It's rarely a good thing when a tweet tells you more about the personality of the tweeter than the ostensible subject at hand. Calm down, jesus.
These screens will probably be around for at least twenty years. They aren't good, a waste of money from an institution that is already cash-strapped.
That's only on the older trains that are being replaced. Every new train on the main red, green, and blue lines have recorded stop announcements for at least the last 10 years.
Everybody knows that NYC subway isn't as modern as a lot of systems but don't forget that it's by far the largest subway system in the world and also one of the oldest.
They've installed LTE in almost all the stations now and are rolling out contactless payments. Progress is being made to modernize but it's not happening overnight.
Why won't NYS put more money into the subways? Why can't we get driverless trains? Why do trains have to slow to a crawl if there is work being done two tracks over? Do we need barriers between tracks? So many low hanging fruits blocked by greed and short term self interest.
You are absolutely right but I still think USD 20B is not enough.
How much of the GDP of the Empire State is (directly or indirectly) because of the city? Everyone benefits when the city has better infrastructure. Yes, the unions are at fault. I too get angry when the overtime nonsense making the news.
> After five hours, all he could explain is what a chief measurement operator does — claiming Caputo is “one of only a few people” who can operate an “advanced track-geometry car,” which examines the rails for defects.
> Overtime payments surged to more than $1.3 billion across the entire MTA last year, up from $1.2 billion the year before, according to Empire Center data.
> Tarek justified the megasized paychecks by saying the agency has been working on major upgrades recently, and claiming it is often cheaper to pay an existing employee overtime than to hire another person for the same work.
Yes, this is a problem. I don't have a solution but I think the solution to this problem will also help introduce self-driving trains and help us reduce head count at the agency which should help decrease costs.
And it’s not just unions either. The capital spending side is also monstrously inefficient.
As far as solutions go, I don’t blame you for not having one but I can’t support any additional money when we can’t even have confidence that it will actually improve anything rather than just going directly into even waste, fraud, and abuse.
A quick Google search suggests that homeowners are recommended to budget 1% of home value for maintenance.
In the case of Silicon Valley, I would think that service prices would tend to make up some of the difference.
How much more of an outlier can you get than a subway system in NYC when the rule of thumb is for single family homes nationwide?
Citation needed.
It is the 6th largest after Shanghai (676km) , Beijing (628km), Guangzhou (478km), London (402km), and Moscow (397km). NYC has 380km in comparison. (1)
NYC has a lot of stations compared to anywhere else, but in my experience the stations in NYC are often basically just a set of steps going down from the street, then a simple turnstile leading to a bare platform with basically no other facilities at all (e.g. you cant even change directions without going outside back up to the street and back down through the turnstile again ... I don't think there are even chairs?). Compare this to some stations elsewhere in the world where there is significantly more going on in a station (e.g. Paris, Moscow, London etc). Some of the larger stations in NYC have a bit more in terms of infrastructure, but a lot of the time there is basically nothing down there (and they are dark and filthy) and I can't imagine they are a significant source of strain on the system compared to the other parts (tracks, signals, rolling-stock, bridges, tunnels etc)
It isn't that they didn't have chairs, it's that they can't have chairs.
The homeless problem in NYC is bad and it just happens subway stops are better than above ground when winter comes. Call it mean, I think it is unfair but homeless people belong elsewhere than the subway.
It isn’t really possible to “invest” in affordable housing without significant government subsidies. After all the difference between a normal apartment building and a luxury one is what? Hiring a concierge and buying a brushed metal fridge for the apartments instead of a white one? It doesn’t actually cost much different.
Please, tell me of a city in the US that's built affordable housing within the past 50 years - not luxury housing that got old and thus cheap, and not luxury housing that gets subsidized by the city.
Homelessness in the broader sense which includes people sleeping on couches or in shelters is much more linked to affordability. Unfortunately there’s not great vocabulary to distinguish among rather distinct issues.
The station spacing is not really an issue in New York since the subway has express overlays and the commuter rail system for even faster travel. This is compared to Beijing or Shanghai, which have no/poor regional rail to speak of.
This is really curious. Cities like Beijing and Shanghai have subways optimized for short-distance travels, and high speed trains for long-distance travels, but not much for medium-distance trips like from one edge of the city to another. How so? Is it just because the demand isn't there?
New York was able to avoid this choice starting out, because its CBD and densest areas (Manhattan) are built out in a long, narrow fashion, and you can only thread so many lines through such a narrow space before you start having to make choices between missing transfers and building stops way too close, so the decision was made early on to concentrate tracks for easier transfer facilities.
It's also worth noting that New York's subway was built in an era where buses and streetcars were slow (and in some cases still pulled by horses), so the local lines have stop spacing optimized for walking to a station and the express lines are the actual commuting heavyweights that people transfer to. More modern subways are designed with medium-speed last mile transfers in mind, either via cars that park and ride (BART, DC) or lots of bus transfers (Asian systems.)
Eventually a city becomes too big and belatedly realizes that it needs to have regional express service running through the city, because transferring rail+subway+rail is way too slow. Cities that developed first during the Industrial Revolution can depend on the dense legacy rail networks that were developed during that time (Paris RER, Berlin S-Bahn, London Crossrail) and this mostly consists of building connecting tunnels through the city center with stops at major activity centers.
Those cities that developed during the 20th century mostly don't have dense legacy rail networks, and have to new build after the fact, which can be hard and difficult. Seoul is building a similar network to the above cities, but it's a massively expensive project that they have to build 50-60m underground because no provisions were made to accommodate it before hand. I suppose Beijing and Shanghai will get around to building such a network when they reach a similar level of wealth and can afford it.
http://www.urbanrail.net/as/cn/shan/shanghai-future.htm
But the undisputed king of subway/regional integration is Tokyo, where virtually every subway line runs directly onto regional rail, often at both ends.
I think it all broke down when the city took over, didn't have any money, and cut the frequency of trains. You stopped being able to rely on a timed transfer, and so you often find yourself stuck on a local train because stopping every 5 seconds is still better than waiting 20 minutes for the express train.
The official policy of the MTA that I read a couple years ago is for trains to always try to make the timed transfer outside of rush hour. But my experience is that they stopped doing this when people started writing newspaper articles about on-time performance. Better to have an empty train arrive at the terminal on time than for it to be a minute late because it waited for the express. The customers got exactly what they asked for ;)
As a result, NYC has 1182 track-km.
This simply isn't accurate.
You can change directions at most stations. Generally the only stations where you can't are local stops, with side platforms, and no lobbies... that configuration certainly exists but definitely isn't the majority.
Regarding chairs, off the top of my head I can't think of a single station that has no seating whatsoever. I mean there's probably a small handful like this, but it's definitely a tiny tiny minority.
> Some of the larger stations in NYC have a bit more in terms of infrastructure, but a lot of the time there is basically nothing down there (and they are dark and filthy)
What's the appeal of large stations? Big lobbies just add significant walking time to get to the platform. Also means more space for the transit authority to maintain, clean, and police.
As for cleanliness, personally I'd rather have a dirty 24/7 system than a clean part-time system.
Heck, we’re only just now starting to replace the last of the trains (R32s) from the early 1960s (the trains that were supposed to replace them ended up instead replacing some ‘70s trains (R44s) that turned out to have a construction or design flaw that caused them to fail earlier than expected).
And the agency is basically bankrupt.
Decades to next upgrade the crappy waste of money digital displays seems to me to be a reasonable conjecture.
Paris and London’s subway are older, around the same size, and much much much more modern than NYC. Underinvestment is what makes NYC’s metro shitty.
Analog life is never going to be what people on HN want it to be.
As far as I'm concerned, low-tech solutions in the meatspace are frequently the best ones. Signage of static assets like subway maps should be paper. We've got these short-and-wide red LED signs for "next train in 4 minutes", those should be digital, but they're getting phased out in favor of LCD screens, that's dumb.
Those big new screens showing ads… I guess if they bring in more revenue, great? But if they double as notices for service interruptions, the serious stuff ought to show up more frequently than once every 12 screens…
All anyone wants is some common sense and not expensive change for change's sake. Low-res digital subway maps are stupid, full stop.
- Bring the NYC subway fully under the control of the city: remove Albany from the decision making process so we can actually fund the system.
- Charge based on distance.
- Stop policing poverty. E.g. stop paying cops to arrest fare jumpers and fix shit instead.
Also don't forget trains physically still cost money to maintain. The longer distance they travel, they sooner they'll need to undergo maintenance. As one example, the wheels they roll on do need to be taken off, inspected, repaired/replaced on fixed schedules.
The longer they travel with less people moved, the more expensive it is.
Another case is that the fuel or electricity they use isn't free either.
It’s bikeshedding.
I think the single fare is one of the city’s most egalitarian features and I hope it stays that way.
In this age of inequality further burdening those with the least is the last thing we should be doing.
I’d much prefer a tax on residences which stay vacant for more than six months a year, and a tax on empty store fronts encouraging store front rents to drop enough to let back in some more Mom and Pop stops to shop instead of yet another bank, Walgreens, or Subway.
Discounting the speed walkers who nearly cartwheel over the turnstile we manage to put millions through a day without bottlenecking the fare/board process even on busses (with the exception of those types of riders who will watch the bus coming down the road blocks away and still forget to have their pass ready)
What’s the point of tinkering with the fare structures if you want to make it pay-what-you-wish anyway?
Great. Then you can also cut out the statewide taxes and the local taxes for counties that are in the MTA region, but where the vast majority of people don’t use the MTA.
You trust City management more than I do. How did NYCHA work out? How about the endless proliferation of agencies being created, hand-outs to City employees and contractors?
Stop policing poverty. E.g. stop paying cops to arrest fare jumpers and fix shit instead.
Putting aside the debate about fare evasion and its relation to budget, being discussed here, how does not 'paying cops' and paying someone else (?) play a role?
Everyone that's been to a shopping mall knows how consistently broken information touch-screens are. Suggesting they put interactive maps in a subway is just silly.
The digital maps are to show realtime service information, train-times, etc. It is an improvement, and if you want a printed map, you can just ask for one at the booth.
The whole "my map isn't in 4k" argument is whiny, at best.
This is a design decision that was neglected, plain and simple
Also, interactive maps aren’t as simple as “make it from D3.js”. You’ll have a screen that thousands of people interact with every day, likely. The UX needs to be thoroughly considered so that it’s non fuck-uppable or breakable. That’s actually the #1 problem with digital maps, they break.
Is it just me, or does anyone else think that the Twitter user is being fairly tame, and everyone here calling him "hysterical" are the ones overreacting?
Having infrastructure for digital maps is a good plan. The service patterns change at night. There is construction during all off hours that results in trains being rerouted. A digital map just shows you exactly what the state is right now. That is a good thing, not a bad thing.
Making the taxpayers buy super high resolution not-yet-invented displays instead of off-the-shelf $200 TVs to make this guy happy doesn't seem worth it. Just make the text a little bit bigger, or even redesign the map completely.
Make sure that when the electronic sign says that the train is "2 minutes away" it actually shows up in 2 minutes.
If a sign says train is "6 minutes away" there should not be a train entering a platform 2 minutes later.
What the MTA needs to fix is when the schedule says that trains run every 3 minutes, the countdown clock needs to look like "1 4 7 10", not "1 4 33 36".
Think of this being an autoscaling group. The argument that people make sounds like "Well, autoscaling group works fine when one instance can handle all the traffic. We are happy with it." It is not an achievement to have an autoscaling group work when it has one instance in it.
https://www.theatlantic.com/technology/archive/2015/11/why-d...
Start with automated lines ( the L ) where you know the location of every train. You know the speed of every train at every point. Now it is a standard algebra problem that 15 year olds were solving 20 years ago in algebra/comp sci exams, using pen and paper - it was pre-calc so they invented how they could divide the track into equal section and compute the time it takes the train it should take the train to get to the station by approximating the unknown time based off the time that it took the train to pass the sections behind it. On non-automated lines start by adding god damn train trackers to the individual trains and start splitting the blind sections of the track by adding cameras and sensors. They don't need to make the lines automatic. They just need to add visibility.
None of this is difficult. MTA and "infrastructure is hard" people are full of excuses. They are the same people who were saying that the L should be fully shutdown for 18 months when now we have MTA announcement that even with the partial shutdown they are way ahead of schedule with the tunnel work so they can start doing the station work.
There should never be the case when it says a train is 8 minutes away and a train behind it is 9 minutes away when a train gets to the station in 2 minutes and the train behind it gets to the station in 5 minutes.
This is not a small thing
https://www.amny.com/transit/mta-to-install-subway-countdown...
I think they rolled this out from prototype to full implementation in less than a year, which is very impressive.
This is like complaining about a car trip taking less time than Google Maps predicted, due to an accident ahead of you getting cleared up before you reach that stretch of road.
Unpredictable things happen on public transit: sick passengers, police activity, signal problems, mechanical failures, cascading delays causing platform crowding and longer board times. These things cause delays, and then on the tail-end of the event, sometimes they cause trains to be "early" when the delay clears up. Even then, though, in my experience it's pretty rare that the train comes before the time on the countdown clock.
All things considered, I find the countdown clocks to be pretty accurate a vast majority of the time. And they're certainly a huge improvement over how things were before they were installed: a loud beeping noise about a minute before the train arrived, sometimes paired with a muffled inaudible announcement from the station attendant over an ancient PA system.
There's an announcement on the loud speaker that there's no service between Broadway Junction and Canarsie in both directions because of police activity. Two minutes later, according to the count down clock there's a train to Lorimer in 6 minutes and to Manhattan in 17 minutes. Followed by 5 - 16 and 3 - 13. There's still an announcement that there's no service between Broadway Junction and Canarsie in both directions. But there are magical countdown clock connected trains that will be now here in 3 minutes and 10 minutes. Oh wait, the magical clock is saying there's a train entering a platform! It is going to Lorimer and it is followed by a train to Manhattan! The clock is at zero and it is blinking! All aboard the subway track! This went on for about 40 minutes. Twitter says there's no service. Countdown clock says there's a train coming! Google transit says there's no train. Countdown clock says there's one! And when the service was restored the countdown clock said the next train was going to Lorimer and it was finally 6 minutes away. Except that a train to Manhattan was pulling into the station.
The countdown clocks are garbage because the only time that they matter is when there are sick passengers, police activity, mechanical failures and cascading delays -- otherwise we know the train will be here in about ~4-5 minutes and that's exactly the time when the countdown clocks cannot provide reliable information.
MTA should finish that project and make it work. After that they can pick their next "digital initiative"
The point is it does not matter if the L is going through the historic reconstruction project -- countdown clock should depend on two pieces of real time data and one piece of historic ( same trip data ) from previous sections:
1. Position of the train which is always known on the L because it is an ATC line.
2. Speed of the train. Also known because it is an ATC line.
The distance between stations is fixed. The data about speed and the distance train covered at t-1 and t-2 is known.
So it is algebra + basic message passing architecture + LCD displays connected to the local systems.
2. Speed of the train
On some level, isn't this just the nature of twitter?