>just use public transpo-HOW AM I SUPPOSED TO GO TO WORK NOW REEEEEEEEEEEEEEEEEE
Sounds like germans need a Railway Labor Act to neuter their unions.
>>1693500cringe wagecuckery defender
>>1693501Nah.I'm a freight conductor.
>>1693440>planes, a 3-dimensional system, were pretty much fully automated for decades already, the studs in the cockpit are just for safety>cars, a 2-dimensional system, made leaps and bounds in terms of autonomous driving in the last 10 years>trains, a 1-dimensional system, still pretend that a driver is neededthose fuckers need a dose of reality, they are only kept around as a courtesy (and because for now it's cheaper to field them than to switch), the moment they try to bite the hand that feeds they shall be ruthlessly disposed of in favour of faster, safer, more efficient autonomous system
>not making sure to work exactly on strike dates and take holidays during regular operationsIt's like enjoy making money for Mr. Nosenberg.
>>1693440Your chariot awaits.
>>1693549True chads commute by space hopper.
>>1693440join in the strike
>>1693440>public sector unions
>>1693508The automated ore train in Australia only saves them the time of crew changes. It also took off on its own and had to be derailed to stop it.
>>1693440this is a perfect opportunity to get a new job driving a train
Get a car bro
Because trains are the best!(At hauling bulk loads. Passenger traffic is highly debatable).
>>1693440Just join the strike
>>1693440If you're not in a position to negotiate days off in these circumstances you need a better union.
>>1693440>three-day strikesince when are 48 hours three days
>>1693602The automated trains in the DLR in London have been running without drivers for 30 years.
>>1693730>start at noon on day 1>24 hours later it's noon on day 2>another 24 hours later it's noon on day 3Successfully ruined three days of commuting with only striking for 48 hours, German efficiency at its best!
They pull that shit every few years. They are some of the most privileged workers to be able to pull it off. Strikes for such basic infrastructure should be illegal and punishable by imprisonment.
>>1693440>have job>dont do itunions r fun
>nooooooo you can’t collectively demand better pay how are we supposed to make money if you refuse to work because you don’t make enough money even though you are apparently critical to the economy as shown by me being upset about this just work for cheap you fucking goyim
>>1693597DB is private company, the state is just the shareholder.
>>1693440The whole GDL (=Union of engine drivers) mess is too complicated to discuss in detail on an English language board and even in German it probably wouldn't work out.That said, on large stretches of the German railway network drivers already largely fulfill a supervision task rather than an operational one nowadays.This probably won't be extended for a while, though, because >Deutschland.>>1693751>They are some of the most privileged workersNot really. They were literally state officials on a comparable level to policemen until the 90s.The nation wanted to save money, so they cut that (turning them into regular employees) and the result is, that they now may go on strike.
>>1693500It's impressive they allowed a union in Germany anyways. Most companies there don't have one because neoliberalism is what gives companies (mostly through backdoors) freedom to do what they want: lol no union for you, more money for me!
>>1693602>It also took off on its own and had to be derailed to stop it>train, apply brakes>I'm sorry Dave, I'm afraid I can't do that
>>1695498>It's impressive they allowed a union in Germany anyways.At the moment there are plans to ban them (only ONE union per company should be allowed to call for a strike - the largest one).This strike is part of an effort to prevent that.The larger union at DB, EVG, has been tamed for the most part - their heads got board positions.
>>1693597well that towel was a whole lot of use
>>1693751>>1693500see >>1695326Train drivers were State employees like policemen State Employees cant be fired and get really good retirement. on the other side the cant strike and the state can say no if they want to resign.but our politicians tough it was a great idea to privatise the railroad. now all small railway lines were closed. and investments into infrastructure was minimised.
>>1695594>now all small railway lines were closedTo be fair: That mostly happened in the late years of Bundesbahn.Since the Regionalisierung of Nahverkehr smaller lines have been reopened using local public money rather than closed.I don't think the current state of DB AG is all too bad. It's in a position to reliably serve its purpose as the backbone of public transit for the foreseeable future.Its largest failure at the moment would probably logistics. Basically ending Einzelwagenverkehre on the rail was a massive mistake.With the """new""" possibilities through Digitalisierung that could have become a massive money printing machine, but if it takes decades to introduce a new automatic coupling mechanism on a European scale (and that still being a fucking mess), then that's probably a little too optimistic.
>>1695556>EVG has tamedi cannot imagine how cucked you need to be to pay them union duesthey literally cut pensions in half and cut wagestheir only purpose is to be bigger than the GDL and they are not good at their job
>>1693440solidarity strike ofc
>>1695594>>1695556>Well you're not really vital to the functioning of the state so we'll give you lower wages, lower job security, lower work benefits and make you compete with each other.>Also infratstructure is vital for the state so you shouldn't be allowed to go on strike. There really shouldn't be any competition so we want only one single union we control.It's always an universal truth huh?
