Babbage

IBM turns 100, Groupon prepares to launch its IPO despite hefty losses and video game manufacturers unveil their latest products

THE scheme to renumber the internet's billions of connected devices will be subject to a significant test of preparedness on June 8th, dubbed World IPv6 Day. IPv6 (internet protocol version 6) will ultimately replace the current IPv4 addressing system, which has nearly run out of addresses for networks to assign to computers and other devices. A complete conversion to the new standard is still a long way off, but the test is the first step on the way. It aims to determine how well big internet firms fare when it comes to providing access to common resources using both old and new addressing conventions at once.

IPv4 has about four billion addresses assigned in blocks of varying size. At one time, a single institution, like Stanford or Apple, might be given an allocation of 16m addresses, and use just a fraction. Others might be allowed just 254 and use one or two. To eke out more time on the IPv4 death watch, owners of idle addresses in the larger blocks had to give up unused portions. IPv6 allows for 3.4x10³⁸ addresses which, boffins reckon, should last until the heat death of the universe. That may seem like overkill, but so did four billion-odd a few decades ago.

The idea is for IPv4 and IPv6 to work side by side indefinitely, as hundreds of millions of pieces of older hardware, including plenty of commercial network equipment and home routers, are incompatible with IPv6. (The idea is that the internet will move from islands of IPv6 in a sea of IPv4, to islands of IPv4 in a sea of IPv6.) The test will reveal whether end users can access websites, the most common public resource on the internet, without too much hassle. But some hassle there certainly will be as operating systems, browsers, home and office broadband gateways, and the internet's backbone of high-capacity routers, all begin talking to each other in a slightly different language. The Internet Society, a think-tank and instigator of the day-long experiment, reckons about one in 1,000 users trying to reach the participating websites could experience a problem.

The Internet Society has championed internet standards for decades, and commands attention. It has beaten the IPv6 drum ever since it became clear that four billion is not such a big number, after all, imploring companies and institutions to take the issue seriously. Some have—Mac OS X, Windows and modern mobile operating systems generally handle IPv6 with gusto—but concern about maintaining pipe pressure along the way from a user to his online destination have kept service providers from hooking up the IPv6 taps in earnest.

To allay providers' fears, as well as their own, big content sites including Facebook, Google and Yahoo! signed on to the World IPv6 Day early. So did content-distribution networks like Akamai and Limelight. Hundreds of other sites large and small are participating, too. Hopefully, any flaws the test reveals—and there are bound to be some—will be easy enough to mend. If so, a world of IPv4 and IPv6 may be schizophrenic; but it ought to prove livable.

DEFENCE companies have been left defenceless. A prominent internet giant has found itself the target of an online plot that allowed outsiders to read some of its users’ emails. And a media organisation has hit the headlines after its own website was vandalised by digital intruders. The cyber attacks against Lockheed Martin and L-3 Communications, two American defence giants, as well as those against Google and America’s Public Broadcasting System (PBS) differ in their details. But they all highlight the fact that hackers are becoming ever more tenacious and creative in their attempts to get their hands on sensitive data.

It’s not just American firms that are under attack. On June 2nd, a group of hackers calling itself “LulzSec” claimed that it had been able to get into the network of Sony Pictures. In a statement, the group said it had accessed details of a million customer accounts, including email addresses and dates of birth. Sony’s executives are investgating the group’s claim. If it turns out to be true, it will be a huge embarrassment for the company, which is only just recovering from another recent attack that forced it to shutter its high-profile PlayStation network for several weeks, costing it millions of dollars.

LulzSec (which also uses the moniker “The Lulz Boat”) claims it is behind the hacking of PBS’s website too. Here the hackers stole passwords and other sensitive data, as well as posting a false story claiming that Tupac Shakur, a rapper who died in 1996, was in fact alive and kicking in New Zealand. They then sent tweets taunting the media outfit. One read: “Anyway, say hello to the insides of the PBS servers, folks.” LulzSec also separately posted a message justifying its attack on Sony, saying the company’s approach to handling sensitive data was “disgraceful and insecure”.

