A lot of you are probably reading this at work and despite that, a lot of you are probably also drunk. That's because most of us have jobs where, if you maybe screw up here and there, it's not the end of the world.
Or at least that's what we'd like to think. It turns out some of the biggest, costliest disasters have resulted from some random employee making a single tiny mistake. Such as ...
Over the course of three months in 2006, AOL compiled search data on over 650,000 of its users. That might sound ominous, but all they wanted was a tool for researchers. Sure, the users didn't know their data was being saved, but what they didn't know couldn't hurt them, right? After all, it's not like they would ever release it to the general public.
Whoops ...
Somebody should have told company researcher Abdur Chowdhury. On Friday August 4, 2006, with a click of a mouse, Chowdhury uploaded a single compressed text file of the search data on an AOL website that was, in fact, open to the public.
But don't fret, the user names weren't listed and AOL officials quickly realized the mistake and took the file down on Monday, the next business day.
Really, What's the Worst That Could Happen?
This is the internet, there is no such thing as the next day. By the time the file was taken down, word of the data leak had spread through blogs far and wide, the search results were posted on mirror sites including one that remains today as a searchable database. The media had already taken to the frighteningly easy task of personally identifying some of the users.
See, despite the absence of user names, a number of people had unknowingly identified themselves by way of "ego searches." That means that, along with searches for pleasant topics like rape, murder, committing rape and murder, hiding rape and murder, and Clay Aiken CDs, they also searched for their own names, addresses and social security numbers.
Within days, The New York Times had released, with consent, the name of a user who they tracked down by cross-checking search keywords with phone books and other public information. After a few weeks, AOL had not only fired the researcher responsible for the leak, but also his supervisor and Chief Technology Officer Maureen Govern.
All because of one click of the mouse.
As a bizarre postscript to all of this, one of the users identified in the file only by number ("User 927") became internet famous for having basically the creepiest search habits imaginable. Searches included "human mold," "dog sex," "child porn," "Disney Beauty and the Beast Porn" and, most frighteningly, "'Sugar, We're Going Down' by Fallout Boy." No, really.
Well, recently, a stage production premiered, based on their life, called User 927.
On the night of July 13, 1977, a system operator sat in New York City's ConEdison electric facility, probably reading a comic book and wishing the internet had been invented.
Then, lightning struck. Three times. It nearly crippled the facility. To make things worse, neighboring facilities then opened their connections to the ConEd system to keep their own from overloading. The details are technical, but let's just say at that point, the system was going to be fucked unless somebody took action.
But no worries, our trusty system operator was on duty. And all he needed to do was flip a few switches and disaster would be averted. What could go wrong?
Whoops ...
Did we mention those switches needed to be flipped quickly? And in the proper order? Someone should have mentioned it to the system operator. One switch flipped out of order and within a few minutes, a 230,000 volt connection with New Jersey closed and the system began to overload. At 9:36 PM, the entire ConEdison system shut down.
Really, What's the Worst That Could Happen?
New York City was suddenly plunged into 25 hours of electricity-free mayhem. With mid-July temperatures sweltering, a deranged serial killer who took his orders from his neighbor's dog on the loose, and 1977's New York City just being a generally unhappy place to be, people lost their shit.
In short order, the raucous, block party-like atmosphere in the streets turned into violent looting. Fires were started, store windows were smashed, electronics were stolen (albeit not used for some time) and the fucking Yankees were well on their way to another World Series title. Son of a bitch.
Above: Why we love New York
After all was said and done, 1,616 stores were damaged, 1,037 fires were set, and 3,776 arrests were made. A Congressional study estimated the total damage to the New York City area at $300 million. Also, as a bizarre side effect: hip-hop was born. Seriously. The looting apparently resulted in the first access to DJ equipment for poor inner city youths, launching the movement.
In the aftermath of the blackout, ConEdison implemented changes to make sure the same problem never happened again (which it totally did in 2003). We're assuming this involved something along the lines of a few sequentially ordered labels above those switches.
The Hubble Telescope was initially conceived and budgeted for in the '70s and planned for launch in 1983. Various mishaps, not the least of which being the Challenger disaster, delayed the project for years. When it launched in 1990, scientists expected the Hubble to take its place among NASA's "great observatories," placing it in the company of, among others, the Compton Gamma Ray Observatory.
Probably not related to the Compton Gamma Ray Observatory.
The Hubble was expected to deliver some pretty kick-ass images thanks to its ability to capture those images with little to no back light (as you'd get with an earthbound telescope). Sounds like a huge task, but the Hubble was equipped with one of the most powerful mirrors ever built.
A team of the best engineers in the world gathered to build that mirror, working 12-hour days for five straight years, grinding the mirror with equipment that would make sure it was perfect to within a millionth of an inch.
