Red Hat CEO: Business of Linux validated

CAMBRIDGE, Mass. -- Red Hat Inc. sits atop the heap of Linux distributors. As the 11-year-old publicly traded Raleigh, N.C.-based company steams toward adolescence, it has nearly $1 billion in cash in the back with which to beat back Novell Inc.'s challenge in the Linux market while continuing to drive innovation and maintain its enterprise focus. "We're at a very early stage of an enormous paradigm shift both technically and I hope socially as well," said Red Hat's dynamic chairman, chief executive officer and president Matthew J. Szulik on Monday at MIT during a stop on Red Hat's current world tour. An entourage of Red Hat engineers and executives will have visited seven cities on four continents in 14 days by the time the tour concludes today in Toronto. Szulik joined the tour yesterday as it made its first stop in the United States and in this interview talked about the competition, Red Hat's strategy and SCO.

What kind of feedback have you heard from customers regarding the protection plan Red Hat offers? Hasn't been an issue. We've generated 4,000 new customers in the last 90 days, 87,000 new subscribers. You'll find a customer here or there will tighten the noose on certain terms they would like, and it's up to us to decide whether we want to do that or not. If you look at the indemnification offerings and look at the fine print and the...

claims many vendors have made, I'm not sure that has reached the levels of customer satisfaction they want.

The good thing about this has been ... the great benefit the customer has gained in the improvements in the licensing relationship with the vendor. That's great.

The greater issue for us is: Where is the innovation going to come from? Operating systems aren't largely a new technology, so you go all the way back to the days of AT&T Unix System V; I think what Kernighan and Ritchie [Brian W. Kernighan, Dennis M. Ritchie] had in mind was to continue to build on the successful ideas of others.

So when you start to see this craziness that's going on right now and the lack of a strong copyright and patent process, you're going to start to see the larger companies dominate the patent application process. Microsoft, in 1991, had 10 patents. Now they've got more than 3,500. It's no accident their revenues have escalated through that process.

We've spent an arm and a leg [on attorney fees] on this SCO case, so that means we're not hiring engineers and we're not adding more value for shareholders and customers and, as of yet, nothing has been resolved. It's foolish. The Chinese, the Indians, the Russians are laughing their tails off at the stupidity of this.

Send your feedback on this interview to the news team.

Could you identify any positives from the SCO saga? Has it identified any issues in the development that warrant more attention?
This really raises the issue of the flaws in our patent process right now within the government. When you look at the root of this issue, [SCO] has allowed itself to achieve levels of financial gain, has been able to throw broken glass on the highway of progress and not have to substantiate any claims. When you look at other products that are applied for through the patent process, you have to have a full submittal of the completed works to receive a patent.

Software patents are relatively new in this country; they've only been around since the beginning of the 1980s. Should the U.S. patent and trade process work in a way that required full submittal of source code and you as a software developer could go to an information base and examine what that application is looking at, there would be no need for this case. There would be no need for the decline in customer acceptance or no need for the massive legal fees that both parties are expending in this, of which as of today there has been no resolution.

In my interactions with the open source community, I have seen nothing but respect with regard to intellectual property and patents. There will be zealots in every community who will say bad things about patents. We don't feel that way. But I think what we've learned from this is the need for the U.S. government to take a much greater scrutiny of the lack of sustainable processes associated with patent applications for software. That threatens innovation long term. What has to happen for that to change?
[Enterprises] are seeing the proof points of Linux on the desktop on the commercial side already. There's a lot of .doc, .ppt, a lot of ActiveX, a lot of Visual Basic, a lot of Exchange, NT 4.0 [in the enterprise]. When you keep going deeper and deeper into this morass, you say, 'I never want to do this again, because I'm never going to get out of this.' The migration costs have very little to do with the presentation layer. It has everything to do with the support or lack of supporting infrastructure -- and that's a really hard problem to solve on an enterprise-wide or government-wide basis. How much attention and resources is Red Hat going to devote to the desktop in the next year?
It's been a big focus. If you look at the guys who are leading the efforts inside of our company, they're the leaders of GNOME. We've been involved in Linux client computing for seven years. I think the hard question is the Hollywood effect against how you monetize it. We're going to choose to stay out of the Hollywood effect and deliver value that solves customer problems. That's where there's been a lot of enthusiasm for the desktop, I just don't see a lot of customers saying this is a top priority.

Red Hat could have done a much better job communicating, planning and preparing for [Red Hat Linux end of life]. Had we done a better job, that transition to Fedora might have been easier.
Matthew J. Szulik
CEORed Hat Inc.
There was some backlash when you made the decision to be strictly enterprise-focused, to end support for Red Hat Linux and to introduce Fedora. Where you prepared for it? Was it worse than you imagined?
It was worse than we imagined. I think part of that had to do with the better job that we could do to communicate what we were doing. I understood the transition was hard, but what surprised me all over the world where I've spoken is the number of people who were running small and medium-sized businesses on Red Hat Linux 5.0, 5.2. There [were] some very deep, deep uses of our technology.

