass transit system in the US. It's dirty and depressing at times, but it's also one of the great equalizers in New York City life.

[0] http://web.mta.info/nyct/paratran/guide.htm#what_is

[1] https://subwayrecord.files.wordpress.com/2015/03/ada-subway-...

[2] http://content.tfl.gov.uk/step-free-tube-guide-map.pdf

[3] http://parisbytrain.com/wp-content/uploads/2008/09/plan_pbs....


I did see someone pooping once. The train car cleared out pretty fast. I also got threatened twice during rush hour. When I first moved to NY, I tried talking to the good people at the counter ... never made that mistake again. After a bit, I just started taking cabs. I did encounter traffic jams periodically but felt a better experience overall. Btw .. I also have respect for NYC cabbies. They are the opposite of Bay area cabbies. I try to not use Uber when I am in NY. Anyways ... the NYC subway sucks when compared to international standards. If you travel a bit internationally, you'll know what I'm talking about.


Travelling internationally has reinforced my positive opinions of NYC's mass transit. Sure, it's occasionally gross, and certainly less polished than London or Paris, but it's 24 hour presence is truly special. It's redundancy and express/local service patterns are also unique. You seem to be able to afford cabs -- that isn't the case for a good portion of NYC's population. That I can go from Williamsburg to Inwood in ~1 hr. 15 min. for $2.50 at 2am Sunday morning....that's something special that other cities can't compare on.

I'm not excusing the cost overruns and budget mishaps the MTA is prone to, but these stereotypes of a cesspool-like NYC mass transit are dated and a disservice to their actual typical operating conditions.


Several cities in Europe that don't run 24h/day every day at least run overnight on Friday and Saturday nights. (London doesn't, since the unions are currently blocking it.)

But your point would work if you'd chosen 2am on Monday.

Copenhagen's metro runs 24 hours a day, every day, but it's small and new (and entirely automated).


I've actually seen it... But that was over 20 years ago. (I really can't get the image out of my head, even after all this time...)

FWIW... I love the NYC Subway, and wish there was more money/attention paid to it. (Especially seeing the metros of other cites in the world.


I guess it is attractive enough as it is?

In other words, subway is efficient because it is always competitive with the congestion over ground, and even if it is creepy, it continues to be used.


A truly american approach. :)

"It works well enough, sort of, so don't touch it."


Not just American (I'm not American, but I recognize the tendency. It's not broken, so don't fix it... and even if it is broken, just fix it, don't try anything fancy.)


Yeah, why not throw a few billion at it; like the "Big Dig". What could go wrong?


The big dig catches a lot of flack, but the value it added to the city is almost incalculable. The North End alone has probably doubled in land value. In the long run, it will more than pay for itself.


Yikes!

Those photos are depressing -- especially the one showing people in a dinky office waiting to pull metal levers manually so trains in the network can pass safely.

Anyone would agree the NYC subway system is mission-critical (without it, NYC would slow down to a crawl), yet its technology infrastructure has obviously been neglected for decades, due to lack of money, lousy management, corruption, or a combination of all these factors.

And no one wants to pay higher fares or greater taxes.


"And no one wants to pay higher fares or greater taxes."

For what exactly? Something almost no one interacts with is new for the sake of new?

Here on HN where its assume high tech is necessary to scale because of high growth, the subway provides a view, however temporary, of another world where once technology scales "enough", new is no longer a relevant goal, aside from maintenance costs.

One big problem with a forklift upgrade of a fifty year old system is the short term maintenance cost of the new system is lower (or, more accurately is unknown, but optimistically assumed to be lower) however the total net systemic cost over the next half century is likely to be much higher for the new system... after all, its very easy to rebuild 1950s and older technology, but near impossible to repair 2010s technology. I'm claiming (cost of maintaining old system 1950 to 2050) < (old system from 1950-2000) + (new system 2000-ish to 2050). We know we can afford the budget cost of maintaining pre-1950s gear in 2015... those costs will only drop with technology, but we know its impossible to internally maintain 2015 gear now and it may never be possible.


That's got to start falling off as a benefit at some point though, right? I mean according to the article at least they're having to machine replacement parts, which has to be multiple times the cost of the original part to do, especially considering the price of American labor.

I don't know enough to say with any certainty if they've reached it yet, but I think they may have.


