> its implementation of a “NextGen” air traffic control system to replace the current version may not be completed until 2034, even though the project was started in 2003.
Governments (and a lot of businesses) like to look at software as a one-time purchase, but it's really better too look at it as a liability and an ongoing cost. It'd be better to have a team make continuous, incremental improvements to the system than have "NextGen" last-gen replacement vaporware.
I can say that at least one extremely large US company successfully retired their transaction processing mainframe system using exactly this pattern.
It wasn't simple, but running old+new in parallel on the same production data stream (then verifying behavior against each other) is a surprisingly powerful approach.
Most critically, because it trades temporary maintenance and infra costs in exchange for decreasing risk and flexibility around cutover date.
Kind of like black box TDD, which helps the new system team be much more aggressive with their delivery timelines.
There are examples of both. Firmware and embedded software usually is a one time purchase. Tax software requires ongoing updates. FAA software would fall somewhere in between.
Lots of firmware is upgradable these days. Some of of it even should be updated regularly, or you end up with weird novelty problems like your bluetooth earbuds being vulnerable to drive-by attacks.
It really is a binary though. Either you need to think about and deal with deploying (system-wide) updates, or you don't.
Even infrequent ~5y update lifecycles tend to be extremely painful unless there is substantial investment in treating it as an essential business process. This leads to a "kick the can" mentality that translates to show-stopping amounts of tech debt.
Five years is almost the worst case. In five years most of the people who did it last time are either in a different post or have left, meanwhile related systems have probably changed, so if they did leave documentation and if you can find it still, that may no longer be the correct process.
When Let's Encrypt was created it could have issued 3 year certificates. But it didn't because they knew that's a bad idea, the whole point of Let's Encrypt was automation, and if the certificates expire in 3 years you will say "Eh, we'll automate later" and never do the work.
They like to give contracts to connected outsiders so that they can take care of their friends and take the blame away. 30 years? quoting the venerable Sir Humphrey Appleby:
"Precisely. Months of fruitful work. Leading to a mature and responsible conclusion."
Since it takes 30 years it must be a responsible piece of software, polished to the bone.
i flew delta out of ewr last week. the pilot announced over speaker that air traffic control was serializing take offs on one runway to avoid problems. we sat on the tarmac for two hours waiting our turn. that was on top of another hour delay for our flight arriving due to this policy.
delta doesn’t reimburse for missed connections claiming air traffic control policies are outside of their control.
it reminds me of a dev team i worked with once which used single threaded memcache as a way to serialize inbound requests to a server with improper locking logic inside.
> delta doesn’t reimburse for missed connections claiming air traffic control policies are outside of their control.
AFAIK if it's one booking on the same airline or a codeshare they are required to rebook you. If you planned a "connection" which is two single flights with different airlines you don't get any legal protections. This isn't just Delta, no airline will reimburse you for missing a flight you didn't book through them
they did book me on a flight the next day. i was scheduled to fly two legs on delta. but see my other replies below on lack of accommodation and rep behavior.
perhaps. what’s more my ticket to cross from east coast to western state cost more than my previous ticket to cross the atlantic into the country. maybe that’s a rational market outcome but it’s hard to fathom sometimes.
they still booked me on next day flight but refused any compensation for hotel or meals or ground transport for the overnight stay due to the missed connection.
>delta doesn’t reimburse for missed connections claiming air traffic control policies are outside of their control.
Reimburse? They'll put you on another connecting flight of your choice as any airline does. I did this on a Delta flight with a missed connection due to delay just last week. They also automatically upgraded me to Comfort+ at no extra charge for the inconvenience.
I'm in no way qualified to comment on the details of the issue outlined here (though I did get stuck babysitting a friend's kids for many extra hours due to a 6hr delay on said friend's flight into EWR a few weeks ago). For anyone who lives far enough north and west of EWR though, I highly recommend trying either the Allentown or Scranton/Wilkes-Barre airports. I've moved my business travel to Scranton 100% and have been loving it. There are more restrictions on flight times, and basically everything needs a connection, but it more than makes up for any time I would have lost due to sinkhole traffic on I-80 and the EWR parking shuttles. It's amazing to park and be at the security line basically 3 minutes later.
> To save money, the FAA elected not to build a new STARS server in Philadelphia to support the move. A new server alone would require tens of millions of dollars, as well as installation of new internet and power infrastructure.
> Instead, it elected to send a “mirror feed” of telemetry from the STARS servers at N90, traveling over 130 miles of commercial copper telecom lines, with fiber optics to follow by 2030.
> The annoyances of traditional cable internet — frequent lag, dropped sessions — are probably familiar to those who stream video or play games online. But for air traffic controllers, even the smallest service disruptions can become dangerous.
So LOL what, they just ... piped it over the Internet? Also can someone make sense of this "new server" costing millions of dollars? Presumably it's not the cost of a server, which is orders and orders of magnitude less than that?
I doubt that 130 miles of copper is the public Internet. It’s presumably some legacy telecom system that depends on a bunch of generally fairly well made but thoroughly obsolete hardware.
Keep in mind that Ethernet over copper is only specified to ~100 meters. Long distance copper networks have been obsolete for a few decades.
Here in my EU state, air traffic control still used leased copper twisted pair for server link over ~150km as of at least 2015 or so. It's pretty reliable and not "public internet" at all - just two modems with known performance and known latency taking to each other.
Might be something like that in this case as well.
Sure, this kind of technology works fine. Except that it might be difficult to find people who how maintain or repair the fancy (pressurized? oil-filled?) copper cables or the modems, and finding parts might be interesting.
I did some enterprise network work ~20 years ago, including fiddling with some inter-building links, rummaging through closets, and visiting the inside of a legacy campus-scale analog phone exchange, and I’ve never even seen the kind of equipment that can send data at an appreciable speed (even by 1990s standards, and even with repeaters) over copper at a range like 100km.
In contrast, single-mode fiber has improved over time, but it’s not obsolete, and it has maintained a remarkable degree of compatibility over the years. New transceivers largely work on old fiber, old transceivers work on new fiber, etc.
It’s called a dry loop. It might be real with repeaters at the various COs, or it might be partially virtualized and sent over a fiber trunk. Usually somebody leading a dry loop does so for the relatively extreme reliability, even though the speed is not very high.
Why not pipe it over the internet? A few redundant ISPs at the endpoints should be as reliable as a private run of fiber or T1 lines or whatever bespoke solution they think they need?
Doing so sticks billions of transistors between A and B, all of which have to work perfectly billions of times a second. If you just run fiber, you need to put an amplifier in every 50km or so, other than that it’s all just buried glass.
The fiber is already there, and it gives you guaranteed capacity, even during a major power outage. And it guarantees no one is changing out a switch or a router at a bad time, etc.
True, saturating all routes to a host is quite feasible, but with sufficient redundancy (say with multiple PTP wireless links, which can be done quite cheaply) you'd have to knock the Internet out in such a huge area we'd be talking about WW3.
There's no argument that a private line is ideal for critical infrastructure, but if they must make do, there are ways to make it work.
People managed to shoot Wikipedia off the Internet as recently as 2019.
Anyone not behind one of the major CDNs can be targeted and taken down. And bandwidth-exhausting a few routers in front of the target is certainly possible, it's just not profitable for cybercriminals - but for a nation-state? No problem at all to muster a few terabit/s of capacity for them.
If it's mainframe, and there is no reason to believe it's not given the strict requirements, the hardware is expensive as fuck and the people who can keep these beasts alive are just as expensive.
I think they are running on old Sun Ultra workstations, not mainframes. Hardware easily replaced and upgraded with much faster, cheaper options. Difficulty is much more on the software side.
An ATCS like STARS needs to feed from multiple different OT and IT sources like radars, weather stations, other TRACONs, etc and is implemented in it's own airgapped environment.
It can get very pricy very quick. On top of that, the FAA's budget has been sclerotic for decades now after the 1980s era union action and the 1990s era national cost cutting.
And finally, it is a political organization, and NATCA is a fairly prominent union within the AFL-CIO, and could make the lives of NJ representatives hell for pushing reassignment out of Newark.
I would believe it was hard. And maybe it still is if you’re unwilling or unable to take advantage of modern technology.
Current low-cost equipment can easily send 10 or 100Gbps over long distance fiber links. Depending on how quickly you want to fail over when a link or an entire switch, router or rack fails, there are plenty of options that make various tradeoffs between failover latency and bandwidth, all the way up to completely duplicating all the traffic on redundant routes. I would bet that the entire aggregate traffic needed for air traffic control in a region is well under 10Gbps. And 10Gbps dedicated links or leases or (effective) purchases of dark fiber are not expensive on the scale of the FAA. Air traffic should use a network with a lot of redundancy, so maybe multiple those low costs by something like 5.
Heh, have fun hooking legacy systems to high speed networks without significant testing.
If seen plenty of old stuff crash because you'll have some ancient serial device with a limited buffer and someone jams a faster link in-between. All of a sudden you have a much larger amount of bandwidth delay product and the system doesn't handle a few megabytes of data getting lost on the line when it bursts for some unexpected reason. On the old fixed line that just couldn't happen.
