Washington Metro’s computers crashed twice this past weekend, forcing all trains to stop and stranding passengers for up to 30 minutes. This is just the latest example of how the aging transit system is slowly falling apart.
It is hard to imagine today what kind of computers Metro used in 1976, when it opened DC’s first new rail line. Programming probably used COBOL or some other now-archaic language. (The Antiplanner has heard rumors that the COBOL programmers who wrote the software that runs the San Francisco BART system refuse to ride the trains.) Anyone who has an older computer knows that things go wrong and those cumulative failures add up until eventually the system just does not reliably work.
In any case, the Metro system has roughly a $10 billion maintenance backlog. As a result, rails break; trains fall apart during operation; computers crash; and the agency’s bureaucracy can’t even keep up with the problems.
Rail advocates never mention the cost of maintenance when they propose new rail lines. Instead, they call highways “antiquated” and argue that a higher share of federal transportation dollars–most of which come from highway user fees–should go to transit. In fact, it is rail transit that is antiquated, while the real future of transportation will continue to use highways.
The first distinction in need of clarification is that piriformis syndrome is considered a “functional viagra pills without prescription entrapment syndrome.” The word “functional” describes neurological compression disorders resulting from positional or kinesiological factors that are not merely linked to structural or inflammatory conditions. Moreover, it stays in your system for longer. like it viagra price canada Oil and ghee if the air passages are not blocked and male organ getting what it really needs cialis overnight shipping to be in GREAT shape to be in this division. Physiotherapists can visit worksites to assess the patient’s working posture and physical demands of the job and to provide advice and modifications where necessary. cialis online australia
As Google chair Eric Schmidt recently predicted, self-driving cars will “become the predominant mode of transportation in our lifetime.” Ironically, George Mason University economist Tyler Cowan reports that DC police pulled over the Google driverless car for some apparent infraction.
Cowan, who has written favorably about driverless cars before, should have checked his facts. Slate staff write Will Oremus did, and learned that the car happened to be parked where Cowan spotted it and the police officers came over to learn more about them.
While the Antiplanner has previously stated that the main obstacles to self-driving cars are institutional, some technical obstacles remain. First, the laser-beam device that sits on top of most of the driverless cars that have been tested in the last decade costs $70,000. Significant improvements will be needed to reduce this cost to affordable levels. Second, the on-board software needs to carefully map every street, traffic signal, and traffic law in the country. This means people who travel all over the country will require reliable memory drives in the terabyte, if not petabyte, range. Both of these problems are solvable, and you are likely to see cars fully capable of driving themselves on the market by the end of this decade.
One data correction: Schmidt was quoted as saying that 35,000 people die each year in drunk-driving accidents in the United States, and this number has “remained constant over two or three decades.” Both these statements are wrong. In 2010 and 2011, fewer than 33,000 people per year were killed in all U.S. auto accidents, and less than a third of those accidents involved alcohol. Both auto fatalities and alcohol-related auto fatalities have declined dramatically over the last several decades. Still, self-driving cars can save a lot of lives.
Rail advocates never mention the cost of maintenance when they propose new rail lines. Instead, they call highways “antiquated†and argue that a higher share of federal transportation dollars–most of which come from highway user fees–should go to transit. In fact, it is rail transit that is antiquated, while the real future of transportation will continue to use highways.
Absolutely true. But for rail fanatics like msetty, nothing matters except getting the government to build a wasteful empire of rail.
Though how do you expect robots to buy cars if they have no money?
Partly- or fully-autonomous vehicles can indeed save lives and time. But their efficiency will be greatly reduced if they’re programmed to obey the present catalog of speed limits, stop signs, and other traffic-control devices installed with no system or science. Imagine an arterial highway operating at 45 mph with today’s mix of human-controlled traffic, but posted with a 35-mph speed limit — an entirely common situation in most states. Such a road might operate at 50 mph or more with cyber-assisted vehicles, in greater safety than today. It would be a grotesque waste to use a sophisticated, multi-billion-dollar system to hold traffic to an existing speed limit, which is frequently a random number pulled out of a politician’s butt. Perhaps the growth of autonomous vehicles will provide the motivation to apply science to traffic law.
Interestingly, turning an autonomous vehicle loose on public streets does not seem to violate any law in most places: existing laws regulate the behavior of the driver, and the law is silent when the “driver” is a computer program. States like Nevada and Florida are not the first to allow use of autonomous vehicles, they are the first to RESTRICT it.
“States like Nevada and Florida are not the first to allow use of autonomous vehicles, they are the first to RESTRICT it.”
Indeed. Good point.
The problem with robocars is the world outside them.
Just because the physical properties of a road allow 50 mphspeeds doesn’t mean the human and other factors around it – kids playing, people walking, wild animals like deer permit fulluse of its engineered speed. Robo cars do nothing to handle incidents like your robo car running over my kid on a bike who accidentally runs out of a driveway or sidewalk while he is learning to ride because it can’t stop in time – which is why speed limits will continue to be a political issue.
Of course, typical of highway fanatics, they always fail to acknowledge the existence of a built enviornment outside of cars, curb lines, and garages that impacts on the use of cars, and to which the cars must conform.
“Robo cars do nothing to handle incidents like your robo car running over my kid on a bike who accidentally runs out of a driveway or sidewalk while he is learning to ride because it can’t stop in time…”
First of all, teach your kid to ride a bike somewhere other than your driveway or the sidewalk (bicycling on sidewalks is often illegal and definitely unsafe). If your kid is killed in these scenarios, it’s your fault for being an idiot.
Secondly, the Stanford prototype “Junior” has an “object recognition system [that] now includes recognition of pedestrians, bicyclists, and cars.” Once objects are recognized, speed is adjusted accordingly.
This technology will only improve.
Frank:
First of all, teach your kid to ride a bike somewhere other than your driveway or the sidewalk (bicycling on sidewalks is often illegal and definitely unsafe). If your kid is killed in these scenarios, it’s your fault for being an idiot.
I own my driveway, my sidewalk, AND the street in front of my house. It is most certainly not illegal for my child to bike in any of those places. Where do you propose he learn to ride if not around my house? And I truly loathe people like you who suggest that my property and right to peacably enjoy it is subordinate to your ability to misbehave while driving.
As to your technology learning to recognize my chiildren (when? after it runs over three of them it will figure out what it is doing and spare the other two?) it can go screw.
You’ve focused on legality while conveniently turning a blind eye to the danger of biking on a sidewalk.
You don’t own the entire length of the sidewalk, and it is entirely possible that it’s illegal to bike on the sidewalk in your city. Regardless, it’s still stupid given the dangers.
“And I truly loathe people like you who suggest that my property and right to peacably enjoy it is subordinate to your ability to misbehave while driving.”
And I loathe people who misrepresent my statements. You were talking about automated cars hitting your kid because you allowed your kid to bike in an unsafe area, not about my “ability to misbehave while driving”; try to stay on point, chap.
As for learning to ride, since you asked, I suggest a bike trail or park where there are no cars. I saw a kid on a bike with training wheels yesterday at just such a place.
Or let ’em ride on the sidewalk and let natural selection work.
Frank:
There is something really wrong with you and ypur perception of the world if all the area immediately outside my house is an unsafe no-go-zone for my children. It is not illegal for kids to bike on sidewalks where I live, and as I said, I object to your proposed behavior on the street outside my house, which is also my deeded property.
I don’t want to live in a world where I need to drive 5 miles to find a place for my kid to ride his bike in a safe area away from your technology or the likes of you claiming undisputed right of way to drive over anything not ensconced in 2 tons of steel or behind a cinder block wall.
It might not be illegal, but it is still dangerous. No need to blame me for the danger.
From the link you undoubtedly failed to read:
“Bicyclists on a sidewalk or bicycle path incur greater risk than those on the roadway (on average 1.8 times as great), most likely because of blind conflicts at intersections.”
I said nothing about riding on the street in front of your house. Try to keep up, chap. That is indeed where biking riding should occur, but not for learners for safety reasons.
If you live five miles from a park or a bike path, I feel sad for you. I’m guessing you’re just exaggerating for rhetorical purposes.
And if you don’t want to live in such a world, by all means, don’t.
“I don’t want to live in a world where I need to drive 5 miles to find a place for my kid to ride his bike in a safe area away from your technology…”
Five miles? Exaggerate much? You have a track right across the street from your house, where you kid can safely learn to ride. And you’re on a cul de sac, but watch out for those high school students entering the parking lot! They represent a greater threat to your child on a bike than the imaginary scenario you presented.
I don’t think one needs to worry so much about autonomous cars in a nieghborhood. The issue is on highways and major surface streets where one should hope children are not playing.
Get a load of this objection to “robocars” in the comments on the linked photo in Antiplanner’s article:
The Antiplanner wrote:
It is hard to imagine today what kind of computers Metro used in 1976, when it opened DC’s first new rail line. Programming probably used COBOL or some other now-archaic language. (The Antiplanner has heard rumors that the COBOL programmers who wrote the software that runs the San Francisco BART system refuse to ride the trains.) Anyone who has an older computer knows that things go wrong and those cumulative failures add up until eventually the system just does not reliably work.
I don’t know what high-level language was used to write the original software to control the Washington Metrorail trains in the early 1970’s (it could well have been FORTRAN), but I do know what kind of computer was used – it was a Xerox Data Systems Sigma-series machine, not an especially good choice, since Xerox shut its XDS subsidiary down in 1975 (the year before the first part of the Metrorail system opened) after losing a lot of money.
