In August 1999 I ran into Miguel de Icaza and Nat Friedman in a gift shop not far from the LinuxWorld Conference & Expo (LWCE) in San Jose. Among other things, Friedman and de Icaza were kicking around the idea of forming a company. Later the same year they did exactly that in founding Helix Code, a company whose sole mission has been to bring the GNOME desktop to new levels of completeness, polish, and ease of use. In August 2000 I ran into the dynamic duo again at the LWCE in San Jose. This time it was at the GNOME party.

Miguel de Icaza is normally the press-magnet. His natural exuberance, demonstrative speech and mannerisms, and his brilliant work have made him one of the stars of the Linux and free software worlds. But de Icaza told me at the party that night that the idea to form a company to help make GNOME -- and by extension Linux -- easier to use was Friedman's, not his. He suggested that I do an interview with Friedman.

Well, I'm slow but I do eventually follow up on things. Here we are on the eve of another LWCE, this one in New York City, and I've just spent an enjoyable half-hour chatting on the phone with Nat Friedman, CEO of Ximian, formerly Helix Code. In fact, it was the name change that finally got me off the dime and on the phone with Friedman.

One of the more interesting things I learned during the call was that although de Icaza may get all the press, Friedman is quite impressive in his own right. He is an MIT graduate (1999) and the CEO of one the free software world's best-known commercial ventures. Not only that, he's just successfully raised US$15 million of venture capital, and did it in an anything but "easy touch" market for start-ups seeking financial backing. Not bad for a 23-year-old just a year or two out of school.

I also learned that Friedman and de Icaza met at Microsoft Corp. Friedman told me that they met while de Icaza was interviewing for the Internet Explorer team in Redmond. Friedman was -- hold on to your hats -- a Microsoft employee at the time, working on the IIS (Microsoft's Web server) project. It boggles the mind. They also met online in a network set up by and for Linux developers. Soon de Icaza was mentoring Friedman in the intricacies of open source software and the Linux kernel. Not that Friedman especially needed help with computers. He got his first computer, an Apple, when he was six. Not long afterward he began "playing around" with the Logo programming language. A year or two later a friend taught him BASIC and they began writing video games. At age 11 he learned Pascal, and a couple of years later added C to his repertoire. When he was 14 he and another friend formed a software company. Friedman said they wrote a program something like the Norton Commander utility, as well as a bunch of video games. He added that while they didn't make very much money, they did have a lot of fun doing it.

Listening to Friedman, you quickly become aware that you are talking to a very bright person. That led me to ask him this question: "Between you and Miguel, who is the smart one?" Friedman replied: "We're smart in different ways. Miguel's intelligence is the kind that is never satisfied with staying in one place. He likes to look at questions from a thousand different angles -- all at once sometimes -- so fast the words kind of like pile up in his throat and they come very quickly and all confused.... He's just never satisfied with the way things are understood now. He has this sort of ruthlessly honest intelligence, so he is really bright that way. I like to think of Miguel as our idea machine. He comes up with an idea every 20 minutes or so, and it's a new and incredible idea."

While Friedman appreciates that intelligence and finds it both inspiring and charismatic, he notes that you sometimes need a buffer between it and the tasks the company is trying to accomplish. He noted that, "You can't change tack every 20 minutes." His own intelligence, Friedman said, is different. And he left it at that.

Since I interviewed Friedman a week before the LWCE was scheduled to kick off in New York City, I wondered what surprise announcements might be coming from GNOME or Ximian. Friedman told me that although we haven't heard much from Ximian recently -- other than the name-change announcement -- we shouldn't be misled by the lull. Ximian may have a big announcement at LWCE, he explained, if the code is ready in time. But he wouldn't say whether he was referring to the appearance of StarOffice inside GNOME as Bonobo components or something else entirely. Is Ximian still working on the StarOffice components? Yes, he could say that much. He added, "That's a big deal to us, because StarOffice allows us to provide a much more complete solution to people."

About the name change: A lot of people, myself included, liked Helix Code just fine. So why the change? Friedman said: "Helix Code we couldn't trademark. Very simply, we thought we could, but we couldn't. It was a temporary name which we ended up being stuck with after a Wall Street Journal article was written about us, and then it was just too late to change it."

OK. But why Ximian? Friedman explained that basically it is the word "simian," with the leading s replaced with an X. If nothing else, it is unique enough to trademark. Like Bonobo, the new name's root (simian) conjures up the evolution of humans from the apes, except of course in this case it is about the evolution of software.

The pronunciation of words beginning with x always gives me pause, so I had to ask, "Is Ximian pronounced 'zim-ee-un?'" Friedman said yes. Then he added, "Actually, Miguel and I are thinking about putting up a funny sound file on the Website. He would say 'This ees Miguel de Icaza, and I pronounce Ximian as "zeem-ee-un."' And I would say 'No, no! Miguel! It's not "zeem-ee-un," it's "zim-ee-un!"' Miguel would say, 'That's what I said: "zeem-ee-un!"' To allow Friedman to gauge the progress of Ximian's efforts to make Linux more accessible, I asked him when Linux would challenge Windows on the desktop. He replied: "A good question.... In the operating system desktop space, there is always an incumbent and a challenger. The incumbent is Windows, and we are the challenger, essentially. Apple is no longer a challenger; they don't have momentum or anything like that.... You cannot beat Microsoft, you cannot gain market share over them, by chasing taillights."

So how do you do it? As Friedman explained his view to me, I felt as if I were caught in an eerie remake of "The Graduate." But instead of an older man giving Dustin Hoffman advice about going into plastics, it was the recent graduate Friedman explaining to the old-timer about appliances.

He told me: "We are going to do an end run around them, sort of. And I think we are going to come at them through a lot of things, like Internet access devices, the notion of appliances. If you look at the industry generally, you can say with some confidence, I think, that in five years we're going to be looking at a world with many more appliancelike computing devices, low-cost sort of specific-purpose communications appliances. And that's what we're gearing GNOME up for in many ways."

Finally, I asked Friedman if -- as the Ximian CEO -- he was worried about the number of open source companies floundering on the bottom line. He told me that what concerned him more than anything was that so many of the companies were "blind to begin with." Ximian is not yet profitable, though it is producing revenue. It won't be profitable this year, either, but it may be in 2002. He added that the downturn in the sector "does make it harder to raise money when the capital money gets a little soft. But we just raised US$15 million and we are in a very strong position cashwise."

Whether you like the recent name change or not, you have to like the chances of a company led by not one, but two brilliant and motivated young men. Personally, I like the chances of Linux displacing Windows from the desktop a lot more with those two around.