Red Hat founder maps Linux game plan

Marc Ewing seems happy in his job as Red Hat founder and chief technology officer, a position that has made him a near billionaire in the aftermath of the company's torrid IPO. But he's not complacent for one big reason: Microsoft.

"They are our competition," Ewing said in an interview today, the first day after the company's quiet period following its initial public offering.

Red Hat and the rest of the Linux community undeniably have made some inroads against Microsoft, the world's largest software company.

Years of work on Red Hat's Linux product have paid off financially. Red Hat's IPO last month netted $84 million for the company. In the weeks since, Ewing's more than 9 million shares have given him a net worth, on paper, of more than $900 million. Red Hat's shares edged up to $100 apiece today.

But while some are amazed at Red Hat's market capitalization of more than $6 billion, Ewing points out that it's still about one percent of Microsoft's.

In an interview with CNET News.com, Ewing discussed facing Microsoft, the company's public offering ups and downs, and Red Hat's relationship with Linus Torvalds.

CNET News.com: How big a motivator is Microsoft for Red Hat?
Ewing: It's huge. They're the guys we worry about. Any time you're sort of slacking off or saying you're thinking of taking a day off our president says, "You know, I'll bet Bill Gates is working today." He's probably right--that's the scary thing. They've got amazing marketing. They've got people all over the place going into these corporate accounts, and those are the places we need to go in order to grow our business. They're our prime motivator in keeping everyone here hungry.

We really don't worry too much about a lot of these other Linux players, and we wish a lot of them well. Microsoft is the competitor, not SuSE and Caldera Systems and TurboLinux.

Do you see the IPO as a vindication of Linux or open-source programming?
I certainly think it clearly represents the interest both of the computing community and the investment community in this new model and in Linux in particular that we're able to have an IPO that went very well.

It did legitimize...Red Hat. All our books are now open. And the public sentiment is evidenced every day in the stock price. In that sense people can get a more comfortable feeling. Private companies can always be a little bit mysterious. That mystery, coupled with this totally bizarre development and software distribution model, made a lot of people pretty nervous. Now I think what we've done has helped alleviate that concern.

What was the mood during the IPO?
Anxious. It was a little bit nerve-racking given the state of the market, but we felt we had a unique and compelling story and that we could go out anyway, regardless of what the market was doing. But it was certainly a little bit of a risk, and we were all pretty anxious.

When the first trade happened at 46, with the offering at 14, we were both elated and there was a sigh of relief at the same time. We didn't do a 1-800-Flowers kind of thing [1-800-Flowers fell below its offer price in its IPO]. But at this point, everybody is triply motivated to execute on our plan than they were before.

What did you learn during the affinity program problem, when Red Hat was criticized for its imperfect effort to include open-source programmers in the IPO? (In a new twist on public offerings, Red Hat offered the chance to participate in its IPO to about 5,000 people who have supported the open-source programming. Because of last-minute changes, however, only about 1,150 of those qualified to participate.)

You mentioned that 88 percent of the people who wanted to participate did, and I understand regulators stepped in at the last minute.

In the last 24 hours [before the IPO], I think it was the National Association of Securities Dealers required E*Trade to reconfirm with all the people who had expressed their interest that they were still interested, given that our price range went up from 10-to-12 to 12-to-14. I'm told that in the past, increases of 20 percent have been explicitly allowed by these regulatory bodies, so you don't need to go back and reconfirm people's interest on that level of a price increase. But I guess this one just got extra scrutiny.

So E*Trade really scrambled and put a lot of people on it to try to reconfirm all of them. They all did get reconfirmed. All those people who had expressed interest at 10-to-12 got confirmed at 12-to-14 and got their shares. If you look at the whole process, the end result was what we wanted to have happen. We look at it as a success as something that nobody had really ever done before. We knew at the outset there were probably issues that were going to come up. In general, we're pleased. Other IPOs at E*Trade hadn't [gotten] that high [88 percent participation].

How do you see the interaction between the commercial sphere and the open-source community changing now that you're a publicly traded company?
In a sense, being a public company is the perfect thing for an open-source company. All these guys get to own a piece of us, and they get to see our inner workings and our results every quarter. But it doesn't change a lot for us. It gives a certain sense of legitimacy to a lot of our partners and gives us a ready supply of cash and currency to do a lot of the execution we need to do over the next couple of years. It's sort of business as usual, but we sort of have a lot more ammunition behind us.

What do you think of the theory that Linux has the potential to be as influential as the Internet was for stock market issues?
I think there's a lot of correlation between the Internet and Linux. Open source and open, common standards are what helped drive this highly decentralized development model of both Linux and the Internet. That let it be decentralized but at the same time maintain a coherency. That's sort of the magic mix. The Internet seems to have done that, and Linux seems to have done that.

The position in the past has been that you pay the core programmers and you let them go their own way. Is that still the case?
Yes and no. If [lead Linux developers] Alan [Cox] or Dave Miller [say] this is the sort of thing that needs to be done in the kernel in order to get great performance, very few people at Red Hat can actually question that opinion. So in that sense they get to direct themselves.

At the same time...these guys realize that there's a lot of stuff left to be done in the Linux kernel. There are a lot of things that aren't quite being addressed right now that corporate users, real high-volume transaction users or high availability users, need in the Linux kernel and Linux system utilities.

Our customers are coming back to us and saying these are things they need. That feedback is very interesting to these developers. They really get motivated by it. It really gives them real direction. They don't feel like they're wasting their time.

Have you tried to hire Linus Torvalds onto your staff?
Not recently, no. A long time ago, I'm guessing about three years ago... [Red Hat CEO] Bob [Young] and I...talked to him about it. One, for us, so we could say we had Linus on our staff, and two, so he could devote all his time to Linux. But Linus has always made the right decisions.

His leadership of Linux development can't be questioned. He decided he didn't want to align himself with any company, no matter how much independence he would have. At the same time, he's a very smart guy, and he wanted to exercise his mind outside of doing pure Linux development. At Transmeta he does things other than just Linux and gets exposed to new challenges.