A "Server" is just layman's speak for that big room with all the noise and blinky lights. Including all the hardware, networking and software that goes with it. Of course you're not going to run critical infra on a single server :)
Though I have to admit I have seen government operations where the "server" was an old dell optiplex desktop lying on its side in a broom closet without ventilation, a post-it with "IT SERVER DON'T TURN OFF", a spiderweb of cables running through the closet and the "server" fans screaming for air trying to keep everything cool in the enclosed space. I'm not kidding.
I mean, I know, government. Small local welfare-related org. Shoestring budget. Sure, that sucks. But at least you can make sure it's tidy and the cabling doesn't look like shit. Jeez. I didn't imagine I'd still see that in this century. Do people no longer take pride in their job? They hadn't even activated the "AC Power on" in the BIOS so after electrical maintenance they had to wait for the "engineer" to press the on button again.
There's no budget to fix whatever you messed up when redoing the cabling, which we also didn't really have the budget for. Everyone who knows what the box actually does has retired or died, so nobody wants you in the broom closet shifting dust motes near a working system we can't replace.
Old tech. You’d literally lease a line from ATT or whatever Bell company covers your area. A reserved private connection. Modem at each end. They ran at about 56Kb/s IIRC.
It's amazing, really. You hear about "government overspending" all the time. You actually look into something in any detail and what you discover is a consistent pattern of underspending. Call it mismanagement if you want, but it is consistently what we, as votes, and the executive, ask them to do.
Meanwhile, back at the ranch, I was a contractor for HHS for a year or so. When I started, there was a server farm built for a contract that was powered up and maintained; never used, it was broken up and sold for surplus. Just before I left, HHS thought they would like that program again, so us US taxpayers bought another tranche of computers to power up. I don't know the end of the story.
I could type more, but it would be a long and boring story.
So true. The state of Washington has a great example: the outgoing Secretary of Transportation was very clear that if the state doesn’t change the budget to be primarily about maintenance, and much less about new highways or lane expansion, then our infrastructure will quickly begin crumbling.
Even the people in charge of our highways want us to switch to operations & maintenance oriented projects, but the representatives have not done so.
The incentives in government are really fucked — you get visibility and wins through cool projects, not by keeping the lights on and things running smoothly. Honestly true in big companies as well.
Here in Texas the new hotness is just building more and more toll roads and express lanes. Which are not toll roads, but even more expensive somehow.
It's to a point some major parts of DFW like Plano and Frisco are only accessible via tollway. That's the only highway - the other "highway" has a light every half mile. Meanwhile, we have some of the most dangerous interstates and highways in the country.
Where's all that toll money going? I have no idea, but maybe it'll pan out. It just sucks because it's already quite expensive to commute in DFW.
Newark’s overworked controllers might argue differently. Of course in this forum the general suggestion will be to replace tired controllers with sleepless machines, and the technologists here have strong incentives to advocate for such solutions.
(Shrug) ATC is no job for humans, and I'm tired of pretending it is.
If we were building our aviation infrastructure from scratch starting today, you would get some really strange looks if you suggested employing humans to manage air traffic.
En-route ATC is already mostly automated, with humans supervising the system and talking to the pilots.
Arrival/departure/ground ATC has to deal with much more complex traffic, emergency situations and edge cases in general. Technologically, we're nowhere near fully automating this.
When's the last time someone in a control tower was actually surprised by anything genuinely new? (Anything that didn't involve ancient equipment breaking down in a new and unusual way, that is.)
ATC is a solved problem. Nothing is going to happen that hasn't happened before, or otherwise can't be anticipated. There's nothing about ATC that inherently demands human involvement in real time.
> Nothing is going to happen that hasn't happened before, or otherwise can't be anticipated.
By the way, this mentality is at odds with safety. If you design a safety-critical system under the assumption that nothing unexpected will ever happen, then whenever it happens (and it's a question of when, not if) people will die.
The first and most important reaction to anything unexpected is simple: localize the trouble spot and reroute traffic away from it.
What can't be handled by doing that? You've listed quite a few examples (and thanks for taking the time to do so!) but all but one of them seem like perfectly reasonable scenarios for automation.
Another point I'd raise is that most ATC screwups don't involve anything weird happening except failures to follow existing ATC procedures. Any list of Things That Make For Bad Days at the Airport needs to include that.
Happy you asked! Here are a few interesting cases that the respective ATCs probably hadn't seen before or will ever see again:
- Plane loses radio communication on final approach and ignores go-around orders, lands on occupied runway as ATC gets others out of the way [1]
- Guy steals a plane and takes off without permission [2]
- Someone left some cones on a runway at JFK [3]
- Door blows off aircraft [4]
- Some dude runs across the runways [5]
- Pregnant woman giving birth on flight [6]
- Student pilot freaks out, ATC calms her down and gets her instructor on frequency [7]
- Earthquake [8]
- Pilot hits a deer [9]
- Bomb threat on board, pilots decide to evacuate on the tarmac after no help arrives in ~1h [10]
Some situations have probably happened before, somewhere. Some others are completely new. I would highly recommend that you watch these videos (they're all relatively short) and genuinely ask yourself whether our current state-of-the-art AI models would be able to successfully handle these situations in the short timeframe required to do so. Let alone the fact that by AI we mean text models, so I'm not sure how they would integrate with terrain information, real-time radar data, arrival/departure routes, etc.
I think you have a fundamental misunderstanding of what control towers do. They're not there to simply observe planes do their thing and intervene if they get too close. They actively handle the traffic, and this task requires human pattern recognition and cross-domain reasoning skills in a matter of seconds, and the technology to replace this is simply not there yet. If you still disagree, I'd love to learn which technologies you'd apply to this problem and how they would compose with each other in order to achieve the same outcomes as in the cases I linked.
Of course, this is not to say that ATC shouldn't be made safer and more automated wherever possible. Particularly in the US, where equipment is severely outdated and some dubious regulations allow their ATCs to handle runways and give clearances in a way that would not be allowed in Europe and have already resulted in more than a few close calls. These are all valid concerns, but IMO they can't be extrapolated to "all ATC services can and should be automated".
I fly often, and I for one feel safer (at least, with our current technology) knowing that there are humans in the cockpit and in the control tower who can react and take control when unexpected stuff happens.
Plane loses radio communication on final approach and ignores go-around orders, lands on occupied runway as ATC gets others out of the way [1]
Trivially handled by video game-grade AI. Is an aircraft somewhere it's not supposed to be, or doing something it's not supposed to be doing (most likely because a human has screwed up somewhere?) Route everything else out of the area, signal authorities. Human intervention not required.
Guy steals a plane and takes off without permission [2]
See above. What can ATC possibly do about this, besides alert other aircraft to the situation and signal law enforcement?
- Someone left some cones on a runway at JFK [3]
Foreign object detected or reported on runway, an everyday occurrence. Alert other aircraft and facilities personnel. What else is the controller going to do, go downstairs and pick up the cone?
Door blows off aircraft [4]
What is the controller supposed to do about this, other than accept the crew's request for an emergency landing and (guess what) route everybody else out of their way?
Some dude runs across the runways [5]
Like other 'surprises', two-legged FOD might have surprised the Wright brothers, but that's about it. Alert law enforcement and warn aircraft.
Pregnant woman giving birth on flight [6]
Just another routine 'surprise' covered by standard procedure. Handled primarily by the air crew rather than ATC. Handle emergency landing permission if requested by crew, alert medical personnel to meet the aircraft when it lands, done deal.
Student pilot freaks out, ATC calms her down and gets her instructor on frequency [7]
Honestly not sure what should be done to automate this kind of situation. I suppose an LLM could handle it as well as anyone else. :-P
See also the edge cases where the pilot(s) are incapacitated and an untrained civilian needs to be talked through the landing. How often does that happen in real life (and how often does it actually work?)
Earthquake [8]
Not really something ATC would be involved with except at the purely-tactical level. Ground outgoing traffic and reroute incoming flights until all-clear given by officials.
Pilot hits a deer [9]
Reroute traffic and alert medics.
Bomb threat on board, pilots decide to evacuate on the tarmac after no help arrives in ~1h [10]
Not an ATC issue except for the need to (guess what) reroute traffic.
They actively handle the traffic, and this task requires human pattern recognition and cross-domain reasoning skills in a matter of seconds, and the technology to replace this is simply not there yet. If you still disagree, I'd love to learn which technologies you'd apply to this problem and how they would compose with each other in order to achieve the same outcomes as in the cases I linked.
You could still have a valid overall point regarding the need for humans in the real-time loop, but I disagree that most of the scenarios you mention support that point. All of those cases can be (and are) anticipated, all of them have happened before except possibly [10], and all of them can be handled by computers at least as well as humans. Except possibly the situation with the pilot needing real-time psych support.
Let alone the fact that by AI we mean text models, so I'm not sure how they would integrate with terrain information, real-time radar data, arrival/departure routes, etc.
I definitely don't mean LLM-style text models. As I suggested in the first answer, nothing on this list except (again) #7 would have flummoxed a game programmer ten or twenty years ago. If some people want to try an ELIZA-like LLM to deal with #7, fine, but that would be a research problem. Nothing else on your list requires any new research.
You have listed the correct actions to take in each scenario, but it's still not at all obvious to me how each case would be fully automated, start to finish. I would like more specifics, besides simply
Since I took the time to compile a list of examples that you sadly didn't find surprising enough, I would appreciate it if you returned the courtesy and provided a more concrete design of a fully automated ATC system using current technology.
> Trivially handled by video game-grade AI.
I think it's fairly likely I have played more flight simulator games than you have. If you know of a single one of them where the ATC AI isn't utterly stupid, please do let me know, I'd love to try it. There's a reason VATSIM exists and remains very popular :)
First of all, to the extent controllers are "surprised" by any of those events, you're making my point for me.
Second, I wasn't referring to ATC AI in flight sims; as you suggest I have no experience with that. I doubt anyone ever put any serious effort into flight sim ATC AI compared with AI for realtime strategy or even FPS games. Years ago, a primitive neural net dragged a 9-dan Go master up and down the ban, so I imagine our current tech can handle a few planes.
I'm not paid to redesign the ATC system, or qualified to do so (although that won't stop Musk, I'm sure.) But as I've made clear, I believe it can and probably should be done. In practice it would probably look more like
if (AircrewOrEmployeesReportForeignObjectOnRunway)
{
closeRunwayAndRedirectTraffic();
alertFacilityPersonnel();
leaveRunWayClosedUntilSomeoneInAuthorityReopensIt();
}
So many of the things you mention are simply not in ATC's wheelhouse to begin with. Their job is to keep things moving (or not) while other people in authority deal with those situations. That part wouldn't change, as I see it.
Thanks for replying. I won't continue this conversation as it seems to me you don't actually have any specific evidences of your claim besides "I believe it can be done", and don't actually know much about the responsibilities and areas of authority of ATC.
I'd say the burden of proof belongs with those who argue that it's impossible to automate ATC while maintaining at least the current safety record. That's an extraordinary claim at this point.
I asked for examples of situations that couldn't be handled through automation, you provided some that I consider invalid or inapplicable, and... well, there we are. GG
It's very easy to make a flippant comment that disagrees with another, with no supporting evidence, but demands multiple examples from any future replies. A reply doing that could take a huge time investment to do properly (just look at agubelu's incredible reply to you). Sometimes the only feasible solution to that imbalance is a down vote.
It’s like I summoned you! Just be honest about your incentives if you care to make these arguments. Then be prepared to answer the accountability question, for when the system inevitably fails.
My "incentive" is that I fly somewhere every once in a while, as do people that I care about, and I want the system to be as safe, reliable, efficient, resilient, and cost-effective as possible.
> Then be prepared to answer the accountability question, for when the system inevitably fails.
Airplanes have gotten increasingly automated. Who is responsible when Airbus' excellent automations that have prevented countless upsets and accidents fail? Nobody, if it was an honest mistake, and lessons learned are applied to improve even further.
The problem with modern ATC is that a lot of the safety systems are bolted and backported on top of existing extremely legacy tech. Ffs, the communications still happen over radio where transmissions are missed if more than one person talks at the same time. And people have died because of this, as well as controllers making a mistake or pilots and controllers misunderstanding each other.
There's no reason to continue bolting more stuff on top. A very large part of ATC can be fully automated and made safer.
All true, except that a key reason they still use AM for voice communications is precisely because it's obvious if multiple users are trying to transmit at once.
AM is obviously not the way a "CSMA/CD" system would be designed today, but it does get the job done, and has for a long time.
AM is good for instructions because its broadcast nature allows everyone to hear everyone, but arguably coordinates-over-voice-over-AM is dangerously error-prone. My sense is that the ideal system might look something like "AM with a digital side band" where ATC can press buttons to transmit data (e.g. authorized altitudes and vectors) to a plane but metadata is still carried over AM.
A lot of fatalities have been avoided thanks to a pilot overhearing ATC giving takeoff or landing clearance to another plane, but quite a few incidents could also have been avoided if plane cockpits had a big red light "authorized to enter runway" which could only be turned on by ATC. In an industry which is designed around so many redundant systems, it's rather astonishing that an error in a single communication channel can lead to disaster.
In a pretty remarkable statement a month or so ago Trump actually said we were overspending on the military and suggested cuts. Then his budget comes out and proposes military expansion and Medicare cuts so like usual he was just bullshitting I guess.
It's like tech debt. It's an ongoing cost in a one and done environment, it's hard to see problems from the outside until there's catastrophic failure, and if there's a slow niggling annoyance of things getting worse over time the point where people notice enough to care is usually past the point of needing a refactor. So we get underspending where it matters, overspending where it doesn't, and the solution is always a redo.
Came across this post, which perfectly encapsulates a sort of pathological inversion of sunk-cost fallacy, where some people I encounter would happily let infrastructure rot and countless people suffer if it saves them pennies on their taxes, and are incapable of re-evaluating their position because “taxes bad” has somehow become such an ingrained/religious value that anything else is unconsciously and immediately rejected no matter the consequences.
>>I posted this somewhere else ages ago, feel it's relevant
>I remember having a conversation with my ex's sister and their mum a few years ago, around election time. I try to not talk politics with people because it's a fast way to lose friends, but the topic came up between my ex and them over dinner and I just listened in.
>I remember them saying that the only thing they're interested in is tax cuts. More money for them. I had to chime in and ask what about the NHS, what about funding schools? They said they didn't care because they had private healthcare through their jobs (finance), so they don't need the NHS. The mum said her kids are through school so she doesn't care about funding schools, and the sister said she'll be sending her kids to private school one day. I was pretty gobsmacked at the brazen selfishness of it, and asked what if they lose their jobs - and therefore their private health care - or become unable to work, what if when you have kids you can't afford private school? Neither of them could grasp this hypothetical... it was as if I was speaking another language to them. They were just like 'but we do have jobs.' And what if you didn't? 'But we do.' It was just circular and they couldn't see themselves in any situation other than the one they were currently in.
>I can quite see why empathy is a hard concept for right wingers to grasp, it was like they just simply couldn't understand the concept. They weren't stupid either, and nor were they rich - the mum worked in admin for a finance company in the city, and the sister was being paid by the same company (mum got her in the door) to train as an accountant.
>I think about those two every now and then when I can't understand how the other side thinks. Because it seems we do literally think very differently.
Have been many billions of dollars spent over the years to "modernize" air traffic control. If my recollection is correct, most of it ended up being wasted.
The current administration is asking for a lot of money to try to fix it again
"Most people see a number $6 Trillion and something breaks inside them.
What they don't realize is maintaining infrastructure is expensive."
It breaks all our brains too, because that $6 Trillion has very little to do with maintaining infrastructure. The bulk of it is just direct payouts (Social Security, Medicare and defense contracts).
Pretty frustrating that this big number means the government is politically forced to do drastic austerity for things like keeping planes flying safely.
Also, makes DOGE starting at USAID (<1% of budget) look especially incompetent.
When people think wastage, they think bureaucrats wasting away in offices. Social security admin, air traffic control, and teachers are more like Amazon warehouses at this time in terms of utilization. Less than 1% of social security admin funding is overhead at this time. Any charity would be exhilarated to hit numbers like that.
With 3 million employees and about one cent a flush, the US government spends 100,000 or so a day just to flush. Assuming three bathroom breaks a working day.
It doesn't, but if it did the decline wouldn't just be relative but absolute because active military personnel is actually down almost 50% since its cold war peak. The US military is significantly leaner nowadays across the board. In terms of spending too, something like 8% of GDP at its peak compared to 3.x now.
> you discover is a consistent pattern of underspending
No, what you discover is a pattern of wasted spending. Then they ask for more money to actually get the job done despite all the waste.
People try to solve this by privatizing certain things, figuring that competition will help efficiency. Sometimes it works, sometimes it doesn't.
Maybe we need competing governments, and whichever government is more efficient gets to rule. Seriously: Add a second FAA at some test airports, see if they can do better, with the understanding that if they can't, they get shut down.
> Maybe we need competing governments, and whichever government is more efficient gets to rule. Seriously: Add a second FAA at some test airports, see if they can do better, with the understanding that if they can't, they get shut down.
And you would be willing to be personally responsible if people die in this experiment?
It's funny how people here always complain that any money government spends is wasted, but if you look at big companies they are "wasting" money as well. Just look at the number of projects that google killed. It's simply a function of large (and small) organizations that they don't get it right all the time, it's difficult to predict the future.
>It's funny how people here always complain that any money government spends is wasted, but if you look at big companies they are "wasting" money as well. Just look at the number of projects that google killed. It's simply a function of large (and small) organizations that they don't get it right all the time, it's difficult to predict the future.
I worked at a Fortune 50 financial services company back in the 1990s and they designed, built and deployed a brand, spanking new customer service platform.
They spent USD$200,000,000.00 building the system. It was ready for production when someone realized they'd spend USD$50,000,000+ per annum supporting the platform. The product roll out was scrapped, and tens of millions of dollars of equipment sat in warehouses.
When trying to redeploy said equipment across the enterprise, group heads would purchase new equipment rather than using the equipment sitting in warehouses as it was "cheaper" on their budget lines to spend real money rather than put the depreciation of already purchased equipment on those budget lines.
So yeah, big companies can be incredibly wasteful. Often more wasteful than government (e.g., US Medicare has ~3% overhead -- show me any private insurer that can top that) as well.
It's all about the incentives. People (and consequently, organizations) respond to the incentives inherent in a situation.
Sometimes those incentives promote efficient outcomes and sometimes not so much. The trick is to maximize the former and minimize the latter. Something easier said than done.
> Instead, it elected to send a “mirror feed” of telemetry from the STARS servers at N90, traveling over 130 miles of commercial copper telecom lines, with fiber optics to follow by 2030.
This does not make any sense. If they would really transmit data over a 130 miles copper line (which I doubt even still exist, especially not commercial ones), we would be talking rates in the low Mbit/s. I suspect the situation is that the "last mile" of the center is served by copper connections, not good either but by far not as bad as a 130 miles copper connection.
EDIT: I should add if they really would have a link running on copper lines it would have repeaters, which would be sitting in datacenters. In New Jersey there would by 1000s of km of dark fiber floating around, so it would be trivial to convert at least the majority of the link to fiber.
Telephone providers lease "dry loop" lines for point-to-point signalling.
I've used them for telemetry systems with acoustic modems on both ends.
Also for sending audio between broadcast studios. I recall that it was priced by bandwidth (in the literal, analog, sense), e.g.: 5kHz (~AM radio) was less expensive than 15kHz (~FM radio). For comparison: A normal phone call is 3kHz.
So yes, copper and repeaters. But very inexpensive and quick to provision. :)
I am guessing that the data feed requirements of ATC telemetry are fairly modest, well within the range that a pair of analog modems could handle. You don't have to send a raw radar feed to the airport, just put the intelligence (object detection, delta computation, etc) at the endpoints and transmit the results.
I'm also guessing that they do this every day at hundreds of other airports, and it works just fine.
Moving to fiber gives you greater bandwidth, reliability, multipath redundancy, etc. But I bet the real motivation is access to newer and more general communications equipment. Maintaining the old, perfectly adequate but increasingly unusual hardware/software is more of a pain than replacing with new stuff that exceeds requirements but is maintained just like all the other modern stuff everywhere.
All guesses. I do not work in ATC, but I've seen the pattern in other industries with similarish requirements.
Air travel is far too subsidized by the public and I can think of few worse applications of public funds save maybe sports stadiums. It's a huge waste of energy, pollution, human labor...
Rail is incredibly efficient, and there's a reason China has been building high speed rail as fast as it can.
To all the "it would never work here" people: we used to be a nation of rail travel, where you could walk or bike or take a taxi to the local trolley/train/bus station, take a train to where you needed to go.
All that was systematically ripped apart by the auto industry either directly or indirectly. There is no reason whatsoever we can't work our way back, especially given how much faster and easier construction of a railway line is now.
I'd LOVE some proper high speed rail across the US but EWR is also huge for international flights. A good rail network would at least help getting to/from their for it but that still leaves a good amount of flights that make no sense to dump.
I think it was the wrong choice for a number of reasons, but the farm towns in question are just first lot of the whole network, starting with the (supposedly) easiest part. Instead of building in the densest parts which would be even more complicated and expensive.
As a European, I don't understand the fetishization of railway. These are two different means of transportation with two different use cases, with tiny overlap between them.
There is a case to be made for enhancing rail transit in the eastern seaboard and maybe parts of the Midwest, but America is too large and sparse to justify rail transit at scale.
It makes more sense to concentrate on rail infra for freight transit and work on revamping our existing rail freight infra.
> there's a reason China has been building high speed rail as fast as it can
China stopped subsidizing HSR during the COVID recession. It costs the exact same as a flight ticket now [0] due to high debt [1] (excluding the Beijing-Shanghai track, which actually can justify usage).
Most Chinese use normal rail for intercity transit, but this is easier to justify given the density and ease of land acquisition.
But even then, China began slowing down railway investment and construction since 2018 [2][3], and started calibrating towards air transit [4] as part of a commercial aviation push [5]
An interim solution would be to force carriers to upsize their jets to reduce total flights per day. Force any carrier who has wide-body aircraft to use them on domestic routes. heavily limit or ban the use of smaller commuter jets (replace them with 737s with fewer total flights). I'm sure they won't be happy about it, but it's a solution that could be implemented overnight and the larger carriers would have no problem adjusting. smaller ones could sell their slots or lease larger aircraft.
What's stopping us from implementing holo style 3d displays like in the movies? Star-Trek etc. Are we not there yet?? alternatively, what about VR? we could virtually project the traffic controllers out into space like silver-surfer or ironman right(pov of course)? I'm not an expert, but it seems we need a better UX/UI right?
It's the fact that the FAA has $5.2B in outstanding repairs but only $1.7B allocated for repair.
On top of that, the GS pay scale penalizes federal employees in high CoL areas.
Both are very difficult problems to solve, as the former means dramatically increasing the FAA's budget (which is tiny for the scope of responsibility it has across North America), and the latter means completely reforming the General Schedule.
On top of that, Congress constantly meddles with the FAA and DoT in general because it's the easiest way to get some quick wins for constituents.
The FAA has been working on modernizing air traffic control, but that project won't be completed til 2030 at the earliest.
Furthermore, the Northeast is a uniquely congested airspace with the massive number of airports and passengers.
The budgetary questions are unrelated to UX/UI stagnation though.
It's annoying how many modern web sites change their entire design framework once every two years, yes. But ATC? Aeronautics in general? Most of maritime? Once it's certified, it's practically ossified - and for good reason. Bad UI/UX can literally kill [1].
Nevertheless, I think it's worth having the debate - and that led by actual air traffic controllers, please - if and if yes, how, UX/UI can be improved.
I don't believe the user's primary complaint is the UI of their devices. Why should they take time out of their overworked schedules to have a debate they aren't already having?
On the hierarchy of needs, it's a much lower priority than actually investing in solving maintenance related problems that are the primary cause of the Newark ATC related issues.
> its implementation of a “NextGen” air traffic control system to replace the current version may not be completed until 2034, even though the project was started in 2003.
Governments (and a lot of businesses) like to look at software as a one-time purchase, but it's really better too look at it as a liability and an ongoing cost. It'd be better to have a team make continuous, incremental improvements to the system than have "NextGen" last-gen replacement vaporware.
+1, the "strangler" pattern is the best (and often only) way to replace/update systems.
0) no new logic in the old system (stop the bleeding)
1) incrementally wrap old system interfaces with new system
2) proxy to old system and dark launch new system
3) monitor / confirm correctness
4) cut-over
I can say that at least one extremely large US company successfully retired their transaction processing mainframe system using exactly this pattern.
It wasn't simple, but running old+new in parallel on the same production data stream (then verifying behavior against each other) is a surprisingly powerful approach.
Most critically, because it trades temporary maintenance and infra costs in exchange for decreasing risk and flexibility around cutover date.
Kind of like black box TDD, which helps the new system team be much more aggressive with their delivery timelines.
There are examples of both. Firmware and embedded software usually is a one time purchase. Tax software requires ongoing updates. FAA software would fall somewhere in between.
Lots of firmware is upgradable these days. Some of of it even should be updated regularly, or you end up with weird novelty problems like your bluetooth earbuds being vulnerable to drive-by attacks.
It really is a binary though. Either you need to think about and deal with deploying (system-wide) updates, or you don't.
Even infrequent ~5y update lifecycles tend to be extremely painful unless there is substantial investment in treating it as an essential business process. This leads to a "kick the can" mentality that translates to show-stopping amounts of tech debt.
Five years is almost the worst case. In five years most of the people who did it last time are either in a different post or have left, meanwhile related systems have probably changed, so if they did leave documentation and if you can find it still, that may no longer be the correct process.
When Let's Encrypt was created it could have issued 3 year certificates. But it didn't because they knew that's a bad idea, the whole point of Let's Encrypt was automation, and if the certificates expire in 3 years you will say "Eh, we'll automate later" and never do the work.
5 years is far from the worst case. 50 years is a better upper bound.
Fifty years is often so different that it counts as a novel problem each time, so there's little value in knowing how you did it last time.
They like to give contracts to connected outsiders so that they can take care of their friends and take the blame away. 30 years? quoting the venerable Sir Humphrey Appleby:
"Precisely. Months of fruitful work. Leading to a mature and responsible conclusion."
Since it takes 30 years it must be a responsible piece of software, polished to the bone.
Software is a liability, a product is an asset.
This is a classic system at 99.9% capacity - there’s no slack to take up issues or do anything but run at, well, 99.9%.
And without something like a major disaster, it’ll likely continue to get worse and worse.
i flew delta out of ewr last week. the pilot announced over speaker that air traffic control was serializing take offs on one runway to avoid problems. we sat on the tarmac for two hours waiting our turn. that was on top of another hour delay for our flight arriving due to this policy.
delta doesn’t reimburse for missed connections claiming air traffic control policies are outside of their control.
it reminds me of a dev team i worked with once which used single threaded memcache as a way to serialize inbound requests to a server with improper locking logic inside.
> delta doesn’t reimburse for missed connections claiming air traffic control policies are outside of their control.
AFAIK if it's one booking on the same airline or a codeshare they are required to rebook you. If you planned a "connection" which is two single flights with different airlines you don't get any legal protections. This isn't just Delta, no airline will reimburse you for missing a flight you didn't book through them
they did book me on a flight the next day. i was scheduled to fly two legs on delta. but see my other replies below on lack of accommodation and rep behavior.