>>1693440I went on holiday to France last year in January and every single trainline was on strike. The only train I got to ride was an automated one in Toulouse which was pretty cool.
>>1693440>blocks your commute
>>1693440the best transport strike I ever heard about was when tokyo bus drivers wanted to strike to protest conditions. They agreed it wouldn't work to not do their job as they'd stop transport in the city and they would look like the bad guys, so instead they did their jobs as normal, but refused to collect fares. Apparently it worked well for them. Too bad they can't do that in Germany I guess.
>>1697303That would be blatant theft, imagine being on strike as a warehouse employee and giving away the merchandise without collecting the payments.Bus, fuel, insurance, maintenance, etc, the transport isn't free to operate.It's basically giving away the inventory unless they're compensating it themselves.
>>1697314The transport companies only work as sub contractors, so they get payed by the Verkehrsverbund. It could work with bus drivers, but obviously not with trains. But most people have montly passes and they wouldn't loose enough.
>>1697259It's even more convoluted than that.The state went as far as lowering the standards for becoming an engine driver (originally limited to "Eisenbahner im Betriebsdienst"="railwayman in operational service", which is a proper qualification) in Germany.Refugees and long-term-unemployed take extremely slimmed down courses to cover for the lack of domestic engine drivers.Fortunately, technology has advanced far enough, that the safety risk only increases slightly through this.
>>1697449I mean it's clear theft and stealing if you're giving away the company assets without charging, so that is most likely not allowed as it would be ridiculous. (Aka the warehouse employee giving away the store inventory without collecting payment).Every kilometer they drive they incure costs of depreciation of the vehicle and they are burning fuel directly related driving, those are company assets they would be giving away 'for free', which is theft. Also complicated things like insurance. If I drive a company vehicle on company time, and I hit a pedestrian by accident, my employers insurance will pay for the damage. But if I take a company vehicle without permission and I start driving it around in my own private 'strike' time and I hit a pedestrian, is my employer still responsible? Even if the insurance of the company would pay out, which I doubt, it would still be a form of stealing financial assets from the company who is paying for the insurance.I don't think a german judge would allow this in any form because it's far too much trouble.
>>1697474Yeah, exact same in the Netherlands. They only need a few key supervisors who oversee the network digitally, the rest really can't fuck up being on the train they just need to be there for formality.The same is going on for everything they can save a few bucks and employ some 'minorites'. Same is now going on for the police force, trying to replace them with cheaper 'ordnungsbeambte/civil security guards' that require no education or training other than a written test with 2 weeks of studying. They want to give them weapons also so they can replace police on the streets fully. I believe the teachers will be up next, they are already playing that 'life experience' should be able to get a certificate too. 'ervaringsdeskundige'
>>1697478>teachers will be up next"Funny" story from Berlin: Over there they are so starved for teachers, they will take fucking anyone that thinks he will be eventually qualified for the job.Keyword: "Quereinsteiger"The thing is: Berlin as a federal state is in such a sorry state, that they don't fucking find anyone. Teachers in Berlin have to practically fear for their life in everyday classes.All their teachers are apparently fleeing to Brandenburg.
>>1697475It was more an act that made the company look bad than the bus drivers. The bus drivers are the good guys for showing up to work and doing their job, and the company is the bad guys for not capitulating to their demands. The legality is rather unimportant unless you consider your entire fleet workforce to be instantly expendable, which would also fail in Japan as it's a bad look for a company to fire an employee, and to fire everyone would be suicide as no one would want to work there. Maybe it's the sort of thing that would only work due to JP culture.
>>1697480Why do teachers fear for their lives in Berlin classrooms?
>>1697578Diversity, but that is not really an /n topic so just skip it.
>>1697577Yeah, definitely didn't work out for the ATC union in the US when they tried to strike and Reagan fired over 10k of them.
>>1697475>I mean it's clear theft and stealing if you're giving away the company assets without charging, so that is most likely not allowed as it would be ridiculous. (Aka the warehouse employee giving away the store inventory without collecting payment).Every kilometer they drive they incure costs of depreciation of the vehicle and they are burning fuel directly related driving, those are company assets they would be giving away 'for free', which is theft.Legally their would be a difference between using assets with the risk of depreciation and actual theft, which is taking control of a foreign movable asset.Striking temporarily disrupts duties of the employer and the employee and I argue that this does not have to be absolute on the side of an employee. (Do nurses have to turn of the ECMO when they strike?)In many strikes a part of the duties are kept by the employees, such as the duty to drive the bus to a certain place at a certain time. This can be interpreted as either doing what they are ordered to do or the employer having to mirror the duty of the employee by fulfilling his duty to provide adequate work and work materials. Collecting the fares is a different task and a different duty that is covered by the strike. As long as they do not stop anyone from paying or take the money for themselve, it's unlikely to be illegal.The passengers not paying would be guilty of sureption. Theft can't apply in this case as the transport is a service and not a good.This would also cover the insurance scenario you mentioned, as employers have to provide workplace insurance even if the relationship is suspended by a strike. Vehicle insurance has to cover all cases of use of the vehicle otherwise you wouldn't be allowed to posses a license plate for the car. (There are fringe cases, but they only apply if a person would generally not be part of the users of the vehicle.)
>>1698257>Legally their would be a difference between using assets with the risk of depreciation and actual theft, which is taking control of a foreign movable asset.There is no such thing as a 'risk' of depreciation, that always occures. every mile you drive with a car is causing wear and tear no matter how small. And yes, using up fuel/electric would be a tangible moveable good that is defined as such by law. (where I live anyway).>Vehicle insurance has to cover all cases of use of the vehicle otherwise you wouldn't be allowed to posses a license plate for the carI don't know where you live, but that's not true at all insurance only covers aspects previously defined in contract, with differences between who is allowed to operate vehicle, when it is operated and for what purpose it may be operated. Your private insurance for your private car bares no resemblance to that of a train or bus, lol.
>>1693500>>1693501State employed workers are already represented via the elected government, they don't need Unions.
>>1698353That was the thing, they were privatized so they're no longer state employed workers but a commercial firm and allowed to strike. Now government as usual want best of both worlds, you can't strike, and no you don't need job security and benefits either you can just 'compete' on the 'free' market.
>>1693508interesting enough, it's exactly the single-dimension nature of the railroads that makes them hard to automate, as even minor malfunctions can come with harsh consequences, and if something fails to go as planned, it requires complex work to correct iti.e. we have some rail lines here that have centralized traffic control (as opposed to having a red hat guy at every station), they're efficient if everything goes with the plan, but if something fails locally, and you have to send down a guy for 100 kms with a car just to correct the mess that'd only require a little bit of complex thinking from a local traffic operator, then the whole line crumbles
>>1698494You can't talk trains past signals and have conductors throw switches by hand?
>>1698701Good luck doing that in the states with 1 man crews. You'll wait an hour for the "super conductor", or scab utility man, to show up and throw the switches. Even longer during the winter and or during a traffic meltdown.
>>1698796>In my hypothetical scenario, you can't do that!No one cares
>>1698701long story short, nolong story long, conductors don't have proper ground to touch anything besides tickets (they're already whining if they have to get in the cab if they have to act as watchmen during bad weather), their retraining would be a nightmare and possibly most of them would fail it, and they're so understaffed that such a move prolly would increase the number of those who leave the railwayand that's only the human part, technologically you can't even throw switches by hand on a station that has local operation but motorized or distanced (push-pull rods) switches, which is like two thirds of all switches in my country, without the aid of a technician who makes it possible, who is present on locally operated stations, but not on centralized control onesalso also, automated trains require pre-sent signals: it's pretty common on manned main lines too, you have a big flashing box over your head with the projected signal ahead, and the big flashing light on the signpost is just the confirmation of it; however, while manned trains can easily get past of it not working and just concentrating on what the post will say, automated trains basically die without iti think that frequent and less complex systems, like metro lines do have a future in automatization, but i seriously cannot imagine the hell that an automated national system would cause
>>1699396Lol fuck. I see the problem though, your conductors aren't really 'railroaders,' they're basically station staff (just onboard). Interesting to find out, I thought they were basically like conductors in the US and Canada. It's a shame about the switches and signalling being so complex. Offhand it looks like that hinders operations more than it helps, but again that is my outside view.
>>1699413yeah, conductors in Europe are pretty much more like human resourcesup until like 15 years earlier (at least here in Hungary), we had conductors for freight trains too, who were like conductors in the US, now they're a rarity and only occur on such rarely used branch lines where there are no staff on the stations anymore and you have to bring someone who controls the traffic, most of the freight trains are only using shunting operators who belong to the stations, not to specific trains
>>1699413also, our switches on modern lines basically look like this, you can clearly see that it needs a lot of work to be able to thrown manually
>>1698280>There is no such thing as a 'risk' of depreciation, that always occures. every mile you drive with a car is causing wear and tear no matter how small. And yes, using up fuel/electric would be a tangible moveable good that is defined as such by law. (where I live anyway).I'm speaking about german law. Use outside of what was agreed could incur civil damages, but that doesn't apply for this case. The employee is not using it for a private matter or using his employers tools for the strike, but to fullfil an order of the employer even if he is relieved of his duty to fullfil it. Rescheduling tasks in any way is banned during a strike.>I don't know where you live, but that's not true at all insurance only covers aspects previously defined in contract, with differences between who is allowed to operate vehicle, when it is operated and for what purpose it may be operated. That's seen as a breach of contract against the insurance company, for which you are liable to pay a penalty relative to the costs of insurance that you would pay for a contract that covered that use. It has no impact on the liability of the insurance company.