Whoops ...
A guy named Lou Montagnino was in charge of testing the thing, using equipment so sensitive they had to do it in the middle of the night--(the vibration of a car driving three miles away would throw it off).
Unknown to Lou, a microscopic chip of paint flecked off a measuring rod that was supposed to make sure the mirror was the right shape. It started giving back false readings as a result, and the mirror wound up being off by four microns.
That was their mistake. Four microns. Twenty-five times smaller than the width of a human fucking hair. Smaller than a mosquito's flaccid penis.
Really, What's the Worst That Could Happen?
When the first images were returned from the Hubble, the quality was drastically less than what NASA expected, and nowhere close to powerful enough for what NASA needed it to do.
Of course the real problem was that by the time they discovered the flaw, the damned thing was already out in space. So say goodbye to a few billion more dollars, which is what it cost for a series of Space Shuttle missions to fix the thing's mirror (the repairs got so costly that there was debate as to whether it wouldn't be better to just build a new one). We're surprised they didn't just strap Lou Montagnino to a rocket and send him up there with some really fine grit sandpaper in his hand.
At an air force base in Guam, during a routine check of a Stealth Bomber (aka The Most Expensive Fucking Plane Ever Built) somebody on a maintenance crew noticed the humidity was screwing up the air pressure sensors. Not a big deal, it's just a $1.4 billion aircraft, not like they could have ever guessed it would be flown in a place where there was humidity. We always go to war with dry countries.
Anyway, they just made sure to dry off the sensor before calibrating it. Problem solved. Good thing they worked that out before anything went wrong!
Whoops...
Communication is a beautiful thing. As simple as the sensor fix was, the maintenance crew overlooked one minor detail, which was telling other maintenance crews to do the same thing. But seriously, it's just an air sensor, with some droplets of water on it. Do those things really even serve any purpose? It's not like it's an engine or a flux capacitor or something.
Really, What's the Worst That Could Happen?
When another bomber pulled into Guam earlier this year, on presumably an equally humid day, a different maintenance crew left the wet sensors the way they were. As it turns out, those air sensors feed data to the Stealth Bomber's flight control system. Important data. The kind that keeps Stealth Bombers in the air.
See, that's what makes a $1.4 billion plane cost $1.4 billion--it takes hundreds of pounds of sophisticated computers to fly the thing. The malfunctioning sensors resulted in a premature take off, a 30-degree nose-towards-the-sky ascent, and ... well let's just show you:
Fortunately for the pilots, they were able to safely eject. And on the bright side, the next time a problem like this arises, they'll know how to fix it!
Along with The Mars Polar Lander, The Mars Climate Orbiter was one part of the Mars Survey '98 project. The ambitious project was intended to study weather and climate patterns on Mars, presumably so we can all move there one day when things finally go completely off the rails here on earth.
With two separate, unmanned aircraft designed to work together from completely different points on a (probably) uninhabited planet, an obvious question is raised. Who the hell is working on our flying cars?!?! Whoever it is, we can only hope that as much careful attention and detail goes into our airborne Prius as was put into the Mars Climate Orbiter.
Whoops ...
Or not. When Maryland-based contractor Lockheed Martin was tapped to help build the Orbiter, they made as assumption that many of us probably would also. They're in the United States, NASA is in the United States, and 'round these parts, we don't deal with no stinking metric system. Thus some unnamed engineers installed software in the craft's thrusters that operated on the good ol' American units.
Nobody told NASA this, and they continued doing business in the same fruity metric system way they always have. But shit man, isn't there a checklist somewhere in the billion dollar orbiter building process that confirms these things?
We like to think there was one lone intern in mission control who, upon seeing some odd readouts on a screen, got the urge to ask his supervisor, "Is this like, in metric or American here?" but was afraid it was a stupid question.
Really, What's the Worst That Could Happen?
There are no stupid questions when it comes to $330 million spacecraft. You would really be hard pressed to find a NASA project that went more horribly wrong that didn't involve multiple fatalities. While neither machine performed particularly well, The Mars Polar Lander at least lived up to the promise of its name and, you know, landed.
The Mars Climate Orbiter, on the other hand, just couldn't be bothered. Initially expected to enter orbit at an altitude of 140 kilometers above Mars, the whole metric system misunderstanding caused the thrusters to fire incorrectly, causing the Orbiter to come in as low as 57 kilometers. At that height, the Orbiter was perilously close to the Martian atmosphere. Or at least that's where it was the last time anyone saw it.
The thing vanished, the most likely explanation being that atmospheric pressures and friction caused it to burst into flames and disintegrate. But, as a website that thrives on geekery, we're unwilling to rule out alien intervention. Whatever the case, it proves that countless disasters can be prevented by simply assuming everyone you're working with is a moron.
No comments:
Post a Comment