If the US had stronger regulation of airlines they would require compensation in these circumstances. In the EU you would be covered.
Flying in the US is a hellscape now thanks to deregulation and union-busting.
perhaps. what’s more my ticket to cross from east coast to western state cost more than my previous ticket to cross the atlantic into the country. maybe that’s a rational market outcome but it’s hard to fathom sometimes.
What do you mean delta doesn’t reimburse connections? I believe they would still be responsible for getting you to your final destination
they still booked me on next day flight but refused any compensation for hotel or meals or ground transport for the overnight stay due to the missed connection.
>delta doesn’t reimburse for missed connections claiming air traffic control policies are outside of their control.
Reimburse? They'll put you on another connecting flight of your choice as any airline does. I did this on a Delta flight with a missed connection due to delay just last week. They also automatically upgraded me to Comfort+ at no extra charge for the inconvenience.
no upgrades for me; not even an apology. imo the ground team leader was practically gleeful at not having to provide any compensation for the delay.
Sounds like it's time to drop Delta
I'm in no way qualified to comment on the details of the issue outlined here (though I did get stuck babysitting a friend's kids for many extra hours due to a 6hr delay on said friend's flight into EWR a few weeks ago). For anyone who lives far enough north and west of EWR though, I highly recommend trying either the Allentown or Scranton/Wilkes-Barre airports. I've moved my business travel to Scranton 100% and have been loving it. There are more restrictions on flight times, and basically everything needs a connection, but it more than makes up for any time I would have lost due to sinkhole traffic on I-80 and the EWR parking shuttles. It's amazing to park and be at the security line basically 3 minutes later.
Flew out of ABE most of my life, still an excellent airport, but it’s best to be aware that any connections through EWR or PHL will be by bus.
https://archive.is/9tLpI
> To save money, the FAA elected not to build a new STARS server in Philadelphia to support the move. A new server alone would require tens of millions of dollars, as well as installation of new internet and power infrastructure.
> Instead, it elected to send a “mirror feed” of telemetry from the STARS servers at N90, traveling over 130 miles of commercial copper telecom lines, with fiber optics to follow by 2030.
> The annoyances of traditional cable internet — frequent lag, dropped sessions — are probably familiar to those who stream video or play games online. But for air traffic controllers, even the smallest service disruptions can become dangerous.
So LOL what, they just ... piped it over the Internet? Also can someone make sense of this "new server" costing millions of dollars? Presumably it's not the cost of a server, which is orders and orders of magnitude less than that?
I doubt that 130 miles of copper is the public Internet. It’s presumably some legacy telecom system that depends on a bunch of generally fairly well made but thoroughly obsolete hardware.
Keep in mind that Ethernet over copper is only specified to ~100 meters. Long distance copper networks have been obsolete for a few decades.
Here in my EU state, air traffic control still used leased copper twisted pair for server link over ~150km as of at least 2015 or so. It's pretty reliable and not "public internet" at all - just two modems with known performance and known latency taking to each other.
Might be something like that in this case as well.
Sure, this kind of technology works fine. Except that it might be difficult to find people who how maintain or repair the fancy (pressurized? oil-filled?) copper cables or the modems, and finding parts might be interesting.
I did some enterprise network work ~20 years ago, including fiddling with some inter-building links, rummaging through closets, and visiting the inside of a legacy campus-scale analog phone exchange, and I’ve never even seen the kind of equipment that can send data at an appreciable speed (even by 1990s standards, and even with repeaters) over copper at a range like 100km.
In contrast, single-mode fiber has improved over time, but it’s not obsolete, and it has maintained a remarkable degree of compatibility over the years. New transceivers largely work on old fiber, old transceivers work on new fiber, etc.
Those are all good points. I do wonder if - as you mention - it might just be leased fiber.
I mean its a Verge article.
They dont know the difference between copper and fibre.
And yea fibre was in the 80s too. No reason for new deployments to be copper.
It’s called a dry loop. It might be real with repeaters at the various COs, or it might be partially virtualized and sent over a fiber trunk. Usually somebody leading a dry loop does so for the relatively extreme reliability, even though the speed is not very high.
Why not pipe it over the internet? A few redundant ISPs at the endpoints should be as reliable as a private run of fiber or T1 lines or whatever bespoke solution they think they need?
Doing so sticks billions of transistors between A and B, all of which have to work perfectly billions of times a second. If you just run fiber, you need to put an amplifier in every 50km or so, other than that it’s all just buried glass.
The fiber is already there, and it gives you guaranteed capacity, even during a major power outage. And it guarantees no one is changing out a switch or a router at a bad time, etc.
It’s just a lot less to control and to go wrong.
As soon as it's on the general Internet, any ill-meaning enemy can just force the VPN endpoints offline by blasting them with junk traffic.
That can be mitigated effectively in this case using an IP whitelist. All unrecognized traffic can simply be dropped.
...Until they switch to a bandwidth exhaustion attack.
True, saturating all routes to a host is quite feasible, but with sufficient redundancy (say with multiple PTP wireless links, which can be done quite cheaply) you'd have to knock the Internet out in such a huge area we'd be talking about WW3.
There's no argument that a private line is ideal for critical infrastructure, but if they must make do, there are ways to make it work.
People managed to shoot Wikipedia off the Internet as recently as 2019.
Anyone not behind one of the major CDNs can be targeted and taken down. And bandwidth-exhausting a few routers in front of the target is certainly possible, it's just not profitable for cybercriminals - but for a nation-state? No problem at all to muster a few terabit/s of capacity for them.
Nah, probably on leased lines like t1s/t3s. Airport telecom infra isnt always the best.
When they say "server" it means the (cheap) hardware and the (very expensive) software to drive these complex and critical systems.
If it's mainframe, and there is no reason to believe it's not given the strict requirements, the hardware is expensive as fuck and the people who can keep these beasts alive are just as expensive.
I think they are running on old Sun Ultra workstations, not mainframes. Hardware easily replaced and upgraded with much faster, cheaper options. Difficulty is much more on the software side.
> I think they are running on old Sun Ultra workstations
It looks like it, though in this brochure there is a bunch of what look like Sun rackmount servers in the background as well:
https://web.archive.org/web/20120930072126/http://www.raythe...
Very interesting. Looks like the LITE STAR runs on the Workstation and the full STAR deployment runs on the servers.
But can you even buy these now? A new STAR server setup must be x86 and virtualized,no? Maybe even cloud?
Agree, I didn't mean 10k servers, I mean 100k servers (eg); and I also meant $10M persons. Trying to (poorly) state a two orders spread.
Aerospace is hard.
An ATCS like STARS needs to feed from multiple different OT and IT sources like radars, weather stations, other TRACONs, etc and is implemented in it's own airgapped environment.
It can get very pricy very quick. On top of that, the FAA's budget has been sclerotic for decades now after the 1980s era union action and the 1990s era national cost cutting.
And finally, it is a political organization, and NATCA is a fairly prominent union within the AFL-CIO, and could make the lives of NJ representatives hell for pushing reassignment out of Newark.
> Aerospace is hard.
I would believe it was hard. And maybe it still is if you’re unwilling or unable to take advantage of modern technology.
Current low-cost equipment can easily send 10 or 100Gbps over long distance fiber links. Depending on how quickly you want to fail over when a link or an entire switch, router or rack fails, there are plenty of options that make various tradeoffs between failover latency and bandwidth, all the way up to completely duplicating all the traffic on redundant routes. I would bet that the entire aggregate traffic needed for air traffic control in a region is well under 10Gbps. And 10Gbps dedicated links or leases or (effective) purchases of dark fiber are not expensive on the scale of the FAA. Air traffic should use a network with a lot of redundancy, so maybe multiple those low costs by something like 5.
Heh, have fun hooking legacy systems to high speed networks without significant testing.
If seen plenty of old stuff crash because you'll have some ancient serial device with a limited buffer and someone jams a faster link in-between. All of a sudden you have a much larger amount of bandwidth delay product and the system doesn't handle a few megabytes of data getting lost on the line when it bursts for some unexpected reason. On the old fixed line that just couldn't happen.
A "Server" is just layman's speak for that big room with all the noise and blinky lights. Including all the hardware, networking and software that goes with it. Of course you're not going to run critical infra on a single server :)
Though I have to admit I have seen government operations where the "server" was an old dell optiplex desktop lying on its side in a broom closet without ventilation, a post-it with "IT SERVER DON'T TURN OFF", a spiderweb of cables running through the closet and the "server" fans screaming for air trying to keep everything cool in the enclosed space. I'm not kidding.
I mean, I know, government. Small local welfare-related org. Shoestring budget. Sure, that sucks. But at least you can make sure it's tidy and the cabling doesn't look like shit. Jeez. I didn't imagine I'd still see that in this century. Do people no longer take pride in their job? They hadn't even activated the "AC Power on" in the BIOS so after electrical maintenance they had to wait for the "engineer" to press the on button again.
There's no budget to fix whatever you messed up when redoing the cabling, which we also didn't really have the budget for. Everyone who knows what the box actually does has retired or died, so nobody wants you in the broom closet shifting dust motes near a working system we can't replace.
What are leased lines?