>>1699420That looks like a regular power switch (can't tell for sure b/c the machine isn't totally visible)... they can't be taken off power and manually operated at all?
>>1699426no, since there's no manual attachment, basically the switching rod is connected to the electric engine and the electric engine onlywhat you have to do is disconnect the rod from the engine and install a manual switcher (fun fact, we call the weight that you throw over "weight pear")
>>1699492That's unreal to me. But I suppose they designed it like that for a reason.
>>1699499yes, it's not just an attribute of the system, but a requirement from the transportation board tooin older systems, where the "weight pears" are kept, they're painted grey, and they're switched directions, so that even if unauthorized hands want to touch it, the pulling lever hits itself into the column and can't get past it (pic)
>>1699500I see. Here's a US/Canadian dual control switch for comparison. The second you take it off power into hand throw mode the signal drops to red, so if a crew has to fool around with one the dispatcher will give them permission to hand operate the switch.
>>1699420That crossover doesn't even have movable point frogs.
>>1699534Good. Less shit to fuck up
>>1699535Also means it's a slower crossover
>>1699542That's fine. Not everything needs to be made for high speeds
>>1699534we don't have any here, they are usually good for any speed in straight direction, and you can get up to 80 in diverting direction i thinkthere's basically no need for them to be continuous
>>1699396>conductors don't have proper ground to touch anything besides ticketsThat's simply completely wrong.While the driver is responsible for the engine, the conductor is responsible for the train.What you meant to say, is probably, that the conductor is not involved in actually driving the train in any capacity.Except if the driver also happens to be the conductor, of course.>while manned trains can easily get past of it not working and just concentrating on what the post will say, automated trains basically die without itAren't regular signals on lines fit for pre-sent signals generally switched off, when a train fit for that passes by? That's at least how I know it.Having a fall-back mode is certainly a good idea, but you can't hope to operate remotely as efficiently as with pre-sent signals.
>>1693440>Unions for public employeesTHESE COCKSUCKERS ARE ALREADY SHELTERED FROM THE BRUTAL MARKET FORCES AND CHINK LABOUR, HOW IS THIS LEGAL?!
>>1699631Privatization. There is competition in the regional market.
>>1699631They aren't "public employees" anymore.At least those of them, that are allowed to unionize and go on strike.Deutsche Bahn was turned into a stock company in the 90s in preparation for privatization.Now, the state still owns 100% of those stocks to this day, but other than that "DB AG" largely operates like a private company.
https://www.bahn.de/p/view/service/aktuell/index.shtml>Vom 2. September um 2:00 Uhr bis zum 7. September um 2:00 Uhr>Die Gewerkschaft Deutscher Lokomotivführer (GDL) hat bundesweite Streikmaßnahmen im Personenverkehr angekündigt. Die Deutsche Bahn bietet dennoch ein verlässliches Grundangebot. Trotzdem können wir in dieser Situation nicht garantieren, dass alle Reisenden an ihr Ziel kommen.>Bitte verschieben Sie Ihre vom 02. bis 06.09. geplanten Fernverkehrsreisen, wenn Sie nicht zwingend fahren müssen. Die Ticket-Gültigkeit wird ausgeweitet.>Gültigkeit gekaufter Tickets>Für den Zeitraum des GDL-Streiks gelten besondere Kulanzregelungen für die Gültigkeit bereits gekaufter Fernverkehrstickets:>Alle Fahrgäste, die im Zeitraum vom Donnerstag, den 02.09.2021 bis Dienstag, den 07.09.2021 ihre Reise aufgrund des Streiks der GDL verschieben möchten, können ihr bereits gebuchtes Ticket ab Montag, 30.08.2021 bis einschließlich Freitag, den 17.09.2021 entweder flexibel nutzen oder kostenfrei stornieren. Auch Sitzplatzreservierungen können kostenfrei umgetauscht werden.>Zudem gelten bei Verspätung oder Zugausfall die allgemeinen Fahrgastrechte.
>>1693503And you're against unions? Fucking idiot. Bootlicker scum. You're the human equivalent of shooting yourself in the foot.
>>1693440Yesterday, my 1 hour commute by train home from work took 3 hours because someone decided to jump onto the tracks to kill themselves.
>>1702475Worst part is you can't say 'f*ck that guy' because you're supposed to feel sad for him, being a human life and all. Do they stop the train 2 hours everytime they hit a deer or stray cat? No
>>1698172>reaganwell color ME surprised
>>1693440You must be a boomer, only boomers don't take public transport. No excuses, "strikes" are just cagetroll propaganda
>>1693440man I wish I could migrate to Germany and become a conductor.I'm not brown enough sadly.
Any news on the next strike?I was planning to put that "Abo Upgrade" to the test this weekend.
>>1693440Buy a car.
>>1693440>Live in poorland>Train drivers never go on strike because they are carefully selected railfags
>>1693602One poorly implemented system doesn't mean you can't do it. As far as automating machines goes, a train is fairly straightforward. Most of the complexity is handled by external systems.
>>1709268>t. no railroad experience
>>1709272Yes. How often do the scenarios where the driver has to get out of the train and operate the switches actually come up? I've been commuting on trains for most of my life (ex-USSR elektrichka) and can count the number of times the train has had to stop on the tracks on one hand. And all of those were due to some freight train mishap. Face it, commuter trains could run automated. The extreme edge cases can be handled by stopping the train and having someone drive out to take a peek. The vast majority of the time the system around the train just werks.
>>1709282If it was as easy as you think it is, most trains would have already been automated.
>>1709305Automated trains already exist. It just takes a lot of capital up-front for something that is bound to be politically unpopular, in an industry that is usually heavily unionized, monopolistic/oligopolistic and tied deeply with local and regional governments.
>>1709310It's useless arguing with the train lobbyists on here. They will fight you tooth and nail when arguing about train automation.
>>1709312I'm interested in hearing some arguments that don't boil down to "but what would you do if shit breaks down in the middle of nowhere". Because most commuter rail systems, especially in Europe, are in densely populated urban areas and their immediate hinterland, and operate on well-developed infrastructure that is already heavily automated and centralized.
>>1709310>>1709312It's easy to build AGTs from the ground up. Automating existing lines and equipment is very difficult. If It was as easy as you think it is, it would have been rolled out by now because eliminating train crews is a significant cost savings.
>>1709315Because station agents don't know how to fix mechanical problems, and driverless trains cannot handle people and obstructions on the track and other unforeseen conditions like broken rails. Again, this would be well underway by now if it was possible because of the cost savings. It's easy to come to erroneous conclusions when you have zero experience working on trains.
>>1709314>If It was as easy as you think it is, it would have been rolled out by now because eliminating train crews is a significant cost savings.But that's my point, it's mainly difficult for political reasons, not technological ones. Rather than talking in vague terms you could mention some specific challenges faced by would-be train automators?Automating a fleet of locomotives is straightforward. Expensive, but straightforward. As far as I know most switching and signalling equipment is already tied into extensive DCS/SCADA systems, which makes integration with the newly automated locomotives also straightforward - also expensive, but also straightforward.
>>1709318>Because station agents don't know how to fix mechanical problemsHow often does a commuter train have to fix mechanical problems on the fly? Again, the vast majority of the time the train and the system around it just works. If something comes up the train has to stop anyway. Increased costs in low-frequency failure events because you have to wait an hour for a technician to come down don't necessarily offset massive cost savings in removing the ever-present human operator on board.>obstructions on the trackTrivial to deal with in 2021. Even before the current CNN-powered computer vision boom there were systems capable of straightforward obstruction detection through video and lidar/radar.>broken railsAgain, relatively simple to detect in this day and age, and since the train has to stop for this until the condition can be fixed there is no point in having an operator on board. It's also such a ridiculously low probability edge case that I am surprised you brought it up.
>>1709319>But that's my point, it's mainly difficult for political reasonsNot in the US. I don't know about Europe, but I imagine carriers would love to reduce their operating costs by eliminating train crews. There is no driverless train technology that's ready to go right now except for new passenger systems that were built with it in mind and integrated into the design from day one.>Automating a fleet of locomotives is straightforward.Nope, and remember that you've never worked in the industry a single shift and are wholly ignorant about the challenges of automation in it.
