SCIENCE:ALL GEEK TO ME

Twentysomething, computer-mad and prodigiously talented, Microsoft's youthful programmers may prove to be a danger to their company. Charles Arthur reports on the kids who throw tantrums in Windows

NOBODY who reads a newspaper can have missed the August launch of Windows 95, the most overhyped software package in history. Nobody who watches television can be unfamiliar with the image of Microsoft's corporate headquarters in Redmond, near Seattle, a sprawl of low buildings, neat paths and manicured lawns that richly deserves its nickname, "the campus".

The analogy with student life doesn't end here. Living near the campus, in shared houses that resemble student halls of residence, are thousands of young men and women who write, design, test and market the programs that run 80 per cent of the world's computers. Mostly in their twenties, they are a new breed of ambitious college graduate - too geek-ish to be grunge or movie stars, but with the talent to be computer millionaires. These youngsters are Microsoft's main asset - and its most serious commercial threat.

So much attention has focused on Microsoft's products that its people, an equally fascinating phenomenon, have been largely overlooked until now. In his latest novel, Microserfs, Douglas Coupland (author of Generation X) satirises the new breed. In I Sing the Body Electronic, another new book, journalist Fred Moody of the Seattle Weekly tells how his unprecedented access to the "Microsofters" - he spent a whole year in their company - exposed them as the corporation's Achilles' heel.

The Microsofters relative youth and inexperience has created an atmosphere where coherent planning is almost impossible (which helps explain why Windows 95 was eventually launched nine months late). A combination of immaturity, driving ambition, brilliant minds and unrealistic deadlines has resulted in a pressure-cooker of talent, held in by claustrophobic offices whose occupants prefer to "talk" via electronic mail - even when they are only feet apart.

In researching his book, Moody was given unlimited access to a team working on a new multimedia CD-ROM. This helped him put into context a side to Microsoft that the company is less than keen to thrust into the limelight: its failures. "What happened with Windows 95," he told me, "was almost certainly the same as happened with the CD-ROM. They set themselves deadlines that were impossible to meet. I wasn't at all surprised when the release date kept slipping, because they were trying to incorporate too much."

Yet staff are devastated when deadlines are missed, and there are heavy penalties for doing so. "At the same time," says Moody, "they know Microsoft has never shipped a product on time - so there's also the understanding that the deadlines aren't reasonable." The process is like George Orwell's doublethink: everyone believes both ideas simultaneously. The creative process at Microsoft is fuelled by a mixture of fear (of missing deadlines) and total confidence (that the deadline will be extended).

Moody's book is a journey of exploration, an odyssey of shifting beliefs. He initially thinks that Microsoft, because of its success, must be the kind of company to have structures and systems that outclass any other commercial organisation, but concludes that it is flying by the seat of its pants. Yet it still gets results, as its domination of the computer industry has shown.

Microsoft operates on a very flat management structure: hundreds of small teams work separately, designing and programming products. The exclusive feel is heightened by the company's physical separation from anywhere else - "there isn't anywhere to go for lunch except on the campus," says Moody - and the gruelling recruitment process, in which thousands of applicants are whittled down in rigorous interviews. They are required to spot errors in programs and write solutions almost instantly. For Microsoft, being bright isn't enough; you have to be exceptional.

The combination of extraordinary talent and youth may be the company's worst problem. The average age of its 16,000 employees is 31, but its programmers are almost all under 25. The company also uses freelances, many of whom are still undergraduates. They earn good money, up to $50,000 (pounds 32,000) a year doing something for which they have exceptional aptitude. As one employee puts it, Microsoft "offers us the life we had in school, except we get paid to do our work. And, unlike school, we get to do the really cool stuff."

But aptitude does not imply maturity. Designers, creators of the graphic images on screen, sulk like children when their work is rejected because it does not fit the product's "feel". In his book, Moody tells of one who sat in silence in meetings, often curled up on the floor. Programmers, who turn designs into lines of machine code, have tantrums when changes forced by shifting deadlines generate more work for them.

Sometimes freelances, annoyed at not being offered full-time jobs, deliberately insert bugs into programs. Moody relates a last-minute dash to remove one such bug, in an encyclopaedia CD-ROM, which displayed the message: "Slayer [a heavy metal band] sucks like a vacuum". The offending programmer was quickly traced and told he would never work for Microsoft again. In a gesture that demonstrates Microsoft's tolerance of student-style japes, he was still given a good job reference.