The rise of “hacktivism”, which involves groups of hackers not necessarily driven by financial gain (though this can be a handy by-product of their nefarious activities), poses a growing challenge to companies and governments. Often the motive is revenge. LulzSec claimed its attack on PBS was motivated by the media organisation’s decision to air an investigative report that included criticism of WikiLeaks, the organisation that has been publishing leaked diplomatic cables. Anonymous, a hacker collective that has gained global notoriety for penetrating the networks of credit-card companies and other organisations, has also justified some of its actions by saying they are protests at the way in which Julian Assange, the founder of WikiLeaks, has been persecuted by governments and courts.

Other hackers are launching what many reckon are government-backed intrusions over the internet. This week Google revealed that a cyber attack originating from China had used a technique known as “spear phishing” to extract Gmail passwords from unwary users. This was used to read the emails of senior American officials, journalists, Chinese political activists and government officials in several Asian countries, most notably South Korea. The Chinese government denied it had anything to do with the attacks, but some experts note that Chinese hackers often operate independently, but with the tacit approval of the state.

There is certainly evidence that some online intrusions are the result of a very different approach to the random assaults mounted by hacktivists. Earlier this year, RSA, the security division of EMC, a data-storage firm, admitted that it had been the victim of “an extremely aggressive cyber attack” that gave the hackers information about RSA products designed to protect customers’ systems. It is possible that some of that information may have been used in the attack on Lockheed Martin. 

The details of the intrusion are not fully known, but in Lockheed Martin’s case it is clear that hackers were able to install so-called “keylogger” software on a remote computer accessing Lockheed’s network. This allowed them to see the user’s password and any security codes entered from an RSA SecurID token, which companies such as Lockheed use to generate unique passcode numbers every 30 seconds for their employees as an additional level of protection. 

Harry Sverdlove, the chief technology officer of Bit9, a security company, thinks that the hackers may have used these data in combination with information gleaned during the RSA intrusion to penetrate Lockheed’s network. And he reckons that one consequence of this and other cyber attacks will be to get companies to tighten controls on vulnerable “endpoints”—portable computers and other devices that access networks from a distance. Another will be to stiffen official resolve in America and elsewhere to tackle cyber aggression more forcefully. It is surely no coincidence that Lockheed’s news came out at pretty much the same time that the Pentagon in America was floating the notion that cyber attacks launched by another nation constitute an act of war that justifies a riposte involving traditional military force.

JUNE 5th marks 30 years since America’s Centres for Disease Control and Prevention reported a cluster of unusual infections in Los Angeles. These were the first cases of acquired immune deficiency syndrome, better known as AIDS, to be recognised by medical science. Since then it has killed 25m people (with another 34m currently infected). Now, as we argue in this week's print edition, it appears that the thirty years war against one of the scourges of the late 20th century can be won, if the world has the will to do so.

The Economist first mentioned AIDS in October 1982, 16 months after that first clutch of infections, portentously acknowledging that "there has never been anything quite like this." Nevertheless, at the time no one could have anticipated the extent of the damage AIDS would wreak in the ensuing decades. Readers may bear that in mind as they read the article's (somewhat incongruous) title, its first sentence and the picture caption (move the cursor over the photo to see it). With hindsight these may seem rather crude:

This has been a bad year for people who like to sleep around. The old venereal diseases—syphilis, gonorrhea and chla­mydial infections—do not make them unduly anxious. Those diseases are fairly easy to treat with antibiotics, though too few people are treated to keep them from spreading (there are 2m-3m new cases of chlamydial infections in America each year, 1m-2m of gonorrhea and 50,000­-100,000 of syphilis). Much more frighten­ing are two relative newcomers: genital herpes, a known disease for which there is no known cure; and, worse though much rarer, a lethal disease of obscure origin that causes its victims' immune systems to fail and strikes most heavily at promiscuous homosexual men. It is called acquired immune deficiency syndrome (Aids).