Red Hat as a company could have done a much better job communicating that, planning and preparing for that. Had we done a better job, that transition to Fedora might have been easier. We didn't do that. I don't think the full acceptance of Fedora has been reached. I've been surprised how popular Fedora has been received outside the United States. I was just in India and spoke to more than 400 people at InfoSys in Bangalore, and many of them were running Fedora. That was pretty interesting. It's going to be about change and transition. Fedora is a very good thing for the community and for developers, but the pain still exists. You've mentioned a couple of times the economic viability of Linux. Was there a turning point when this became a reality for Red Hat?
When we opened a Boston office. Absolutely, 18 months ago when we opened a Boston office and made the decision to focus on the enterprise. I'm not saying that because I'm here, but what happened was a cultural and financial shift for the company. We brought in guys that had helped build the DEC [Digital] VAX cluster environment. We brought in senior professional software developers to help scale the whole Linux computing environment -- the kernel, the tools, the libraries. That helped us build the first enterprise-class operating environment that would allow an Oracle, Veritas, IBM DB2 [to work on Linux]. That was the first time [we could] go into the enterprise with a complete offering. Could something like that happen to Red Hat someday, when a giant comes in with a big check?
I would hope not. We just finished raising $585 million, so the company has almost $1 billion on the balance sheet. We're profitable. It would be odd to think of a traditional proprietary software company buying us, but when you look at where the growth is coming domestically, and in China, India, Russia, open source is the watchword in those technology communities. You can never say never, but I think our goal is to create a great company. What was your reaction to Novell's acquisition of SuSE? Is it a good thing for enterprise Linux?
It helps validate what we've been saying for 11 years. I've been spending a fair amount of my life trying to convince people that Red Hat and open source could be an economically viable alternative. So [the acquisition] continues to validate it. Competition is good. We think one of the great things about open source computing is choice. If you're good and you're trying to build a great company, there's nothing better [for validating] that than to have more competition in the market. When one vendor has 96% of anything, you see what the result of that is. We think it's good; it will challenge us. As you talk to enterprise decision makers, CIOs, how has that conversation changed now, compared with two years or even a year ago?
Two years ago, there was a great deal of cynicism, suspicion about economic models. Now it's rare that I speak to a CIO who doesn't have a pilot going. In some cases, they've gone way beyond pilots. We've got customers on Wall Street who are running six and seven thousand servers. You've got companies like VeriSign that literally run the Internet on Linux, running 4,000 Red Hat Linux servers. So you're seeing those first movers not only get performance advantages, but also getting economic advantages. They're pushing deeper and deeper into what we call an open source architecture. They're never going backwards.

And it's helped that companies like IBM and Hewlett-Packard have thrown their weight behind this. Seven hundred applications now run on Red Hat Linux, which is fantastic. That total ecosystem has made it comfortable for [CIOs at large companies] to deploy.

So you're seeing those first movers not only get performance advantages, but also getting economic advantages.
Matthew J. Szulik
CEORed Hat Inc.
How often do you talk to the community about the business end of Linux? What's that dialogue like?
I attend a fairly large number of Linux user group meetings, which I learn a lot from. I spend a lot of time with customers. There's a big difference between the people who are developing an open source component in the community and how it's deployed at a Morgan Stanley, Putnam or Fidelity.

I spend a fair amount of time on the site looking at the lists and paying attention to what people say. Of course, we have great guys who are luminaries in the open source community who work at Red Hat, and I hear from them whether I like it or not. A lot of people still hold dear the grass-roots aspect of open source and believe that Linux shouldn't be commercialized.
There are those who believe that. But if you were to sit down with Alan Cox, Steven Tweedy and Matt O'Keefe, luminaries in the community, they understand the core value of open source software is to produce a better product for the developer. The fact that we put 21% of our revenues back into the general public under the GNU Public License, we've produced better products like Fedora, all of that is positive reinforcement that we haven't lost our way in the last 11 years.

Matthew J. Szulik
In setting Red Hat's corporate strategy, how do you balance your responsibility to your shareholders and customers with your responsibilities to the open source community? Is it a difficult juggling act?
It used to be. This is all we know, unlike some of our competitors, who may have spent a couple of bucks recently and call themselves an open source company, when they've been building proprietary software for the last 20 years. We've been an open source developer for the last 11 years.

In the mid-'90s to the late '90s, when we started to develop an economic model around open source software, there were some members of the community who [had] a hard time with that. But then you saw companies like VALinux, at the time, go public, you saw Caldera getting ready to go public. I think right now [the fact] that all of our technology continues to be placed back into the general public under the GPL license keeps us front and center and leading a lot of the important movements around Linux.

Dig deeper on Linux news and updates



Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to: