To prove the point, they actually did such a reversal in the case of telephone-queue waiting times. Traditionally, these have been assumed to follow a Poisson distribution, but some recent research suggests they actually follow a power law. Analysing the participants’ responses suggests that a power law, indeed, it is.
SCIENCE, being a human activity, is not immune to fashion. For example, one of the first mathematicians to study the subject of probability theory was an English clergyman called Thomas Bayes, who was born in 1702 and died in 1761. His ideas about the prediction of future events from one or two examples were popular for a while, and have never been fundamentally challenged. But they were eventually overwhelmed by those of the “frequentist” school, which developed the methods based on sampling from a large population that now dominate the field and are used to predict things as diverse as the outcomes of elections and preferences for chocolate bars.
Recently, however, Bayes’s ideas have made a comeback among computer scientists trying to design software with human-like intelligence. Bayesian reasoning now lies at the heart of leading internet search engines and automated “help wizards”. That has prompted some psychologists to ask if the human brain itself might be a Bayesian-reasoning machine. They suggest that the Bayesian capacity to draw strong inferences from sparse data could be crucial to the way the mind perceives the world, plans actions, comprehends and learns language, reasons from correlation to causation, and even understands the goals and beliefs of other minds.
These researchers have conducted laboratory experiments that convince them they are on the right track, but only recently have they begun to look at whether the brain copes with everyday judgments in the real world in a Bayesian manner. In research to be published later this year in Psychological Science, Thomas Griffiths of Brown University in Rhode Island and Joshua Tenenbaum of the Massachusetts Institute of Technology put the idea of a Bayesian brain to a quotidian test. They found that it passes with flying colours.
Prior assumptions The key to successful Bayesian reasoning is not in having an extensive, unbiased sample, which is the eternal worry of frequentists, but rather in having an appropriate “prior”, as it is known to the cognoscenti. This prior is an assumption about the way the world works—in essence, a hypothesis about reality—that can be expressed as a mathematical probability distribution of the frequency with which events of a particular magnitude happen.
The best known of these probability distributions is the “normal”, or Gaussian distribution. This has a curve similar to the cross-section of a bell, with events of middling magnitude being common, and those of small and large magnitude rare, so it is sometimes known by a third name, the bell-curve distribution. But there are also the Poisson distribution, the Erlang distribution, the power-law distribution and many even weirder ones that are not the consequence of simple mathematical equations (or, at least, of equations that mathematicians regard as simple).
With the correct prior, even a single piece of data can be used to make meaningful Bayesian predictions. By contrast frequentists, though they deal with the same probability distributions as Bayesians, make fewer prior assumptions about the distribution that applies in any particular situation. Frequentism is thus a more robust approach, but one that is not well suited to
THE co-founder, chairman and “chief software architect” of Microsoft, the world’s largest software company, would deny it on his life, but the one person Bill Gates admires most for his geeky prowess—and might have chosen to succeed him as software architect—is almost certainly Steve Jobs. Unfortunately, Mr Jobs, the co-founder of Apple Computer and victim of Mr Gates’s predatory business instincts during the 1980s and 1990s, cannot be considered available, since he is busy leading Apple’s renaissance as a builder of gadgets and software that, in the opinion of his fans, put Microsoft to shame. So Mr Gates spent years courting the geek he admires second most, a software pioneer named Ray Ozzie.
After many overtures, Microsoft last year bought Mr Ozzie’s company, Groove Networks, and thus brought Mr Ozzie and his brother Jack inside the Microsoft tent. Mr Gates then groomed Mr Ozzie to take the lead in defining Microsoft’s direction at the highest (ie, software) level, alongside Steve Ballmer, the chief executive, who will continue to use his prodigious energies to bash heads together in the name of implementation, and Craig Mundie, who will oversee Microsoft’s research efforts and policy lobbying. On June 15th Mr Gates made the transition official, announcing that he would hand over his role as chief software architect to Mr Ozzie in two years’ time, in order to concentrate on becoming the world’s greatest philanthropist.
Mr Ozzie, at 50, is one month younger than Mr Gates and nine months younger than Mr Jobs. Like the other two, he has been admitted to the Computer History Museum’s “hall of fame” for his role in bringing about the PC era. As a kid in suburban Chicago, Mr Ozzie was already soldering all sorts of dangerous circuits together in a guest bedroom, but it was at college in the 1970s that he discovered his passion, which was, as he once put it, “to augment relationships” among human beings through technology. The catalyst was his encounter with PLATO, a cluster of a thousand dumb terminals that were connected to a mainframe and that Mr Ozzie and his friends playfully used to communicate, by exchanging what would today be called e-mails and instant messages. This experience so captured his imagination that he devoted his next three decades to writing software that enables “collaboration”.
His single biggest breakthrough came in the 1980s, when Mr Ozzie personally wrote a million of the first 3.5m lines of code for the first successful collaboration software, Lotus Notes. At a time when nobody had heard of the world wide web, Lotus Notes already offered “workspaces” not unlike today’s wikis. Mr Gates watched with his usual mixture of emotions towards innovations by others—envy and grudging admiration. These probably gave way to anxiety when IBM, Microsoft’s partner-turned-enemy, bought Lotus in 1995 for $3.5 billion.
Mr Ozzie soon parted with IBM and again set out to take collaboration to the next level, this time by using new ideas about decentralised networks between people and their computers. This approach would subsequently be called “peer-to-peer” and has since been made famous by Napster, a method for “sharing” music, and Skype, a service for free internet telephony. Mr Ozzie’s company, Groove, was not a commercial success this time, but Mr Gates and others in the industry nonetheless saw the idea and recognised its potential. Last April Messrs Gates and Ozzie joined forces.
CRITICS have derided a 1998 extension of American copyrights as the “Mickey Mouse Protection Act” because it stopped early images of the Disney company’s mascot from entering the public domain. But such laws, they argue, are no joke. Extending and strengthening copyrights, they claim, will help a handful of big corporations crush creativity in the digital age. On the contrary, say Hollywood studios and big record companies. Without stronger copyright protection, a wave of piracy will destroy their industries, depriving consumers everywhere of a broad choice of movies, music and books.
Last week America’s Supreme Court weighed into what is rapidly becoming a nasty worldwide battle about the scope and enforcement of copyrights, by rejecting a challenge to the 1998 law on constitutional grounds. But even as it upheld the law, the court expressed misgivings. Blistering dissents from two justices dismissed the 20-year extension of copyright as unwarranted, and even the majority’s opinion hinted that Congress’s decision may have been “unwise”.
The court’s ambivalence is understandable. The growing quarrel over copyright is just one of the many difficult issues thrown up by the spread of the internet and related technologies (see our survey of the internet society in this issue). But of all these issues, the copyright battle is becoming one of the most urgent, and bitterly fought, because it could yet determine the future character of cyberspace itself.
Both sides have a point. Digital piracy does indeed threaten to overwhelm so-called “content” industries. As the power and reach of the internet continue to grow, the illicit trading of perfect copies may well devastate the music, movie and publishing industries. The content industries want to protect themselves with anti-copying technology, backed by stronger laws. So far, they have been at loggerheads with technology firms about how to implement such schemes (see article). But a deal between Hollywood and Silicon Valley is likely eventually. Critics are right to fear that, when such a deal is struck, it will be in the interests of big firms, not the public.
A grand new bargain The alternative is to return to the original purpose of copyright, something no national legislature has yet been willing to do. Copyright was originally the grant of a temporary government-supported monopoly on copying a work, not a property right. Its sole purpose was to encourage the circulation of ideas by giving creators and publishers a short-term incentive to disseminate their work. Over the past 50 years, as a result of heavy lobbying by content industries, copyright has grown to such ludicrous proportions that it now often inhibits rather than promotes the circulation of ideas, leaving thousands of old movies, records and books languishing behind a legal barrier. Starting from scratch today, no rational, disinterested lawmaker would agree to copyrights that extend to 70 years after an author’s death, now the norm in the developed world.
Digital technologies are not only making it easier to copy all sorts of works, but also sharply reducing the costs of creating or distributing them, and so also reducing the required incentives. The flood of free content on the internet has shown that most creators do not need incentives that stretch across generations. To reward those who can attract a paying audience, and the firms that support them, much shorter copyrights would be enough. The 14-year term of the original 18th-centur
WHAT weighty issues were preoccupying America’s Senate this week? Soaring health-care costs perhaps, or the fiscal consequences of retiring baby-boomers? Dream on. The country’s top lawmakers argued about a constitutional amendment to ban gay marriage, which had no hope of passing; a daft law to give native Hawaiians their own government; and repeal of the estate tax, a levy that affects only the richest 0.5% of Americans.
These skewed priorities may be blamed on Bill Frist, the Senate majority leader, who sets the agenda and is an aspiring presidential candidate. To bolster his appeal to the Republican Party’s conservative base, he has been pushing for votes on issues the right cares about: flag-burning, gay marriage and, above all, getting rid for ever of the estate tax.
America has had this tax, a levy on the transfer of large fortunes between generations, since 1916. Like Britain, America taxes the estate of the deceased; most other countries which impose such a tax target heirs and their inheritances. The estate tax affects very few Americans (never much over 2% in recent decades, and now far fewer). In the late 1990s it generated just over 1.5% of all tax revenue, or around 0.3% of GDP. Conservatives long for its abolition, which is now tantalisingly close.
George Bush’s 2001 tax cuts created a paradox: the estate tax is in terminal decline, but only temporarily so. Under current law, only estates worth more than $7m for a couple will be taxed in 2009 (compared with a joint exemption of $1.35m in 2000). In 2010 the tax goes altogether. But since all Mr Bush’s tax cuts technically expire in 2011, the abolition lasts only one year. In 2011 the estate tax returns to its pre-Bush form.
This mess clearly needs fixing, if only to avoid a surge in suspicious deaths in Palm Beach in 2010. But how? The House of Representatives has voted several times to make repeal permanent. The Senate meant to do so last year, until Hurricane Katrina’s havoc made it look too callous to give a big tax break to multi-millionaires.
As The Economist went to press, it was not clear whether Mr Frist would succeed this time. Several Democratic senators support the tax’s abolition, but a handful of Republicans think full repeal is too expensive. By one estimate, repealing the estate tax would reduce federal revenues by some $776 billion, or around 0.4% of GDP, between 2012 and 2021. (The costs of repeal are more than the tax actually raises because the bean-counters assume that rich people will shift their assets in a way that reduces income taxes.) One possibility is a “compromise” proposal that would increase the exemption per couple to $10m and cut rates to 15%, a combination that would, in fact, cost almost as much as repeal. Even ardent Democrats agree that a big chunk of estate-tax relief must be kept permanently, perhaps by setting the rates and exemptions of 2009 in stone. That would cost “only” 40% as much as full repeal.
The eve of the baby-boomers’ retirement seems an odd time to abandon a small, but significant, source of tax revenue. If the tax permanently raised 0.3% of GDP, for instance, it would fix about half the hole in Social Security, the public pension plan. And with income becoming ever more concentrated among America’s richest—since 1980, the share of overall income going to the top 1% has doubled from 8% to 16%—it seems an odd time to abandon the country’s
I'm all for limiting how much money is passed on directly from one generation to the next to avoid the Paris Hiltons of the world, but for the Joes and Bobs, there should be a floor $ amount below which the government (oops, I mean "society") doesn't see a dime.
See here, specifically this part that seems directly applicable to yourself: “Over 70% of Americans support the abolition of the estate tax (inheritance tax), even though only one household in 100 pays it.”
Don’t tell me you’ve bought into the right-wingers’ rhetoric on this issue. Not with that signature line.
“Although the United States is seen as a world of opportunity, the reality may be different. Some studies have shown that it is easier for poorer children to rise through society in many European countries than in America. There is a particular fear about the engine of American meritocracy, its education system. Only 3% of students at top colleges come from the poorest quarter of the population. Poor children are trapped in dismal schools, while richer parents spend ever more cash on tutoring their offspring.”
MORE than any other country, America defines itself by a collective dream: the dream of economic opportunity and upward mobility. Its proudest boast is that it offers a chance of the good life to everybody who is willing to work hard and play by the rules. This ideal has made the United States the world’s strongest magnet for immigrants; it has also reconciled ordinary Americans to the rough side of a dynamic economy, with all its inequalities and insecurities. Who cares if the boss earns 300 times more than the average working stiff, if the stiff knows he can become the boss?
Look around the world and the supremacy of “the American model” might seem assured. No other rich country has so successfully harnessed the modern juggernauts of technology and globalisation. The hallmarks of American capitalism—a willingness to take risks, a light regulatory touch and sharp competition—have spawned enormous wealth. “This economy is powerful, productive and prosperous,” George Bush boasted recently, and by many yardsticks he is right. Growth is fast, unemployment is low and profits are fat. It is hardly surprising that so many other governments are trying to “Americanise” their economies—whether through the European Union’s Lisbon Agenda or Japan’s Koizumi reforms.
Yet many people feel unhappy about the American model—not least in the United States. Only one in four Americans believes the economy is in good shape. While firms’ profits have soared, wages for the typical worker have barely budged. The middle class—admittedly a vague term in America—feels squeezed. A college degree is no longer a passport to ever-higher pay. Now politicians are playing on these fears. From the left, populists complain about Mr Bush’s plutocratic friends exporting jobs abroad; from the right, nativists howl about immigrants wrecking the system.
A global argument The debate about the American model echoes far beyond the nation’s shores. Europeans have long held that America does not look after its poor—a prejudice reinforced by the ghastly scenes after Hurricane Katrina. The sharp decline in America’s image abroad has much to do with foreign policy, but Americanisation has also become synonymous with globalisation. Across the rich world, global competition is forcing economies to become more flexible, often increasing inequality; Japan is one example (see article). The logic of many non-Americans is that if globalisation makes their economy more like America’s, and the American model is defective, then free trade and open markets must be bad.
This debate mixes up three arguments—about inequality, meritocracy and immigration. The word that America should worry about most is the one you hear least—meritocracy.
Begin with inequality. The flip-side of America’s economic dynamism is that it has become more unequal—but in a more complex way than fir
IT IS arguably the most useful gadget in the space-farer’s toolkit. In “The Hitchhiker’s Guide to the Galaxy”, Douglas Adams depicted it as a “small, yellow and leech-like” fish, called a Babel fish, that you stick in your ear. In “Star Trek”, meanwhile, it is known simply as the Universal Language Translator. But whatever you call it, there is no doubting the practical value of a device that is capable of translating any language into another.
Remarkably, however, such devices are now on the verge of becoming a reality, thanks to new “statistical machine translation” software. Unlike previous approaches to machine translation, which relied upon rules identified by linguists which then had to be tediously hand-coded into software, this new method requires absolutely no linguistic knowledge or expert understanding of a language in order to translate it. And last month researchers at Carnegie Mellon University (CMU) in Pittsburgh began work on a machine that they hope will be able to learn a new language simply by getting foreign speakers to talk into it and perhaps, eventually, by watching television.
Within the next few years there will be an explosion in translation technologies, says Alex Waibel, director of the International Centre for Advanced Communication Technology, which is based jointly at the University of Karlsruhe in Germany and at CMU. He predicts there will be real-time automatic dubbing, which will let people watch foreign films or television programmes in their native languages, and search engines that will enable users to trawl through multilingual archives of documents, videos and audio files. And, eventually, there may even be electronic devices that work like Babel fish, whispering translations in your ear as someone speaks to you in a foreign tongue.
This may sound fanciful, but already a system has been developed that can translate speeches or lectures from one language into another, in real time and regardless of the subject matter. The system required no programming of grammatical rules or syntax. Instead it was given a vast number of speeches, and their accurate translations (performed by humans) into a second language, for statistical analysis. One of the reasons it works so well is that these speeches came from the United Nations and the European Parliament, where a broad range of topics are discussed. “The linguistic knowledge is automatically extracted from these huge data resources,” says Dr Waibel.
Statistical translation encompasses a range of techniques, but what they all have in common is the use of statistical analysis, rather than rigid rules, to convert text from one language into another. Most systems start with a large bilingual corpus of text. By analysing the frequency with which clusters of words appear in close proximity in the two languages, it is possible to work out which words correspond to each other in the two languages. This approach offers much greater flexibility than rule-based systems, since it translates languages based on how they are actually used, rather than relying on rigid grammatical rules which may not always be observed, and often have exceptions.
Examples abound of the ridiculous results produced by rule-based systems, which are unable to cope in the face of similes, ambiguities or bad grammar. In one example, a sentence written in Arabic meaning “The White House confirmed the existence of a new bin Laden tape” was translated using a standard rule-based translator and became “Alpine white new presence tape registered for cof
“The trouble is, the evidence does not back up this litany. First, energy and other natural resources have become more abundant, not less so since the Club of Rome published ‘The Limits to Growth’ in 1972. Second, more food is now produced per head of the world’s population than at any time in history. Fewer people are starving.”
IN 10,000 years, the earth’s population has doubled ten times, from less than 10m to more than six billion now and ten billion soon. Most of the calories that made that increase possible have come from three plants: maize, rice and wheat. The oldest, most widespread and until recently biggest of the three crops is wheat (see chart). To a first approximation wheat is the staple food of mankind, and its history is that of humanity.
Yet today, wheat is losing its crown. The tonnage (though not the acreage) of maize harvested in the world began consistently to exceed that of wheat for the first time in 1998; rice followed suit in 1999. Genetic modification, which has transformed maize, rice and soyabeans, has largely passed wheat by--to such an extent that it is in danger of becoming an “orphan crop”. The Atkins diet and a fashion for gluten allergies have made wheat seem less wholesome. And with population growth rates falling sharply while yields continue to rise, even the acreage devoted to wheat may now begin to decline for the first time since the stone age.
It is time to pay tribute to this strange little grass that has done so much for the human race. Strange is the word, for wheat is a genetic monster. A typical wheat variety is hexaploid--it has six copies of each gene, where most creatures have two. Its 21 chromosomes contain a massive 16 billion base pairs of DNA, 40 times as much as rice, six times as much as maize and five times as much as people. It is derived from three wild ancestral species in two separate mergers. The first took place in the Levant 10,000 years ago, the second near the Caspian Sea 2,000 years later. The result was a plant with extra-large seeds incapable of dispersal in the wild, dependent entirely on people to sow them.
The story actually starts much earlier, around 12,000 years ago. At the time, after several warm millennia, a melting ice sheet in North America collapsed and a gigantic lake drained into the North Atlantic through the St Lawrence seaway. The torrent of cool, fresh water altered the climate so drastically that the ice age, which had been in full retreat, resumed for a further 11 centuries. The Scandinavian ice sheet surged south. Western Asia became not only cooler, but much drier. The Black Sea all but dried out.
People in what is now Syria had been subsisting happily on a diet of acorns, gazelles and grass seeds. The centuries of drought drove them to depend increasingly on wild grass seeds. Abruptly, soon after 11,000 years ago, they began to cultivate rye and chickpeas, then einkorn and emmer, two ancestors of wheat, and later barley. Soon cultivated grain was their staple food. It happened first in the Karacadag Mountains in south-eastern Turkey--it is only here that wild einkorn grass contains the identical genetic fingerprint of modern domesticated wheat.
Who first replanted the seeds and why? For a start, he was probably a she: women have primary responsibilities for plant gathering in hunter-gatherer societies. The time was certainly ripe for agriculture: the ability to make tools and control fire (cooking makes many plants more digestible) was already well established. But was it an act of inspiration or desperation? Did it perhaps happen by accident, as discarded grains germinated around human settlements?
TWO years ago next month space travel underwent its Wright-brothers moment with the first flight of SpaceShipOne. The roles of Orville and Wilbur were played by Burt Rutan, who designed the craft, and Mike Melvill, who flew it—although they were ably assisted by Paul Allen, one of the founders of Microsoft, who paid for it. Of course, history never repeats itself exactly. Unlike the brothers Wright, who were heirs to a series of heroic failures when it came to powered heavier-than-air flight, Messrs Rutan and Melvill knew that manned spaceflight was possible. What they showed was that it is not just a game for governments. Private individuals can play, too.
Now, lots of people want to join in, and most of them have just met up at the International Space Development Conference in Los Angeles, to engage in that mixture of camaraderie and competition that characterises the beginnings of a new technology. And, as might be expected, they are brimming with two of the necessary ingredients of success: ideas and money.
First, the money. So far, more than $1 billion is known to have been committed to building private spaceships and the infrastructure to support them. For example, Mr Rutan’s follow-up vehicle, SpaceShipTwo, is expected to cost its backers, Virgin Galactic, $240m for a fleet of five. The spaceport in New Mexico from which these are intended to fly will account for another $225m, although New Mexico’s government is planning to raise this money itself.
These are not small sums, of course. On the other hand, Virgin Galactic has already banked $14m of deposits towards the $200,000 fare from people who want to travel on SpaceShipTwo, even though it has yet to be built, let alone flown.
All this suggests that spaceflight, if not exactly entering the age of the common man, is at least entering the age of the moderately prosperous enthusiast. For entrepreneurs, it is no longer necessary to have billions of dollars to get into space; millions will now do. And for those who merely wish to travel there, and have a few hundred thousand in the bank, reality beckons—provided that at least one of the ideas actually works.
Chocks away As with aircraft a century ago, a plethora of designs are competing with each other, and there is no certainty about which will prevail. The initial goal is to build a “suborbital” vehicle. This will not have to develop the tremendous speed needed to go into orbit around the Earth. Instead, it will travel briefly into space, offering a short thrilling ride out of the atmosphere, a few minutes of weightlessness, and a spectacular view of the planet from about 100km. Four important criteria are how you take off, what fuel you use, what your craft is made of, and how you come back.
Most people’s vision of a rocket launch is straight up from the ground. But, of the five vehicles most likely to be developed (see table), two will actually be launched from the air. SpaceShipTwo will be carried to high altitude by a purpose-built aircraft known as Eve before its rocket motor is ignited. And Explorer, a vehicle being designed by Space Adventures, will be launched from the top of a high-altitude Russian research plane called the M-55X, according to Eric Anderson, the firm’s president and chief executive.
As Dennis Jenkins, a consultant engineer at NASA, America’s space agency, points out, this is similar to using a two-stage rocket to get into space, with the aircraft acting as the first stage. However, a plane offers several advantages over a throw-away boos
THE versificator, a machine described in George Orwell’s novel “1984”, automatically generated music for the hapless masses. The idea of removing humans from the creative process of making music, an art form so able to stir the soul, made for a good joke when the book was published in 1949. But today, computer programmers working in a new field called “music intelligence” are developing software capable of predicting which songs will become hits. This surprisingly accurate technology could profoundly change the way pop music is created.
The software uses a process called “spectral deconvolution” to isolate and analyse around 30 parameters that define a piece of music, including such things as sonic brilliance, octave, cadence, frequency range, fullness of sound, chord progression, timbre and “bend” (variations in pitch at the beginning and end of the same note). “Songs conform to a limited number of mathematical equations,” says Mike McCready of Platinum Blue, a music-intelligence company based in New York, that he founded last December. Platinum Blue has compiled a database of more than 3m successful musical arrangements, including data on their popularity in different markets.
To the human ear, music has changed a lot over the years. Music-intelligence software, however, can reveal striking similarities in the underlying parameters of two songs from different eras that, even to a trained ear, seem unrelated. According to Platinum Blue’s software, called Music Science, for example, a number of hit songs by U2 have a close kinship to some of Beethoven’ s compositions. If a song written today has parameters similar to those of a number of past hits, it could well be a hit too.
Carlos Quintero, a producer and remixer at Orixa Producciones in Madrid, recently tried out another music-intelligence system, called Hit Song Science (HSS). “It practically left me in shock, it’s stunning,” he says. Mr Quintero’s production company now has the most promising demo songs it receives from aspiring musicians evaluated by Polyphonic HMI, the Barcelona-based developer of HSS and Platinum Blue’s only serious competitor. (Both companies perform analyses in-house, rather than selling software.) The results—consisting of a graph, numerical scores, computer-generated comments and suggested changes—help Orixa’s managers decide which songs to produce. Then, during the recording and post-production phases, Orixa uses HSS to reanalyse successive versions of each track for fine-tuning.
Belief in music intelligence is spreading, as Polyphonic HMI and Platinum Blue rack up bull’s-eye predictions of success, including “Candy Shop” by 50 Cent, “Be the Girl” by Aslyn, “Unwritten” by Natasha Bedingfield, “She Says” by Howie Day, and “You’re Beautiful” by James Blunt. Still, labels that use music intelligence generally prefer to keep quiet about it, so non-disclosure agreements are common. “No one wants people to think their decisions are coming from a box,” says Ric Wake, an American producer of two Grammy-winning acts who routinely employs Music Science. Even so, the names of many customers have leaked out. They include Capitol Records, Universal Music Group, Sony Music, EMI and Casablanca Records. Labels sometimes don’t tell even their established artists when they use music intelligence to help decide which singles to promote.
The firewall is porous. Imaginative users can find ways of searching for sensitive topics such as news about Falun Gong, a banned spiritual movement. In Google, entering the words “Falun Gong” will cause the entire results page to be blocked, but “FLG movement” will not. Many Chinese internet-users are well practised in configuring their internet browsers to route page requests through unblocked proxy servers outside China. These help bypass the firewall.
“DO YOU know how serious a mistake you’ve made?” Yan Yuanzhang recalls an official asking him not long ago. Mr Yan had been summoned to Beijing’s Internet Propaganda Management Office to talk about his websites. They were causing, he was told, the Communist Party to lose face. They were providing material that foreign media could use to attack China. They were illegal and must be closed down within 24 hours.
“Farewell, worker comrades,” wrote Mr Yan in notices posted that day on his China-based websites, China Workers Net and Communist Net. Visitors could hear a lugubrious rendition of the communist anthem, the Internationale, through their computer speakers as they read. “Whether there is any hope of starting again, heaven knows.” He says now that he will relaunch one of the two sites on May 1st, this time on a server in Taiwan.
It is remarkable that the websites lasted as long as they did. Mr Yan, who is not a party member, launched them on May 1st last year to mark Labour Day. The aim, he says, was to provide platforms for a “leftist” critique of China’s embrace of “Dickensian capitalism”. They did not, as he tried to explain to the city government, attack the party itself or its leaders. But they did provide something the party abhors: uncensored news about worker unrest. In September he launched a bulletin board on which visitors could directly post their comments. Messages complained about corruption, the privatisation of state-owned enterprises and the hardships of unemployed workers.
As Mr Yan talks, he gets a text message on his mobile phone. It is from Tan Jiaming, a university student in southern China who has been running a website of similar outlook, Revolutionary Marxism. It too, the message says, has been closed. The student had posted a notice entitled “Strongly Protest the Snuffing Out of the China Workers Website by the Beijing Authorities”. He was summoned to hear a dozen officials threaten him with expulsion from his university for backing Mr Yan.
Six years ago Bill Clinton described China’s efforts to restrict the internet as “sort of like trying to nail Jell-O to the wall”. But as China’s web-filtering technology has grown more sophisticated, and the ranks of its internet police have swelled, some have begun to wonder. A report in 2003 by the Carnegie Endowment for International Peace suggested that, despite the difficulties the internet posed to authoritarian regimes, it could also be used to fortify them. China, the authors concluded, had been “largely successful at guiding use” of the internet. At a congressional hearing in February on American companies involved in internet business in China, a Republican congressman, Christopher Smith, said the internet there had become “a malicious tool, a cyber sledgehammer of repression”.
Some of the companies testifying at the hearing—Cisco, Google, Microsoft and
“The trouble is, the evidence does not back up this litany. First, energy and other natural resources have become more abundant, not less so since the Club of Rome published ‘The Limits to Growth’ in 1972. Second, more food is now produced per head of the world’s population than at any time in history. Fewer people are starving.”
IN 10,000 years, the earth’s population has doubled ten times, from less than 10m to more than six billion now and ten billion soon. Most of the calories that made that increase possible have come from three plants: maize, rice and wheat. The oldest, most widespread and until recently biggest of the three crops is wheat (see chart). To a first approximation wheat is the staple food of mankind, and its history is that of humanity.
Yet today, wheat is losing its crown. The tonnage (though not the acreage) of maize harvested in the world began consistently to exceed that of wheat for the first time in 1998; rice followed suit in 1999. Genetic modification, which has transformed maize, rice and soyabeans, has largely passed wheat by--to such an extent that it is in danger of becoming an “orphan crop”. The Atkins diet and a fashion for gluten allergies have made wheat seem less wholesome. And with population growth rates falling sharply while yields continue to rise, even the acreage devoted to wheat may now begin to decline for the first time since the stone age.
It is time to pay tribute to this strange little grass that has done so much for the human race. Strange is the word, for wheat is a genetic monster. A typical wheat variety is hexaploid--it has six copies of each gene, where most creatures have two. Its 21 chromosomes contain a massive 16 billion base pairs of DNA, 40 times as much as rice, six times as much as maize and five times as much as people. It is derived from three wild ancestral species in two separate mergers. The first took place in the Levant 10,000 years ago, the second near the Caspian Sea 2,000 years later. The result was a plant with extra-large seeds incapable of dispersal in the wild, dependent entirely on people to sow them.
The story actually starts much earlier, around 12,000 years ago. At the time, after several warm millennia, a melting ice sheet in North America collapsed and a gigantic lake drained into the North Atlantic through the St Lawrence seaway. The torrent of cool, fresh water altered the climate so drastically that the ice age, which had been in full retreat, resumed for a further 11 centuries. The Scandinavian ice sheet surged south. Western Asia became not only cooler, but much drier. The Black Sea all but dried out.
People in what is now Syria had been subsisting happily on a diet of acorns, gazelles and grass seeds. The centuries of drought drove them to depend increasingly on wild grass seeds. Abruptly, soon after 11,000 years ago, they began to cultivate rye and chickpeas, then einkorn and emmer, two ancestors of wheat, and later barley. Soon cultivated grain was their staple food. It happened first in the Karacadag Mountains in south-eastern Turkey--it is only here that wild einkorn grass contains the identical genetic fingerprint of modern domesticated wheat.
Who first replanted the seeds and why? For a start, he was probably a she: women have primary responsibilities for plant gathering in hunter-gatherer societies. The time was certainly ripe for agriculture: the ability to make tools and control fire (cooking makes many plants more digestible) was already well established. But was it an act of inspiration or desperation? Did it perhaps happen by accident, as discarded grains germinated around human settlements?
As it happens, The Economist recently ran an article addressing some of these issues. The article also provides context and perspective that should be of interest to those participating in this discussion. For convenience, the full text is reproduced below; it is also accessible online (may require paid subscription).
----
Computing
From factoids to facts
Aug 26th 2004 | REDMOND, WASHINGTON
From The Economist print edition
At last, a way of getting answers from the web
WHAT is the next stage in the evolution of internet search engines? AltaVista demonstrated that indexing the entire world wide web was feasible. Google's success stems from its uncanny ability to sort useful web pages from dross. But the real prize will surely go to whoever can use the web to deliver a straight answer to a straight question. And Eric Brill, a researcher at Microsoft, intends that his firm will be the first to do that.
Dr Brill's initial crack at the problem is a system called "Ask MSR" (MSR stands for Microsoft Research). This program uses information on web pages to respond to questions to which the answer is a single word or phrase--such as "When was Marilyn Monroe born?" Ask MSR starts by manipulating the question in various ways: by identifying the verb, for example, and then changing its tense or moving it into different positions in the sentence ("Marilyn was Monroe born", "Marilyn Monroe was born" and so on). The resulting phrases are then fed into a search engine, and documents containing matching strings of words are retrieved. It sounds a promiscuous strategy, but gibberish phrases produce few matches, so, as Dr Brill puts it, "being wrong is very cheap."
Once accumulated, the pile of documents is scanned for possible answers, and these are ranked by frequency. In practice, the correct answer appears in one of the first three places around 75% of the time. That might not sound very good, but human intelligence provides a second filter, since wrong answers are often obvious. If you ask how many times Bjorn Borg won Wimbledon, for example, "1980" is not a plausible answer, but "5" is. If in doubt, clicking on an answer produces a list of links to pages which provide support for that answer.
Ask MSR is still a prototype, although Microsoft is trying to improve it and it may be launched commercially under the name AnswerBot. Dr Brill, meanwhile, has moved to a more difficult task. One of his most recent papers, written jointly with Radu Soricut of the University of Southern California, is entitled "Beyond the Factoid". It describes his efforts to build a system capable of providing 50-word answers to questions such as "What are the rules for qualifying for the Academy Awards?" This is harder than finding a single-word answer, but Dr Brill thinks it should be possible using something called a "noisy channel" model.
Such models are already employed in spell-checking and speech-recognition systems. They work by modelling the transformation between what a user means (in spell-checking, the word he intended to type) and what he does (the garbled word actually typed). Just as a telephone line distorts the voice of the person at the other end of the line, this process can be thought of as being a noisy channel that transforms the user's intention into something rather different.
By analysing many pairs of correct and mis-spelled words using statistical techniques, it is possible to predict how such transformations work in general cases. A system can then be designed to work the process backwards. Given a mis-spelled word, it can guess what that word is most likely to be a mis-spelling of.
Dr Brill's question-answering system does something similar. Many question-and-answer pairs exist on the web, in the form of "frequently asked questions" (FAQ) pages. Dr Brill trained his system using a million such pairs, to create a model that, given
Slashdot Mirror
To prove the point, they actually did such a reversal in the case of telephone-queue waiting times. Traditionally, these have been assumed to follow a Poisson distribution, but some recent research suggests they actually follow a power law. Analysing the participants’ responses suggests that a power law, indeed, it is.
——
Science & Technology / Psychology
Bayes rules
Jan 5th 2006
From The Economist print edition
A once-neglected statistical technique may help to explain how the mind works
IMAGE
SCIENCE, being a human activity, is not immune to fashion. For example, one of the first mathematicians to study the subject of probability theory was an English clergyman called Thomas Bayes, who was born in 1702 and died in 1761. His ideas about the prediction of future events from one or two examples were popular for a while, and have never been fundamentally challenged. But they were eventually overwhelmed by those of the “frequentist” school, which developed the methods based on sampling from a large population that now dominate the field and are used to predict things as diverse as the outcomes of elections and preferences for chocolate bars.