Continue reading "From the archive: Next they'll tax it"»

LIKE millions of others, your correspondent’s extended family took to the road over the long weekend (Memorial Day in America, Spring Bank Holiday in Britain) that marks the unofficial start of summer. With warmer weather beckoning, the urge to migrate to the mountains, forests, beaches and deserts seems every bit as much a genetic imperative among humans as the migratory instincts of petrels, cetaceans or wildebeest.

From June to September, Americans have traditionally taken to the roads in droves, clocking up thousands of miles for leisure and pleasure. By all accounts, this year has started off no differently. Certainly, the nose-to-tail traffic leaving Los Angeles at the beginning of the holiday weekend testified to the perennial Whitsun wanderlust.

The surprise was the amount of traffic, given the current price of petrol. While the average pump price is down from the $4.11 peak reached just before the economy tanked in 2008, regular (ie, 87-octane) grade still costs $3.80 a gallon nationally. In California, which has some of the priciest petrol in the country, the average cost of a gallon of regular has fallen over the past week to $4.05 (€0.74/litre).

Forced to pay over twice as much, Europeans snigger at how little it costs even Californians to fill the tank. But people in the United States pay through the nose for many other things in life (health, education, property taxes, wireless services, etc) that foreigners get for far less or even free. For most Americans, having to drop $70 or more at a gas station can still give serious pause for thought about whether the journey is really necessary.

Yet, here’s the conundrum. Following all previous recessions, petrol consumption has been a leading indicator of recovery, bouncing back sharply as people started using their vehicles more to shop, to dine out, to seek the curious and the entertaining, and, above all, to take vacations. Despite the American economy’s belated and still timid recovery—seen in increasing sales of cars, clothing, hospitality, entertainment, and consumer goods generally (though still not housing)—the amount of petrol being consumed across the country has tumbled to 2001 levels, and shows every sign of falling further.

The Bureau of Economic Analysis, the federal agency that churns out monthly reports on how the economy is faring, believes the 2008 spike in petrol prices and the subsequent recession have changed the consumption patterns of American motorists irreversibly. How so? The short answer is that technology and marketing have altered the type of vehicles Americans are now buying.

For a start, the gas-guzzling V-8 engine that once ruled American roads has all but vanished from the showrooms. Starting imperceptibly in the early 1990s, American motorists traded in their eight-cylinder vehicles for the better fuel efficiency of six-cylinder models. Today, eight-cylinder cars and trucks account for less than 10% of new vehicles. In the new era of austerity, the hulking 6,000lb SUV, once the pride of the parking lot, is now ridiculed as obscene.

Meanwhile, sales of even six-cylinder vehicles have plummeted over the past five years, from 40% of the market to 25%. The big winners have been cars with four-cylinder engines, which have increased their share of the market from 48% five years ago to 65% today. In short, the new cars Americans are buying are almost as frugal as those driven by their counterparts in Europe, if not Japan.

Continue reading "The Difference Engine: Four bangs for the buck"»

IT'S 3 am. Do you know where your mobile phone or tablet is? Of course you do. It's in the bedroom somewhere, probably within reach on a bedside bureau. In a recent survey of a few thousand business travellers, 38% of respondents admitted to actually reaching for it occasionally. A disturbing 8% checked at least once in the middle of every night.

The firm that conducted the survey, iPass, points out that these average figures hide significant demographic differences. Business folk in the Asia-Pacific region, for instance, with their indefatigable work style, were twice as likely as Europeans to check in the wee hours (60% against 30%). Americans ranked neatly in the middle at 45%. The young (22 to 34 years of age) were 25% to 40% more likely to wake and tap their phone or tablet than more senior folk. (Despite being American, and not that old, your correspondent is clearly more like an aged European in this regard; his smartphone and tablet are three rooms away when darkness sets in.)

None of this comes as a surprise. But it is rather perturbing to see the numbers in the light of day. Frequent business travellers, often jet-lagged and sleepless, may find psychological succour of sorts in groggily retrieving e-mail or re-reading a forthcoming presentation. Half the survey's respondents expect to spend 30 days or more on the road this year. Yet iPass did not ask specifically about habits away from home, but about general behaviour.