Old tech. You’d literally lease a line from ATT or whatever Bell company covers your area. A reserved private connection. Modem at each end. They ran at about 56Kb/s IIRC.
> pp. lag
> hold all traffic
> rdy?
> gogogo
Need more trains.
It's amazing, really. You hear about "government overspending" all the time. You actually look into something in any detail and what you discover is a consistent pattern of underspending. Call it mismanagement if you want, but it is consistently what we, as votes, and the executive, ask them to do.
Meanwhile, back at the ranch, I was a contractor for HHS for a year or so. When I started, there was a server farm built for a contract that was powered up and maintained; never used, it was broken up and sold for surplus. Just before I left, HHS thought they would like that program again, so us US taxpayers bought another tranche of computers to power up. I don't know the end of the story.
I could type more, but it would be a long and boring story.
There is overspending. It’s a scam.
There is mismanagement.
There is also a misallocation and underfunding of essential services and infrastructure. This is the excuse for ever more funding.
The general pattern I see is that operating spending is too high while capital spending is too low.
So true. The state of Washington has a great example: the outgoing Secretary of Transportation was very clear that if the state doesn’t change the budget to be primarily about maintenance, and much less about new highways or lane expansion, then our infrastructure will quickly begin crumbling.
Even the people in charge of our highways want us to switch to operations & maintenance oriented projects, but the representatives have not done so.
The incentives in government are really fucked — you get visibility and wins through cool projects, not by keeping the lights on and things running smoothly. Honestly true in big companies as well.
Here in Texas the new hotness is just building more and more toll roads and express lanes. Which are not toll roads, but even more expensive somehow.
It's to a point some major parts of DFW like Plano and Frisco are only accessible via tollway. That's the only highway - the other "highway" has a light every half mile. Meanwhile, we have some of the most dangerous interstates and highways in the country.
Where's all that toll money going? I have no idea, but maybe it'll pan out. It just sucks because it's already quite expensive to commute in DFW.
Newark’s overworked controllers might argue differently. Of course in this forum the general suggestion will be to replace tired controllers with sleepless machines, and the technologists here have strong incentives to advocate for such solutions.
My understanding is that ATC controllers would be far less overworked if they had modern (and properly functioning!) equipment.
(Shrug) ATC is no job for humans, and I'm tired of pretending it is.
If we were building our aviation infrastructure from scratch starting today, you would get some really strange looks if you suggested employing humans to manage air traffic.
En-route ATC is already mostly automated, with humans supervising the system and talking to the pilots.
Arrival/departure/ground ATC has to deal with much more complex traffic, emergency situations and edge cases in general. Technologically, we're nowhere near fully automating this.
When's the last time someone in a control tower was actually surprised by anything genuinely new? (Anything that didn't involve ancient equipment breaking down in a new and unusual way, that is.)
ATC is a solved problem. Nothing is going to happen that hasn't happened before, or otherwise can't be anticipated. There's nothing about ATC that inherently demands human involvement in real time.
> Nothing is going to happen that hasn't happened before, or otherwise can't be anticipated.
By the way, this mentality is at odds with safety. If you design a safety-critical system under the assumption that nothing unexpected will ever happen, then whenever it happens (and it's a question of when, not if) people will die.
The first and most important reaction to anything unexpected is simple: localize the trouble spot and reroute traffic away from it.
What can't be handled by doing that? You've listed quite a few examples (and thanks for taking the time to do so!) but all but one of them seem like perfectly reasonable scenarios for automation.
Another point I'd raise is that most ATC screwups don't involve anything weird happening except failures to follow existing ATC procedures. Any list of Things That Make For Bad Days at the Airport needs to include that.
Happy you asked! Here are a few interesting cases that the respective ATCs probably hadn't seen before or will ever see again:
- Plane loses radio communication on final approach and ignores go-around orders, lands on occupied runway as ATC gets others out of the way [1]
- Guy steals a plane and takes off without permission [2]
- Someone left some cones on a runway at JFK [3]
- Door blows off aircraft [4]
- Some dude runs across the runways [5]
- Pregnant woman giving birth on flight [6]
- Student pilot freaks out, ATC calms her down and gets her instructor on frequency [7]
- Earthquake [8]
- Pilot hits a deer [9]
- Bomb threat on board, pilots decide to evacuate on the tarmac after no help arrives in ~1h [10]
Some situations have probably happened before, somewhere. Some others are completely new. I would highly recommend that you watch these videos (they're all relatively short) and genuinely ask yourself whether our current state-of-the-art AI models would be able to successfully handle these situations in the short timeframe required to do so. Let alone the fact that by AI we mean text models, so I'm not sure how they would integrate with terrain information, real-time radar data, arrival/departure routes, etc.
I think you have a fundamental misunderstanding of what control towers do. They're not there to simply observe planes do their thing and intervene if they get too close. They actively handle the traffic, and this task requires human pattern recognition and cross-domain reasoning skills in a matter of seconds, and the technology to replace this is simply not there yet. If you still disagree, I'd love to learn which technologies you'd apply to this problem and how they would compose with each other in order to achieve the same outcomes as in the cases I linked.
Of course, this is not to say that ATC shouldn't be made safer and more automated wherever possible. Particularly in the US, where equipment is severely outdated and some dubious regulations allow their ATCs to handle runways and give clearances in a way that would not be allowed in Europe and have already resulted in more than a few close calls. These are all valid concerns, but IMO they can't be extrapolated to "all ATC services can and should be automated".
I fly often, and I for one feel safer (at least, with our current technology) knowing that there are humans in the cockpit and in the control tower who can react and take control when unexpected stuff happens.
[1] https://www.youtube.com/watch?v=cXNWwKx9c1o
[2] https://www.youtube.com/watch?v=4LLmF9tZoEE
[3] https://www.youtube.com/watch?v=YmywjMQDbos
[4] https://www.youtube.com/watch?v=0ma0JzO43Ig
[5] https://www.youtube.com/watch?v=VZviKoEKAaw
[6] https://www.youtube.com/watch?v=5pZ3VOPlarw
[7] https://www.youtube.com/watch?v=dgc2Wh4cOgo
[8] https://www.youtube.com/watch?v=o80cNJ_XhX0
[9] https://www.youtube.com/watch?v=x04kRUIgXpQ
[10] https://www.youtube.com/watch?v=aAW2JbqxLRM
Plane loses radio communication on final approach and ignores go-around orders, lands on occupied runway as ATC gets others out of the way [1]
Trivially handled by video game-grade AI. Is an aircraft somewhere it's not supposed to be, or doing something it's not supposed to be doing (most likely because a human has screwed up somewhere?) Route everything else out of the area, signal authorities. Human intervention not required.
Guy steals a plane and takes off without permission [2]
See above. What can ATC possibly do about this, besides alert other aircraft to the situation and signal law enforcement?
- Someone left some cones on a runway at JFK [3]
Foreign object detected or reported on runway, an everyday occurrence. Alert other aircraft and facilities personnel. What else is the controller going to do, go downstairs and pick up the cone?
Door blows off aircraft [4]
What is the controller supposed to do about this, other than accept the crew's request for an emergency landing and (guess what) route everybody else out of their way?
Some dude runs across the runways [5]
Like other 'surprises', two-legged FOD might have surprised the Wright brothers, but that's about it. Alert law enforcement and warn aircraft.
Pregnant woman giving birth on flight [6]
Just another routine 'surprise' covered by standard procedure. Handled primarily by the air crew rather than ATC. Handle emergency landing permission if requested by crew, alert medical personnel to meet the aircraft when it lands, done deal.
Student pilot freaks out, ATC calms her down and gets her instructor on frequency [7]
Honestly not sure what should be done to automate this kind of situation. I suppose an LLM could handle it as well as anyone else. :-P
See also the edge cases where the pilot(s) are incapacitated and an untrained civilian needs to be talked through the landing. How often does that happen in real life (and how often does it actually work?)
Earthquake [8]
Not really something ATC would be involved with except at the purely-tactical level. Ground outgoing traffic and reroute incoming flights until all-clear given by officials.
Pilot hits a deer [9]
Reroute traffic and alert medics.
Bomb threat on board, pilots decide to evacuate on the tarmac after no help arrives in ~1h [10]
Not an ATC issue except for the need to (guess what) reroute traffic.
They actively handle the traffic, and this task requires human pattern recognition and cross-domain reasoning skills in a matter of seconds, and the technology to replace this is simply not there yet. If you still disagree, I'd love to learn which technologies you'd apply to this problem and how they would compose with each other in order to achieve the same outcomes as in the cases I linked.
You could still have a valid overall point regarding the need for humans in the real-time loop, but I disagree that most of the scenarios you mention support that point. All of those cases can be (and are) anticipated, all of them have happened before except possibly [10], and all of them can be handled by computers at least as well as humans. Except possibly the situation with the pilot needing real-time psych support.
Let alone the fact that by AI we mean text models, so I'm not sure how they would integrate with terrain information, real-time radar data, arrival/departure routes, etc.
I definitely don't mean LLM-style text models. As I suggested in the first answer, nothing on this list except (again) #7 would have flummoxed a game programmer ten or twenty years ago. If some people want to try an ELIZA-like LLM to deal with #7, fine, but that would be a research problem. Nothing else on your list requires any new research.
You have listed the correct actions to take in each scenario, but it's still not at all obvious to me how each case would be fully automated, start to finish. I would like more specifics, besides simply
Since I took the time to compile a list of examples that you sadly didn't find surprising enough, I would appreciate it if you returned the courtesy and provided a more concrete design of a fully automated ATC system using current technology.> Trivially handled by video game-grade AI.
I think it's fairly likely I have played more flight simulator games than you have. If you know of a single one of them where the ATC AI isn't utterly stupid, please do let me know, I'd love to try it. There's a reason VATSIM exists and remains very popular :)
First of all, to the extent controllers are "surprised" by any of those events, you're making my point for me.
Second, I wasn't referring to ATC AI in flight sims; as you suggest I have no experience with that. I doubt anyone ever put any serious effort into flight sim ATC AI compared with AI for realtime strategy or even FPS games. Years ago, a primitive neural net dragged a 9-dan Go master up and down the ban, so I imagine our current tech can handle a few planes.
I'm not paid to redesign the ATC system, or qualified to do so (although that won't stop Musk, I'm sure.) But as I've made clear, I believe it can and probably should be done. In practice it would probably look more like
So many of the things you mention are simply not in ATC's wheelhouse to begin with. Their job is to keep things moving (or not) while other people in authority deal with those situations. That part wouldn't change, as I see it.Thanks for replying. I won't continue this conversation as it seems to me you don't actually have any specific evidences of your claim besides "I believe it can be done", and don't actually know much about the responsibilities and areas of authority of ATC.
I'd say the burden of proof belongs with those who argue that it's impossible to automate ATC while maintaining at least the current safety record. That's an extraordinary claim at this point.
I asked for examples of situations that couldn't be handled through automation, you provided some that I consider invalid or inapplicable, and... well, there we are. GG
[flagged]
It's very easy to make a flippant comment that disagrees with another, with no supporting evidence, but demands multiple examples from any future replies. A reply doing that could take a huge time investment to do properly (just look at agubelu's incredible reply to you). Sometimes the only feasible solution to that imbalance is a down vote.
It’s like I summoned you! Just be honest about your incentives if you care to make these arguments. Then be prepared to answer the accountability question, for when the system inevitably fails.
My "incentive" is that I fly somewhere every once in a while, as do people that I care about, and I want the system to be as safe, reliable, efficient, resilient, and cost-effective as possible.
Don't you?
> Then be prepared to answer the accountability question, for when the system inevitably fails.
Airplanes have gotten increasingly automated. Who is responsible when Airbus' excellent automations that have prevented countless upsets and accidents fail? Nobody, if it was an honest mistake, and lessons learned are applied to improve even further.
The problem with modern ATC is that a lot of the safety systems are bolted and backported on top of existing extremely legacy tech. Ffs, the communications still happen over radio where transmissions are missed if more than one person talks at the same time. And people have died because of this, as well as controllers making a mistake or pilots and controllers misunderstanding each other.
There's no reason to continue bolting more stuff on top. A very large part of ATC can be fully automated and made safer.
All true, except that a key reason they still use AM for voice communications is precisely because it's obvious if multiple users are trying to transmit at once.
AM is obviously not the way a "CSMA/CD" system would be designed today, but it does get the job done, and has for a long time.
AM is good for instructions because its broadcast nature allows everyone to hear everyone, but arguably coordinates-over-voice-over-AM is dangerously error-prone. My sense is that the ideal system might look something like "AM with a digital side band" where ATC can press buttons to transmit data (e.g. authorized altitudes and vectors) to a plane but metadata is still carried over AM.
A lot of fatalities have been avoided thanks to a pilot overhearing ATC giving takeoff or landing clearance to another plane, but quite a few incidents could also have been avoided if plane cockpits had a big red light "authorized to enter runway" which could only be turned on by ATC. In an industry which is designed around so many redundant systems, it's rather astonishing that an error in a single communication channel can lead to disaster.
The US overspends on military and underspends on most other things. Not sure when this will ever change. Only when it has to I guess.
In a pretty remarkable statement a month or so ago Trump actually said we were overspending on the military and suggested cuts. Then his budget comes out and proposes military expansion and Medicare cuts so like usual he was just bullshitting I guess.
Exactly. The military-industrial complex is a very real thing, unfortunately.
It's like tech debt. It's an ongoing cost in a one and done environment, it's hard to see problems from the outside until there's catastrophic failure, and if there's a slow niggling annoyance of things getting worse over time the point where people notice enough to care is usually past the point of needing a refactor. So we get underspending where it matters, overspending where it doesn't, and the solution is always a redo.
Came across this post, which perfectly encapsulates a sort of pathological inversion of sunk-cost fallacy, where some people I encounter would happily let infrastructure rot and countless people suffer if it saves them pennies on their taxes, and are incapable of re-evaluating their position because “taxes bad” has somehow become such an ingrained/religious value that anything else is unconsciously and immediately rejected no matter the consequences.
https://old.reddit.com/r/goodnews/comments/1kuaasx/i_voted_f...
>>I posted this somewhere else ages ago, feel it's relevant
>I remember having a conversation with my ex's sister and their mum a few years ago, around election time. I try to not talk politics with people because it's a fast way to lose friends, but the topic came up between my ex and them over dinner and I just listened in.
>I remember them saying that the only thing they're interested in is tax cuts. More money for them. I had to chime in and ask what about the NHS, what about funding schools? They said they didn't care because they had private healthcare through their jobs (finance), so they don't need the NHS. The mum said her kids are through school so she doesn't care about funding schools, and the sister said she'll be sending her kids to private school one day. I was pretty gobsmacked at the brazen selfishness of it, and asked what if they lose their jobs - and therefore their private health care - or become unable to work, what if when you have kids you can't afford private school? Neither of them could grasp this hypothetical... it was as if I was speaking another language to them. They were just like 'but we do have jobs.' And what if you didn't? 'But we do.' It was just circular and they couldn't see themselves in any situation other than the one they were currently in.
>I can quite see why empathy is a hard concept for right wingers to grasp, it was like they just simply couldn't understand the concept. They weren't stupid either, and nor were they rich - the mum worked in admin for a finance company in the city, and the sister was being paid by the same company (mum got her in the door) to train as an accountant.
>I think about those two every now and then when I can't understand how the other side thinks. Because it seems we do literally think very differently.
Have been many billions of dollars spent over the years to "modernize" air traffic control. If my recollection is correct, most of it ended up being wasted.
The current administration is asking for a lot of money to try to fix it again
Most people see a number $6 Trillion and something breaks inside them.
What they don't realize is maintaining infrastructure is expensive.
Sure there are a lot of inefficiencies with need to be fixed, but you can't throw the baby out with the bathwater.
Sadly, such is society, and this is a problem that happens everywhere - be they democracies or authoritarian states.
"Most people see a number $6 Trillion and something breaks inside them. What they don't realize is maintaining infrastructure is expensive."
It breaks all our brains too, because that $6 Trillion has very little to do with maintaining infrastructure. The bulk of it is just direct payouts (Social Security, Medicare and defense contracts).
Pretty frustrating that this big number means the government is politically forced to do drastic austerity for things like keeping planes flying safely.
Also, makes DOGE starting at USAID (<1% of budget) look especially incompetent.
When people think wastage, they think bureaucrats wasting away in offices. Social security admin, air traffic control, and teachers are more like Amazon warehouses at this time in terms of utilization. Less than 1% of social security admin funding is overhead at this time. Any charity would be exhilarated to hit numbers like that.
With 3 million employees and about one cent a flush, the US government spends 100,000 or so a day just to flush. Assuming three bathroom breaks a working day.
>With 3 million employees
that is, for context, about as high as it was in 1960 when the US population was half as large as it is today.
https://datawrapper.dwcdn.net/HQ8pa/full.png
Does that include military?
It doesn't, but if it did the decline wouldn't just be relative but absolute because active military personnel is actually down almost 50% since its cold war peak. The US military is significantly leaner nowadays across the board. In terms of spending too, something like 8% of GDP at its peak compared to 3.x now.
https://usafacts.org/articles/is-military-enlistment-down/
Close down the toilets. Poof, problem gone. Think about applying to DOGE. They may like your ideas.
Poop-from-home! Serious expense offloading.
[dead]
> you discover is a consistent pattern of underspending
No, what you discover is a pattern of wasted spending. Then they ask for more money to actually get the job done despite all the waste.
People try to solve this by privatizing certain things, figuring that competition will help efficiency. Sometimes it works, sometimes it doesn't.
Maybe we need competing governments, and whichever government is more efficient gets to rule. Seriously: Add a second FAA at some test airports, see if they can do better, with the understanding that if they can't, they get shut down.
> Maybe we need competing governments, and whichever government is more efficient gets to rule. Seriously: Add a second FAA at some test airports, see if they can do better, with the understanding that if they can't, they get shut down.
And you would be willing to be personally responsible if people die in this experiment?
It's funny how people here always complain that any money government spends is wasted, but if you look at big companies they are "wasting" money as well. Just look at the number of projects that google killed. It's simply a function of large (and small) organizations that they don't get it right all the time, it's difficult to predict the future.
>It's funny how people here always complain that any money government spends is wasted, but if you look at big companies they are "wasting" money as well. Just look at the number of projects that google killed. It's simply a function of large (and small) organizations that they don't get it right all the time, it's difficult to predict the future.