For some years after XDS was shut-down, owners of these “orphaned” Xerox computers could purchase maintenance service from Honeywell’s mainframe computer company.
In fairness to Metro, I do not think they upgraded to a computer like the one shown in this Unsuck DC Metro posting.
This is from the Washington Post:
Metrorail’s reputation for speed and reliability is in trouble. Since last December, trains have been breaking down with unusual frequency, stranding passengers on crowded station platforms and slowing other trains on the line.
But this cite is not a recent article. It ran in the Post on August 22, 1982.
Just rename it “City of Ember”
Perhaps highways will continue to dominate over rail. And no doubt, rail requires significant maintenance costs. But then again, highways also require enormous maintenance costs that are also grossly underfunded, sometimes leading to catastrophic failures.
As no link or statistic was provided to bolster this claim, I found this Wikipedia entry. I’m not sure the data support your claim with respect to overall U.S. auto fatalities. From 1982 (three decades ago) until 2007 (five years ago), the number of accidents per year varied from a high of 47,087 (1988) to 39,250 (1992). The average was 42,880 deaths per year, with a standard deviation of ~1925.
While the number of fatalities has decreased significantly since 2007 (indeed, pushing the overall 1982-2010 standard deviation to ~3193 by adding just three years’ data), claiming that auto fatalities “have declined dramatically over the last several decades” seems overstated.
I did include a link to a news report that 2011 fatalities had fallen to 32,310. That’s a pretty dramatic decline from 47,087 in 1988, especially considering Americans drove far more miles in 2011 than in 1988.
The decline is statistically significant at any reasonable level regardless of which set of data you base the calculations on. And I would say that a decline of fully one-third of all fatal crashes over that period matters a great deal in practical terms.
In fairness to Joel, it is a standing subject of great interest as to why auto crashes have declined so dramatically since 2007 as compared to the previous ~40,000 per year level thay had been at forever since the cars took 90% of the travel market circa 1940.
The likely candidates – less overall driving since 2007, less rush hour congestion due to mass unemployment, fewer construction zones due to lagging road maintenance, possibly lower speeds due to high gas prices, fewer overall trips being taken due to high gas prices, all are not promising for a sustained trend in this direction if growth in highway use and employment is to resume as auto advocates like Randall would like.
I recall a report a few years ago by the TEXAS Department of Transportation pointing out that “user fees” usually covered less than 50% of the ongoing cost of highway maintenance, let alone amortizing the original cost.
Of course, the report is not available on-line, since it probably offended someone in the pro-highway lobby.
As for robocars, in some ways the Google people are geniuses, in other ways they lack a lot of common sense.
As for Metrosuck’s inane comments, beware those who have an unthinking, naive faith believing technology solves all problems (ignoring his moronic insult). Technology can, but human factors are much more important.
The problem with robocars are NOT technological, but human, e.g., the way people use cars. The criticism that Frank poo-poos is exactly right: people buy cars for many emotional reasons and manufacturers differentiate them for these reasons: the biggest are (1) the sense of “control” many people gain by driving an automobile, if ultimately disingenuous; (2) status; and (3) an ultimate sense of “freedom.” Given that robocars have none of these attributes, motor vehicle sales would plummet.
I also seriously doubt the vast majority of current drivers would willingly accept mandatory conversion of all vehicles and roadways to automation (except perhaps for dedicated new robocar roadways, if we see fit to waste many more hundreds of billons on such things). Such a mandatory conversion goes against EVERYTHING that outspoken auto advocates hold dear, as well as the resistance against even more government monitoring of everything individuals do. Sounds like a libertarian dystopia to me.
My final objection to robocars as Randal et al envisage is that the whole thing goes against what I’ve written about “Organization before Electronics before Concrete” linked at http://www.publictransit.us/ptlibrary/whitepapersHomePage.htm. That essay talked mainly about transit, but the same principles apply to other forms of transportation and many other types of activities as well.
I share similar thoughts. I don’t think people will give up the opportunity to drive themselves. That turns the fun of driving into something as mundane as an escalator. But the advantage would be that automated vehicles will replace trains as the principal means of transit.
I find it ironic how you say you hate “fixed guideway” transportation, yet you want to mimic “fixed guideway” transportation.
Just for the record, there is absolutely nothing wrong with COBOL programs per se. COBOL was and is a very capable programming language. Blaming old computer programs for DC Metro’s problems is just silly. They could have been written in C++ and, given the political considerations of the day, failed just as much.
Stick to the politics.
I do not think that there was any support for C or C++ or any other variant of C on the XDS Sigma family of computers.