The firm also wondered why people act the way they do (respondents were allowed to give more than one answer). And so, an absent-minded 36% either forget to change the alert settings and "it pings"—or worse, want a noise to alert them while slumbering. A third blamed working with people outside their country; the same proportion claimed they were working on something important that requires their attention. As many as 36% were simply unable to sleep. Most worryingly, perhaps, over a fifth check "out of habit". 

The surveyed group exhibit disturbing proclivities during daytime, too. Nine in ten fiddle with their smartphone during so-called downtime, and 7% check for e-mail or use apps ten times an hour when they should be doing something else. Compared with previous surveys, iPass notes an uptick in such obsessive behaviours.

These habits spill over into other areas. Asked whether mobile technology interfered with their personal relationships, 29% of the respondents said yes. This being a highly personal question, one expects the honest number to be even higher.

Many jobs now demand constant connectedness, leaving little space for life outside of work. Indeed, while 64% of those asked said it was unacceptable to take a call in a public toilet, 29% confessed to talking on the loo. The line between work and private life has to be drawn somewher. At the toilet door, perhaps?

BABBAGE is no motoring journalist. He knows little of cars, perhaps a little bit more about how they work in theory. Although he has owned vehicles in the past, he rarely drives, especially since he moved to relatively central London where possessing a vehicle mostly causes grief. But he loves them. And after surviving Britain's Society of Motor Manufacturers and Traders (SMMT) test day on May 26th, he loves them with a reinvigorated passion.

Every year, the SMMT invites hacks whose work brings them into contact with the motor industry and its products for what ultimately amounts to a boys' day out.* The producers it managed to corral at the Millbrook test facility north of London did not skimp. Many brought along some of their top models, many worth well over £100,000 ($160,000) apiece, and handed the keys over to grovelling reporters without so much a flich. (Though some did insist that a minder be present in the fanciest machines.)

To Babbage, four cars stood out. These were, in driving order: Jaguar E-Type, Fiat 500 Abarth, Mercedes SLS AMG and Audi R8 Spyder (see above). Babbage will leave engine sizes, piston counts, 0-60mph times (0-100kph for those of a metric persuasion), waggish quips about handling, ride, etc, to the experts. The impressions that follow are just that, impressions.

Start with the E-Type. Babbage has been pining to drive one ever since a colleague waxed lyrical about the beast in a Difference Engine column a while back. On the technical front, all your correspondent can say is that, unlike all the modern cars he has driven, this one had the indicator switch on the right-hand side of the steering wheel, where the lever for operating the windscreen-wipers would normally be; the latter were activated by flipping a switch on the dashboard. (A more experienced colleague tells Babbage that set-up was typical in cars from the 1960s.) It also, discomfittingly, lacked outside rear-view mirrors; the one inside was useless, too, given the blur afforded by the plastic rear window on the soft top. In brief, it was impossible to shake off the thought of driving a bit of automotive history—history being the operative term.

The Mercedes, meanwhile, offered a glimpse of the future; or the present, if you have £170,000 to spare. Aesthetically, it may well be the single most pulchritudinous automobile Babbage has ever set his eyes on. A happy marriage of Art-Déco chic with an utterly contemporary sveltness, it is the 21st-century answer to the graceful E-Type (and, more directly, Mercedes's 300SL Gullwing coupé from 1954). Indeed, the two look very much alike: the same priapic bonnet attached to a roundish canopy, almost boatlike when viewed from behind the steering wheel—an effect strengthened by the rain spluttering intermittently over the windscreen during both rides. A few tweaks and the thing would be fit for Batman. Needless to say, it goes like the clappers.

However, the chasm between the cars of yesteryear and those of today was made starkest by the cheap (well, cheapish, at just over £12,000) and cheerful Fiat Abarth. On the track this rocket-powered skateboard, as one colleague put it, would, of course, be blown to kingdom come by the Mercedes. Yet it would run rings round the apotheosis of top-notch engineering from the 1960s. Readers will forgive Babbage the trite comment, but the discrepancy, experienced first hand, was quite remarkable.