I worked at a Fortune 50 financial services company back in the 1990s and they designed, built and deployed a brand, spanking new customer service platform.
They spent USD$200,000,000.00 building the system. It was ready for production when someone realized they'd spend USD$50,000,000+ per annum supporting the platform. The product roll out was scrapped, and tens of millions of dollars of equipment sat in warehouses.
When trying to redeploy said equipment across the enterprise, group heads would purchase new equipment rather than using the equipment sitting in warehouses as it was "cheaper" on their budget lines to spend real money rather than put the depreciation of already purchased equipment on those budget lines.
So yeah, big companies can be incredibly wasteful. Often more wasteful than government (e.g., US Medicare has ~3% overhead -- show me any private insurer that can top that) as well.
It's all about the incentives. People (and consequently, organizations) respond to the incentives inherent in a situation.
Sometimes those incentives promote efficient outcomes and sometimes not so much. The trick is to maximize the former and minimize the latter. Something easier said than done.
Don't generalize. There was 0 underspending at the Department of Education or USAID.
I don't understand this sentence:
> Instead, it elected to send a “mirror feed” of telemetry from the STARS servers at N90, traveling over 130 miles of commercial copper telecom lines, with fiber optics to follow by 2030.
This does not make any sense. If they would really transmit data over a 130 miles copper line (which I doubt even still exist, especially not commercial ones), we would be talking rates in the low Mbit/s. I suspect the situation is that the "last mile" of the center is served by copper connections, not good either but by far not as bad as a 130 miles copper connection.
EDIT: I should add if they really would have a link running on copper lines it would have repeaters, which would be sitting in datacenters. In New Jersey there would by 1000s of km of dark fiber floating around, so it would be trivial to convert at least the majority of the link to fiber.
Telephone providers lease "dry loop" lines for point-to-point signalling.
I've used them for telemetry systems with acoustic modems on both ends.
Also for sending audio between broadcast studios. I recall that it was priced by bandwidth (in the literal, analog, sense), e.g.: 5kHz (~AM radio) was less expensive than 15kHz (~FM radio). For comparison: A normal phone call is 3kHz.
So yes, copper and repeaters. But very inexpensive and quick to provision. :)
Not sure I understand your point, using a 15 kHz carrier gives you nowhere near enough bandwidth to transmit even just 100 Mb/s.
I am guessing that the data feed requirements of ATC telemetry are fairly modest, well within the range that a pair of analog modems could handle. You don't have to send a raw radar feed to the airport, just put the intelligence (object detection, delta computation, etc) at the endpoints and transmit the results.
I'm also guessing that they do this every day at hundreds of other airports, and it works just fine.
Moving to fiber gives you greater bandwidth, reliability, multipath redundancy, etc. But I bet the real motivation is access to newer and more general communications equipment. Maintaining the old, perfectly adequate but increasingly unusual hardware/software is more of a pain than replacing with new stuff that exceeds requirements but is maintained just like all the other modern stuff everywhere.
All guesses. I do not work in ATC, but I've seen the pattern in other industries with similarish requirements.
It would have been easier and cheaper to raise the salaries of the Long Island employees.
Air travel is far too subsidized by the public and I can think of few worse applications of public funds save maybe sports stadiums. It's a huge waste of energy, pollution, human labor...
Rail is incredibly efficient, and there's a reason China has been building high speed rail as fast as it can.
To all the "it would never work here" people: we used to be a nation of rail travel, where you could walk or bike or take a taxi to the local trolley/train/bus station, take a train to where you needed to go.
All that was systematically ripped apart by the auto industry either directly or indirectly. There is no reason whatsoever we can't work our way back, especially given how much faster and easier construction of a railway line is now.
I'd LOVE some proper high speed rail across the US but EWR is also huge for international flights. A good rail network would at least help getting to/from their for it but that still leaves a good amount of flights that make no sense to dump.
> Air travel is far too subsidized by the public
Don't most of FAA's funds come from taxes on air travel? And around half of Americans travel by air every year, so it's not a niche service.
> I can think of few worse applications of public funds save maybe sports stadiums.
Californians can.
Subsidies to the movie industry, plumbing run to waterless urinals, bullet trains between farm towns, ...
> bullet trains between farm towns,
This is disingenuous and you know it.
I think it was the wrong choice for a number of reasons, but the farm towns in question are just first lot of the whole network, starting with the (supposedly) easiest part. Instead of building in the densest parts which would be even more complicated and expensive.
> Rail is incredibly efficient
As a European, I don't understand the fetishization of railway. These are two different means of transportation with two different use cases, with tiny overlap between them.
> Rail is incredibly efficient
If you have the density to justify it.
There is a case to be made for enhancing rail transit in the eastern seaboard and maybe parts of the Midwest, but America is too large and sparse to justify rail transit at scale.
It makes more sense to concentrate on rail infra for freight transit and work on revamping our existing rail freight infra.
> there's a reason China has been building high speed rail as fast as it can
China stopped subsidizing HSR during the COVID recession. It costs the exact same as a flight ticket now [0] due to high debt [1] (excluding the Beijing-Shanghai track, which actually can justify usage).
Most Chinese use normal rail for intercity transit, but this is easier to justify given the density and ease of land acquisition.
But even then, China began slowing down railway investment and construction since 2018 [2][3], and started calibrating towards air transit [4] as part of a commercial aviation push [5]
[0] - https://www.nytimes.com/2024/05/13/business/china-bullet-tra...
[1] - https://www.caixinglobal.com/2019-01-29/zhao-jian-whats-not-...
[2] - https://www.caixinglobal.com/2018-01-02/china-railway-corp-s...
[3] - https://www.caixinglobal.com/2021-03-30/china-looks-to-slow-...
[4] - https://www.caixinglobal.com/2021-03-25/smaller-cities-reach...
[5] - https://www.reuters.com/business/aerospace-defense/comac-jet...
Going from Qingdao to Beijing is much cheaper on train vs airplane. Takes roughly the same time as well.
Also, China does both new airports and new rail lines.
Aviation in China is uncompetitive mostly because a ton of Chinese airspace (more than half) is closed to civilian traffic.
I love trains, but the problem in the US is mostly that it's so damn hard for the government to acquire land for useful rail projects.
It turns out trying to run an airport like twitter is a bad idea, who knew
An interim solution would be to force carriers to upsize their jets to reduce total flights per day. Force any carrier who has wide-body aircraft to use them on domestic routes. heavily limit or ban the use of smaller commuter jets (replace them with 737s with fewer total flights). I'm sure they won't be happy about it, but it's a solution that could be implemented overnight and the larger carriers would have no problem adjusting. smaller ones could sell their slots or lease larger aircraft.
We had a 9/11's worth of death EVERY DAY for the first TWO YEARS of covid
Then a 9/11's worth of death EVERY WEEK for the next TWO YEARS of covid
We still have a 9/11's worth of death EVERY MONTH in 2025
Personally I don't want anyone to die in a plane crash
But apparently a hundred million other people do not care anymore about others dying needlessly
So factor that into your next flight if you are taking your life into your own hands?
What's stopping us from implementing holo style 3d displays like in the movies? Star-Trek etc. Are we not there yet?? alternatively, what about VR? we could virtually project the traffic controllers out into space like silver-surfer or ironman right(pov of course)? I'm not an expert, but it seems we need a better UX/UI right?
Are you aware that many things depicted in movies do not actually exist?
> I'm not an expert, but it seems we need a better UX/UI right?
This isn't where the problem is. It's system reliability, increased air traffic, and increased controller workload.
> but it seems we need a better UX/UI right?
Wrong.
It's the fact that the FAA has $5.2B in outstanding repairs but only $1.7B allocated for repair.
On top of that, the GS pay scale penalizes federal employees in high CoL areas.
Both are very difficult problems to solve, as the former means dramatically increasing the FAA's budget (which is tiny for the scope of responsibility it has across North America), and the latter means completely reforming the General Schedule.
On top of that, Congress constantly meddles with the FAA and DoT in general because it's the easiest way to get some quick wins for constituents.
The FAA has been working on modernizing air traffic control, but that project won't be completed til 2030 at the earliest.
Furthermore, the Northeast is a uniquely congested airspace with the massive number of airports and passengers.
The budgetary questions are unrelated to UX/UI stagnation though.
It's annoying how many modern web sites change their entire design framework once every two years, yes. But ATC? Aeronautics in general? Most of maritime? Once it's certified, it's practically ossified - and for good reason. Bad UI/UX can literally kill [1].
Nevertheless, I think it's worth having the debate - and that led by actual air traffic controllers, please - if and if yes, how, UX/UI can be improved.
[1] https://uxmovement.com/buttons/how-an-interface-mode-killed-...
I don't believe the user's primary complaint is the UI of their devices. Why should they take time out of their overworked schedules to have a debate they aren't already having?
On the hierarchy of needs, it's a much lower priority than actually investing in solving maintenance related problems that are the primary cause of the Newark ATC related issues.
Sure, sure, but this can and should be done in parallel.
>but it seems we need a better UX/UI right?
No, you sound like a middle-manager.
Talk to cashiers. They all want terminal DOS based systems where keyboard is king. It's the fastest once you learn it.