And the Sigma computer system that ran the D.C. Metrorail system has been replaced by something else (I don’t know what).
There is nothing wrong with running old computers and old software. If it worked back then it can work now.
Many military applications work on old software and computers. I was on an AWAC at an airshow and they said their air traffic identification computer systems and software predate Apple and MSDOS.
That’s the problem. Government’s don’t modernize very well. The thing about cronyism is it puts people in a position; to ensure job security, any technology that may undercut staff is “overlooked” which is a fancy way of saying suppressed or ignored. The White House admits governments technology is outdated. We tend to think of government having a vast array of devices at their disposal. It’s sort of true. We think of national laboratories and supercomputers. But sitting in warehouses all across the country are antiquated computers and data storage the government holds simply because they’re afraid some classified bit of information may be stored they’ve yet to find themselves. The Defense Department admit’s it’s been trying for the better part of over a decade to release a credible financial record. If you look at the lines of code they have, it’s overwhelming. They may have equipment they don’t even realize they have, and if that’s the case if they don’t know that they have it, they may accidentally buy it again; everything from bullets to spare parts.
In the 70’s supercomputers were used to forecast the weather. By the 80’s they were being used for probability analysis and by the 90s they mastered the art of cracking (Brute Force Code breaking) and today used for Molecular simulations. Despite this they’ve yet to even replace networks for places like the White House. Imagine if the government bought a new computer for each employee every two years. Just think about how much it would cost, think of how much waste would be generated both physical and financial. The U.S. still has some of the lowest download speeds of any industrialized nation with an average of a few megabits per second. Where as South Korea is 20 times faster. An entire high quality Blu-Ray movie can be downloaded online in mere minutes where in the U.S. it would take nearly an hour. Everything from antiquated record keeping to 30 year old computers. Of course when the private sector modernizes they do so to remain competitive. Even an hours worth of Youtube loads faster in countries like South Africa or France. Don’t think for an instant that it’s meerly a technical issue. It’s also a bureaucratic issue.
Two opposite comments.
1. We rely way too much on computing power. NASA got to the moon and back with slide rules and less computing power than your TV remote. I doubt NASA could go to the moon today.
2. Government is the worst at upgrading because nothing forces them to change, and in fact change is risk for government.
The U.S. still has some of the lowest download speeds of any industrialized nation with an average of a few megabits per second. Where as South Korea is 20 times faster. An entire high quality Blu-Ray movie can be downloaded online in mere minutes where in the U.S. it would take nearly an hour.
Not if you are on Verizon’s FiOS network, which is quite fast (and getting faster).
The only reason the XDS Sigma computer at WMATA’s control center got my attention is because I worked on several of them a long, long time ago (but after 1975, when XDS was shut-down by its parent company) at NASA’s Goddard Space Flight Center in Greenbelt, Maryland.
The federal government held on to clunker technology like the Sigma computers because it could afford to do so, and because contractors working for NASA had written a huge amount of implementation-dependent computer code that could not easily be ported to more-modern technology platforms of the 1970’s and early 1980’s like an IBM mainframe computer or a DEC minicomputer.
A legend at NASA (note: never verified by me) was that they ended up with XDS computers because a NASA manager became aware that an entire Xerox system was sitting unused in a barn (!) at the U.S. Department of Agriculture’s Beltsville Agricultural Research Center, which is adjacent to Goddard Space Flight Center.
According to the legend, Xerox had given the Agriculture Department one of its systems in the hope that it would lead to a large sale of Sigma computers to Agriculture, but they were not interested, and the computer ended up stored in that barn. Richard Nixon did not care much for NASA (because it was so identified with President Kennedy), and cut its budget significantly after the Apollo landings on the Moon, and when NASA was starting up its Landsat remote-sensing program (which I was to work on after Nixon was safely out of office), it had no computer power to support the program. So NASA took the “free” Sigma computer(s) [there may actually have been more than one Sigma computer in that barn] and used them for operational control of the spacecraft, and to do much of the processing of the imagery from Landsat 1 and 2 and some of 3.
Because the Xerox computers were orphaned technology after 1975, getting computer programmers and support staff willing to work on them was expensive, and the programmers were not exactly the most-competent people out there. The money that NASA “saved” with “free” Xerox computer hardware was probably wasted many times over.
Note that none of the above is directly related to XDS Sigma computers used by WMATA to run its trains. I don’t know why WMATA procured Sigma computers, what it paid for them, nor do I know when or why WMATA got rid of them.
WTOP Radio: Metro says system-stalling glitch fixed
Unsuck D.C. Metro blog: Hot Cars: ‘Worse than Ever’
I’m not worried about the computer language used. I am concerned that tens of thousands of people’s lives depend on some software that was most likely developed in a manner equivalent to cowboy coding.