>>1709320>How often does a commuter train have to fix mechanical problems on the fly?That's only one scenario in which automation would fail. There are others: signal malfunction, communication malfunctions, trespassers, and the like are others. You wouldn't get it as you've never worked in railroading and think running a train is as easy as pressing a button to start and another to stop.>Trivial to deal with in 2021.Hitting people, cars, trees, and rocks isn't trivial.>Again, relatively simple to detect in this day and ageYes, and getting across it is another issue entirely. Your options are wait until it's' fixed (could be short, could be hours) or, if an engineer is present, run at a reduced speed through the block on the lookout for a broken rail and walk the train over it, and have maintenance fix it afterward.If all of this is so easy, why haven't driverless trains exploded across the world? Muh unions? Grow up. There's no technology available today that can handle it, except in the case of systems like AGTs that are designed/built to use it and have completely separated ROWs
>>1709322>There is no driverless train technology that's ready to go right nowStop treating trains as if they were consumer goods that you buy and get working out of the box. It's a large industrial system. Automating such systems involves automating and integrating their many disparate elements in part. Even if some big company purports to sell "out of the box" automation solutions they're still going to have to design and build your particular system from many parts, tailored for your particular application.The technology is already all there. It's just a matter of how much it would cost and what the regulatory/political barriers to it are.>Nope, and remember that you've never worked in the industry a single shift and are wholly ignorant about the challenges of automation in it.And you haven't automated anything - which shows because you clearly don't understand what can be automated and how. You listed a bunch of specific challenges you could think of, and I refuted them. Can you do away with train drivers entirely? Probably not for a long time to come, because somebody has to be there to jockey the locomotives around the depot, attaching/detaching wagons and whatnot. Either from the locomotive itself or remotely, a human operator is still required for these sorts of less-structured tasks. But are they really needed for routine operation on the actual lines? I don't think so.
>>1709324>There are others: signal malfunction, communication malfunctionsLow probability event. If your signalling and comms system is ancient, shitty and never maintained, that's a separate issue. All the control system has to be able to do is stop and signal for help when it detects a condition it can't handle. Millions of factories worldwide operate complex, wholly automated machinery using exactly this approach. I've built systems that do this myself. >, trespassers, and the like are others.Obstructions on track are trivial to deal with, as already discussed.
>>1709326>Stop treating trains as if they were consumer goods>you haven't automated anythingI was a railroader. I know more about them than you do. Every point you've mentioned has already been considered by actual industry researchers and engineers to solve. Eventually the technology will arrive, but it's a long way off. You jumped to the conclusions that because a train doesn't need to be steered, it can be automated easily, which is not true.>The technology is already all there.Where?>But are they really needed for routine operation on the actual lines? I don't think so.Well, that's what you get for being ignorant about railroad operations.
>>1709324>Hitting people, cars, trees, and rocks isn't trivial.A computer can avoid collisions far better than any human in 2021. >Your options are wait until it's' fixedYes. This is how it can work, and work very well. Again, you're assuming extreme edge cases are the norm. A human driver does not necessarily pay for himself just because once in several million miles he might be able to save a few hours of delay for some edge case failure mode the automated system couldn't handle.>why haven't driverless trains exploded across the world? Muh unions?In part, yes. In part also the fear of automation among normoids like you. And in much bigger part the big up front capital investment that needs to be justified in the face of concerns like those mentioned before.
>>1709327>Low probability eventSource?>Millions of factories worldwide operate complex, wholly automated machinery using exactly this approach.Straw man. Trains aren't factories where only authorized and trained personnel are near the equipment inside them.>Obstructions on track are trivial to deal with, as already discussed.Just go slow bro someone else will pick up the body bro boulders can't derail trains bro
>>1709328>I was a railroader. I know more about them than you doYou don't know more about automating arbitrary types of machinery than me, because unlike me you were not a controls engineer.>WhereIdiotic spoiling question. On the market. Control systems are the integration of many separate technologies. Give me a locomotive, a set of operating parameters and I will build an automatic control system for it using off-the shelf components and a bit of high-level code tying it all together - just like for any other piece of machinery. It's just going to cost you an arm and a leg.>Well, that's what you get for being ignorant about railroad operations.Your entire argument against it boils down to "muh obstructions". Which are trivial to detect (using computer vision) and deal with (just stop, if it's big enough to cause trouble).
>>1709330>A computer can avoid collisions far better than any human in 2021.Okay, show me a system that can handle that in sue in the railroad industry. You can't because it doesn't exist.>Yes. This is how it can work, and work very well.Not when there are passengers on board. A trained crew can circumvent many unusual circumstances with minimal delay. Automating trains and waiting for relevant personnel to assemble on location then fix the problem will work for freight, but not passenger operations.>Again, you're assuming extreme edge cases are the norm. Any automated train operating system will have to deal with unusual circumstances. If automation was capable of handling them all, trains would be automated.>In part also the fear of automation among normoids like you. Incorrect. The technology hasn't arrived and won't in the near future.>And in much bigger part the big up front capital investment that needs to be justified in the face of concerns like those mentioned before.You are clueless. Crew costs are the largest expenditure of North American rail systems, and I imagine it's the same story elsewhere in the world. The case for automation has been made, but the technology hasn't arrived.
