Q: What do you make of Oracle's hostile takeover bid for PeopleSoft? Do you think Ellison is serious?
A: Absolutely serious. It's been one of Larry's regrets over the last few years that he didn't buy PeopleSoft earlier. When PeopleSoft was in a weakened state, about five years ago, Oracle could probably have picked it up for a 2% dilution. But it was probably an even bigger mistake that Siebel Systems didn't buy PeopleSoft, because Siebel badly needed an ERP play. Larry believes the industry is going through phases quite rapidly. He views PeopleSoft as a company without a great future on its own, but with a fantastically attractive and solid base from Oracle's point of view. He sees it as a way of growing Oracle's maintenance income, which is the most important thing these days for software companies. As software develops into a service and subscription business model, maintenance revenue could matter most. Larry believes he's going to pull off this [PeopleSoft] deal. He thinks there will be clearance on the antitrust concerns within the next month or two.

Q: You show how Ellison likes to "bet the company" every so often on a new technology, such as the ill-fated network computer. Now it seems to be grid computing. The Web site announces "Get on the Grid," and Ellison focused on grid computing in his recent Oracle World speech. Yet, grid technology is hardly mainstream. What do you make of this?
A: I don't think it's what you'd call a bet-the-company move. The E-Business suite was a bet-the-company product, because of the amount of resources and the change in direction that went with it. Regarding grid computing, the idea of computing as a utility has been around for awhile. I think IBM was the first to start talking about it. The difference now is that Oracle has its own technology enabling that. It's drawing attention to its ability to offer a new kind of outsourcing model based on subscriptions and software services, and at the same time it's playing to its core technological strengths. It's an incredibly big deal when you see it demo'd at customers running Oracle online using this stuff. It's amazing-you can get the same computing power using racks of cheap Dell boxes as you get with a big Sun machine.

Q: What about that $1 billion reduction in operating costs Oracle claimed to have gotten by using its own E-Business suite? You wrote that the claim wasn't entirely accurate, didn't you?
A: All I meant was that when Oracle was achieving its major cost savings, it wasn't doing it with the E-Business suite, because the suite was still in prototype form. Oracle was using itself as a lab to understand more about how the E-Business suite should be designed. So when the company's advertising campaign said Oracle had saved itself a billion dollars by using its business suite, that wasn't quite right. The E-Business suite came slightly later and was more of a consequence than the cause of that. But in some ways, Oracle saved much more. One of the reasons it's been able to maintain its margins during a difficult few years for the industry is because it's gone on cost cutting. And it's been able to do that because it's understood more and more about how to automate its business. A lot of people said, "Oracle has succeeded in cutting cost because it's reduced head count and done a lot of things that had nothing to do with the software." But that's not true. Yes, the head count has gone down, but it's the software that enables you to reduce head count in key areas. In terms of the way Oracle reengineered its business processes, it was to make sure the software it uses works better and to make Oracle employees use the software.

Q: Your book's subtitle is "an intimate portrait of Larry Ellison and Oracle," as opposed to being either a biography of the man or a profile of the company. To what degree is Oracle a reflection of just one man?
A: In many ways the company is a reflection of him. When you have these fantasy CEOs who also provide the vision and the business strategy for the company to a large measure, they have a huge impact on everything that goes on within those companies. That's not to say that companies like Microsoft and Oracle won't have a life after Gates and Ellison have left the scene. They will, because they both have huge management strength and depth. But they'll change and evolve into something different.

Q: Does Ellison view himself as a business warrior? You call your book "Softwar," not "Software," and I keep thinking of two lines from Ellison that you quote: one, when he remembers thinking that "the entire computer industry was heading in the wrong direction," and the other, when he says he's never so happy as when everyone thinks he's wrong.
A: Larry's a contrarian. He doesn't believe you should be out on a limb all the time, every time, because if you are, chances are you're going to fall off badly. But nonetheless, he thinks you need to take an opposite view every few years, because being right and unconventional is what actually makes you money.

Q: So when Ellison makes his public pronouncements—the ones that have people rolling their eyes—he's not just being provocative for the sake of being provocative?
A: No, he absolutely believes it. He thinks very deeply about the industry and its products, and how it hasn't always delivered. It's consistent with his belief that what matters more than anything is data integrity. Obviously, coming from a database expert, that's what you'd expect. But it gives Larry a different perspective that leads him to a view of the nature of computing and the kind of architecture you need to support it. It's not that he wants to say, "Everybody is wrong, and I'm right," but, rather, to express things in a vivid way and keep on repeating them, so people will look at ideas they didn't get the first time around and pay more attention to them. It doesn't always work, because some people do just roll their eyes and say, "Larry's at it again." But he's been right often enough for people to take notice.

Q: Over the years it seems some very talented people have found it more or less impossible to work with Ellison.
A: Oracle has a deep bench in terms of the abilities of the people who work there. Most of the important companies in Silicon Valley have people who came through Oracle. The company has consistently provided a large part of the leadership in the software industry. These people didn't necessarily leave because they couldn't work with Larry. Some remain close friends and allies, though there are a number who, it can be said, butted heads with Larry in such a way that they had to leave the company. What's more interesting, in a way, is how Oracle has consistently managed to recruit talent.

Q: What about Gary Bloom, whom you describe as being Ellison's "right hand" during 1998 and 1999, but who left Oracle in 2000 to become CEO of Veritas?
A: Inevitably, when you have very talented people, they look for opportunities to run their own shows. Gary Bloom is an example of that. The main thing for Gary was that he had been working in a No. 2 or 3 position to Larry, and he knew Larry was going to carry on for quite a long time. Gary wanted to be a CEO, so he left. That happens. There's no question that Larry is a strong personality, and there are times when he will become difficult for some people to work with. When you have two strong personalities and they disagree about something over time, that's not going to work. But I also saw that Oracle is an open place. There's plenty of argument and no false deference. I've been in plenty of meetings where people made it clear they disagreed with Larry, and they weren't afraid to argue with him. Finally, it's a fact that the most important people at Oracle are the people who've been in the core database team. They're the people who've been responsible for product superiority--for making Oracle money--and it's pretty much the same people who've been there for 20 years. They're a very close-knit team of people, and there have been very few departures from there. On the whole, the people who have come in and out have been on the sales side. They tend to have a higher profile outside the company, but that doesn't necessarily mean they're the most important people in the company.

Q: You write that Ellison periodically gets distracted by sailing and other interests. He gets bored with the business, loses focus, and then has to scramble and adjust. What are the chances he might soon lose interest in Oracle altogether and leave the company?
A: I don't think there's a chance of that. He seems much more interested in running the company than ever before. He used to find it dull; he used to think applications were dull. He was, until fairly recently, interested only in the core database business because that was the math and the physics stuff that he found intellectually challenging. It was only when he got involved in the applications--because of the problems Oracle had in that area--that he actually began to realize that applications were interesting. He saw applications' ability to automate a company and get real-time management information out of the business. I think he gets more job satisfaction out of the way Oracle is running now than he did previously. That said, he's got a lot of other interests, and I think at some point he would like to ease himself away from Oracle. But he can only do that if he thinks Oracle has "made it," in whatever terms he applies that judgment. I suspect he will find it very hard to do. I think he may have a retirement plan in mind for a little over five years from now, when he will be 65 years old. But whether it will happen is anybody's guess.