2015: Regent’s Park International Airport

A line of limousines and taxis snakes its way into the Royal Park to deliver 300 well-heeled passengers and their smart luggage to the discreet air terminal. They are in no rush because the flight they are about to board to New York will take two days.

Moored on the grass outside the terminal is a 600ft long behemoth, a vast Hybrid Air Vehicle. A cross between a balloon and an aircraft wing, this new-wave blimp is filled with non-flammable helium and air. Slung beneath is a vast passenger cabin akin to a miniature first-class cruise ship with dining rooms, a ballroom, bars and a casino.

For the same price as a club-class plane ticket, these 300 discerning travellers will eat, sip cocktails and dance as they float serenely across the Atlantic.

There is no runway; there is no need. Once clearance is given for take-off, the captain disengages the hover cushions that suck the craft to the ground, directs the thrust of four 8,000hp engines down, and powers the ship up to 9,000ft.

In 48 hours they will touch down in New York harbour, having burned just a fifth of the fuel used by an aeroplane. It’s a stress-free hop from central London to the centre of Manhattan, with no lengthy airport connections at either end, and no icebergs either.

The doomed R101 in one of the hangers

Airship travel has been a distant dream ever since a catastrophic fire in 1937 ripped through the  LZ-129 Hindenburg as it neared its mooring mast in New Jersey, killing thirty-five people on board and one man on the ground.

Reporter Herbert Morrison’s vivid eye-witness testimony would become the industry’s epitaph: ‘It’s a terrific crash, ladies and gentlemen. It’s smoke, and it’s in flames now; and the frame is crashing to the ground… Oh the humanity!’

Could an industry dogged by tragedy and belonging to a bygone era finally have found the technology to cruise back into the mainstream?

The American Department of Defense thinks so. They have just handed a £315 million contract to design and build the world’s largest flying object to a small British company based in Bedfordshire. Having beaten aviation giants Lockheed Martin, Hybrid Air Vehicles have just four months to build the belly and bones of the craft – the payload module, the fuel tanks, the four engines, the propulsion ducts and bow thrusters (the prototype is pictured on the previous pages).

If all goes to plan these parts will leave its secure manufacturing facility in May, be loaded on a vast Antonov cargo plane, and flown to Arizona where they will join up with the ‘envelope’ (ie, the balloon).

Once assembly is complete, military technology giant Northrop Grumman will add the top-secret surveillance equipment and the vehicle will travel on its own power to a U.S. army base on the east coast of the United States. Once there the U.S. military will put the fully assembled 300ft long craft through its places, flying it with pilots and without.

When it finally completes testing and trials in January 2012, it will leave the US and fly back across the Atlantic to the UK, the first time this has happened since the heyday of Zeppelins in the Thirties.

Guided by a three-man crew, the giant ship will stay at a U.S. Army base here, ready to be deployed. It will be available for use in Afghanistan where it can be flown remotely, climbing to 20,000ft and circling for 21 days, an omniscient god perpetually surveying the battlefield and giving advance warnings of IED attacks and ambushes.

The Cardington airship hangars in Bedfordshire

A zeppelin in a war zone?

Testing has shown that bullets, even missiles pass directly through the envelope because of the incredibly low pressure. Reassuringly, the company insists it has come a long way from the technology of the Thirties.

The 60 per cent helium and 40 per cent air mix replaces flammable hydrogen. And where the classic cigar-shaped Zeppelins struggled against the wind, hybrids use it in combination with their aerodynamic shape  to get more lift. They are helped by vectored thrust, like a Harrier jet, which directs the engine output downwards to provide vertical lift and allows them to take off carrying heavy payloads, even in high winds. They also burn less fuel than a plane while hauling more cargo and, with hovercraft-style landing gear, they don’t require an airport. They can even touch down on water.

The vast 800ft-long Cardington Airship Hangars in Bedfordshire are an eerie sight, dominating the skyline for miles around. Here history looms large.

In 1916 about 800 people worked at Cardington for Shorts Brothers, producing their first airship in 1918. In hard times after the war, the station was closed and construction abandoned, reopening again in 1924 as part of the Imperial Airship Service.

It was in Cardington that the 777ft-long R101, the then biggest airship in the world, was built, and from here that it began its ill-fated final voyage at 6.24pm on Saturday October 4, 1930 bound for India; first planned stop Egypt.

R101 reached London by 8pm, crossed the Channel in two hours, and at midnight a final message went out: ‘15 miles SW of Abbeville speed 33 knots. Wind 243 degrees (West South West) 35 miles an hour. Altimeter height 1,500ft. Air temperature 51 Fahrenheit. Weather – intermittent rain. Cloud nimbus at 500 feet. After an excellent supper our distinguished passengers smoked a final cigar and having sighted the French coast have now gone to bed to rest after the excitement of their leave-taking. All essential services are functioning satisfactorily.’

Two hours later, R101 went into a steep dive, the nose hitting the ground at just 13.8mph. Then fire broke out, from which only eight of the 56 passengers and crew survived. Plans for more advanced and bigger airships were scrapped. After a brief resurgence during World War II when they made barrage balloons for the war effort, the Cardington sheds and the industry slid into decline.

Now, Cardington shed No 2 acts as a temporary home to Warner Brothers’ technicians. The cavernous space was just the job for a full-sized mock-up of Gotham  City for Christopher Nolan’s epic Batman series. The other largely derelict shed is out of bounds, a reminder of the industry’s capricious history.

How the new breed of Hybrid Air Vehicles would look over London’s Olympic complex

But just as cruise ships survived the Titanic disaster, so some enthusiasts never gave up hope for the airship. Among them was Roger Munk, the epitome of a charismatic British engineering visionary. The idea for the Hybrid Air Vehicle was his; he spent much of his  40-year career designing and building airships, completing a number of ‘lighter than air’ projects for the American military.

Yet his own work was haunted by the inherent danger of airships going up in flames. In 1995, a fire apparently caused accidentally during welding work set alight the Weeksville hangar in North Carolina. At half-a-mile long, it was the largest wood-construction building in the world. Supports for the 180-ton doors were being rebuilt when the fire took hold, burning the hangar to the ground and destroying his Sentinel 1000 blimp.

Munk refused to give up. He decided to begin a new project creating a vehicle that would solve some of the problems inherent in airships, especially ground handling and ballast issues. He based his 15-man team in portable huts in the shadow of the Cardington sheds, and went back to the drawing board.

With a small beer tent as a hangar, Munk created the concept of a hybrid. The first prototype was flown in 2000. Though Munk was able to oversee the final perfection of his vision, he died of a heart attack in February 2010 – before the team heard news that they had won the U.S. military contract.

The team now has 100 engineers and designers and the firm has ditched its draughty sheds for two brand new office buildings nearby. But if Hybrid Air Vehicles’ potential is taken up then the team hopes to begin manufacturing and storing the vehicles again in Cardington.

The 50ft long prototype itself seems otherworldly. Almost as wide as it is long, it is surprisingly balloon-like to the touch. Even the most cynical observer cannot disguise the thrill of childlike wonder on feeling just how light this huge craft is. The pressure inside it is just 0.1 psi – a car tyre is between 20 and 40 psi.

CEO Gary Elliott, the man largely responsible for putting together the Northrop Grumman deal, says: ‘We took existing technologies and the concept of an airship, took a step back and thought – why don’t we do this and this differently, so that it projects itself through the air?’

The Hybrid Air Vehicles’ flight simulator

In a nearby office a team of flight-control specialists occupies a meeting room. In the corner of another office sits a full-size mock-up of the cockpit, constructed entirely from cardboard. The cabinetry is the work of the team’s 70-year-old handyman.

Pilots sit here and try out all possible instrumentation combinations to find the most practical configuration. Who needs  virtual reality when you have a few old computer boxes and some photocopied instruments?

A few footsteps away, though, there is a concession to technology – a large simulator which operates using four screens linked to four networked, high-end gaming PCs. Veteran airship pilots, recruited from across the industry, with experience flying blimps and seaplanes, are teaching the computers how to react to various flying situations, so that when a remote operator issues the ship with a command, the automated system will be able to move the controls in the same way as a human pilot; in other words, they are teaching it to fly itself.

The system has been designed by another UK company, Blue Bear Systems Research. It designed the flight-control system of the Harrier jump jet and also designs UAVs (unmanned aerial vehicles) that can be launched and fly themselves autonomously along a pre-programmed route.

Although every Hybrid Air Vehicle (HAV) will be capable of being flown remotely as a military surveillance platform, it will also be able to operate with a three-man crew – a pilot, co-pilot and load master. It takes about 100 hours of flight training to convert a pilot, though they don’t all make the switch easily, often because they aren’t used to stopping in mid-air.

Dave Burns is a pilot with thousands of hours experience flying passenger airliners for BA and Monarch. He is the company’s test pilot and chief flight training officer, and also the man who will fly the HAV 304 back across the Atlantic.

‘It doesn’t respond like a plane at all,’ says Burns. ‘You move the stick, telling the ship to move, and nothing happens for three or four seconds – and then it responds, which can be a little disconcerting. Plus, the mass underneath it acts like a pendulum, always trying to make it come level again.

‘The difficult thing is landing and take-off. In the past airships had ropes and ground crew waiting; we don’t need those so now what you have to do is present the vehicle so it comes down very slowly.’

A German Graf Zeppelin visiting Britain in 1931

Although the first 300ft version of the craft has been commissioned by the U.S. military, the real commercial potential of the vehicles could be for heavy lifting, says director of sales Gordon Taylor who has been living and breathing the things through multiple prototypes since joining his friend Roger Munk in 1997.

‘Our hybrids are based on a blend of technologies, in the same way that a Toyota Prius is a hybrid because it runs on electricity and petrol,’ he says.

‘Firstly it uses aerodynamics. The shape is like a big wing – air moves over it, lower air pressure is created across the top of the wing and it creates lift. Only if it’s fully loaded does it need a runway, and even then, with a 20 knot headwind they can land in three hull lengths.

‘Secondly we use “lighter-than-air” technology. With a normal airship you moor it on the ground to a mast. In order to fly anywhere it has to take off ballast, then it floats up. In a hybrid we push ourselves forward and that immediately generates lift.

‘Thirdly we have vectored thrust: our propulsion ducts rotate like a jump jet. Finally, we have hovercraft-style landing gear – a cushion of air that means that you can land on any reasonably flat surface, including water. This also works in reverse to secure the vehicle to the ground by suction.’

The company has calculated that it would take only 20 minutes to move a shipping container from Milton Keynes to London by HAV – a journey that presently takes hours thanks to traffic. Add a road network that grinds to a halt after a seasonal dusting of snow and you suddenly find an application for a cheaper, faster form of transport.

The Hindenburg disaster at Lakehurst, New Jersey in 1937 which marked the end of the era of passenger-carrying airships

‘You can forget ice road truckers too in places with more extreme cold,’ he adds.

‘They can carry the same load that goes on the back of those trucks and they love the cold because you get more lift in the denser air. We have a version with a 20-ton payload, which is what a Lockheed C-130 Hercules carries. We have plans for craft to eventually carry up to 1,000 tons.’

The team is already in formal discussions with oil companies that routinely spend hundreds of millions of dollars on roads and airports every time they find a new supply of oil or gas. By using HAVs the oil companies would simply be able to touch down without need of an airport.

‘Some of these companies are paying a million dollars a day in the development of infrastructure.

‘You could run these hybrids in convoy too, of course. The price difference between air freight and shipping is huge – so what if you could move freight by air but for a similar price as a ship? It could mean a whole new market in transport.’

Later this year the full-scale version of the current prototype will become the largest flying object in the world. After its initial use in military surveillance and heavy lifting, it could be just a few years before passengers are floating around beneath them. Need to be in New York fast? Take a plane. Don’t mind being in New York a day later? Then take an HAV.

And precisely how long will it take after that  for us to see a fleet of orange easyBalloons hauling budget passengers to and from Malaga?

Robbie Cooper for the New York Times

The end of a 12-hour shift at Donghua Networks in Jinhua, China.