Intentional bug-writing is the likely explanation for another Microsoft embarrassment this summer: in July the company published 5,500 copies of a CD-ROM containing a novel computer virus that could infect word-processing files. Microsoft did not publicly announce the fact until companies writing anti-virus software alerted the media - by which time the virus had spread worldwide. Microsoft still refers to it as a "prank macro" rather than a virus, and says officially: "Nobody from Microsoft wrote the prank macro." But it could very easily have been a contractor who was either trying to impress, or get revenge.

Arrogance comes easily to the young, especially in successful companies. The CD-ROM team charged headlong into the project - a "multimedia encyclopaedia" dubbed Sendak, aimed at boys and girls aged between six and 11 - without consulting a developmental psychologist, a teacher or even a children's encyclopaedia to find out what sort of things their target audience would find interesting, how they approached learning, or what they would find boring. Instead, they began designing a product that could not, in the end, be programmed as conceived. Moody, who is 46 and has three children, was surprised by how uninterested everyone was in his knowledge or viewpoint, even though his children were their potential audience.

Predictably, the programmers' approach yielded few results and a lot of frustration. "I watched and listened in growing amazement," notes Moody, "as work on Sendak kept moving on a variety of fronts, in a direction deemed by all but me to be forward. I was struck at every turn by how young these people were, and how inexperienced. Their greenness seemed to render them terrified of making a mistake - with the result that they were unable to commit to a decision and then move on."

Others displayed arrogance that bordered on breathtaking philistinism. At one stage the head of the multimedia project - a medium which offers intriguing possibilities for combining video, sound and text - offered his opinion on films. "I don't see what people say is so great about Citizen Kane," he says. "That is a long movie." Then he recalls reading Homer's Iliad (in English) at school, and loving it: "I couldn't believe how graphic that book was when I read it! They should make a movie out of it." This man, remember, plans to make the software to put your children into raptures.

But those are the happy, relaxed moments. More common are the tensions as the young workers strive to mature. Microsoft's ethos promotes a culture of conflict between designers and developers, refereed occasionally by their "program managers". Meanwhile, they design the software of the future, working from indefinite aims. Unsurprisingly, all three groups fall into recurrent bouts of despair. Should the "guide" who shows the user around the electronic encyclopaedia be an alien? A child? An adult? An animal? Should there be a guide at all? Should the user have some "mission" to fulfil? If so, how difficult should it be? Moody's book notes how this question goes unresolved throughout almost the entire project.

I asked Moody if he thought other businesses could use Microsoft as a model. He thought not: "The one thing that makes Microsoft able to withstand these sorts of internal pressures is the rise in its stock price. The stock options [allowing employees to buy shares cheaply for resale when the market price is far higher] are really valuable. People can be worth hundreds of thousands of dollars overnight. [They] are always checking to see how the stock is doing When they get really depressed they start talking about their stock to cheer themselves up. No other corporation has seen such a meteoric rise in its stock."

To find the ultimate source of these people's drive, and for a large part of the aura around Microsoft, one must look to the personality of Bill Gates, its chairman and co-founder (along with the less conspicuous Paul Allen). His youthful, nerdish appearance is legendary, heightened in the past by his preference for conspicuously large, round spectacles; on paper, he is the richest man in the US (this month, Forbes magazine estimated his wealth at $14.8 billion, nearly pounds 10 billion). But these well-publicised facts pale beside others that become clear only when you meet him or talk to people who have worked with him.

Employees express admiration and fear in roughly equal measure. Presenting one's work to Gates is a terrifying prospect, known internally as "having a Bill meeting". He is famed for his ability to seek out the weak point in an argument or a piece of work. Teams spend the week before such a meeting rehearsing, with someone playing agent provocateur in true Gates fashion.

The real Gates can explode in fury over work he dislikes. Even when talking in normal situations, he shows his underlying impatience. I have met him on a number of occasions in the past 10 years, and his demeanour always reminds me of John McEnroe after a bad game. In his post-match press conferences, McEnroe could be by turns edgy, vocal and highly argumentative, with a tendency to digress along a path entirely his own. If someone interrupted him, he almost always carried on talking, ignoring them until they shut up. Psychologists call this "overtalking": it is a method of establishing dominance over others. Bill Gates uses the same technique; there are very few people he defers to.

Moody recalls a meeting where Gates was talking about publishing deals. The aim was to work with book publishers, putting versions of well-known books on computer. Gates wanted the project to include not just PCs, but Nintendo machines, video arcade systems, and even products that had not yet been invented. "The default position is that it should be broad platforms!" Gates almost yelled. "But if we can't get that, screw 'em! I mean, it's just not fair!" Not fair? This, the familiar idiom elsewhere in the organisation, is the language of the playground.