Recently, however, Bayes’s ideas have made a comeback among computer scientists trying to design software with human-like intelligence. Bayesian reasoning now lies at the heart of leading internet search engines and automated “help wizards”. That has prompted some psychologists to ask if the human brain itself might be a Bayesian-reasoning machine. They suggest that the Bayesian capacity to draw strong inferences from sparse data could be crucial to the way the mind perceives the world, plans actions, comprehends and learns language, reasons from correlation to causation, and even understands the goals and beliefs of other minds.
These researchers have conducted laboratory experiments that convince them they are on the right track, but only recently have they begun to look at whether the brain copes with everyday judgments in the real world in a Bayesian manner. In research to be published later this year in Psychological Science, Thomas Griffiths of Brown University in Rhode Island and Joshua Tenenbaum of the Massachusetts Institute of Technology put the idea of a Bayesian brain to a quotidian test. They found that it passes with flying colours.
Prior assumptions
The key to successful Bayesian reasoning is not in having an extensive, unbiased sample, which is the eternal worry of frequentists, but rather in having an appropriate “prior”, as it is known to the cognoscenti. This prior is an assumption about the way the world works—in essence, a hypothesis about reality—that can be expressed as a mathematical probability distribution of the frequency with which events of a particular magnitude happen.
The best known of these probability distributions is the “normal”, or Gaussian distribution. This has a curve similar to the cross-section of a bell, with events of middling magnitude being common, and those of small and large magnitude rare, so it is sometimes known by a third name, the bell-curve distribution. But there are also the Poisson distribution, the Erlang distribution, the power-law distribution and many even weirder ones that are not the consequence of simple mathematical equations (or, at least, of equations that mathematicians regard as simple).
With the correct prior, even a single piece of data can be used to make meaningful Bayesian predictions. By contrast frequentists, though they deal with the same probability distributions as Bayesians, make fewer prior assumptions about the distribution that applies in any particular situation. Frequentism is thus a more robust approach, but one that is not well suited to
Business / Face value
Ozzie the wizard
Jun 22nd 2006
From The Economist print edition
Bill Gates replaces himself as Microsoft’s software boss with Ray Ozzie, his top choice but one
IMAGE (AP)
THE co-founder, chairman and “chief software architect” of Microsoft, the world’s largest software company, would deny it on his life, but the one person Bill Gates admires most for his geeky prowess—and might have chosen to succeed him as software architect—is almost certainly Steve Jobs. Unfortunately, Mr Jobs, the co-founder of Apple Computer and victim of Mr Gates’s predatory business instincts during the 1980s and 1990s, cannot be considered available, since he is busy leading Apple’s renaissance as a builder of gadgets and software that, in the opinion of his fans, put Microsoft to shame. So Mr Gates spent years courting the geek he admires second most, a software pioneer named Ray Ozzie.
After many overtures, Microsoft last year bought Mr Ozzie’s company, Groove Networks, and thus brought Mr Ozzie and his brother Jack inside the Microsoft tent. Mr Gates then groomed Mr Ozzie to take the lead in defining Microsoft’s direction at the highest (ie, software) level, alongside Steve Ballmer, the chief executive, who will continue to use his prodigious energies to bash heads together in the name of implementation, and Craig Mundie, who will oversee Microsoft’s research efforts and policy lobbying. On June 15th Mr Gates made the transition official, announcing that he would hand over his role as chief software architect to Mr Ozzie in two years’ time, in order to concentrate on becoming the world’s greatest philanthropist.
Mr Ozzie, at 50, is one month younger than Mr Gates and nine months younger than Mr Jobs. Like the other two, he has been admitted to the Computer History Museum’s “hall of fame” for his role in bringing about the PC era. As a kid in suburban Chicago, Mr Ozzie was already soldering all sorts of dangerous circuits together in a guest bedroom, but it was at college in the 1970s that he discovered his passion, which was, as he once put it, “to augment relationships” among human beings through technology. The catalyst was his encounter with PLATO, a cluster of a thousand dumb terminals that were connected to a mainframe and that Mr Ozzie and his friends playfully used to communicate, by exchanging what would today be called e-mails and instant messages. This experience so captured his imagination that he devoted his next three decades to writing software that enables “collaboration”.
His single biggest breakthrough came in the 1980s, when Mr Ozzie personally wrote a million of the first 3.5m lines of code for the first successful collaboration software, Lotus Notes. At a time when nobody had heard of the world wide web, Lotus Notes already offered “workspaces” not unlike today’s wikis. Mr Gates watched with his usual mixture of emotions towards innovations by others—envy and grudging admiration. These probably gave way to anxiety when IBM, Microsoft’s partner-turned-enemy, bought Lotus in 1995 for $3.5 billion.
Mr Ozzie soon parted with IBM and again set out to take collaboration to the next level, this time by using new ideas about decentralised networks between people and their computers. This approach would subsequently be called “peer-to-peer” and has since been made famous by Napster, a method for “sharing” music, and Skype, a service for free internet telephony. Mr Ozzie’s company, Groove, was not a commercial success this time, but Mr Gates and others in the industry nonetheless saw the idea and recognised its potential. Last April Messrs Gates and Ozzie joined forces.
One reason why Mr Gat
Leaders / Copyrights
A radical rethink
Jan 23rd 2003
From The Economist print edition
The best way to foster creativity in the digital age is to overhaul current copyright laws
IMAGE (Reuters)
CRITICS have derided a 1998 extension of American copyrights as the “Mickey Mouse Protection Act” because it stopped early images of the Disney company’s mascot from entering the public domain. But such laws, they argue, are no joke. Extending and strengthening copyrights, they claim, will help a handful of big corporations crush creativity in the digital age. On the contrary, say Hollywood studios and big record companies. Without stronger copyright protection, a wave of piracy will destroy their industries, depriving consumers everywhere of a broad choice of movies, music and books.
Last week America’s Supreme Court weighed into what is rapidly becoming a nasty worldwide battle about the scope and enforcement of copyrights, by rejecting a challenge to the 1998 law on constitutional grounds. But even as it upheld the law, the court expressed misgivings. Blistering dissents from two justices dismissed the 20-year extension of copyright as unwarranted, and even the majority’s opinion hinted that Congress’s decision may have been “unwise”.
The court’s ambivalence is understandable. The growing quarrel over copyright is just one of the many difficult issues thrown up by the spread of the internet and related technologies (see our survey of the internet society in this issue). But of all these issues, the copyright battle is becoming one of the most urgent, and bitterly fought, because it could yet determine the future character of cyberspace itself.
Both sides have a point. Digital piracy does indeed threaten to overwhelm so-called “content” industries. As the power and reach of the internet continue to grow, the illicit trading of perfect copies may well devastate the music, movie and publishing industries. The content industries want to protect themselves with anti-copying technology, backed by stronger laws. So far, they have been at loggerheads with technology firms about how to implement such schemes (see article). But a deal between Hollywood and Silicon Valley is likely eventually. Critics are right to fear that, when such a deal is struck, it will be in the interests of big firms, not the public.
A grand new bargain
The alternative is to return to the original purpose of copyright, something no national legislature has yet been willing to do. Copyright was originally the grant of a temporary government-supported monopoly on copying a work, not a property right. Its sole purpose was to encourage the circulation of ideas by giving creators and publishers a short-term incentive to disseminate their work. Over the past 50 years, as a result of heavy lobbying by content industries, copyright has grown to such ludicrous proportions that it now often inhibits rather than promotes the circulation of ideas, leaving thousands of old movies, records and books languishing behind a legal barrier. Starting from scratch today, no rational, disinterested lawmaker would agree to copyrights that extend to 70 years after an author’s death, now the norm in the developed world.
Digital technologies are not only making it easier to copy all sorts of works, but also sharply reducing the costs of creating or distributing them, and so also reducing the required incentives. The flood of free content on the internet has shown that most creators do not need incentives that stretch across generations. To reward those who can attract a paying audience, and the firms that support them, much shorter copyrights would be enough. The 14-year term of the original 18th-centur
United States / Death and taxes
Gilding the elite
Jun 8th 2006 | WASHINGTON, DC
From The Economist print edition
Whatever its outcome, the battle in Congress over the estate tax bodes badly for America’s fiscal future
IMAGE (Corbis)
WHAT weighty issues were preoccupying America’s Senate this week? Soaring health-care costs perhaps, or the fiscal consequences of retiring baby-boomers? Dream on. The country’s top lawmakers argued about a constitutional amendment to ban gay marriage, which had no hope of passing; a daft law to give native Hawaiians their own government; and repeal of the estate tax, a levy that affects only the richest 0.5% of Americans.