>>1709331>Source?If your signalling system can't support an automatic train running on it within acceptable failure rates, you need to pay for a better signalling system as part of the upgrade. This is a non-issue. Low probability of failure signalling systems exist and you rely on them every day.>Straw man. Trains aren't factories where only authorized and trained personnel are near the equipment inside them.The strawman is assuming this matters. Set a target failure rate, tweak your system until it operates below it. Lots of manufacturing processes are much more finnicky and tricky to deal with than a track that might occasionally have an obstruction on it.>Just go slow bro someone else will pick up the body bro boulders can't derail trains broLiterally just stop.>boulders on trackNow you're off into retard land. I get that you're some sort of cowboy operating trains in the American hinterland where your track might have disappeared under a landslide from one day to the next, but this is not the kind of operating environment we're talking about when discussing European urban-suburban commuter rail systems.
>>1709333>Okay, show me a system that can handle that in sue in the railroad industry. You can't because it doesn't exist.no true scotsman. Computer vision systems don't care what industry they are in. Set it up to classify events and trigger commands. Train and tweak it until it works for your application. That's it.>many unusual circumstances with minimal delayI keep asking for these unusual circumstances but so far all I've been getting is "obstructions on track". Which you can deal with by stopping. And which are rare enough in the commuter rail environment for this to be entirely acceptable.
>>1709332>You don't know more about automating arbitrary types of machinery than me, because unlike me you were not a controls engineer.I don't care about automating arbitrary types of machinery, I'm talking about trains.>On the market.Okay, which ones?>Give me a locomotive, a set of operating parameters and I will build an automatic control system for it using off-the shelf components and a bit of high-level code tying it all together - just like for any other piece of machinery. It's just going to cost you an arm and a leg.Yes I'm sure you've got the answers that have stumped every other engineer who already works on rail automation.>Your entire argument against it boils down to "muh obstructions". Nope, they are one condition any automated train handling system would have to deal with though.>Which are trivial to detect (using computer vision) Okay, which railroads are using systems can do that?>and deal with (just stop, if it's big enough to cause trouble)Already explained why that's not feasible with passenger trains. Troubleshooting and working around problems on-the-spot is what the human mind can do better than any automated system can.
>>1709335>If your signalling system can't support an automatic train running on itOkay, then, no source.>Lots of manufacturing processes are much more finnicky and tricky to deal with than a track that might occasionally have an obstruction on it.How many of them carry hundreds of people over hundreds of km of track?>Literally just stop.Already explained why this isn't a solution>Now you're off into retard land.Ad hom attack. Btw, you can find articles about rock slides affecting train traffic in Europe with a cursory Google search.
>>1709338>Computer vision systems don't care what industry they are in.If it's as easy as you think it is, it would have been done by now. Must not be as easy as you think it is.>I keep asking for these unusual circumstances but so far all I've been getting is "obstructions on track". Which you can deal with by stopping. And which are rare enough in the commuter rail environment for this to be entirely acceptable.And your only rebuttal is "uhhh just hope the theoretical automated driver stops and then wait idk." Btw, I already said signal and comm failures and track defects, and of course you are moving and dealing with humans. If you had even a month of experience working for a railroad you'd know that it's not as easy as stop/start. That's probably why you don't work in rail automation.
>>1709341>If it's as easy as you think it is, it would have been done by now.No, this is a bullshit fallacious argument. Jerking off is easy but I haven't done it today.Human-level computer vision became so cheap you can put it in toys in the mid to late 2010s. Train infrastructure gets built out on timescales of decades.
>>1709376Just let us know when such a system is in use. If it's as easy and cheap as you say it is, you should have no trouble finding plenty of examples
>>1709377Same retarded argument. You can buy an off the shelf cognex camera and train it to detect obstructions on rail under extremely varied conditions in a few weeks. But the camera itself is not the system, somebody actually needs to shell out the millions to actually set up the entire system and drag it through the paperwork hell it would take to get permission for it to operate.
>>1709379So it's neither easy nor cheap. Got it
>>1709377>>1709341>hurr durr technology is a black box that magically appears out of nowhereThe subsystems required to make an automated train are all there. Hardware is not the issue, any of your local industrial automation contractors could handle the locomotive conversion. The problem, until recently, was the lack of software capability to do easy computer vision at a human level - but this capability has been around since 2015 or so, and is now being widely adopted in tons of different industries. Rail is just slow. Everything else - specifically robust comms networking and DCS/SCADA for centralized control, which take over the role of track-side signalling entirely in both high speed and automated rail lines - has been around for ages already.>but muh edge casesHere's how you handle edge cases for commuter rail in a flat, mild-climate metropolitan area in Europe:>what if there's an obstruction on track?stop, signal for help and wait. If the obstruction clears, move on.>what if the network goes downstop, signal for help (until the network is back up) and wait>what if there's a broken railstop, signal for help and wait>what if what if what ifstop. signal for help. wait. Maintain your infrastructure properly with the funds saved by cutting redundant personnel and you will still have perfectly acceptable reliability and timeliness even with all of the above edge cases.