When you meet Bill Gates, you get some idea of the force, even anger, that drives Microsoft onwards. But Moody's year spent among staff has convinced him that the juggernaut cannot continue forever. There are two reasons for this, and the first is Gates himself. Later this month, he turns 40; his company is already 20 years old. He owns a third of the company's stock, and the organisation revolves around him.

"It's like a charismatic religious organisation," Moody says. "If you took Gates out, you couldn't fill the void. The expectations that he sets and the belief that his vision is right fuel the company." But Gates has said he will retire when he is 50, and give away 95 per cent of his fortune. "I don't think people will ever have the same devotion after that," says Moody.

There is another, bigger hurdle looming. "Microsoft is growing at a phenomenal rate," Moody observes, "faster than the computer industry overall. But eventually, everyone who wants a Microsoft product for their PC will have one. Sales will drop, stock value will fall, and that incentive for new recruits will vanish. The stock markets here are brutal when your growth slows."

Could it be that the sound and fury, the glitz and glamour of the Windows 95 launch was the beginning of Microsoft's swansong? Many have written the company off in the past, yet still it survives - and thrives. Yet, in similar fashion, IBM once thought its position unassailable.

Perhaps, when the Microsoft story's all over, someone will make a movie out of it.

! 'I Sing the Body Electronic', by Fred Moody, is published by Hodder & Stoughton at pounds 17.99

MICROSOFT: THE RELENTLESS RISE

January 1975: An article on the Altair 8800 minicomputer in the magazine Popular Elec-tronics inspired Paul Allen and Bill Gates to develop a BASIC language program for the Altair. "We realised that our life would focus," Bill Gates said.

February 1975: Gates and Allen complete BASIC and sell it to Microsoft's first customer, the manufacturers of the Altair. This is the first computer language program for a PC.

4 April 1975: Microsoft officially set up as a partnership.

18 June 1979: Microsoft introduces BASIC for the Intel 8086 microprocessor. This is the first release of a high-level language for a 16-bit machine and marks the beginning of widespread use of the processor.

12 August 1981: IBM introduces its PC, which uses Microsoft's 16-bit operating system, MS-DOS version 1.0, plus other Microsoft products.

1 April 1982: Microsoft establishes a subsidiary in England, the beginning of its foreign sales effort.

10 November 1983: Micro-soft announces Windows, an extension to MS-DOS that provides a universal operating environment for developing bit-mapped application programs.

20 November 1985: Microsoft ships (publishes) the retail version of Windows.

13 March 1986: Microsoft stock goes public at $21 per share, rising by the end of the first day to $28 per share. Initial offering raises $61 million.

1 January 1987: Over 20 million MS-DOS systems installed.

April 1987

January 1988: Microsoft becomes the biggest-selling PC software company.

1 January 1989: MS-DOS installations near 40 million.

September 1989: Company headquarters at One Microsoft Way, Redmond, near Seattle. 13 March 1990: Microsoft announces a 100 per cent stock dividend. They will issue one additional share for each share held on 26/3/90.

22 May 1990: Microsoft an-nounces immediate worldwide availability of Windows 3.0.

17 September 1990: Micro-soft launches the Windows marketing programme, the largest single marketing campaign in the company's 15-year history.

22 May 1991: A year after it was announced, Windows 3.0 is actually available in 12 languages in 24 countries.

March 1992: Microsoft kicks-off its first television advertising campaign. The TV ads are designed to build on the success of Windows and Windows- based applications.

6 April 1992: Microsoft ships Windows 3.1 with more than 1,000 enhancements. The new version created unprecedented user demand with over one million advance orders placed throughout the world.

20 May 1992: Microsoft announces that three million copies of Windows 3.1 were shipped in the first six weeks after the product's release.

23 January 1993: Microsoft becomes the world's biggest computer-industry company based on the total value of its stock, a measure known as market value.

31 March1993: 12 years after MS-DOS was first launched, MS-DOS 6 ships.

14 April 1993: The licensed number of Windows users now totals more than 25 million, making it the most popular graphical operating system in the whole world.

19 January 1994: Shipments of Windows reach more than 40 million units.

4 April 1994: Microsoft Windows 3.11 ships.

December 1994: Windows 95 misses its internal release date.

August 1995: Windows 95 ships amid extraordinary hype.