These skewed priorities may be blamed on Bill Frist, the Senate majority leader, who sets the agenda and is an aspiring presidential candidate. To bolster his appeal to the Republican Party’s conservative base, he has been pushing for votes on issues the right cares about: flag-burning, gay marriage and, above all, getting rid for ever of the estate tax.
America has had this tax, a levy on the transfer of large fortunes between generations, since 1916. Like Britain, America taxes the estate of the deceased; most other countries which impose such a tax target heirs and their inheritances. The estate tax affects very few Americans (never much over 2% in recent decades, and now far fewer). In the late 1990s it generated just over 1.5% of all tax revenue, or around 0.3% of GDP. Conservatives long for its abolition, which is now tantalisingly close.
George Bush’s 2001 tax cuts created a paradox: the estate tax is in terminal decline, but only temporarily so. Under current law, only estates worth more than $7m for a couple will be taxed in 2009 (compared with a joint exemption of $1.35m in 2000). In 2010 the tax goes altogether. But since all Mr Bush’s tax cuts technically expire in 2011, the abolition lasts only one year. In 2011 the estate tax returns to its pre-Bush form.
This mess clearly needs fixing, if only to avoid a surge in suspicious deaths in Palm Beach in 2010. But how? The House of Representatives has voted several times to make repeal permanent. The Senate meant to do so last year, until Hurricane Katrina’s havoc made it look too callous to give a big tax break to multi-millionaires.
IMAGE
As The Economist went to press, it was not clear whether Mr Frist would succeed this time. Several Democratic senators support the tax’s abolition, but a handful of Republicans think full repeal is too expensive. By one estimate, repealing the estate tax would reduce federal revenues by some $776 billion, or around 0.4% of GDP, between 2012 and 2021. (The costs of repeal are more than the tax actually raises because the bean-counters assume that rich people will shift their assets in a way that reduces income taxes.) One possibility is a “compromise” proposal that would increase the exemption per couple to $10m and cut rates to 15%, a combination that would, in fact, cost almost as much as repeal. Even ardent Democrats agree that a big chunk of estate-tax relief must be kept permanently, perhaps by setting the rates and exemptions of 2009 in stone. That would cost “only” 40% as much as full repeal.
The eve of the baby-boomers’ retirement seems an odd time to abandon a small, but significant, source of tax revenue. If the tax permanently raised 0.3% of GDP, for instance, it would fix about half the hole in Social Security, the public pension plan. And with income becoming ever more concentrated among America’s richest—since 1980, the share of overall income going to the top 1% has doubled from 8% to 16%—it seems an odd time to abandon the country’s
Don’t tell me you’ve bought into the right-wingers’ rhetoric on this issue. Not with that signature line.
“Although the United States is seen as a world of opportunity, the reality may be different. Some studies have shown that it is easier for poorer children to rise through society in many European countries than in America. There is a particular fear about the engine of American meritocracy, its education system. Only 3% of students at top colleges come from the poorest quarter of the population. Poor children are trapped in dismal schools, while richer parents spend ever more cash on tutoring their offspring.”
——
Leaders / The United States
Inequality and the American Dream
Jun 15th 2006
From The Economist print edition
The world’s most impressive economic machine needs a little adjusting
IMAGE
MORE than any other country, America defines itself by a collective dream: the dream of economic opportunity and upward mobility. Its proudest boast is that it offers a chance of the good life to everybody who is willing to work hard and play by the rules. This ideal has made the United States the world’s strongest magnet for immigrants; it has also reconciled ordinary Americans to the rough side of a dynamic economy, with all its inequalities and insecurities. Who cares if the boss earns 300 times more than the average working stiff, if the stiff knows he can become the boss?
Look around the world and the supremacy of “the American model” might seem assured. No other rich country has so successfully harnessed the modern juggernauts of technology and globalisation. The hallmarks of American capitalism—a willingness to take risks, a light regulatory touch and sharp competition—have spawned enormous wealth. “This economy is powerful, productive and prosperous,” George Bush boasted recently, and by many yardsticks he is right. Growth is fast, unemployment is low and profits are fat. It is hardly surprising that so many other governments are trying to “Americanise” their economies—whether through the European Union’s Lisbon Agenda or Japan’s Koizumi reforms.
Yet many people feel unhappy about the American model—not least in the United States. Only one in four Americans believes the economy is in good shape. While firms’ profits have soared, wages for the typical worker have barely budged. The middle class—admittedly a vague term in America—feels squeezed. A college degree is no longer a passport to ever-higher pay. Now politicians are playing on these fears. From the left, populists complain about Mr Bush’s plutocratic friends exporting jobs abroad; from the right, nativists howl about immigrants wrecking the system.
A global argument
The debate about the American model echoes far beyond the nation’s shores. Europeans have long held that America does not look after its poor—a prejudice reinforced by the ghastly scenes after Hurricane Katrina. The sharp decline in America’s image abroad has much to do with foreign policy, but Americanisation has also become synonymous with globalisation. Across the rich world, global competition is forcing economies to become more flexible, often increasing inequality; Japan is one example (see article). The logic of many non-Americans is that if globalisation makes their economy more like America’s, and the American model is defective, then free trade and open markets must be bad.
This debate mixes up three arguments—about inequality, meritocracy and immigration. The word that America should worry about most is the one you hear least—meritocracy.
Begin with inequality. The flip-side of America’s economic dynamism is that it has become more unequal—but in a more complex way than fir
Technology Quarterly
How to build a Babel fish
Jun 8th 2006
From The Economist print edition
Translation software: The science-fiction dream of a machine that understands any language is getting slowly closer
IMAGE
IT IS arguably the most useful gadget in the space-farer’s toolkit. In “The Hitchhiker’s Guide to the Galaxy”, Douglas Adams depicted it as a “small, yellow and leech-like” fish, called a Babel fish, that you stick in your ear. In “Star Trek”, meanwhile, it is known simply as the Universal Language Translator. But whatever you call it, there is no doubting the practical value of a device that is capable of translating any language into another.
Remarkably, however, such devices are now on the verge of becoming a reality, thanks to new “statistical machine translation” software. Unlike previous approaches to machine translation, which relied upon rules identified by linguists which then had to be tediously hand-coded into software, this new method requires absolutely no linguistic knowledge or expert understanding of a language in order to translate it. And last month researchers at Carnegie Mellon University (CMU) in Pittsburgh began work on a machine that they hope will be able to learn a new language simply by getting foreign speakers to talk into it and perhaps, eventually, by watching television.
Within the next few years there will be an explosion in translation technologies, says Alex Waibel, director of the International Centre for Advanced Communication Technology, which is based jointly at the University of Karlsruhe in Germany and at CMU. He predicts there will be real-time automatic dubbing, which will let people watch foreign films or television programmes in their native languages, and search engines that will enable users to trawl through multilingual archives of documents, videos and audio files. And, eventually, there may even be electronic devices that work like Babel fish, whispering translations in your ear as someone speaks to you in a foreign tongue.
This may sound fanciful, but already a system has been developed that can translate speeches or lectures from one language into another, in real time and regardless of the subject matter. The system required no programming of grammatical rules or syntax. Instead it was given a vast number of speeches, and their accurate translations (performed by humans) into a second language, for statistical analysis. One of the reasons it works so well is that these speeches came from the United Nations and the European Parliament, where a broad range of topics are discussed. “The linguistic knowledge is automatically extracted from these huge data resources,” says Dr Waibel.
Statistical translation encompasses a range of techniques, but what they all have in common is the use of statistical analysis, rather than rigid rules, to convert text from one language into another. Most systems start with a large bilingual corpus of text. By analysing the frequency with which clusters of words appear in close proximity in the two languages, it is possible to work out which words correspond to each other in the two languages. This approach offers much greater flexibility than rule-based systems, since it translates languages based on how they are actually used, rather than relying on rigid grammatical rules which may not always be observed, and often have exceptions.
Examples abound of the ridiculous results produced by rule-based systems, which are unable to cope in the face of similes, ambiguities or bad grammar. In one example, a sentence written in Arabic meaning “The White House confirmed the existence of a new bin Laden tape” was translated using a standard rule-based translator and became “Alpine white new presence tape registered for cof
“The trouble is, the evidence does not back up this litany. First, energy and other natural resources have become more abundant, not less so since the Club of Rome published ‘The Limits to Growth’ in 1972. Second, more food is now produced per head of the world’s population than at any time in history. Fewer people are starving.”
--
The story of wheat
Ears of plenty
Dec 20th 2005
From The Economist print edition
The story of man’s staple food
[Image] (Still Pictures)
IN 10,000 years, the earth’s population has doubled ten times, from less than 10m to more than six billion now and ten billion soon. Most of the calories that made that increase possible have come from three plants: maize, rice and wheat. The oldest, most widespread and until recently biggest of the three crops is wheat (see chart). To a first approximation wheat is the staple food of mankind, and its history is that of humanity.