Twelve hours a night, seven nights a week, with only two or three nights off per month, this is what Li does — for a living. On this summer night in 2006, the game on his screen was, as always, World of Warcraft, an online fantasy title in which players, in the guise of self-created avatars — night-elf wizards, warrior orcs and other Tolkienesque characters — battle their way through the mythical realm of Azeroth, earning points for every monster slain and rising, over many months, from the game’s lowest level of death-dealing power (1) to the highest (70). More than eight million people around the world play World of Warcraft — approximately one in every thousand on the planet — and whenever Li is logged on, thousands of other players are, too. They share the game’s vast, virtual world with him, converging in its towns to trade their loot or turning up from time to time in Li’s own wooded corner of it, looking for enemies to kill and coins to gather. Every World of Warcraft player needs those coins, and mostly for one reason: to pay for the virtual gear to fight the monsters to earn the points to reach the next level. And there are only two ways players can get as much of this virtual money as the game requires: they can spend hours collecting it or they can pay someone real money to do it for them.

At the end of each shift, Li reports the night’s haul to his supervisor, and at the end of the week, he, like his nine co-workers, will be paid in full. For every 100 gold coins he gathers, Li makes 10 yuan, or about $1.25, earning an effective wage of 30 cents an hour, more or less. The boss, in turn, receives $3 or more when he sells those same coins to an online retailer, who will sell them to the final customer (an American or European player) for as much as $20. The small commercial space Li and his colleagues work in — two rooms, one for the workers and another for the supervisor — along with a rudimentary workers’ dorm, a half-hour’s bus ride away, are the entire physical plant of this modest $80,000-a-year business. It is estimated that there are thousands of businesses like it all over China, neither owned nor operated by the game companies from which they make their money. Collectively they employ an estimated 100,000 workers, who produce the bulk of all the goods in what has become a $1.8 billion worldwide trade in virtual items. The polite name for these operations is youxi gongzuoshi, or gaming workshops, but to gamers throughout the world, they are better known as gold farms. While the Internet has produced some strange new job descriptions over the years, it is hard to think of any more surreal than that of the Chinese gold farmer.

The market for massively multiplayer online role-playing games, known as M.M.O.’s, is a fast-growing one, with no fewer than 80 current titles and many more under development, all targeted at a player population that totals around 30 million worldwide. World of Warcraft, produced in Irvine, Calif., by Blizzard Entertainment, is one of the most profitable computer games in history, earning close to $1 billion a year in monthly subscriptions and other revenue. In a typical M.M.O., as in a classic predigital role-playing game like Dungeons & Dragons, each player leads his fantasy character on a life of combat and adventure that may last for months or even years of play. As has also been true since D. & D., however, the romance of this imaginary life stands in sharp contrast to the plodding, mathematical precision with which it proceeds.

Players of M.M.O.’s are notoriously obsessive gamers, not infrequently dedicating more time to the make-believe careers of their characters than to their own real jobs. Indeed, it is no mere conceit to say that M.M.O.’s are just as much economies as games. In every one of them, there is some form of money, the getting and spending of which invariably demands a lot of attention: in World of Warcraft, it is the generic gold coin; in Korea’s popular Lineage II, it is the “adena”; in the Japanese hit Final Fantasy XI, it is called “gil.” And in all of these games, it takes a lot of this virtual local currency to buy the gear and other battle aids a player needs to even contemplate a run at the monsters worth fighting. To get it, players have a range of virtual income-generating activities to choose from: they can collect loot from dead monsters, of course, but they can also make weapons, potions and similarly useful items to sell to other players or even gather the herbs and hides and other resources that are the crafters’ raw materials. Repetitive and time-intensive by design, these pursuits and others like them are known collectively as “the grind.”

For players lacking time or patience for the grind, there has always been another means of acquiring virtual loot: real money. From the earliest days of M.M.O.’s, players have been willing to trade their hard-earned legal tender — dollars, euros, yen, pounds sterling — for the fruits of other players’ grinding. And despite strict rules against the practice in the most popular online games, there have always been players willing to sell. The phenomenon of selling virtual goods for real money is called real-money trading, or R.M.T., and it first flourished in the late 1990s on eBay. M.M.O. players looking to sell their virtual armor, weapons, gold and other items would post them for auction and then, when all the bids were in and payment was made, arrange with the highest bidder to meet inside the game world and transfer the goods from the seller’s account to the buyer’s.

Until very recently, in fact, eBay was a major clearinghouse for commodities from every virtual economy known to gaming — from venerable sword-and-sorcery stalwarts EverQuest and Ultima Online to up-and-comers like the Machiavellian space adventure Eve Online and the free-form social sandbox Second Life. That all came to an official end this January, when eBay announced a ban on R.M.T. sales, citing, among other concerns, the customer-service issues involved in facilitating transactions that are prohibited by the gaming companies. But by then the market had long since outgrown the tag-sale economics of online auctions. For years now, the vast majority of virtual goods has been brought to retail not by players selling the product of their own gaming but by high-volume online specialty sites like the virtual-money superstores IGE, BroGame and Massive Online Gaming Sales — multimillion-dollar businesses offering one-stop, one-click shopping and instant delivery of in-game cash. These are the Wal-Marts and Targets of this decidedly gray market, and the same economic logic that leads conventional megaretailers to China in search of cheap toys and textiles takes their virtual counterparts to China’s gold farms.

Indeed, on the surface, there is little to distinguish gold farming from toy production or textile manufacture or any of the other industries that have mushroomed across China to feed the desires of the Western consumer. The wages, the margins, the worker housing, the long shifts and endless workweeks — all of these are standard practice. Like many workers in China today, most gold farmers are migrants. Li, for example, came to Nanjing, in the country’s industry-heavy coastal region, from less prosperous parts. At 30, he is old for the job and feels it. He says he hopes to marry and start a family, he told me, but doesn’t see it happening on his current wages, which are not much better than what he made at his last job, fixing cars. The free company housing means his expenses aren’t high — food, cigarettes, bus fare, connection fees at the local wang ba (or Internet cafe) where he goes to relax — but even so, Li said, it is difficult to set aside savings. “You can do it,” he said, “but you have to economize a lot.”

This is the quick-sketch picture of the job, however, and it misses much. To sit at Li’s side for an hour or two, amid the dreary, functional surroundings of his workplace, as he navigates the Technicolor fantasy world he earns his living in, is to understand that gold farming isn’t just another outsourced job.

When the night shift ends and the sun comes up, Li and his co-workers know it only by the slivers of daylight that slip in at the edges of the plastic sheeting taped to the windows against the glare. As Li clocks out, another worker takes his seat, takes control of his avatar and carries on with the same grim routines amid the warrior monks of Azeroth. On most days Li’s replacement is 22-year-old Wang Huachen, who has been at this gold farm for a year, ever since he completed his university course in law. Soon, Wang told me, he will take the test for his certificate to practice, but he seems in no particular hurry to.

“I will miss this job,” he said. “It can be boring, but I still have sometimes a playful attitude. So I think I will miss this feeling.”

Two workstations away, Wang’s co-worker Zhou Xiaoguang, who is 24, also spends the day shift massacring monks. To watch his face as he plays, you wouldn’t guess there was anything like fun involved in this job, and perhaps “fun” isn’t exactly the word. As anyone who has spent much time among video-gamers knows, the look on a person’s face as he or she plays can be a curiously serious one, reflective of the absorbing rigors of many contemporary games. It is hard, in any case, for Zhou to say where the line between work and play falls in a gold farmer’s daily routines. “I am here the full 12 hours every day,” he told me, offhandedly killing a passing deer with a single crushing blow. “It’s not all work. But there’s not a big difference between play and work.”

I turned to Wang Huachen, who remained intent on manipulating an arsenal of combat spells, and asked again how it was possible that in these circumstances anybody could, as he put it, “have sometimes a playful attitude”?

He didn’t even look up from his screen. “I cannot explain,” he said. “It just feels that way.”

In 2001, Edward Castronova, an economist at Indiana University and at the time an EverQuest player, published a paper in which he documented the rate at which his fellow players accumulated virtual goods, then used the current R.M.T. prices of those goods to calculate the total annual wealth generated by all that in-game activity. The figure he arrived at, $135 million, was roughly 25 times the size of EverQuest’s R.M.T. market at the time. Updated and more broadly applied, Castronova’s results suggest an aggregate gross domestic product for today’s virtual economies of anywhere from $7 billion to $12 billion, a range that puts the economic output of the online gamer population in the company of Bolivia’s, Albania’s and Nepal’s.

Not quite the big time, no, but the implications are bigger, perhaps, than the numbers themselves. Castronova’s estimate of EverQuest’s G.D.P. showed that online games — even when there is no exchange of actual money — can produce actual wealth. And in doing so Castronova also showed that something curious has happened to the classic economic distinction between play and production: in certain corners of the world, it has melted away. Play has begun to do real work.

This development has not been universally welcomed. In the eyes of many gamers, in fact, real-money trading is essentially a scam — a form of cheating only slightly more refined than, say, offering 20 actual dollars for another player’s Boardwalk and Park Place in Monopoly. Some players, and quite a few game designers, see the problem in more systemic terms. Real-money trading harms the game, they argue, because the overheated productivity of gold farms and other profit-seeking operations makes it harder for beginning players to get ahead. Either way, the sense of a certain economic injustice at work breeds resentment. In theory this resentment would be aimed at every link in the R.M.T. chain, from the buyers to the retailers to the gold-farm bosses. And, indeed, late last month American WoW players filed a class-action suit against the dominant virtual-gold retailer, IGE, the first of its kind.

But as a matter of everyday practice, it is the farmers who catch it in the face. Consider, for example, a typical interlude in the workday of the 21-year-old gold farmer Min Qinghai. Min spends most of his time within the confines of a former manufacturing space 200 miles south of Nanjing in the midsize city of Jinhua. He works two floors below the plywood bunks of the workers’ dorm where he sleeps. In two years of 84-hour farming weeks, he has rarely stepped outside for longer than it takes to eat a meal. But he has died more times than he can count. And last September on a warm afternoon, halfway between his lunch and dinner breaks, it was happening again.

The World of Warcraft monsters he faces down — ferocious, gray-furred warriors of the Timbermaw clan of bearmen — are no match for his high-level characters, but they do fight back and sometimes they get the better of him. And so it appeared they had just done. Distracted from his post for a moment, Min returned to find his hunter-class character at the brink of death, the scene before him a flurry of computer-animated weapon blows. It wasn’t until the fight had run its course and the hunter lay dead that Min could make out exactly what had happened. The game’s chat window displayed a textual record of the blows landed and the cost to Min in damage points. The record was clear: the monsters hadn’t acted alone. In the middle of the fight another player happened by, sneaked up on Min and brought him down.

Min leaned back and stretched, then set about the tedious business of resurrecting his character, a drawn-out sequence of operations that can put a player out of action for as long as 10 minutes. In farms with daily production quotas, too much time spent dead instead of farming gold can put the worker’s job at risk. And in shops where daily wages are tied to daily harvests, every minute lost to death is money taken from the farmer’s pocket. But there are times when death is more than just an economic setback for a gold farmer, and this was one of them. As Min returned to his corpse — checking to make sure his attacker wasn’t waiting around to fall on him again the moment he resurrected — what hurt more than the death itself was how it happened, or more precisely, what made it happen: another player.

It isn’t that WoW players don’t frequently kill other players for fun and kill points. They do. But there is usually more to it when the kill in question is a gold farmer. In part because gold farmers’ hunting patterns are so repetitive, they are easy to spot, making them ready targets for pent-up anti-R.M.T. hostility, expressed in everything from private sarcastic messages to gratuitous ambushes that can stop a farmer’s harvesting in its tracks. In homemade World of Warcraft video clips that circulate on YouTube or GameTrailers, with titles like “Chinese Gold Farmers Must Die” and “Chinese Farmer Extermination,” players document their farmer-killing expeditions through that same Timbermaw-ridden patch of WoW in which Min does his farming — a place so popular with farmers that Western players sometimes call it China Town. Nick Yee, an M.M.O. scholar based at Stanford, has noted the unsettling parallels (the recurrence of words like “vermin,” “rats” and “extermination”) between contemporary anti-gold-farmer rhetoric and 19th-century U.S. literature on immigrant Chinese laundry workers.