>>1709382I never said it was cheap. Automating large physical systems that can potentially put human lives in danger is never cheap. Easy is a meaningless bullshit term. I said it was expensive up-front, but straightforward. No new technology needs to be developed, only integrations using existing technology. You can do it all with off-the-shelf components, but integration to fit requirements is still an expensive process. One that would take long enough and cost enough to make the additional hassle of having to deal with firing butthurt rentseekers and political fallout over "muh computers can't be trusted" seem like too much to deal with.
>>1709384>but this capability has been around since 2015 or so, and is now being widely adopted in tons of different industries.I know that. Irrelevant, though.>Rail is just slow.Because the challenges of integrating them into an extant rail system isn't as easy or cheap as you think it is.>stop>stop>stop>stopNot practical for passenger rail as I've said already. With a human engineer, those problems can often be cleared with a minimal delay. The technology to troubleshoot and solve problems isn't there yet, which is why we still have engineers.>>1709386>I never said it was cheap.Oh?>>1709376Human-level computer vision became so cheapCan you just admit being a commuter on a rail line doesn't make you knowlegeable about railroad hardware, operations, safety, or problems and you only have a surface-level understanding of how it works? Glad you're not in charge of any rail industry automation projects.
>>1693733except for the person required to be in the train that can drive it.
>>1709318>fix mechanical problemsYou hear engine drivers complaining about everything having become too complex to fix anything themselves in case of breakage all the time these days.There are many completely automated *isolated* railway lines around the world, which normal humans ride on a daily basis.The thing those have usually in common, which you can't really realize on a national scale, is, that the tracks are physically inaccessible from the outside.That appears to be the singular thing preventing full nation-wide automation in my eyes.Practically all major railway accidents in the recent past, that come to my mind, were caused by a human circumventing safety equipment, that would usually have prevented him from doing stupid shit.Humans are fucking retarded. All of them. Get them out of the driver's seat already.
https://www.youtube.com/watch?v=BDbfDUqPm8ELmao, these guys don't even bother with trying to stop based on visual confirmation, probably because you can't really stop a train this size in sight distance anyway.
>>1709387>Human-level computer vision became so cheapIs not the same as saying>creating an integrated railroad control system that uses computer vision as one of its many many elements is cheapyou illiterate manual laborer twat>stopping is not a valid solutionYes it is. Commuter trains don't stop for the crew to get out and fix shit day in day out, the vast majority of the time they either run on time or get held up by something the crew can't fix either.
>>1709695>Humans are fucking retarded. All of them.Yes, you're retarded
>>1709553And how often does that person have to do anything? How many times have they gone and resolved some issue that made their presence worthwhile? You're just reaffirming the point that this is primarily a political hurdle and a matter of institutional inertia, not a technical requirement.
>>1709749Get your story straight: it's either cheap and easy, or it's costly and difficult. You keep flipping between both.>Yes it is. Nope
>>1709751>You're just reaffirming the point that this is primarily a political hurdleNope. The technology to automate extant railroads doesn't exist and won't for some time. That's why you keep dancing around the fact that there are no off-the-shelf systems that you keep describing.
>>1709750You would literally not be allowed to run computer systems with failure rates comparable to humans.They have to perform way above humans to be allowed.And in case of operating railways they outperform humans in terms of safety by several orders of magnitude. It's not even close. Not even remotely.Humans are fucking unfit to operate trains, cars, ships or planes with their own hands. (And it pains me to say that, since I love doing those things.)The only reason they still do it, is because we need to attribute responsibility in case something goes wrong and the """loss of control""".We effectively prefer unpredictable, unreproducible human behavior, that with high probability will lead to harm, over predictable, reproducible machine behavior, that with significantly smaller likelihood will lead to harm.
>>1697314>imagine being on strike as a warehouse employee and giving away the merchandise without collecting the payments.Were that a likely outcome, perhaps employers would be less tempted to provoke strikes in the first place. You certainly wouldn't find amazon cramming people into a wage-cage full of pissbottles.
>>1709773>Humans are fucking unfit to operate trains, cars, ships or planes with their own hands. Autistic screeching intensifiesThere's not much of a case for full automation when ATO and similar systems already exist (like on TGV and the Shinkansen), both of which have drivers and superb safety records.
>>1697314>theftI think the technical term is embezzlement.