Yet today, wheat is losing its crown. The tonnage (though not the acreage) of maize harvested in the world began consistently to exceed that of wheat for the first time in 1998; rice followed suit in 1999. Genetic modification, which has transformed maize, rice and soyabeans, has largely passed wheat by--to such an extent that it is in danger of becoming an “orphan crop”. The Atkins diet and a fashion for gluten allergies have made wheat seem less wholesome. And with population growth rates falling sharply while yields continue to rise, even the acreage devoted to wheat may now begin to decline for the first time since the stone age.
It is time to pay tribute to this strange little grass that has done so much for the human race. Strange is the word, for wheat is a genetic monster. A typical wheat variety is hexaploid--it has six copies of each gene, where most creatures have two. Its 21 chromosomes contain a massive 16 billion base pairs of DNA, 40 times as much as rice, six times as much as maize and five times as much as people. It is derived from three wild ancestral species in two separate mergers. The first took place in the Levant 10,000 years ago, the second near the Caspian Sea 2,000 years later. The result was a plant with extra-large seeds incapable of dispersal in the wild, dependent entirely on people to sow them.
The story actually starts much earlier, around 12,000 years ago. At the time, after several warm millennia, a melting ice sheet in North America collapsed and a gigantic lake drained into the North Atlantic through the St Lawrence seaway. The torrent of cool, fresh water altered the climate so drastically that the ice age, which had been in full retreat, resumed for a further 11 centuries. The Scandinavian ice sheet surged south. Western Asia became not only cooler, but much drier. The Black Sea all but dried out.
People in what is now Syria had been subsisting happily on a diet of acorns, gazelles and grass seeds. The centuries of drought drove them to depend increasingly on wild grass seeds. Abruptly, soon after 11,000 years ago, they began to cultivate rye and chickpeas, then einkorn and emmer, two ancestors of wheat, and later barley. Soon cultivated grain was their staple food. It happened first in the Karacadag Mountains in south-eastern Turkey--it is only here that wild einkorn grass contains the identical genetic fingerprint of modern domesticated wheat.
Who first replanted the seeds and why? For a start, he was probably a she: women have primary responsibilities for plant gathering in hunter-gatherer societies. The time was certainly ripe for agriculture: the ability to make tools and control fire (cooking makes many plants more digestible) was already well established. But was it an act of inspiration or desperation? Did it perhaps happen by accident, as discarded grains germinated around human settlements?
Science & Technology / Private spaceflight
Rocket renaissance
May 11th 2006 | LOS ANGELES
From The Economist print edition
The era of private spaceflight is about to dawn
IMAGE (Mary Evans)
TWO years ago next month space travel underwent its Wright-brothers moment with the first flight of SpaceShipOne. The roles of Orville and Wilbur were played by Burt Rutan, who designed the craft, and Mike Melvill, who flew it—although they were ably assisted by Paul Allen, one of the founders of Microsoft, who paid for it. Of course, history never repeats itself exactly. Unlike the brothers Wright, who were heirs to a series of heroic failures when it came to powered heavier-than-air flight, Messrs Rutan and Melvill knew that manned spaceflight was possible. What they showed was that it is not just a game for governments. Private individuals can play, too.
Now, lots of people want to join in, and most of them have just met up at the International Space Development Conference in Los Angeles, to engage in that mixture of camaraderie and competition that characterises the beginnings of a new technology. And, as might be expected, they are brimming with two of the necessary ingredients of success: ideas and money.
First, the money. So far, more than $1 billion is known to have been committed to building private spaceships and the infrastructure to support them. For example, Mr Rutan’s follow-up vehicle, SpaceShipTwo, is expected to cost its backers, Virgin Galactic, $240m for a fleet of five. The spaceport in New Mexico from which these are intended to fly will account for another $225m, although New Mexico’s government is planning to raise this money itself.
These are not small sums, of course. On the other hand, Virgin Galactic has already banked $14m of deposits towards the $200,000 fare from people who want to travel on SpaceShipTwo, even though it has yet to be built, let alone flown.
All this suggests that spaceflight, if not exactly entering the age of the common man, is at least entering the age of the moderately prosperous enthusiast. For entrepreneurs, it is no longer necessary to have billions of dollars to get into space; millions will now do. And for those who merely wish to travel there, and have a few hundred thousand in the bank, reality beckons—provided that at least one of the ideas actually works.
Chocks away
As with aircraft a century ago, a plethora of designs are competing with each other, and there is no certainty about which will prevail. The initial goal is to build a “suborbital” vehicle. This will not have to develop the tremendous speed needed to go into orbit around the Earth. Instead, it will travel briefly into space, offering a short thrilling ride out of the atmosphere, a few minutes of weightlessness, and a spectacular view of the planet from about 100km. Four important criteria are how you take off, what fuel you use, what your craft is made of, and how you come back.
Most people’s vision of a rocket launch is straight up from the ground. But, of the five vehicles most likely to be developed (see table), two will actually be launched from the air. SpaceShipTwo will be carried to high altitude by a purpose-built aircraft known as Eve before its rocket motor is ignited. And Explorer, a vehicle being designed by Space Adventures, will be launched from the top of a high-altitude Russian research plane called the M-55X, according to Eric Anderson, the firm’s president and chief executive.
As Dennis Jenkins, a consultant engineer at NASA, America’s space agency, points out, this is similar to using a two-stage rocket to get into space, with the aircraft acting as the first stage. However, a plane offers several advantages over a throw-away boos
Technology Quarterly
Sounds good?
Jun 8th 2006
From The Economist print edition
Software: “Music intelligence” systems that can distinguish hits from misses could change the way pop music is made and marketed
IMAGE
THE versificator, a machine described in George Orwell’s novel “1984”, automatically generated music for the hapless masses. The idea of removing humans from the creative process of making music, an art form so able to stir the soul, made for a good joke when the book was published in 1949. But today, computer programmers working in a new field called “music intelligence” are developing software capable of predicting which songs will become hits. This surprisingly accurate technology could profoundly change the way pop music is created.
The software uses a process called “spectral deconvolution” to isolate and analyse around 30 parameters that define a piece of music, including such things as sonic brilliance, octave, cadence, frequency range, fullness of sound, chord progression, timbre and “bend” (variations in pitch at the beginning and end of the same note). “Songs conform to a limited number of mathematical equations,” says Mike McCready of Platinum Blue, a music-intelligence company based in New York, that he founded last December. Platinum Blue has compiled a database of more than 3m successful musical arrangements, including data on their popularity in different markets.
To the human ear, music has changed a lot over the years. Music-intelligence software, however, can reveal striking similarities in the underlying parameters of two songs from different eras that, even to a trained ear, seem unrelated. According to Platinum Blue’s software, called Music Science, for example, a number of hit songs by U2 have a close kinship to some of Beethoven’ s compositions. If a song written today has parameters similar to those of a number of past hits, it could well be a hit too.
Carlos Quintero, a producer and remixer at Orixa Producciones in Madrid, recently tried out another music-intelligence system, called Hit Song Science (HSS). “It practically left me in shock, it’s stunning,” he says. Mr Quintero’s production company now has the most promising demo songs it receives from aspiring musicians evaluated by Polyphonic HMI, the Barcelona-based developer of HSS and Platinum Blue’s only serious competitor. (Both companies perform analyses in-house, rather than selling software.) The results—consisting of a graph, numerical scores, computer-generated comments and suggested changes—help Orixa’s managers decide which songs to produce. Then, during the recording and post-production phases, Orixa uses HSS to reanalyse successive versions of each track for fine-tuning.
Belief in music intelligence is spreading, as Polyphonic HMI and Platinum Blue rack up bull’s-eye predictions of success, including “Candy Shop” by 50 Cent, “Be the Girl” by Aslyn, “Unwritten” by Natasha Bedingfield, “She Says” by Howie Day, and “You’re Beautiful” by James Blunt. Still, labels that use music intelligence generally prefer to keep quiet about it, so non-disclosure agreements are common. “No one wants people to think their decisions are coming from a box,” says Ric Wake, an American producer of two Grammy-winning acts who routinely employs Music Science. Even so, the names of many customers have leaked out. They include Capitol Records, Universal Music Group, Sony Music, EMI and Casablanca Records. Labels sometimes don’t tell even their established artists when they use music intelligence to help decide which singles to promote.
Revenues at Polyphonic HMI will exce
The firewall is porous. Imaginative users can find ways of searching for sensitive topics such as news about Falun Gong, a banned spiritual movement. In Google, entering the words “Falun Gong” will cause the entire results page to be blocked, but “FLG movement” will not. Many Chinese internet-users are well practised in configuring their internet browsers to route page requests through unblocked proxy servers outside China. These help bypass the firewall.