The article focuses on an interlocking at West 4th St. It's manually controlled because all normal service goes straight through the interlocking without conflicting with other trains. During construction, you can switch the 8th Ave. trains onto the 6th Ave. line and vice versa. That's all that interlocking does.

There is absolutely no need to upgrade this particular piece of infrastructure; it won't increase the throughput of trains through the interlocking, and it won't be easier to maintain. This stuff is _simple_ and works fine.

The upgrade to CBTC seems like a scam from the contractors to me. The Canarise Line was upgraded to CBTC, which can handle 26 trains per hour as opposed to the 12 trains per hour of the ABS signalling. 12 trains per hour is one train every 5 minutes. But they run trains every 6 minutes anyway -- there is no period in the timetable where they run 26 trains per hour, or even 2 individual trains at that spacing. (There is a period where they claim "every 4-6 minutes", for a maximum of 15tph if indeed there are two trains 4 minutes apart. I've just missed a lot of L trains, and waiting 6 minutes seems like "a good day" to me. As I understand it, the L gets most of its load from one station, and so it takes longer than the best headway just to load passengers there. Then, they can't turn trains quickly enough at the terminal. They blame the track layout but it's actually their relaying procedure that's the problem.)


"due to lack of money..."

There's actually billions of dollars being spent on projects like the 2nd Avenue line and the recently completed extension of the #7 train. For some reason, the powers that be consider this more essential than updating the ancient signalling infrastructure and maintaining the crumbling stations.


The problem with maintenance is, that it cannot be seen in public. What I imply is, that 2nd Ave or #7-extension are highly visible. Press writes about this and politicians that are associated with this leave a legacy.

The legacy-part would also happen with maintenance (be the one who brings the system into 21. century) - but only after many years would this leak into history-books. Not fast enough to win the next election.

Esp. if you are a local politician, who got the local construction for the extension, this might make a difference in incentives.


Part of the work done on the 7 was upgrading signaling, which you could count as maintenance.


Perhaps not different than software world -- people want to be known for "created and built ..." on their resume, rather than for "mainted and supported ...".

It is of course fun and exciting if we are talking about spotify for dogs, not as fun when talking about critical infrastructure.


Agreed, but also nobody wants to be known for "the whole system fell apart while I was in charge".


Then just reword: "Doubled the expected lifespan with preventive maintenance and upgrades"


The federal government kicks in a lot of money for big new capital projects but not much if anything for station rehabilitation. As the issuer of fiat currency, and given that we've been at the zero-bound interest rate for a while now the federal government has a lot more money to throw around than the city or state governments.

The MTA also has an arcane governance structure that makes for poor accountability. Four votes on its board are picked by the NYC mayor, 1 each by the county executives of Nassau, Suffolk, and Westchester, 1/4 each by the county executives of Dutchess, Orange, Rockland, and Putnam counties and six by the Governor of the state.


A new station is very beneficial politically - it sways voting, whereas nobody gets that excited by maintenance.


A new station is only useful to the small minority of riders who use it, while system-wide infrastructure problems affect everyone.

Upgrading signalling on a train line would make life better for many more people than building one new station: trains would be able to run more often (which would make them less crowded) and the system would break down less often (the ancient signalling hardware is currently the least reliable part of the system). I heard that the new signalling on the L line is economically reviving a large part of Brooklyn now that it has reliable train service to Manhattan.

And wouldn't people who suffer daily from late and crowded trains and have to deal with leaky stations with perpetually broken elevators be likely to vote against the current politicians when they run for re-election?


sounds like every programming job ever :(



This is comparing apples to oranges. Would Shanghai Metro use 100-year old infrastructure in 2115? That's what you should be considering.


No, the bottom line is that China races ahead and the US crawls along, yet spends a fortune.


Maybe it just works, has been PROVEN to work over a century, has known failure states, and "upgrades" would simply be a massive waste of time and money?

I know there is a huge obsession with new and shiny amongst programmers, but when it comes to massive real world infrastructure projects the last thing you want is unknown factors and "churn", especially when the benefits are unclear and dubious at best.

Just because something is old, dosent mean it should be upgraded, infact if its worked well for as long as it has you should consider that very carefully first.


This is amazing and I hope it stays that way for another 100 years!


Note the acting-for-the-camera fail with the spanner at 8:28-8:32.5.


Much of the technology in US infrastructure hasn't been updated in over 100 years




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

Search: