At the Helm: LARRY ELLISON
By Julie Gibbs

ORACLE MAGAZINE: You've said that Oracle9i is the last database. What does that mean, and what do you think is next for databases?
LARRY J. ELLISON: This isn't the first time I said Oracle was the last database. What I mean is that Oracle9i introduces a new standard for data management. Oracle9i is an unbreakable system. You can't break it, and you can't break in. It's very secure.

We've passed 14 different certifications to prove our database is secure. IBM DB2 has none of those certifications. Microsoft only has one. We paid a lot of attention to data security, which is extremely important in the age of the internet, as we put more and more of our precious and private data into computer systems.

We have also become more and more reliant on these systems. They can't ever go down. They have to work 7 days a week, 24 hours a day. It's been a long time since we built systems like that. In the past, there were companies, like Tandem, that were dedicated to building nonstop, fault-tolerant systems . Those systems have gone by the wayside, because they were very expensive and it was difficult to write programs for them. They required customized programs and specialized hardware to deliver fault-tolerance.

By fault-tolerant , we mean tolerate failure, so if your server goes down, the application keeps running. If your whole site fails—if for some reason there's an earthquake and your data center loses power or the floor cracks and the machines fall over—another site should be able to take off and keep running. Your users should never experience an outage, your companies should keep going, and your government agencies should keep going. There should never ever be a failure.

You should tolerate site failures—in other words, server failure, software failures. If one of the machines is bugging in the Oracle database and it brings the system down, it should bring one of the servers down, but the other servers should keep going. This is the notion of the Oracle9i fault-tolerant database clusters. And the great thing about the Oracle9i fault-tolerant database clusters is that they take any Oracle application. Any existing Oracle application will run faster and fault-tolerantly if you just move it over to an Oracle9i database cluster. So it's very easy to make your applications run faster: just install the clusters. It's very easy to make your applications fault-tolerant: just install the clusters. You don't change a single line of code.

As we start to automate information, we have databases everyplace. We have files on our PC hard disks. They really belong on a big centralized database, professionally managed, shared when you want them to be shared and kept private when you don't want anyone to see them. All of the different customer databases people have, all of the different ID databases governments maintain, can be consolidated into simple comprehensive databases, and we can make sure because of new technology that those databases run 24 hours a day, 7 days a week. They will never fail, because they are running on unbreakable software. And no one will get information they're not allowed to see, because they can't break in.

OM: Can you talk about Oracle9i Application Server? Tell customers why they should use it.
LJE: If you're looking for a very fast J2EE system, Oracle9iAS is dramatically faster and more standards-compliant than what BEA or IBM offers. The reason you picked Java in the first place is so you wouldn't be locked into any one supplier. So you can be using BEA or IBM WebSphere today, and you can move these applications literally in an hour or two. And Oracle9iAS works much better with the Oracle database, because we made sure the pieces fit together.

OM: What's the difference between Oracle E-Business Suite and what other application vendors claim are suites?
LJE: We want to give you a complete set of applications with the E-Business Suite. But we built the E-Business Suite so it's open and extensible. That's unlike SAP or PeopleSoft or Siebel. We designed Oracle E-Business Suite so it's easy for you to add cool new features to our applications. And the great thing is that when you add cool new features, you use our shared schema, and you use standard Java development tools based on standards such as Java, HTML, and XML to build extensions.

That's very different from the other applications vendors. If you try to extend PeopleSoft, you've got to use PeopleTools—a completely proprietary approach. Siebel uses everything: Visual Basic, Siebel Tools—a long list—because Siebel is a big collection of acquisitions. And with SAP, you don't have XML APIs; you've got something called BAPIs [business application programming interfaces]. You don't use Java to add to SAP; you use a proprietary language called ABAP/4. So we have the only set of applications that are designed to be extended by partners or by customers using standard internet tools.

OM: Where do you see the most creative development happening today, at Oracle and in general?
LJE: It's very hard to keep a developer from being creative. In fact, sometimes developers are too creative. I can't think of a single area where we build software that there aren't a lot of really creative people. The primary thrust of our database is to make it extremely fast, which we've done with clustering; extremely reliable, again with clustering; and extremely secure. And the primary thrust of our applications is to make the system much more complete. So rather than our customers having to buy a lot of component parts and somehow stitch those together, the developer's job is to make complete flows of business information. Automate these complete flows so that companies can stop buying components and trying to glue the components together and instead buy a complete system that works out of the box.

OM: Let's talk about the state of the global economy. People have referred to it as a global economy for a while now, but if you look at places like Africa and Afghanistan, it doesn't seem that we're really there yet. How do you think technology can affect those economies?
LJE: We have a global economy; what we don't have is equality. If you're a very poor nation, you're not included in the global economy. The global economy doesn't market to nations that can't afford to buy.