Min’s English is not good enough to grasp in all its richness the hatred aimed his way. But he gets the idea. He feels a little embarrassed around regular players and sometimes says he thinks about how he might explain himself to those who believe he has no place among them, if only he could speak their language. “I have this idea in mind that regular players should understand that people do different things in the game,” he said. “They are playing. And we are making a living.”

It is a distinction that game companies understand all too well. Like the majority of M.M.O. companies, Blizzard has chosen to align itself with the customers who abhor R.M.T. rather than the ones who use it. A year ago, Blizzard announced it had identified and banned more than 50,000 World of Warcraft accounts belonging to farmers. It was the opening salvo in a continuing eradication campaign that has effectively swept millions in farmed gold from the market, sending the exchange rate rocketing from a low of 6 cents per gold coin last spring to a high of 35 cents in January.

Of course, nobody expected the farmers’ equally rule-breaking customers to be punished too. Among players, the R.M.T. debate may revolve around questions of fairness, but among game companies, the only question seems to be what is good for business. Cracking down on R.M.T. buyers makes poorer marketing sense than cracking down on sellers, in much the same way that cracking down on illegal drug suppliers is a better political move than cracking down on users. (Only a few companies have found a way to make R.M.T. part of their business model. Sony Online Entertainment, which publishes EverQuest, has started earning respectable revenues from an experimental in-game auction system that charges players a small transaction fee for real-money trades.) As Mark Jacobs, vice president at Electronic Arts and creator of the classic M.M.O. Dark Age of Camelot, put it: “Are you going to get more sympathy from busting 50,000 Chinese farmers or from busting 10,000 Americans that are buying? It’s not a racial thing at all. If you bust the buyers, you’re busting the guys who are paying to play your game, who you want to keep as customers and who will then go on the forums and say really nasty things about your company and your game.”

The cost to farmers of being expelled from WoW can be steep. At the very least, it means a temporary drop in productivity, because the character has to be to built up all over again, as well as the loss of all the loot accumulated in that character’s account. Given the stakes, some Chinese gold farms have found that the best way to get around their farmers’ pursuers is to make it hard to distinguish professionals from players in the first place. One business that specializes in doing just that is located a few blocks from the gold farm where Min Qinghai works. The shop floor is about the same size, with about the same number of computers in the same neat rows, but you can tell just walking through the place that it is a more serious operation. For one thing, there are a lot more workers: typically 25 on the day shift, 25 on the night shift, each crew punching in and out at a time clock just inside the entrance. Nobody works without a shirt here; quite a few, in fact, wear a standard-issue white polo shirt with the company initials on it. There is also a crimson version of the shirt, reserved for management and worn at all times by the shift supervisor, who, when he isn’t prowling the floor, sits at his desk before a broad white wall emblazoned with foot-high Chinese characters in red that spell: unity, collaboration, integrity, efficiency.

The name of the business is Donghua Networks, and its specialty is what gamers call “power leveling.” Like regular gold farming, power leveling offers customers an end run around the World of Warcraft grind — except that instead of providing money and other items, the power leveler simply does the work for you. Hand over your account name, password and about $300, and get on with your real life for a while: in a marathon of round-the-clock monster-bashing, a team of power levelers will raise your character from the lowest level to the highest, accomplishing in four weeks or less what at a normal rate of play would take at least four months.

For Donghua’s owners — 26-year-old Fei Jianfeng and 36-year-old Bao Donghua, both former gold-farm wage workers themselves — moving the business out of farming and into leveling was an easy call. Among other advantages, they say, power leveling means fewer banned accounts. Because the only game accounts used are the customers’ own, there is much less risk of losing access to the virtual work site. For their workers, however, the advantages are mixed. Though there is a greater variety of quests and quarries to pursue, the pay isn’t any better, and some workers chafe at the constraints of playing a stranger’s character, preferring the relative autonomy of farming gold.

As one Donghua power leveler said of his old gold-farming job, “I had more room to play for myself.”

It may seem strange that a wage-working loot farmer would still care about the freedom to play. But it is not half as strange as the scene that unfolded one evening at 9 o’clock in the Internet cafe on the ground floor of the building where Donghua has its offices. Scattered around the stifling, dim wang ba, 10 power levelers just off the day shift were merrily gaming away. Not all of them were playing World of Warcraft. A big, silent lug named Mao sat mesmerized by a very pink-and-purple Japanese schoolgirls’ game, in which doe-eyed characters square off in dancing contests with other online players. But the rest had chosen, to a man, to log into their personal World of Warcraft accounts and spend these precious free hours right back where they had spent every other hour of the day: in Azeroth.

Such scenes are not at all unusual. At the end of almost any working day or night in a Chinese gaming workshop, workers can be found playing the same game they have been playing for the last 12 hours, and to some extent gold-farm operators depend on it. The game is too complex for the bosses to learn it all themselves; they need their workers to be players — to find out all the tricks and shortcuts, to train themselves and to train one another. “When I was a worker,” Fan Yangwen, who is now 21 and in Donghua’s main office providing technical support, told me, “I loved to play because when I was playing, I was learning.” But learning to play or learning to work? I asked. Fan shrugged. “Both.”

Fan himself is a striking case of how off-hours play can serve as a kind of unpaid R. and D. lab for the farming industry. He is that rarest of World of Warcraft obsessives, a Chinese gold farmer who has actually bought farmed gold. (“Sure, I bought 10,000 once,” he said, “I don’t have time to farm all that!”) When Fan shows up at the wang ba after work, it is a minor event; the other Donghua workers pull their chairs over to watch him play — his top-level warlock character is an unbelievable powerhouse that no amount of money, real or virtual, can buy.

What makes Fan’s dominance so impressive to his peers is that he achieved it in regions of the game that are all but inaccessible to the working gold farmer or power leveler. Therein lies what is known as the end game, the phase of epic challenges that begins only when the player has accumulated the maximum experience points and can level up no more. The rewards for meeting these challenges are phenomenal: rare weapons and armor pieces loaded with massive power boosts and showy graphics. And the greatest cannot be traded or given away; they can only be acquired by venturing into the game’s most difficult dungeons. That requires becoming part of a tightly coordinated “raid” group of as many as 40 other players (any fewer than that, and the entire group will almost certainly “wipe” — or die en masse without killing any monsters of note). Each player has a shot at the best items when they drop, and players must negotiate among themselves for the top prizes. These end-game hurdles have some subtle but significant effects. For one thing, they force the growth of “guilds” — teams of dozens, sometimes hundreds, of players who join together to hit high-end dungeons on a regular basis. For another, they shut farmers out from an entire class of virtual goods — the most marketable in the game if only they could be traded.

For a long time the Donghua bosses, Fei and Bao (known even to employees as Little Bai and Brother Bao), could do no more than nurse their envy of the raiding guilds’ access to the end game. But Fan’s prowess pointed to another way of looking at it: raiding guilds weren’t the competition, they realized; they were the solution. Donghua would put together a team of 40 employees. They would train the team in all the hardest dungeons. And then, for a few hundred dollars, the team would escort any customer into the dungeon of his or her choice. And when the customer’s longed-for item dropped, the team would stand aside and let the customer take it, no questions asked. Thus would the supposedly unmarketable end-game treasures find their way into the R.M.T. market. And thus would gold farming, of a sort, find its way at last into the end game.

When Brother Bao and Little Bai put their team together in April of last year, Min Qinghai, a veteran Donghua employee at the time, was among the first to make the roster.

“Before I joined the raiding team, I’d never worked together with so many people,” Min told me. They were 40 young men in three adjoining office spaces, and it was chaotic at first. Two or three supervisors moved among them, calling out orders like generals. A dungeon raid is always a puzzle: figuring out which tactics to use to kill each boss is the main challenge; doing so while coordinating 40 players can be dizzying. But members of the team raided just as diligently as they had power-leveled: 12 hours a day, 7 days a week, making their way through the complexities of a different dungeon every day.

There was a lot of shouting involved, at least in the beginning. Besides the orders called out by the supervisors, there were loud attempts at coordination among the team members themselves. “But then we developed a sense of cooperation, and the shouting grew rarer,” Min said. “By the end, nothing needed to be said.” They moved through the dungeons in silent harmony, 40 intricately interdependent players, each the master of his part. For every fight in every dungeon, the hunters knew without asking exactly when to shoot and at what range; the priests had their healing spells down to a rhythm; wizards knew just how much damage to put in their combat spells.

And Min’s role? The translator struggled for a moment to find the word in English, and when I hazarded a guess, Min turned directly to me and repeated it, the only English I ever heard him speak. “Tank,” he said, breaking into a rare, slow smile, and why wouldn’t he? The tank — the heavily armored warrior character who holds the attention of the most powerful enemy in the fight, taking all its blows — is the linchpin of any raid. If the tank dies, everybody else will soon die too, as a rule.

“Working together, playing together, it felt nice,” Min said. “Very . . . shuang.” The word means “open, clear, exhilarating.” “You would go in, knowing that you were fighting the bosses that all the guilds in the world dream of fighting; there was a sense of achievement.”

The end arrived without warning. One day word came down from the bosses that the 40-man raids were suspended indefinitely for lack of customers. In the meantime, team members would go back to gold farming, gathering loot in five-man dungeons that once might have thrilled Min but now presented no challenge whatsoever. “We no longer went to fight the big boss monsters,” Min said. “We were ordered to stay in one place doing the same thing again and again. Everyday I was looking at the same thing. I could not stand it.”

Min quit and took the farming job he works at still. The new job, with its rote Timbermaw whacking, could hardly be less exciting. But it is more relaxed than Donghua was, less wearying — “Working 12 hours there was like working 24 here” — and he couldn’t have stayed on in any case, surrounded by reminders of the broken promise of tanking for what might have been the greatest guild on Earth.

In the meantime, Min is doing his best to forget that his work has anything at all to do with play or that he ever let himself believe otherwise. But even with a job as monotonous as this one, it isn’t easy. On his usual hunt one day, he accidentally backed into combat with a higher-level monster. Losing life fast, he grabbed his mouse and started to flee. He hunched over his keyboard, leaning into his flight, flushed now by the chase. His boss, 26-year-old Liu Haibin, an inveterate gamer himself, wandered by and began to cheer him on: “Yeah, yeah, yeah . . . go!”

Finally the monster quit the chase, and Min got away with no consequence more untoward than having to explain himself. “It’s instinctual — you can’t help it,” he said. “You want to play.”

From Wired Magazine

It’s in the middle of a grassy field, shiny sedans behind gleaming glass walls. Right next door is another luxury car dealership selling a variety of other high-end European rides. It’s as if the sheer magic of wealth has shimmered the glass-and-steel buildings into being.

In fact, expensive cars choke the streets of Râmnicu Vâlcea’s bustling city center—top-of-the-line BMWs, Audis, and Mercedes driven by twenty- and thirtysomething men sporting gold chains and fidgeting at red lights. I ask my cab driver if these men all have high-paying jobs, and he laughs. Then he holds up his hands, palms down, and wiggles his fingers as if typing on a keyboard. “They steal money on the Internet,” he says.

Among law enforcement officials around the world, the city of 120,000 has a nickname: Hackerville. It’s something of a misnomer; the town is indeed full of online crooks, but only a small percentage of them are actual hackers. Most specialize in ecommerce scams and malware attacks on businesses. According to authorities, these schemes have brought tens of millions of dollars into the area over the past decade, fueling the development of new apartment buildings, nightclubs, and shopping centers. Râmnicu Vâlcea is a town whose business is cybercrime, and business is booming.

At a restaurant in a neighborhood of apartment buildings and gated bungalows, I meet Bogdan Stoica and Alexandru Frunza, two of just four local cops on the digital beat. Stoica, 32, is square-shouldered and stocky, with a mustache and prominent stubble. His expression rarely changes. Frunza, 29, is tall and clean shaven. He’s the funny one. “My English will improve after I have a few beers,” he says. We sit at a table on the edge of a big courtyard, piped-in Romanian pop music blaring.

Stoica and Frunza grew up in Râmnicu Vâlcea. “The only cars on the streets were those made by Dacia,” Stoica says, referring to the venerable Romanian carmaker. Access to information was limited, too: Weekday television consisted of two hours of state-run programming, mostly devoted to covering the dictator, Nicolae Ceauşescu. “We had half an hour of cartoons on Sunday,” Stoica says.

In 1989, a revolution that began with anti-government riots ended with the execution of Ceauşescu and his wife, and the country began the switch to a market economy. By 1998, when Stoica finished high school and went off to the police academy in Bucharest, another revolution was beginning: the Internet. Râmnicu Vâlcea was better off than many towns in this relatively poor country—it had a decades-old chemical plant and a modest tourism industry. But many young men and women struggled to find work.

No one really knows how or why those kids started scamming people on the Internet. “If you find out, you let us know,” says Codruţ Olaru, head of Romania’s Directorate for Investigation on Organized Crime and Terrorism. Whatever the reason, online crime was widespread by 2002. Cybercafés offered cheap Internet access, and crooks in Râmnicu Vâlcea got busy posting fake ads on eBay and other auction sites to lure victims into remitting payments by wire transfer. Eventually, FBI agents in the US and Bucharest started to get interested.

In the early days, the perpetrators weren’t exactly geniuses. One of the first cases out of the region involved a team based in the neighboring town of Piteşti. One crook would post ads for cell phones; the other picked up the wired money for orders that would never ship. The two men had made a few hundred dollars from victims in the US, and the guy receiving the cash hadn’t even bothered to use a fake ID. “I found him sitting in an Internet café, chatting online,” says Costel Ion, a Piteşti cop who had been working the cybercrime beat. “He just confessed.”

But as in any business, the scammers innovated and adapted. One early advance was establishing fake escrow services: Victims would be asked to send payments to these supposedly trustworthy third parties, which had websites that made them look like legitimate companies. The scams got better over the years, too. To explain unbelievably low prices for used cars, for example, a crook would pose as a US soldier stationed abroad, with a vehicle in storage back home that he had to sell. (That tale also established a plausible US contact to receive the money, instead of someone in Romania.) In the early years, the thieves would simply ask for advance payment for the nonexistent vehicle. As word of the scam spread, the sellers began offering to send the cars for inspection—asking for no payment except “shipping.”

The con artists got even sneakier. “They learned to create scenarios,” says Michael Eubanks, an FBI agent in Bucharest. “We’ve seen email between criminals with instructions on how to respond to different questions.” The scammers started hiring English speakers to craft emails to US targets. Specialists emerged to occupy niches in the industry, designing fake websites or coordinating low-level confederates.

Internet scammers and their underlings have turned Râmnicu Vâlcea into a hub of international organized crime.
Photo: Nick Waplington

By 2005, Romania had become widely known as a haven for online fraud, and buyers became wary of sending money there. The swindlers adapted again, arranging for payments to be wired to other European countries, where accomplices picked up the cash. A new entry level evolved, people who’d act as couriers and money launderers for a cut of the take. These money mules were called arrows, and their existence elevated Râmnicu Vâlcea to a hub of international organized crime.

Many arrows were Romanians living in Western Europe and the US; some were youngsters from Râmnicu Vâlcea who had moved overseas expressly for the job. They’d go to wire transfer offices to collect remittances from victims, then turn around and wire that money—minus a commission—to Râmnicu Vâlcea or to other arrows in the network. The system served as a kind of firewall, making it much more difficult for law enforcement to track the masterminds.

Back home, the local police were starting to realize they needed people on the cybercrime beat full-time. Frunza, who’d studied informatics in high school before attending the police academy, was working drug cases in Bucharest when he decided to come home. He ended up joining Stoica on the hunt for online con artists. The two learned that suspects expect leniency from the police because their crimes target only foreigners. “The guys will often say, ‘I am not stealing from our countrymen,’” Frunza says. “But a crime is a crime. You have to pay for it.”

Nowadays, Stoica and Frunza occasionally find themselves investigating a childhood acquaintance or, conversely, running into known criminals in social situations. Frunza used to play on the same soccer team as a suspect who was under surveillance. Those connections have helped the two cops pose a formidable challenge to the industry.

A little after 11 pm, Stoica hushes our conversation and tells me to turn around and check out a table across the courtyard, where a small group of flashily dressed young men has just arrived with two blond women who seem barely out of their teens. The men are all under investigation. “It’s a small city,” Stoica says.

The sudden appearance of luxury car dealerships among the grass fields marks the entrance into Râmnicu Vâlcea.
Photo: Nick Waplington

Defining the town center of Râmnicu Vâlcea is a towering shopping mall that looks like a giant glass igloo. The streets are lined with gleaming storefronts—leather accessories, Italian fashions—serving a demand fueled by illegal income. Near the mall is a nightclub, now closed by police because its backers were shady. New construction grinds ahead on nearly every block. But what really stands out in Râmnicu Vâlcea are the money transfer offices. At least two dozen Western Union locations lie within a four-block area downtown, the company’s black-and-yellow signs proliferating like the Starbucks mermaid circa 2003.

Driving past a block of low-rise buildings with neatly trimmed hedges, Stoica notes a couple of apartments owned by people currently under investigation. “I don’t know if the people of Râmnicu Vâlcea are too smart or too stupid,” Stoica says grimly. “They talk a lot to each other. One guy learns the job from another. They ask their high school friends: ‘Hey, do you want to make some money? I want to use you as an arrow.’ Then the arrow learns to do the scams himself.”

It’s not so different from the forces that turn a neighborhood into, say, New York’s fashion district or the aerospace hub in southern California. “To the extent that some expertise is required, friends and family members of the original entrepreneurs are more likely to have access to those resources than would-be criminals in an isolated location,” says Michael Macy, a Cornell University sociologist who studies social networks. “There may also be local political resources that provide a degree of protection.”

Online thievery as a ticket to the good life spread from the early pioneers to scores of young men, infecting Râmnicu Vâlcea’s social fabric. The con artists were the ones with the nice cars and fancy clothes—the local kids made good. And just as in Silicon Valley, the clustering of operations in one place made it that much easier for more to get started. “There’s a high concentration of people offering the kinds of services you need to build a criminal scheme,” says Gary Dickson, an FBI agent who worked in Bucharest from 2005 to 2010. “If your specialty is auction frauds, you can find a money pick-up guy. If you’re a money pick-up guy, you can find a buyer for your services.”

Stoica and Frunza both complain that they’re fighting an unstoppable tide with limited resources. But they haven’t been entirely unsuccessful—in fact, the 2008 case that first revealed the anatomy of Râmnicu Vâlcea’s fraud networks stemmed from Stoica’s investigation of a young entrepreneur named Romeo Chita.

Stoica says Chita started out as an arrow in the UK, and he was good. He moved up the ranks and eventually hired a few friends to establish his own ring. The Romanian authorities began investigating him in 2006, when he started buying new cars every few months and going to clubs every night with no apparent source of legitimate income. Chita launched an Internet service provider called NetOne, which authorities believe he was using as a shelter for fraudulent activity. When cops wanted to identify his customers, Stoica says, Chita usually told them that NetOne didn’t keep records.

Western Union signs have multiplied downtown like the Starbucks mermaid circa 2003.
Photo: Nick Waplington

In January 2008, an informant gave Stoica the cell numbers of two men working for Chita. The police tapped the phones, and the next day one of the men sent Chita a text message with money transfer control numbers—unique numeric sequences required to pick up cash. Stoica and his team followed up with surveillance of Chita and his associates, which established what Stoica calls “the money circuit,” the route through which the funds flowed from victims in the US to Chita and others. Prosecutors now allege that the operation turned into something a little more sophisticated than the usual Râmnicu Vâlcea scam. For example, the case against them details a con known as spear phishing—sending email to US companies that appeared to be from the IRS, the Department of Justice, or some other agency. Through Trojan horses attached to these emails, Chita’s group could obtain the companies’ bank account numbers and passwords. Allegedly, they even hired people in Las Vegas—Stoica says some were homeless—to open fake corporate bank accounts and receive the money.

The same month that Stoica began pursuing Chita, a police officer stopped a car for speeding in the Westlake suburb of Cleveland, Ohio. About to write a ticket, the cop noticed some drug paraphernalia in the car and arrested the two men inside. A further search turned up eight cell phones, two computers, fake IDs, two dozen money transfer receipts, and $63,000 in cash. The pair turned out to be Romanian and eventually confessed to being arrows for an organization authorities traced back to Chita. They had spent most of January driving around the Midwest, picking up money from various Western Union and MoneyGram locations. Their confessions led to more wiretaps and surveillance in the US and Romania over the following months, uncovering a network of at least two dozen accomplices.

That summer, Romanian authorities and FBI agents conducted a series of raids on both sides of the Atlantic. Chita spent 14 months in custody before being granted a provisional release pending the completion of his trial, still pending. On an org chart filed in Stoica’s office, Chita’s photo remains at the top.

Class Café is an inviting coffee shop with a terrace that overlooks a quiet street. It’s nearly empty when I walk in—just the owner behind the counter and a young couple at a corner table.

Stoica discouraged me from attempting this meeting, but I wanted to know what an alleged kingpin looks like. I ask the owner if he knows where Chita is, and he offers to call him. After a brief phone conversation, he hangs up and tells me that Chita is in Bucharest. I remind him that Chita isn’t allowed to leave Râmnicu Vâlcea under the terms of his release, and the owner smiles. He spends a few more minutes on the phone, then hangs up again and asks me to sit. Chita is on his way.

I take a table on the terrace. During our tour of town, Stoica had pointed out Chita’s silver Mercedes on the road, so I ignore the green Jaguar that drives up until a man in Bermuda shorts, canvas shoes, and a white T-shirt climbs out, enters the café, and approaches my table. He introduces himself as Chita’s brother, Marian. He licks his lips nervously and fidgets with an iPhone. “Chita’s coming,” he says after lighting a cigarette and making some phone calls. “But he’s a little drunk.”

A few minutes later, Chita walks around the corner and ambles into the café. Boyish, dressed in shorts, a light-blue polo shirt, and flip-flops, he looks more like a college student than a criminal mastermind. Despite the reputation of Râmnicu Vâlcea’s underworld as relatively free of violence, he has brought along some muscle—a young man in dark glasses with a big tattoo on his arm. The bodyguard slams a beer bottle down on the table and flexes his hand, as if getting ready for a boxing match.

Chita shakes my hand dourly and sits down next to me, looking away. Two other men join us. The young couple from the corner comes over to greet Chita with fawning smiles and handshakes. They clearly recognize him, too. The café owner gets up and leaves. As he walks away, he looks at me gravely and says, “Good luck.”

Râmnicu Vâlcea has become the Silicon Valley of online thievery— a place where the clustering of operations makes boot-strapping a criminal start-up easier.
Photo: Nick Waplington

The tattooed man leans toward me ominously. “Were you sent by Barack Obama?” he asks. I say that I wasn’t, and everyone but me lights cigarettes. Marian, getting increasingly jumpy, demands to know my true agenda. Finally, I spell my name and tell him to search for my stories on his iPhone. He Googles me and shows the screen to his brother. Everybody relaxes a bit, and I silently give thanks for wireless broadband.

Marian asks the young couple to translate for Chita, and they agree to stay. Chita has them tell me to stand, then he pats me down, asking if I’m wearing a wire.

“What do you say to the charges against you?” I ask.

“They are fake,” Chita says, in English.

Marian adds, “It’s all bullshit.” For clarification.

Chita continues with his defense in Romanian, and the couple translates enthusiastically. “He doesn’t even know how to speak English, so it is impossible for him to post ads or exchange email with buyers,” the young woman says. “He doesn’t even have an email address,” she says. “How can he do fraud on the Internet?”

I press Chita about the wiretapped conversations, but his tattooed bodyguard interrupts loudly. “You go back to your hotel room, we send you some nice pussy,” he says, raising his hand for a high five that I feel obligated to meet. The two men beside him laugh, and Chita takes a final drag from his cigarette before rising from his chair. He’s in no mood to discuss the evidence. “This interview is over,” Marian says.

They saunter out of the café and onto the sidewalk, looking surprisingly banal for guys accused of organized cybercrime, enjoying the good life with little effort or risk. Officials have dismantled a few fraud rings in recent years—there were just 188 arrests in all of Romania in 2010—but scores remain in business.

I am left with the friendly couple that helped with the translating. The young man says he’s heard about Chita from his friends and has seen his name in the papers. He tells me he has just received a diploma in engineering from an institution in Bucharest and is now looking for a job here in Râmnicu Vâlcea, his hometown. “I haven’t found anything yet,” he says. Thinking about Marian’s Jag and Chita’s Mercedes, I wonder if he’ll consider a job as an arrow. It’s like Frunza told me at the restaurant: “You arrest two of them and 20 new ones take their place,” he said. “We are two police officers, and they are 2,000.”

Yudhijit Bhattacharjee (yudhijit@gmail.com) is a staff writer at Science. He wrote about decoding a spy’s messages in issue 18.02.

From Wired Magazine by Steven Levy

But the warehouses aren’t meant to be understood by humans; they were built for bots. Every day, hundreds of robots course nimbly through the aisles, instantly identifying items and delivering them to flesh-and-blood packers on the periphery. Instead of organizing the warehouse as a human might—by placing like products next to one another, for instance—Diapers.com’s robots stick the items in various aisles throughout the facility. Then, to fill an order, the first available robot simply finds the closest requested item. The storeroom is an ever-shifting mass that adjusts to constantly changing data, like the size and popularity of merchandise, the geography of the warehouse, and the location of each robot. Set up by Kiva Systems, which has outfitted similar facilities for Gap, Staples, and Office Depot, the system can deliver items to packers at the rate of one every six seconds.

The Kiva bots may not seem very smart. They don’t possess anything like human intelligence and certainly couldn’t pass a Turing test. But they represent a new forefront in the field of artificial intelligence. Today’s AI doesn’t try to re-create the brain. Instead, it uses machine learning, massive data sets, sophisticated sensors, and clever algorithms to master discrete tasks. Examples can be found everywhere: The Google global machine uses AI to interpret cryptic human queries. Credit card companies use it to track fraud. Netflix uses it to recommend movies to subscribers. And the financial system uses it to handle billions of trades (with only the occasional meltdown).

This explosion is the ironic payoff of the seemingly fruitless decades-long quest to emulate human intelligence. That goal proved so elusive that some scientists lost heart and many others lost funding. People talked of an AI winter—a barren season in which no vision or project could take root or grow. But even as the traditional dream of AI was freezing over, a new one was being born: machines built to accomplish specific tasks in ways that people never could. At first, there were just a few green shoots pushing up through the frosty ground. But now we’re in full bloom. Welcome to AI summer.

Today’s AI bears little resemblance to its initial conception. The field’s trailblazers in the 1950s and ’60s believed success lay in mimicking the logic-based reasoning that human brains were thought to use. In 1957, the AI crowd confidently predicted that machines would soon be able to replicate all kinds of human mental achievements. But that turned out to be wildly unachievable, in part because we still don’t really understand how the brain works, much less how to re-create it.

So during the ’80s, graduate students began to focus on the kinds of skills for which computers were well-suited and found they could build something like intelligence from groups of systems that operated according to their own kind of reasoning. “The big surprise is that intelligence isn’t a unitary thing,” says Danny Hillis, who cofounded Thinking Machines, a company that made massively parallel supercomputers. “What we’ve learned is that it’s all kinds of different behaviors.”

AI researchers began to devise a raft of new techniques that were decidedly not modeled on human intelligence. By using probability-based algorithms to derive meaning from huge amounts of data, researchers discovered that they didn’t need to teach a computer how to accomplish a task; they could just show it what people did and let the machine figure out how to emulate that behavior under similar circumstances. They used genetic algorithms, which comb through randomly generated chunks of code, skim the highest-performing ones, and splice them together to spawn new code. As the process is repeated, the evolved programs become amazingly effective, often comparable to the output of the most experienced coders.

MIT’s Rodney Brooks also took a biologically inspired approach to robotics. His lab programmed six-legged buglike creatures by breaking down insect behavior into a series of simple commands—for instance, “If you run into an obstacle, lift your legs higher.” When the programmers got the rules right, the gizmos could figure out for themselves how to navigate even complicated terrain. (It’s no coincidence that iRobot, the company Brooks cofounded with his MIT students, produced the Roomba autonomous vacuum cleaner, which doesn’t initially know the location of all the objects in a room or the best way to traverse it but knows how to keep itself moving.)

The fruits of the AI revolution are now all around us. Once researchers were freed from the burden of building a whole mind, they could construct a rich bestiary of digital fauna, which few would dispute possess something approaching intelligence. “If you told somebody in 1978, ‘You’re going to have this machine, and you’ll be able to type a few words and instantly get all of the world’s knowledge on that topic,’ they would probably consider that to be AI,” Google cofounder Larry Page says. “That seems routine now, but it’s a really big deal.”

Even formerly mechanical processes like driving a car have become collaborations with AI systems. “At first it was the automatic braking system,” Brooks says. “The person’s foot was saying, I want to brake this much, and the intelligent system in the middle figured when to actually apply the brakes to make that work. Now you’re starting to get automatic parking and lane-changing.” Indeed, Google has been developing and testing cars that drive themselves with only minimal human involvement; by October, they had already covered 140,000 miles of pavement.

In short, we are engaged in a permanent dance with machines, locked in an increasingly dependent embrace. And yet, because the bots’ behavior isn’t based on human thought processes, we are often powerless to explain their actions. Wolfram Alpha, the website created by scientist Stephen Wolfram, can solve many mathematical problems. It also seems to display how those answers are derived. But the logical steps that humans see are completely different from the website’s actual calculations. “It doesn’t do any of that reasoning,” Wolfram says. “Those steps are pure fake. We thought, how can we explain this to one of those humans out there?”

The lesson is that our computers sometimes have to humor us, or they will freak us out. Eric Horvitz—now a top Microsoft researcher and a former president of the Association for the Advancement of Artificial Intelligence—helped build an AI system in the 1980s to aid pathologists in their studies, analyzing each result and suggesting the next test to perform. There was just one problem—it provided the answers too quickly. “We found that people trusted it more if we added a delay loop with a flashing light, as though it were huffing and puffing to come up with an answer,” Horvitz says.

But we must learn to adapt. AI is so crucial to some systems—like the financial infrastructure—that getting rid of it would be a lot harder than simply disconnecting HAL 9000’s modules. “In some sense, you can argue that the science fiction scenario is already starting to happen,” Thinking Machines’ Hillis says. “The computers are in control, and we just live in their world.” Wolfram says this conundrum will intensify as AI takes on new tasks, spinning further out of human comprehension. “Do you regulate an underlying algorithm?” he asks. “That’s crazy, because you can’t foresee in most cases what consequences that algorithm will have.”

In its earlier days, artificial intelligence was weighted with controversy and grave doubt, as humanists feared the ramifications of thinking machines. Now the machines are embedded in our lives, and those fears seem irrelevant. “I used to have fights about it,” Brooks says. “I’ve stopped having fights. I’m just trying to win.”

Senior writer Steven Levy (steven_levy@wired.com)

From the NewScientist

A STORM of scepticism has greeted experimental results emerging from the lab of a Nobel laureate which, if confirmed, would shake the foundations of several fields of science. “If the results are correct,” says theoretical chemist Jeff Reimers of the University of Sydney, Australia, “these would be the most significant experiments performed in the past 90 years, demanding re-evaluation of the whole conceptual framework of modern chemistry.”

Luc Montagnier, who shared the Nobel prize for medicine in 2008 for his part in establishing that HIV causes AIDS, says he has evidence that DNA can send spooky electromagnetic imprints of itself into distant cells and fluids. If that wasn’t heretical enough, he also suggests that enzymes can mistake the ghostly imprints for real DNA, and faithfully copy them to produce the real thing. In effect this would amount to a kind of quantum teleportation of the DNA.

Many researchers contacted for comment by New Scientist reacted with disbelief. Gary Schuster, who studies DNA conductance effects at Georgia Institute of Technology in Atlanta, compared it to “pathological science”. Jacqueline Barton, who does similar work at the California Institute of Technology in Pasadena, was equally sceptical. “There aren’t a lot of data given, and I don’t buy the explanation,” she says. One blogger has suggested Montagnier should be awarded an IgNobel prize.

Yet the results can’t be dismissed out of hand. “The experimental methods used appear comprehensive,” says Reimers. So what have Montagnier and his team actually found?

Full details of the experiments are not yet available, but the basic set-up is as follows. Two adjacent but physically separate test tubes were placed within a copper coil and subjected to a very weak extremely low frequency electromagnetic field of 7 hertz. The apparatus was isolated from Earth’s natural magnetic field to stop it interfering with the experiment. One tube contained a fragment of DNA around 100 bases long; the second tube contained pure water.

After 16 to 18 hours, both samples were independently subjected to the polymerase chain reaction (PCR), a method routinely used to amplify traces of DNA by using enzymes to make many copies of the original material. The gene fragment was apparently recovered from both tubes, even though one should have contained just water.

DNA was only recovered if the original solution of DNA – whose concentration has not been revealed – had been subjected to several dilution cycles before being placed in the magnetic field. In each cycle it was diluted 10-fold, and “ghost” DNA was only recovered after between seven and 12 dilutions of the original. It was not found at the ultra-high dilutions used in homeopathy.

Physicists in Montagnier’s team suggest that DNA emits low-frequency electromagnetic waves which imprint the structure of the molecule onto the water. This structure, they claim, is preserved and amplified through quantum coherence effects, and because it mimics the shape of the original DNA, the enzymes in the PCR process mistake it for DNA itself, and somehow use it as a template to make DNA matching that which “sent” the signal (arxiv.org/abs/1012.5166).

“The biological experiments do seem intriguing, and I wouldn’t dismiss them,” says Greg Scholes of the University of Toronto in Canada, who last year demonstrated that quantum effects occur in plants. Yet according to Klaus Gerwert, who studies interactions between water and biomolecules at the Ruhr University in Bochum, Germany, “It is hard to understand how the information can be stored within water over a timescale longer than picoseconds.”

“The structure would be destroyed instantly,” agrees Felix Franks, a retired academic chemist in London who has studied water for many years. Franks was involved as a peer reviewer in the debunking of a controversial study in 1988 which claimed that water had a memory (see “How ‘ghost molecules’ were exorcised”). “Water has no ‘memory’,” he says now. “You can’t make an imprint in it and recover it later.”

Despite the scepticism over Montagnier’s explanation, the consensus was that the results deserve to be investigated further. Montagnier’s colleague, theoretical physicist Giuseppe Vitiello of the University of Salerno in Italy, is confident that the result is reliable. “I would exclude that it’s contamination,” he says. “It’s very important that other groups repeat it.”

In a paper last year (Interdisciplinary Sciences: Computational Life Sciences, DOI: 10.1007/s12539-009-0036-7), Montagnier described how he discovered the apparent ability of DNA fragments and entire bacteria both to produce weak electromagnetic fields and to “regenerate” themselves in previously uninfected cells. Montagnier strained a solution of the bacterium Mycoplasma pirumthrough a filter with pores small enough to prevent the bacteria penetrating. The filtered water emitted the same frequency of electromagnetic signal as the bacteria themselves. He says he has evidence that many species of bacteria and many viruses give out the electromagnetic signals, as do some diseased human cells.

Montagnier says that the full details of his latest experiments will not be disclosed until the paper is accepted for publication. “Surely you are aware that investigators do not reveal the detailed content of their experimental work before its first appearance in peer-reviewed journals,” he says.

How ‘ghost molecules’ were exorcised

The latest findings by Luc Montagnier evoke long-discredited work by the French researcher Jacques Benveniste. In a paper inNature (vol 333, p 816) in 1988 he claimed to show that water had a “memory”, and that the activity of human antibodies was retained in solutions so dilute that they couldn’t possibly contain any antibody molecules (New Scientist, 14 July 1988, p 39).

The British government has been training a Bangladeshi paramilitary force condemned by human rights organisations as a “government death squad”, leaked US embassy cables have revealed.

By Fariha Karim and Ian Cobain for the Guardian

Members of the Rapid Action Battalion (RAB), which has been held responsible for hundreds of extra-judicial killings in recent years and is said to routinely use torture, have received British training in “investigative interviewing techniques” and “rules of engagement”.

Details of the training were revealed in a number of cables, released by WikiLeaks, which address the counter-terrorism objectives of the US and UK governments in Bangladesh. One cable makes clear that the US would not offer any assistance other than human rights training to the RAB – and that it would be illegal under US law to do so – because its members commit gross human rights violations with impunity.

Since the RAB was established six years ago, it is estimated by some human rights activists to have been responsible for more than 1,000 extra-judicial killings, described euphemistically as “crossfire” deaths. In September last year the director general of the RAB said his men had killed 577 people in “crossfire”. In March this year he updated the figure, saying they had killed 622 people.

The RAB’s use of torture has also been exhaustively documented by human rights organisations. In addition, officers from the paramilitary force are alleged to have been involved in kidnap and extortion, and are frequently accused of taking large bribes in return for carrying out crossfire killings.

However, the cables reveal that both the British and the Americans, in their determination to strengthen counter-terrorism operations in Bangladesh, are in favour of bolstering the force, arguing that the “RAB enjoys a great deal of respect and admiration from a population scarred by decreasing law and order over the last decade”. In one cable, the US ambassador to Dhaka, James Moriarty, expresses the view that the RAB is the “enforcement organisation best positioned to one day become a Bangladeshi version of the US Federal Bureau of Investigation”.

In another cable, Moriarty quotes British officials as saying they have been “training RAB for 18 months in areas such as investigative interviewing techniques and rules of engagement”. Asked about the training assistance for the RAB, the Foreign Office said the UK government “provides a range of human rights assistance” in the country. However, the RAB’s head of training, Mejbah Uddin, told the Guardian that he was unaware of any human rights training since he was appointed last summer.

The cables make clear that British training for RAB officers began three years ago under the last Labour government.

However, RAB officials confirmed independently of the cables that they had taken part in a series of courses and workshops as recently as October, five months after the formation of the coalition government. Asked whether ministers had approved the training programme, the Foreign Office said only that William Hague, the foreign secretary, and other ministers, had been briefed on counter-terrorism spending.

The US ambassador explains in the cables that the US government is “constrained by RAB’s alleged human rights violations, which have rendered the organisation ineligible to receive training or assistance” under laws which prohibit American funding or training for overseas military units which abuse human rights with impunity.

Human rights organisations say the RAB cannot be reformed, noting that its human rights record has deterioriated still further in the last 12 months. Human Rights Watch has repeatedly described the RAB as a government death squad.

Brad Adams, the organisation’s Asia director, said: “RAB is a Latin American-style death squad dressed up as an anti-crime force. The British government has let its desire for a functional counter-terrorism partner in Bangladesh blind it to the risks of working with RAB, and the legitimacy that it gives to RAB inside Bangladesh. Furthermore, it is not clear that the British government has ever made it a priority at the highest levels to tell RAB that if it doesn’t change, it will not co-operate with it.”

Amnesty International has also repeatedly condemned the RAB, while the Bangladeshi human rights organisation Odhikar has painstakingly documented the RAB’s involvement in extra-judicial killings and torture since the creation of the force in March 2004.

Asked to comment on the rights groups’ concern about the RAB, the Foreign Office said: “We do not discuss the detail of operational counter-terrorism cooperation. Counter-terrorism assistance is fully in line with our laws and values.” At least some of the British training has been conducted by serving British police officers, working under the auspices of the National Policing Improvement Agency (NPIA), which was established in 2007 to build policing capacity and standards. Recent courses for RAB have been provided by officers from West Mercia and Humberside Police.

Asked whether it believed it was appropriate for British officers to be training members of an organisation condemned as “a government death squad”, and whether courses in investigative interviewing techniques might not render torture more effective, an NPIA spokesman said the courses had been approved by the government and by the Association of Chief Police Officers.

“The NPIA has given limited support to the Bangladeshi Police and the RAB in technical areas of policing such as forensic awareness, management of crime scenes and recovery of evidence. Throughout the training we have emphasised the importance of respecting the human rights of witnesses, suspects and victims.”The purpose of our sanctioned engagement is to support the development and improvement of professional policing that supports democratic, human rights-based practices linked to the rule of law in countries that may have different laws, faiths and policing practices from our own.”

It is understood that there have been disagreements within the Foreign Office about the British government’s involvement with the RAB. Some officials have argued that the partnership with the RAB is an essential component of the UK’s counter-terrorism strategy in the region, while others have expressed concern that the relationship could prove damaging to Britain’s reputation.

Successive Bangladeshi governments have promised to end the RAB’s use of murder. The current government promised in its manifesto that it would end all extra-judicial killings, but they have continued following its election two years ago.In October last year, the shipping minister, Shahjahan Khan, speaking in a discussion organised by the BBC, said: “There are incidents of trials that are not possible under the laws of the land. The government will need to continue with extra-judicial killings, commonly called crossfire, until terrorist activities and extortion are uprooted.”

In December last year the high court in Dhaka ruled that such killings must be brought to a halt following litigation by victims’ familes and human rights groups, but they continue on an almost weekly basis. Most of the victims are young men, some are alleged to be petty criminals or are said to be left-wing activists, and the killings invariably take place in the middle of the night.

In the most recent “crossfire” killings, the RAB reported that it had shot dead Mohammad Mamun, 25, in the town of Tangail, shortly after midnight on Monday, and that 90 minutes later its officers in Dhaka, 50 miles to the south, had shot dead a second man, Taku Alam, 30. Today the RAB announced it had shot dead a 45-year-old man, Anisur Rahman, said to be a member of the Communist party in the west of the country.

See the paper: State and Terrorist Conspiracies

For additional analysis, see here.

By Michel Bauwens for the P2P Foundation

Analysis:

(nearly quoted in full)

“Most of the news media seems to be losing their minds over Wikileaks without actually reading these essays, even though he describes the function and aims of an organization like Wikileaks in pretty straightforward terms. But, to summarize, he begins by describing a state like the US as essentially an authoritarian conspiracy, and then reasons that the practical strategy for combating that conspiracy is to degrade its ability to conspire, to hinder its ability to “think” as a conspiratorial mind. The metaphor of a computing network is mostly implicit, but utterly crucial: he seeks to oppose the power of the state by treating it like a computer and tossing sand in its diodes.

He begins by positing that conspiracy and authoritarianism go hand in hand, arguing that since authoritarianism produces resistance to itself — to the extent that its authoritarianism becomes generally known — it can only continue to exist and function by preventing its intentions (the authorship of its authority?) from being generally known. It inevitably becomes, he argues, a conspiracy:

Authoritarian regimes give rise to forces which oppose them by pushing against the individual and collective will to freedom, truth and self realization. Plans which assist authoritarian rule, once discovered, induce resistance. Hence these plans are concealed by successful authoritarian powers. This is enough to define their behavior as conspiratorial.

The problem this creates for the government conspiracy then becomes the organizational problem it must solve: if the conspiracy must operate in secrecy, how is it to communicate, plan, make decisions, discipline itself, and transform itself to meet new challenges? The answer is: by controlling information flows. After all, if the organization has goals that can be articulated, articulating them openly exposes them to resistance. But at the same time, failing to articulate those goals to itself deprives the organization of its ability to process and advance them. Somewhere in the middle, for the authoritarian conspiracy, is the right balance of authority and conspiracy.

His model for imagining the conspiracy, then, is not at all the cliché that people mean when they sneer at someone for being a “conspiracy theorist.” After all, most the “conspiracies” we’re familiar with are pure fantasies, and because the “Elders of Zion” or James Bond’s SPECTRE have never existed, their nonexistence becomes a cudgel for beating on people that would ever use the term or the concept. For Assange, by contrast, a conspiracy is something fairly banal, simply any network of associates who act in concert by hiding their concerted association from outsiders, an authority that proceeds by preventing its activities from being visible enough to provoke counter-reaction. It might be something as dramatic as a loose coalition of conspirators working to start a war with Iraq/n, or it might simply be the banal, everyday deceptions and conspiracies of normal diplomatic procedure.

He illustrates this theoretical model by the analogy of a board with nails hammered into it and then tied together with twine:

First take some nails (“conspirators”) and hammer them into a board at random. Then take twine (“communication”) and loop it from nail to nail without breaking. Call the twine connecting two nails a link. Unbroken twine means it is possible to travel from any nail to any other nail via twine and intermediary nails…Information flows from conspirator to conspirator. Not every conspirator trusts or knows every other conspirator even though all are connected. Some are on the fringe of the conspiracy, others are central and communicate with many conspirators and others still may know only two conspirators but be a bridge between important sections or groupings of the conspiracy… Conspirators are often discerning, for some trust and depend each other, while others say little. Important information flows frequently through some links, trivial information through others. So we expand our simple connected graph model to include not only links, but their “importance.” … Return to our board-and-nails analogy. Imagine a thick heavy cord between some nails and fine light thread between others. Call the importance, thickness or heaviness of a link its weight. Between conspirators that never communicate the weight is zero. The “importance” of communication passing through a link is difficult to evaluate apriori, since its true value depends on the outcome of the conspiracy. We simply say that the “importance” of communication contributes to the weight of a link in the most obvious way; the weight of a link is proportional to the amount of important communication flowing across it. Questions about conspiracies in general won’t require us to know the weight of any link, since that changes from conspiracy to conspiracy.

Such a network will not be organized by a flow chart, nor would it ever produce a single coherent map of itself (without thereby hastening its own collapse). It is probably fairly acephalous, as a matter of course: if it had a single head (or a singular organizing mind which could survey and map the entirety), then every conspirator would be one step from the boss and a short two steps away from every other member of the conspiracy. A certain amount of centralization is necessary, in other words (otherwise there is no conspiracy), but too much centralization makes the system vulnerable.

To use The Wire as a ready-to-hand example, imagine if Avon Barksdale was communicating directly with Bodie. All you would ever have to do is turn one person — any person — and you would be one step away from the boss, whose direct connection to everyone else in the conspiracy would allow you to sweep them all up at once. Obviously, no effective conspiracy would ever function this way. Remember Stringer Bell’s “is you taking notes on a criminal fucking conspiracy?” To function effectively, the primary authority has to be disassociated from all other members of the conspiracy, layers of mediation which have to be as opaque as possible to everyone concerned (which a paper trail unhelpfully clarifies). But while the complexity of these linkages shield the directing authority from exposure, they also limit Avon Barksdale’s ability to control what’s going on around him. Businesses run on their paperwork! And the more walls you build around him, the less he might be able to trust his lieutenants, and the less they’ll require (or tolerate) him.

This, Assange reasons, is a way to turn a feature into a bug. And his underlying insight is simple and, I think, compelling: while an organization structured by direct and open lines of communication will be much more vulnerable to outside penetration, the more opaque it becomes to itself (as a defense against the outside gaze), the less able it will be to “think” as a system, to communicate with itself. The more conspiratorial it becomes, in a certain sense, the less effective it will be as a conspiracy. The more closed the network is to outside intrusion, the less able it is to engage with that which is outside itself (true hacker theorizing).

His thinking is not quite as abstract as all that, of course; as he quite explicitly notes, he is also understanding the functioning of the US state by analogy with successful terrorist organizations. If you’ve seen The Battle of Algiers, for example, think of how the French counter-terrorist people work to produce an organizational flow chart of the Algerian resistance movement: since they had overwhelming military superiority, their inability to crush the FLN resided in their inability to find it, an inability which the FLN strategically works to impede by decentralizing itself. Cutting off one leg of the octopus, the FLN realized, wouldn’t degrade the system as a whole if the legs all operated independently. The links between the units were the vulnerable spots for the system as a whole, so those were most closely and carefully guarded and most hotly pursued by the French. And while the French won the battle of Algiers, they lost the war, because they adopted the tactics Assange briefly mentions only to put aside:

How can we reduce the ability of a conspiracy to act?…We can split the conspiracy, reduce or eliminating important communication between a few high weight links or many low weight links. Traditional attacks on conspiratorial power groupings, such as assassination, have cut high weight links by killing, kidnapping, blackmailing or otherwise marginalizing or isolating some of the conspirators they were connected to.

This is the US’s counterterrorism strategy — find the men in charge and get ’em — but it’s not what Assange wants to do: such a program would isolate a specific version of the conspiracy and attempt to destroy the form of it that already exists, which he argues will have two important limitations. For one thing, by the time such a conspiracy has a form which can be targeted, its ability to function will be quite advanced. As he notes:

“A man in chains knows he should have acted sooner for his ability to influence the actions of the state is near its end. To deal with powerful conspiratorial actions we must think ahead and attack the process that leads to them since the actions themselves can not be dealt with.”

By the time a cancer has metastasized, in other words, antioxidents are no longer effective, and even violent chemotherapy is difficult. It’s better, then, to think about how conspiracies come into existence so as to prevent them from forming in the first place (whereas if you isolate the carcinogen early enough, you don’t need to remove the tumor after the fact). Instead, he wants to address the aggregative process itself, by impeding the principle of its reproduction: rather than trying to expose and cut particular links between particular conspirators (which does little to prevent new links from forming and may not disturb the actual functioning of the system as a whole), he wants to attack the “total conspiratorial power” of the entire system by figuring out how to reduce its total ability to share and exchange information among itself, in effect, to slow down its processing power. As he puts it:

Conspiracies are cognitive devices. They are able to outthink the same group of individuals acting alone Conspiracies take information about the world in which they operate (the conspiratorial environment), pass through the conspirators and then act on the result. We can see conspiracies as a type of device that has inputs (information about the environment), a computational network (the conspirators and their links to each other) and outputs (actions intending to change or maintain the environment).

Because he thinks of the conspiracy as a computational network, he notes in an aside that one way to weaken its cognitive ability would be to degrade the quality of its information:

Since a conspiracy is a type of cognitive device that acts on information acquired from its environment, distorting or restricting these inputs means acts based on them are likely to be misplaced. Programmers call this effect garbage in, garbage out. Usually the effect runs the other way; it is conspiracy that is the agent of deception and information restriction. In the US, the programmer’s aphorism is sometimes called “the Fox News effect”.

I’m not sure this is what he means, but it’s worth reflecting that the conspiracy’s ability to deceive others through propaganda can also be the conspiracy’s tendency to deceive itself by its own propaganda. So many people genuinely drink the Kool-Aid, after all. Would our super-spies in Afghanistan ever have been so taken in by the imposter Taliban guy if they didn’t, basically, believe their own line of propaganda, if they didn’t convince themselves — even provisionally — that we actually are winning the war against Talibothra? The same is true of WMD; while no one in possession of the facts could rationally conclude that Saddam Hussein then (or Iran now) are actually, positively in pursuit of WMD’s, this doesn’t mean that the people talking about ticking time bombs don’t actually believe that they are. It just means they are operating with bad information about the environment. Sometimes this works in their favor, but sometimes it does not: if Obama thinks Afghanistan is winnable, it may sink his presidency, for example, while the belief of his advisors that the economy would recover if the government rescued only the banks almost certainly lost the midterm elections for the Democrats (and was the death-knell for so many of the Blue Dogs who were driving that particular policy choice). Whether this actually hurts the conspiracy is unclear; those Blue Dogs might have lost their seats, but most of them will retire from public service to cushy jobs supported by the sectors they supported while they were in public service. And lots of successful politicians do nothing but fail.

This is however, not where Assange’s reasoning leads him. He decides, instead, that the most effective way to attack this kind of organization would be to make “leaks” a fundamental part of the conspiracy’s information environment. Which is why the point is not that particular leaks are specifically effective. Wikileaks does not leak something like the “Collateral Murder” video as a way of putting an end to that particular military tactic; that would be to target a specific leg of the hydra even as it grows two more. Instead, the idea is that increasing the porousness of the conspiracy’s information system will impede its functioning, that the conspiracy will turn against itself in self-defense, clamping down on its own information flows in ways that will then impede its own cognitive function. You destroy the conspiracy, in other words, by making it so paranoid of itself that it can no longer conspire:

The more secretive or unjust an organization is, the more leaks induce fear and paranoia in its leadership and planning coterie. This must result in minimization of efficient internal communications mechanisms (an increase in cognitive “secrecy tax”) and consequent system-wide cognitive decline resulting in decreased ability to hold onto power as the environment demands adaption. Hence in a world where leaking is easy, secretive or unjust systems are nonlinearly hit relative to open, just systems. Since unjust systems, by their nature induce opponents, and in many places barely have the upper hand, mass leaking leaves them exquisitely vulnerable to those who seek to replace them with more open forms of governance.

The leak, in other words, is only the catalyst for the desired counter-overreaction; Wikileaks wants to provoke the conspiracy into turning off its own brain in response to the threat. As it tries to plug its own holes and find the leakers, he reasons, its component elements will de-synchronize from and turn against each other, de-link from the central processing network, and come undone. Even if all the elements of the conspiracy still exist, in this sense, depriving themselves of a vigorous flow of information to connect them all together as a conspiracy prevents them from acting as a conspiracy. As he puts it:

If total conspiratorial power is zero, then clearly there is no information flow between the conspirators and hence no conspiracy. A substantial increase or decrease in total conspiratorial power almost always means what we expect it to mean; an increase or decrease in the ability of the conspiracy to think, act and adapt…An authoritarian conspiracy that cannot think is powerless to preserve itself against the opponents it induces.

In this sense, most of the media commentary on the latest round of leaks has totally missed the point. After all, why are diplomatic cables being leaked? These leaks are not specifically about the war(s) at all, and most seem to simply be a broad swath of the everyday normal secrets that a security state keeps from all but its most trusted hundreds of thousands of people who have the right clearance. Which is the point: Assange is completely right that our government has conspiratorial functions. What else would you call the fact that a small percentage of our governing class governs and acts in our name according to information which is freely shared amongst them but which cannot be shared amongst their constituency? And we all probably knew that this was more or less the case; anyone who was surprised that our embassies are doing dirty, secretive, and disingenuous political work as a matter of course is naïve. But Assange is not trying to produce a journalistic scandal which will then provoke red-faced government reforms or something, precisely because no one is all that scandalized by such things any more. Instead, he is trying to strangle the links that make the conspiracy possible, to expose the necessary porousness of the American state’s conspiratorial network in hopes that the security state will then try to shrink its computational network in response, thereby making itself dumber and slower and smaller.

Early responses seem to indicate that Wikileaks is well on its way to accomplishing some of its goals. As Simon Jenkins put it (in a great piece in its own right) “The leaks have blown a hole in the framework by which states guard their secrets.” And if the diplomats quoted by Le Monde are right that, “we will never again be able to practice diplomacy like before,” this is exactly what Wikileaks was trying to do. It’s sort of pathetic hearing diplomats and government shills lament that the normal work of “diplomacy” will now be impossible, like complaining that that the guy boxing you out is making it hard to get rebounds. Poor dears. If Assange is right to point out that his organization has accomplished more state scrutiny than the entire rest of the journalistic apparatus combined, he’s right but he’s also deflecting the issue: if Wikileaks does some of the things that journalists do, it also does some very different things. Assange, as his introductory remarks indicate quite clearly, is in the business of “radically shift[ing] regime behavior.”

If Wikileaks is a different kind of organization than anything we’ve ever seen before, it’s interesting to see him put himself in line with more conventional progressivism. Assange isn’t off base, after all, when he quotes Theodore Roosevelt’s words from his 1912 Progressive party presidential platform as the epigraph to the first essay; Roosevelt realized a hundred years ago that “Behind the ostensible government sits enthroned an invisible government owing no allegiance and acknowledging no responsibility to the people,” and it was true, then too, that “To destroy this invisible government, to befoul this unholy alliance between corrupt business and corrupt politics is the first task of statesmanship.” Assange is trying to shit all over this unholy alliance in ways that the later and more radical Roosevelt would likely have commended.

It’s worth closing, then, by recalling that Roosevelt also coined the term “muckraker,” and that he did so as a term of disparagement. Quoting from Pilgrim’s Progress, he cited the example of the “Muck-Raker” who could only look down, whose perspective was so totally limited to the “muck” that it was his job to rake, he had lost all ability to see anything higher. Roosevelt, as always, is worth quoting:

In Bunyan’s Pilgrim’s Progress you may recall the description of the Man with the Muck-rake, the man who could look no way but downward, with the muckrake in his hand; who was offered a celestial crown for his muck-rake, but who would neither look up nor regard the crown he was offered, but continued to rake to himself the filth of the floor…the Man with the Muck-rake is set forth as the example of him whose vision is fixed on carnal instead of on spiritual things. Yet he also typifies the man who in this life consistently refuses to see aught that is lofty, and fixes his eyes with solemn intentness only on that which is vile and debasing. Now, it is very necessary that we should not flinch from seeing what is s vile and debasing. There is filth on the floor, and it must be scraped up with the muck-rake; and there are times and places where this service is the most needed of all the services that can be performed. But the man who never does anything else, who never thinks or speaks or writes save of his feats with the muck-rake, speedily becomes, not a help to society, not an incitement to good, but one of the most potent forces for evil. There are, in the body politic, economic, and social, many and grave evils, and there is urgent necessity for the sternest war upon them. There should be relentless exposure of and attack upon every evil man, whether politician or business man, every evil practice, whether in politics, in business, or in social life. I hail as a benefactor every writer or speaker, every man who, on the platform, or in book, magazine, or newspaper, with merciless severity makes such attack, provided always that he in his turn remembers that the attack is of use only if it is absolutely truthful…

Roosevelt was many things when he uttered those words, but he was not wrong. There is a certain vicious amorality about the Mark Zuckerberg-ian philosophy that all transparency is always and everywhere a good thing, particularly when it’s uttered by the guy who’s busily monetizing your radical transparency. And the way most journalists “expose” secrets as a professional practice — to the extent that they do — is just as narrowly selfish: because they publicize privacy only when there is profit to be made in doing so, they keep their eyes on the valuable muck they are raking, and learn to pledge their future professional existence on a continuing and steady flow of it. In muck they trust.

According to his essay, Julian Assange is trying to do something else. Because we all basically know that the US state — like all states — is basically doing a lot of basically shady things basically all the time, simply revealing the specific ways they are doing these shady things will not be, in and of itself, a necessarily good thing. In some cases, it may be a bad thing, and in many cases, the provisional good it may do will be limited in scope. The question for an ethical human being — and Assange always emphasizes his ethics — has to be the question of what exposing secrets will actually accomplish, what good it will do, what better state of affairs it will bring about. And whether you buy his argument or not, Assange has a clearly articulated vision for how Wikileaks’ activities will “carry us through the mire of politically distorted language, and into a position of clarity,” a strategy for how exposing secrets will ultimately impede the production of future secrets. The point of Wikileaks — as Assange argues — is simply to make Wikileaks unnecessary”.

The fresh-faced 39-year-old man, in a dark T-shirt and jeans, is talking about travelling to Mars. Not now, but when he’s older and ready to swap life on Earth for one on the red planet. “It would be a good place to retire,” he says in all seriousness. Normally, this would be the time to make one’s excuses and leave the company of a lunatic. Or to smile politely and humour a space nerd’s unlikely fantasies. But this man needs to be taken seriously for one compelling reason: he already has his own spaceship.

From the Guardian by Paul Harris

This is Elon Musk, a brilliant entrepreneur who made a fortune from the internet and has invested vast amounts of it in building his own private space rocket company, SpaceX. Indeed, far from being crazy, Musk is the real-life inspiration for the movie character Tony Stark, the playboy scientist hero of the Iron Man franchise.

There are some similarities. Outside the SpaceX plant in the baking southern California sun, Musk’s sexy electric sports car sits in a reserved parking space (he co-founded Tesla, the firm which makes the vehicle), resembling the sort of motor Stark would drive. Musk is also engaged to the beautiful British actress Talulah Riley, star of St Trinian’s and St Trinian’s 2, and he used to get thrills from flying his own private military jet fighter.

What’s more, like Stark, Musk is on a mission to save the world. But while Stark’s aim was to battle evil-doers and achieve world peace, Musk’s mission is a little grander. He wants to secure humanity’s future by turning the human race into a space-faring people able to colonise other planets. It’s the only way, Musk believes, that we can be saved, either from destroying ourselves or from some outside calamity. To put it mildly, Musk thinks big and takes the long view. “It’s important that we attempt to extend life beyond Earth now,” he says in an accent hinting at his childhood in South Africa. “It is the first time in the four billion-year history of Earth that it’s been possible and that window could be open for a long time – hopefully it is – or it could be open for a short time. We should err on the side of caution and do something now.”

SpaceX is Musk’s attempt to do that something. Its headquarters are situated within earshot of the busy runways of Los Angeles International airport. The company’s logo stands proudly on an otherwise nondescript hangar-sized building. But inside, a revolution in space travel could be taking place.

The factory floor has been roughly organised into an assembly line to make space rockets, part of a process of wresting the future of space travel out of the hands of government bodies, such as Nasa, and into the hands of private businesses. Using its hyper-efficient Merlin engines, SpaceX has successfully flown its first rocket, Falcon 1, up into space, where it put a satellite into orbit. Then it successfully flew the much bigger Falcon 9 rocket earlier this year. Now the company is working on Dragon, a space capsule that will sit on top of a Falcon 9 and deliver first cargo – and then, hopefully, astronauts – to the International Space Station.

That is stunning stuff. SpaceX, which was only founded in 2002, is not even a decade old. Yet it is doing things in space that some countries with their own national space programmes have not yet achieved. “When we launched the initial rocket actually leaving the launch pad, that was awesome,” Musk says, gazing at the Dragon module being built. “Getting into orbit was when a lot of people thought: OK, it’s real. That’s something that South Korea tried a couple of times and they failed. Brazil tried three times and they failed. This is normally something a country does, and only a few countries have succeeded.”

SpaceX is not alone in aiming for the stars. A raft of private firms have joined in a new space race. Jeff Bezos, founder of Amazon, is building a suborbital rocket called the Blue Origin New Shepard. John D Carmack, the man behind video games Doom and Quake, has his eyes on a lunar landing. Virgin Atlantic boss Richard Branson is aiming to kickstart space tourism with his Virgin Galactic project. Yet SpaceX is the most advanced and ambitious. Its rockets have already flown into space and it has won hundreds of millions of dollars worth of business contracts for future voyages.

Incredibly, however, SpaceX does not feel like a huge operation. It defeats the received wisdom that only major world powers, or gigantic corporations such as Boeing, can truly set their sights on leaving the grip of Earth’s gravity. Instead, SpaceX feels like a dotcom company. Inside the factory are all the accoutrements one expects of a booming Silicon Valley enterprise. All the office space is open-plan and even Musk has an open cubicle like everyone else. Employees – who dub themselves SpaceXers – wear casual T-shirts and are not afraid to sport goatee beards and a smattering of tattoos. They often travel around the assembly floor on tricycles and until recently, before SpaceX’s employee roster topped 1,000 people, Musk was personally involved in every single appointment. He believes the “all in it together” work culture of a start-up is vital to achieve the firm’s staggeringly ambitious agenda. “People work better when they know what the goal is and why. It is important that people look forward to coming to work in the morning and enjoy working.”

In fact, SpaceX’s Silicon Valley-style culture springs from Musk’s own background as one of the most successful – and wealthy – figures to emerge from the internet. His interest in technology began early. He bought his first computer at the age of 10 when he was growing up in Pretoria, South Africa, the son of a Canadian model and a South African engineer. Musk taught himself to write computer programs and sold his first commercial software – fittingly, a space game called Blastar – when he was just 12. He left at 17 to work on a relative’s farm in Canada, before going to the University of Pennsylvania. He graduated with two degrees, one in physics and the other in economics, before winning a place in 1995 at Stanford as a graduate student. He stayed there for two days before fleeing to start his first internet company, Zip2, which produced publishing software. In 1999, he sold it for more than $300m (£193m) and co-founded X.com, which eventually turned into PayPal. It was sold to eBay in 2002 for $1.5bn.

All of which left Musk wealthy beyond belief and could have led to a life of idle bliss. But besides being a very rich man, Musk is a determined one. Talking to him is a slightly unsettling experience. He is open and friendly, but there is a sense that – on some level – he is operating on a slightly higher plane. Asked why he does what he does, he gives an answer that seems rehearsed but rings totally sincere. “When I was in college there were three areas that I thought most would affect the future of humanity. Those were the internet, the transition to a sustainable energy economy, and space exploration and ultimately extending life beyond Earth and making it multi-planetary.”

For Musk, the best way to achieve that third goal was to popularise space travel and make it affordable. Thus SpaceX and its fleet of rockets were born. He investigated the science behind rocket launching and concluded that there was no real reason why it was so expensive. He believed the space industry was dominated by inefficient government bodies. By starting afresh, and going back to basics, Musk believed getting into space could be done quickly and cheaply. He was right. SpaceX’s Merlin engines are beautifully engineered and powerful, but simply made. They run on highly refined kerosene that costs less than petrol. The rockets they power – in the shape of the Falcon 1 and Falcon 9 – are also simple. They have fewer stages (where one bit of the rocket separates from the other) than their rivals and are mostly re-usable. Thus they can put cargo into space for a fraction of the cost.

The Dragon module is also a throwback. It looks nothing like the space shuttle, which it essentially hopes to replace as the “taxi” service to the International Space Station. Instead, it resembles something from the 60s, being shaped like a shuttlecock. Not that Musk cares about looks. He just cares about the fact that it is being designed with windows: a sign of his commitment to one day put astronauts, including himself, inside it. “I would like to go up in a Dragon at some point,” he says. A few years after its first flying. I think it would be great, huge amounts of fun. A very life-changing experience.”

Of course, Musk’s life has already changed. You can’t be a real-life Tony Stark with plans to retire to Mars and not generate publicity. But it has not been easy for him. Musk, beneath his shell of otherworldliness, is charming and funny, but he finds being in the public eye difficult. He would prefer to spend his time happily working on his rockets, not giving interviews. “I had to learn to be a little more extroverted,” he says. “Ordinarily, I would sit in design meetings all day, exchanging ideas with people. But if I don’t tell the story then it doesn’t get out, and I want to try and get public support for extending life beyond Earth.”

Unfortunately, Musk has discovered that celebrity has a dark side. In his case, that was a painful divorce from his ex-wife, Canadian author Justine Musk, with whom he has five children. The split generated its fair share of media attention, not least because Justine has blogged extensively about the epic legal tussles over the terms of their settlement. As more details emerged, Musk decided to publish his version of events on the Huffington Post. The lengthy piece, in which he wrote about his finances and his relationship with Talulah Riley, began with the words, “Given the choice, I’d rather stick a fork in my hand than write about my personal life.”

Musk’s desire for privacy is perhaps surprising in a man so driven and successful. “I hate writing about personal stuff,” he says. “I don’t have a Facebook page. I don’t use my Twitter account. I am familiar with both, but I don’t use them.”

Outside work, where he spends up to 100 hours a week, Musk says he devotes nearly all his spare time to being a good dad. His children are the reason he gave up flying his military jet. “I have five kids and Iron Man does not have any kids,” he says. “After having kids and running companies, I had so many responsibilities I decided it was not wise to take personal risks.”

So are Musk and his entrepreneurial kin the future of space travel? As Nasa, the big daddy of the global space business, struggles with reduced budgets and a sceptical public, it seems perfectly possible. SpaceX is getting into orbit for a fraction of the cost of the space shuttle programme. It aims to make money as an ongoing business concern, rather than draining an ever-tightening public purse. It wants to drive the costs down and improve reliability and make space travel something that is open to everyone. Only private business, Musk thinks, can do that. “The fundamental barriers are improving reliability and reducing cost, and the government is not that good at either. Would you prefer to fly Virgin Atlantic or Soviet-era Aeroflot?”

But Musk remains a dreamer, not just a businessman. He did not create SpaceX to get rich for the second time. Instead, he is risking his fortune to start a company in a field most people said could not support a project like SpaceX. Again and again, he returns to the themes that keep him going. He sees what SpaceX is doing as part of humanity’s destiny. “I think life on Earth must be about more than just solving problems… It’s got to be something inspiring even if it is vicarious. When the US landed on the moon it was for all humanity. We count that as a human achievement. Anyone who could get near a TV got near a TV. If there was one TV in an African village and you had to walk 50 miles to get there, you’d do it,” he says.

And through it all is the desire to colonise Mars. Musk insists that his most powerful Falcon 9 rockets could already launch missions to Mars if assembled in Earth’s orbit. He wants SpaceX to help humanity spread into space, just like the first European explorers setting out for the New World. “One of the long-term goals of SpaceX is, ultimately, to get the price of transporting people and product to Mars to be low enough and with a high enough reliability that if somebody wanted to sell all their belongings and move to a new planet and forge a new civilisation they could do so.”

Musk’s belief that this can be achieved in two decades is something that most experts would scoff at but Musk, characteristically, finds it frustratingly slow. “Twenty years seems like semi-infinity to me. That’s a long time,” he says, as if surprised that anyone could doubt his aims. It is certainly tempting to dismiss it as a flight of fancy. Except, behind him on SpaceX’s factory floor, Musk’s nascent fleet of working space rockets are already being built.

Space race: the private firms aiming to fly you to the stars

SpaceX is not alone in aiming for the stars. A raft of private firms, set up by billionaires, most of them former CEOs or founders of dotcom or IT companies, have joined in a new space race. These space-age entrepreneurs include:

■ Jeff Bezos, founder of Amazon, now America’s largest online retailer. He set up his space company, Blue Origin, in 2000, though its existence only became public in 2003 when Bezos started buying land in Texas so that he could build a test site for his spacecraft. Blue Origin’s main project is New Shepard, a vertical take-off and landing rocket, that is designed to take tourists to the edge of Earth’s atmosphere: the edge of space .

■ John Carmack, the man behind video games such as Doom and Quake, has set up a company called Armadillo Aerospace which is developing a series of spacecraft including a lunar landing vehicle and a spacecraft which is also aimed at taking tourists to the edge of Earth’s atmosphere. Fares will cost around $100,000, says Carmack. The Virginia-based travel firm Space Adventures has signed an exclusive deal with Armadillo to sell tourist seats on its spaceships.

■ Richard Branson, is planning to start suborbital space-tourist flights on his Virgin Galactic spaceplanes within the next two years. In 2004 he signed a deal with the US inventor Burt Rutan to use the spaceplane technology that he had just developed. When flights begin, a small craft carrying half a dozen passengers – who will pay up to $200,000 – will be flown to the edge of the atmosphere. After a few minutes, the spacecraft will then spiral back to the ground. Branson says he expects first flights to begin within two years.

■ Jeff Greason’s XCOR Aerospace also aims to start suborbital tourist flights. XCOR is based in California where it designs, builds and operates rocket engines and rocket-powered vehicles to government and private markets. The Lynx spacecraft – fuelled by liquid oxygen and kerosene – is a two-seat rocket plane that can take off and land on a runway. The spacecraft has been designed to make up to four flights a day, carrying a single passenger into space where he or she can briefly experience weightlessness before returning to Earth.

■ Steve Bennett is Britain’s principal space engineer. His company, Starchaser, is developing rockets that are intended to blast paying passengers on 20-minute long suborbital flights that will include several minutes in which they will experience the delights of zero gravity.

■ However, SpaceX is the most advanced and ambitious player in the field. Its rockets have already flown into space and it has won hundreds of millions of dollars worth of business contracts for future payload launches.

Camilla Turner

Tan Le is the head of Emotiv Systems, which is developing the next generation of human-machine interface — a headset that takes input directly from the brain.