I believed the hype. By the time I decided to install Linux on my computer, there was no practical reason to do so whatever. I just wanted to do it in order to be cool. Quite why I thought I might become cool by associating myself with the Linux crowd (the stereotypical fat, pony tailed Quake-playing sysadmins of this world), I am not sure. It certainly wasn�t out of anything as misguided as the belief that there was an element of "radical chic" to the decision - sure, others might have been forgiven for seeing Linux as the idealistic software revolutionaries against Micro$oft hegemony, but I had spent enough time online talking to them to know what a spiteful bunch of Libertarian pricks they actually were. But there you have it. I wanted to be tech-savvy and cool, and nothing was going to stop me.

Of course, had I come to my decision a couple of years earlier, I could have claimed I was doing it for purely rational reasons. In the early days of Linux advocacy, everything revolved round the buggy, slow nature of Microsoft products. Open Source advocates like Eric Raymond and Rob Malda would loudly boast about the "uptime" of their systems (the amount of time they could leave their computer on while not using it without having to turn it off and on to stop Windows from freezing up). They would contrast this unfavourably with Microsoft, which crashed all the time. These criticisms were absolutely accurate, of course; Windows products did crash all the time, mainly because the different products in the Office suite were not well enough integrated into one another. I remember a particularly painful couple of episodes attempting to embed an Excel chart into a Word document in 1997, and swearing to myself at the time that I would change over to Linux and never suffer the problem again. Of course, in 1997 Linux didn't actually have any equivalent programs to Word and Excel, and even today the nearest it can come is Star Office. Which, ironically, has a horrid reputation for crashing all the time and having poor object linking and embedding. But the fact remains that in 1997, there was a genuine technical case to be made, over and above the cool factor, for the superiority of Linux as a piece of software over Windows.

However, by 2001 and the release of Windows ME and Office 2000, these problems had more or less been solved. Even the elitest of the elite Linux gurus had to grudgingly admit that Microsoft had solved its teething problems and produced a reliable, high-quality product. It hurt them to admit it, of course, particularly those who had maintained an ideology of software development under which Microsoft's developing a superior technology was almost a logical impossibility. But by 2001, the Linux battle had moved into the courtroom, where the Netscape browser, assisted by the Department of Justice, appeared to be kicking Bill's ass. The Linux users were pinning their hopes on some sort of break-up of Microsoft which would force its software into the public domain, allowing them to merge the Windows 2K codebase into Linux.

Separately from this, the dot-bomb status of Indrema had more or less put the tombstone on Linux as a serious gaming platform - as I knew from Slashdot and the mailing lists, DirectX support was about ninetieth priority for a team of kernel developers desperately trying to keep pace with innovations in the hardware industry. So my ambition of leaving the rat race and becoming a professional game designer by starting off as a big fish in the small pond of Linux gaming looked pretty much a forlorn hope. Unfortunately, it seemed that being "open" to new suggestions wasn't as much of a help as Eric Raymond had predicted it would be in his "Cathedral and Bazaar" essay -- the number of elite computer programmers or "31337 hackers" who cared to devote their spare time to maintaining and updating Linux just didn't seem to grow as quickly as the requirements of the Linux project. There really was no particular argument on grounds of technical merit for risking damage to my slick, professional ME/2K installation by installing the rushed compromise that was the Linux 2.4.1 kernel.

So why, for heaven's sake, did I do it? Well, part of it was pure and simple contrarianism. There was something in the ornery, cussed nature of the Linux kids that appealed to me; the determination to drag a 1970s operating system kicking and screaming into the 1980s, and to take on the biggest bully in the schoolyard (Microsoft). It was a doomed quest, to be sure, but I've always been a fan of Don Quixote and the Linux "community" appeared to be his modern equivalents.

But personal admiration couldn't have been that much of a factor. Although the original inventor, Linus Torvalds himself, seemed like a nice guy with an understated sense of humour, he seemed to have more or less disappeared from the Linux scene by 2001, leaving Linux dominated by opinionated blowhards like Raymond, Bruce Perens and Malda. These three Libertarian loons in particular destroyed whatever nascent goodwill I might have had for Linux as a social movement. By late 2000, it was more or less impossible to open a computer magazine without seeing one of this unholy trinity grinning out at you in pale malnourished splendour, attached to some ass-kissing profile, or to an ill-thought out manifesto, full of ludicrous grandiose statements about world domination, uninformed slurs about Windows, and head-in-the-sand optimism about Linux capabilities as yet unprogrammed.

We were all "lusers" and "sheeple" for going with a trusted product rather than the latest self-advertised "unstable" Linux release, and Eric Raymond claimed to have a secret development model which was twice as efficient as anything which had gone before, because, in his half-thought out position papers, Linux was based on the free market, whereas Microsoft (at that time the largest company by market capitalisation in the history of capitalism) was more like the old Soviet Union. Appalling twaddle, obviously, but just about then, the computer mags would print anything if it would help to hype the free Linux CD attached to the cover. In their defence, this was in the days pre dot bomb, so the arrogance of Malda and Raymond was underpinned by the extraordinary (and, it turns out, possibly manipulated) market capitalisations given to the companies on whose boards they sat (Perens missed the IPO window, and thus remains the poor man's Eric Raymond).

But if it wasn't the personalities involved, why was it? What possessed an otherwise rational financial analyst to install, on what was then his only working laptop, an operating system which would not support either Word or Excel, the twin six-shooters of the financial cowboy's trade? What made me do it?

In the final analysis, it was, purely and simply, the cool factor. The buzz about Linux had spread from the Net to the specialist press, via a detour into the wildest hinterlands of the NASDAQ. In other words, it had started off among a bunch of nuts and diehards, floated a couple of "story stocks" and was now being touted in the industry as the Next Big Thing This is the exact same path trod by radio in the 1930s, bowling alleys in the 1960s, satellite communications in the 1970s and Charles and Maurice Saatchi in the 1980s. The next stop along this line was general fashionability in the financial markets, last stop en route to General Credibility, Backlash and Decline Into Obscurity. I'd seen it happen a couple of years earlier with micro-scooters and Pokemon, for example. This time, I was going to be in the vanguard. I had a reputation as something of a technical guru in my firm, based on a couple of Excel macros I'd written a few years earlier. I also wanted to project the kind of rebellious, cool image that was associated with Apple Macs a few years ago, and Linux seemed perfect for that (Macs were still the unconventional choice, but their association with the design industry made them seem more gay than rebellious per se). Despite the fact that all the technical arguments militated against, it, it was clear to me that now was the Time For Linux. This time, it was me who was going to be "elite"

Next week : Deep background -- the history of Linux, as I saw it