And then there was the Audi. Your correspondent had the good fortune to get his hands on it during a rare dry spell. This meant being able to drive it with the roof down, ensconced in the cosy cockpit. The car's electronics ensured that it maintained its poise even with as mediocre a driver as Babbage at the wheel. To invoke another hackneyed trope, the machine formed the extension of the driver's body. An odd and wonderful feeling that is. A first for this driver. Mrs Babbage would be forgiven for feeling jealous.

* Yes, true to stereotype, ladies were in short supply—shorter even than in the CERN caffeteria dominated by male physicists. (The overall proportion was boosted, unsurprisingly, by the number minding the producers' booths.) A shame, that. Babbage overheard one diminutive female colleague working for Girl Racer magazine explain how she deals with (increasingly rare) disparaging offers for her test car to come in pink. "If it has 300 horse power under the bonnet, I don't mind that it's pink," she would retort.

A STEADY ocean breeze chilled by a current that sweeps down the coast from Alaska cools your correspondent’s hillside home in southern California better than air-conditioning. The breeze is especially welcome during the summer months, when the temperature soars to 85ºF (29ºC) at the beach, 95ºF downtown and over 105ºF in the valleys. At other times, though, it can make one particular part of his backyard uncomfortably chilly. Unfortunately, that part is the best place to sit, chat, cook and sip a glass or two while watching the sun go down and the sailing boats head back to the marina.

Over the years, various solutions have been tossed around. A hedge would do a good job of diffusing the wind, but would block the spectacular views of the ocean, mountains and city that the house was bought for in the first place. A glass fence might easily make things worse—as the turbulent airflow, spilling off the top, created all manner of eddies and back-currents on the downstream side where folks want to sit. A ramp-like structure, designed to divert the windflow over the top of the house, was deemed impractical. The favoured solution at the moment is a wind turbine to sap surplus energy from the breeze.

The object would be simply to still the wind rather than generate electricity. It would be nice to produce some spare kilowatts on the side. But that is impractical, given a prevailing wind speed of only 5-10mph.

To see why, consider a typical windmill—or what those in the business like to call a “horizontal-axis wind turbine” (to distinguish it from the type that spins about a vertical axis like a Savonius or Darrieus turbine). Determining the power available from a windmill is not rocket science, but the mathematics can quickly becomes arcane. Suffice it to say that the power available for harvesting depends on the cube of the wind speed and the area swept by the rotor—in other words, the square of the blade length.

As a rule of thumb, a turbine with a five-foot (1.5 metre) diameter rotor spinning in a 10mph (16km/h) wind can theoretically generate 100 watts of power. Given the cube effect, a doubling of wind speed means the power generated goes up eight-fold, to 800 watts. Double the length of the blades, and the power increases a further four-fold, to a useful 3.2 kilowatts—over half that needed to power a typical home. Obviously, big rotors and high winds get the job done.

But only up to a point. In the real world, not all the power a wind machine is capable of producing can be harvested. Where the First Law of Thermodynamics says the best you can do (when trying to extract work from a machine) is to break even, the Second Law says, forget it, you can’t even do that. Friction and other losses mop up some of the output. In the absence of such losses, you would have a perpetual-motion machine—something which, despite the patent applications by countless crackpots who believe they can suspend the laws of physics, is demonstrably impossible.

With wind turbines, however, the energy losses do not stop with the inexorable effects of entropy. An additional limit to a turbine’s output was discovered in 1919 by a German physicist called Albert Betz. Again, skipping the arcane maths, Betz’s Law says no turbine can capture more than 59.3% of the kinetic energy in the wind (don’t ask, but the term 4²/3³ drops out of the theory).

For his calculations, Betz assumed that the rotor was infinitely thin, had no mass and no hub; that the airflow through it was purely axial and did not get churned up; and that the air was incompressible, its density constant, and no heat transfer took place. In short, even Betz’s Law defines a theoretical upper limit. The best turbines on the market today convert less than 35% of the wind energy into useful work—and if anyone claims more, they are selling snake-oil.

Continue reading "The Difference Engine: Tilting at turbines"»

A FRIEND has been toying with Twitter, and wonders how best to get her feet wet. Whom should she follow? What should she read? Here are some thoughts.

Viewed as a loudspeaker for others—celebrities, creative types, or just acquaintances, pronouncing on what are all too often fatuous banalities—Twitter does indeed resemble the empty medium derided by those outside its embrace; like sitting around an abandoned and empty pool in a solitary chaise longue, with TVs blaring nearby—an experience that is neither convivial nor informative.

Sticking to the bathing analogy, an alternative (though by no means the only one) would be to attend a crowded ocean beach, with swimmers constantly throwing themselves in and out of the water, some more expertly than others. The beach is likely to be lined with reclining observers listening to the water-borne banter, but they are not the ones having the most fun.

Twitter rewards engagement in a way that few other media do (the sea is such a medium, albeit only in a physical sense). The relationship between users is nearly always asymmetrical: following someone does not entail being followed by them. That leads to enormous mismatches: the glitziest twitterati amass hundreds of thousands, even millions of followers, and yet themselves follow just a few dozen people.

Babbage's friend had thrown her lot in some months ago, following a small number of people, and issuing a few tentative tweets. She was not impressed by the response. Her view of Twitter was of an unplugged toaster where the bread neither browns nor pops out.

Suffer as he does from logorrhoea, this Babbage blathers through the day. He also, like his departed mother, knows no shame in striking up conversations with complete strangers—she, in public; he, on Twitter—that he finds interesting. As a child, he complained of her habit. Now, the grudging inheritor of it, he finds easy sociability indispensable in the digital world.

Enter what is perhaps Twitter's greatest, and inadvertent, innovation: the goad in the form of an at-sign (@) followed by a Twitter handle, known as a mention. Twitter did not invent this referral, which was picked up from older chat-system conventions. But the firm adopted it and uses it to create threaded conversations. The mention is distinct from a direct message (DM), which Twitter created intentionally, and which allows a private bit of text to pass between two parties, so long as the recipient follows the sender. The DM is reserved for those who have established a relationship; the mention, meanwhile, works for anyone.

Mentions may easily be ignored—@stephenfry, say, with 2,647,917 followers as this sentence is being written (likely to rise by the time Babbage is done with the post), would be physically incapable of responding to all. To make life a little bit more manageable, most Twitter software, whether from the company itself or innumerable third parties, can be set either to list all mentions in a main chronological stream mixed in with tweets by those the user follows, or only to include the mentions by the latter.

Continue reading "Rules of engagement"»

ON SEPTEMBER 11th, 2001, Americans' feelings of shock and sadness in the wake of the atrocity that befell New York seemed to turn increasingly into anger. Last year boffins from Johannes Gutenberg University in Germany ran the 575,000 pager messages through a computer that categorised the emotional content of the words that made them up—so, "crying" or "grief" would count as sad, "hate" and "annoyed" as angry, etc [link requires subscription]. They found that wrath rose steadily through the day, apparently in reaction to events and news reports, such as speeches by president George W. Bush, or announcements about the growing number of confirmed casualties.

Now, though, it turns out that the rising fury might not have been caused by the terrorist enormity, after all. Instead, it was probably the result of a malfunctioning computer desperately trying to get the attention of a technician. Cynthia Pury, a psychologist at Clemson University, in America, took another look at the data [again, subscription required], and found that nearly a third of the putative anger words from the original study were contained in an automatic message sent by a device and warning of a critical failure. The message in full read as follows:

Reboot NT machine [name] in cabinet [name] at [location]: CRITICAL:[date and time].

Since the word "critical" was flagged as angry, the message—which went out almost 6,000 times, peaking at 552 times an hour by mid-afternoon—skewed the results dramatically. So Dr Pury crunched the data again, after removing the computer's cry for help. The upshot? Anger spilling from messages that remained now appears to have held relatively steady throughout the day.

The huge increase in the use of social media means that automatic textual analysis of real-time reactions to events is bound to proliferate. Some of it may yet yield useful insights. But, as Dr Pury's findings show, researchers use such tools at their peril.

OF ALL the high-speed train services around the world, only one really makes economic sense—the 550km (340-mile) Shinkansen route that connects the 35m people in greater Tokyo to the 20m residents of the Kansai cluster of cities comprising Osaka, Kobe, Kyoto and Nara. At peak times, up to 16 bullet trains an hour travel each way along the densely populated coastal plain that is home to over half of Japan’s 128m people.

Having worked for many years in Tokyo, with family in Osaka, your correspondent has made the two-and-a-half hour journey on the Tokaido bullet-train many times. It is clean, fast and highly civilised, though far from cheap. It beats flying, which is unbearably cramped by comparison, just as pricey, and dumps you an hour from downtown at either end.

The sole reason why Shinkansen plying the Tokaido route make money is the sheer density—and affluence—of the customers they serve. All the other Shinkansen routes in Japan lose cart-loads of cash, as high-speed trains do elsewhere in the world. Only indirect subsidies, creative accounting, political patronage and national chest-thumping keep them rolling.

California wants a share of that bullet-train hubris. Where Florida, Ohio and Wisconsin have turned down billions of federal dollars for high-speed rail, the Golden State has been pressing on with its $43-billion scheme to build a high-speed rail service from Los Angeles to the San Francisco Bay Area, with spurs eventually to San Diego, Sacramento and San Luis Obispo.

The irony is that California has the highest rate of car-ownership in the country, if not the world. It also, despite years of neglect, has one of the best road networks anywhere—certainly leagues ahead of Japan’s. On top of that, it enjoys a highly competitive network of budget airlines serving its main cities. The Los Angeles Times got it about right when it editorialised on May 16th that “California’s much-vaunted high-speed rail project is, to put it bluntly, a train wreck”.

And an expensive one at that. Between them, the federal government, municipals along the proposed route and an assortment of private investors are being asked to chip in some $30 billion. A further $10 billion is to be raised by a bond issue that Californian voters approved in 2008. Anything left unfunded will have to be met by taxpayers. They could be dunned for a lot. A study carried out in 2008 by the Reason Foundation, the Howard Jarvis Taxpayers Association and Citizens Against Government Waste put the final cost of the complete 800-mile network at $81 billion.

Continue reading "The Difference Engine: Fast track to nowhere"»

LAST week The Economist gave warning in a leader that a new bubble seemed to be forming in private-market valuations of some kinds of internet companies. It also cautioned that this could ultimately spill over into public markets. The initial public offering (IPO) on May 19th of shares in America's LinkedIn, a social network for professionals, looks like a sign that this is already starting to happen.

Before the flotation, LinkedIn's advisers had set the price for the company's shares at $45 each. No sooner had trading begun on the New York Stock Exchange than LinkedIn's shares soared as investors rushed to buy them, ultimately closing the day at almost $94. That propelled the company's market capitalisation to a whopping $8.9 billion—some 578 times its net earnings in 2010.

That is an eye-wateringly high multiple and it is reminiscent of the ones that young start-ups going public were getting in the heyday of the last dotcom boom. Some commentators such as Paul Kedrosky have argued that the company's stratospheric valuation (which is the largest initial price tag for a tech firm since Google staged its IPO in 2004) reflects the fact that LinkedIn is the first big American social network to go public, giving investors their first chance to profit from the social-networking revolution. Others have claimed that the heady valuation the firm has been given may well turn out to be justified.

LinkedIn is undoubtedly a solid company. It is focused on a lucrative demographic—executives looking to manage their careers actively—and, unlike other social networks, it has developed a fairly balanced set of revenue streams. These include fees from users for premium services, advertising revenues and money from companies and recruiters that use its services to find talented people. The firm also has an impressive senior management team led by Jeff Weiner, its chief executive.

But there will be a limit to just how fast LinkedIn can scale up, given the nature of its audience. So far its revenue growth has been steady, not spectacular. The $353m that it has raised from the IPO should help the company to expand more swiftly, but it will still have to navigate tricky markets such as China if it is to succeed. And it will need to come up with many more creative ways to monetise its audience without abusing the trust they have in the service. Mr Weiner and his team have a long, hard road ahead to satisfy the investment community's great—no, make that worryingly inflated—expectations.