——
Special Report / China and the internet
The party, the people and the power of cyber-talk
Apr 27th 2006 | BEIJING
From The Economist print edition
At present the party has the upper hand. It is starting to sweat, though
IMAGE
“DO YOU know how serious a mistake you’ve made?” Yan Yuanzhang recalls an official asking him not long ago. Mr Yan had been summoned to Beijing’s Internet Propaganda Management Office to talk about his websites. They were causing, he was told, the Communist Party to lose face. They were providing material that foreign media could use to attack China. They were illegal and must be closed down within 24 hours.
“Farewell, worker comrades,” wrote Mr Yan in notices posted that day on his China-based websites, China Workers Net and Communist Net. Visitors could hear a lugubrious rendition of the communist anthem, the Internationale, through their computer speakers as they read. “Whether there is any hope of starting again, heaven knows.” He says now that he will relaunch one of the two sites on May 1st, this time on a server in Taiwan.
It is remarkable that the websites lasted as long as they did. Mr Yan, who is not a party member, launched them on May 1st last year to mark Labour Day. The aim, he says, was to provide platforms for a “leftist” critique of China’s embrace of “Dickensian capitalism”. They did not, as he tried to explain to the city government, attack the party itself or its leaders. But they did provide something the party abhors: uncensored news about worker unrest. In September he launched a bulletin board on which visitors could directly post their comments. Messages complained about corruption, the privatisation of state-owned enterprises and the hardships of unemployed workers.
As Mr Yan talks, he gets a text message on his mobile phone. It is from Tan Jiaming, a university student in southern China who has been running a website of similar outlook, Revolutionary Marxism. It too, the message says, has been closed. The student had posted a notice entitled “Strongly Protest the Snuffing Out of the China Workers Website by the Beijing Authorities”. He was summoned to hear a dozen officials threaten him with expulsion from his university for backing Mr Yan.
IMAGE
Six years ago Bill Clinton described China’s efforts to restrict the internet as “sort of like trying to nail Jell-O to the wall”. But as China’s web-filtering technology has grown more sophisticated, and the ranks of its internet police have swelled, some have begun to wonder. A report in 2003 by the Carnegie Endowment for International Peace suggested that, despite the difficulties the internet posed to authoritarian regimes, it could also be used to fortify them. China, the authors concluded, had been “largely successful at guiding use” of the internet. At a congressional hearing in February on American companies involved in internet business in China, a Republican congressman, Christopher Smith, said the internet there had become “a malicious tool, a cyber sledgehammer of repression”.
Some of the companies testifying at the hearing—Cisco, Google, Microsoft and
“The trouble is, the evidence does not back up this litany. First, energy and other natural resources have become more abundant, not less so since the Club of Rome published ‘The Limits to Growth’ in 1972. Second, more food is now produced per head of the world’s population than at any time in history. Fewer people are starving.”
--
The story of wheat
Ears of plenty
Dec 20th 2005
From The Economist print edition
The story of man’s staple food
[Image] (Still Pictures)
IN 10,000 years, the earth’s population has doubled ten times, from less than 10m to more than six billion now and ten billion soon. Most of the calories that made that increase possible have come from three plants: maize, rice and wheat. The oldest, most widespread and until recently biggest of the three crops is wheat (see chart). To a first approximation wheat is the staple food of mankind, and its history is that of humanity.
Yet today, wheat is losing its crown. The tonnage (though not the acreage) of maize harvested in the world began consistently to exceed that of wheat for the first time in 1998; rice followed suit in 1999. Genetic modification, which has transformed maize, rice and soyabeans, has largely passed wheat by--to such an extent that it is in danger of becoming an “orphan crop”. The Atkins diet and a fashion for gluten allergies have made wheat seem less wholesome. And with population growth rates falling sharply while yields continue to rise, even the acreage devoted to wheat may now begin to decline for the first time since the stone age.
It is time to pay tribute to this strange little grass that has done so much for the human race. Strange is the word, for wheat is a genetic monster. A typical wheat variety is hexaploid--it has six copies of each gene, where most creatures have two. Its 21 chromosomes contain a massive 16 billion base pairs of DNA, 40 times as much as rice, six times as much as maize and five times as much as people. It is derived from three wild ancestral species in two separate mergers. The first took place in the Levant 10,000 years ago, the second near the Caspian Sea 2,000 years later. The result was a plant with extra-large seeds incapable of dispersal in the wild, dependent entirely on people to sow them.
The story actually starts much earlier, around 12,000 years ago. At the time, after several warm millennia, a melting ice sheet in North America collapsed and a gigantic lake drained into the North Atlantic through the St Lawrence seaway. The torrent of cool, fresh water altered the climate so drastically that the ice age, which had been in full retreat, resumed for a further 11 centuries. The Scandinavian ice sheet surged south. Western Asia became not only cooler, but much drier. The Black Sea all but dried out.
People in what is now Syria had been subsisting happily on a diet of acorns, gazelles and grass seeds. The centuries of drought drove them to depend increasingly on wild grass seeds. Abruptly, soon after 11,000 years ago, they began to cultivate rye and chickpeas, then einkorn and emmer, two ancestors of wheat, and later barley. Soon cultivated grain was their staple food. It happened first in the Karacadag Mountains in south-eastern Turkey--it is only here that wild einkorn grass contains the identical genetic fingerprint of modern domesticated wheat.
Who first replanted the seeds and why? For a start, he was probably a she: women have primary responsibilities for plant gathering in hunter-gatherer societies. The time was certainly ripe for agriculture: the ability to make tools and control fire (cooking makes many plants more digestible) was already well established. But was it an act of inspiration or desperation? Did it perhaps happen by accident, as discarded grains germinated around human settlements?
As it happens, The Economist recently ran an article addressing some of these issues. The article also provides context and perspective that should be of interest to those participating in this discussion. For convenience, the full text is reproduced below; it is also accessible online (may require paid subscription).
----
Computing
From factoids to facts
Aug 26th 2004 | REDMOND, WASHINGTON
From The Economist print edition
At last, a way of getting answers from the web
WHAT is the next stage in the evolution of internet search engines? AltaVista demonstrated that indexing the entire world wide web was feasible. Google's success stems from its uncanny ability to sort useful web pages from dross. But the real prize will surely go to whoever can use the web to deliver a straight answer to a straight question. And Eric Brill, a researcher at Microsoft, intends that his firm will be the first to do that.
Dr Brill's initial crack at the problem is a system called "Ask MSR" (MSR stands for Microsoft Research). This program uses information on web pages to respond to questions to which the answer is a single word or phrase--such as "When was Marilyn Monroe born?" Ask MSR starts by manipulating the question in various ways: by identifying the verb, for example, and then changing its tense or moving it into different positions in the sentence ("Marilyn was Monroe born", "Marilyn Monroe was born" and so on). The resulting phrases are then fed into a search engine, and documents containing matching strings of words are retrieved. It sounds a promiscuous strategy, but gibberish phrases produce few matches, so, as Dr Brill puts it, "being wrong is very cheap."
Once accumulated, the pile of documents is scanned for possible answers, and these are ranked by frequency. In practice, the correct answer appears in one of the first three places around 75% of the time. That might not sound very good, but human intelligence provides a second filter, since wrong answers are often obvious. If you ask how many times Bjorn Borg won Wimbledon, for example, "1980" is not a plausible answer, but "5" is. If in doubt, clicking on an answer produces a list of links to pages which provide support for that answer.
Ask MSR is still a prototype, although Microsoft is trying to improve it and it may be launched commercially under the name AnswerBot. Dr Brill, meanwhile, has moved to a more difficult task. One of his most recent papers, written jointly with Radu Soricut of the University of Southern California, is entitled "Beyond the Factoid". It describes his efforts to build a system capable of providing 50-word answers to questions such as "What are the rules for qualifying for the Academy Awards?" This is harder than finding a single-word answer, but Dr Brill thinks it should be possible using something called a "noisy channel" model.
Such models are already employed in spell-checking and speech-recognition systems. They work by modelling the transformation between what a user means (in spell-checking, the word he intended to type) and what he does (the garbled word actually typed). Just as a telephone line distorts the voice of the person at the other end of the line, this process can be thought of as being a noisy channel that transforms the user's intention into something rather different.
By analysing many pairs of correct and mis-spelled words using statistical techniques, it is possible to predict how such transformations work in general cases. A system can then be designed to work the process backwards. Given a mis-spelled word, it can guess what that word is most likely to be a mis-spelling of.
Dr Brill's question-answering system does something similar. Many question-and-answer pairs exist on the web, in the form of "frequently asked questions" (FAQ) pages. Dr Brill trained his system using a million such pairs, to create a model that, given