Actually, what we really do have is a very discriminating global economy. It's not necessarily terribly humane, however. So we look to governments and philanthropic organizations to show a sense of humanity. Our economically driven institutions and organizations will just cull out those nations and those people that are not market opportunities.

The great thing is, in response to that, you just tax your for-profit organizations to provide money for your nonprofit organizations. That has proven to be the most efficient way to do things.

A corporation's primary goal is to make money. Government's primary role is to take a big chunk of that money and give it to others. The richer corporations become, the more money that flows from corporations to government, and the government then redistributes that money to people who are not so fortunate.

OM: How do you think technology affects economic cycles? Can it help us move in and out of recessions faster?
LJE: It should make us much more efficient. It's very interesting: When you have a global economy, the whole world goes into recession at once. This is uncharted territory. We've never really gone into a recession all at once since 1929. I don't think we are headed for a depression or anything remotely like it. In fact, this economy has in some sense never been better equipped to deal with a recession. We have basically no inflation; the banking system is incredibly strong; and we have government surpluses, not deficits. Technology has also yielded huge benefits in productivity—to the United States primarily, but also to Western Europe.

OM: You're interested in biotechnology. What key advances do you predict in that field in the next five to ten years?

LJE: Drug personalization. Currently there are lots of drugs that make it through the design phase, the animal trial phase, and to clinical trials, which is the human phase. All of a sudden you find that while this drug is pretty effective, unfortunately it kills one half of one percent of the population and therefore can't go out.

In the future, we're going to be able to look at the individual genomes—the DNA signatures of individual people. What are the genetic characteristics that made these people vulnerable to the negative side effects of a drug? If we know your DNA signature, we'll be able to warn you in advance not to take this drug. And with that, we might find a different drug that's effective for 20 percent of the people.

A lot of drugs that have been thrown out of clinical trials will be reexamined and prescribed on a personal basis.

OM: Do you think that will apply to things like cancer or diabetes?
LJE: Across the board.

OM: It certainly seems right now the approach to disease is cut it, burn it, or poison it instead.
LJE: Yes, but there are all sorts of other therapies. We're getting our first effective gene therapies—and the different vectors of the rhinoviruses, which are the classic vector for a lot of the genetic engineering.

It's basically splicing—adding a gene. Let's say you're diabetic and we want to fix the diabetes. We'll have to go in and change your genetics. If it's Type 1 diabetes, something you inherited, you have to change your genetic makeup. There are basically DNA editors, called viruses, that go in and actually splice in DNA. We know how to do it. It's kind of hit and miss and we haven't had too many successes in that area, but we're getting better at it.

And then there are stem cells. There are two kinds of stem cells: adult stem cells and embryonic stem cells. All the controversy is around embryonic stem cells. Irving Weissman over at Stanford is doing a lot of work looking at adult stem cells, which, we think, will also be very effective (and much less controversial) against diseases like Alzheimer's, Parkinson's, and Type 1 diabetes.

OM: You've said Oracle is pretty much the gold standard for clinical trials. It seems that our database applications and server are critical for biotech applications as well. Is that a fair statement?
LJE: Yes, most DNA databases and gene banks are maintained in Oracle. After all, you certainly don't want a data error to start costing people their lives. And you certainly don't want the database to be so slow that you just can't get an answer.

This really is the information age. As Don Rumsfeld said recently, the key to fighting terrorism is not a cruise missile or stealth bomber or a submarine. It's a piece of information. You can't fight cancer, you can't fight terrorism, you can't fight any of that without good information.

OM: You established the Ellison Medical Foundation to support specific areas of research. Tell us about the foundation.
LJE: It's an extraordinary medical foundation led by Nobel laureate Dr. Joshua Lederberg. We actually have five Nobel Prize winners on our advisory board. So when the advisory board meets and I'm there, I'm the only one without a Nobel Prize. It's an amazing group of people.

The foundation funnels money to researchers in two areas. One is infectious disease. Specifically, Third World infectious diseases: tuberculosis, parasitosis, malaria. There is no financial incentive for drug companies to produce drugs in these areas—they can't make any money because they can't sell them. The foundation allocated a fund of a quarter of a billion dollars for research in this area.

The other area is diseases of aging: Alzheimer's, Parkinson's, and cancer. Cancer is a disease of aging. You're much more vulnerable to cancer as you get older. And as our population ages, we've got good news and bad news.

The good news is you're going to live longer. The bad news is the medical costs you incur during your lifetime are going to go up dramatically. That's going to put tremendous economic pressure on the country to deal with our aging population, especially as the baby boomers and people my age go into retirement.

It's also infinitely more humane to cure a couple of these diseases than it is to provide care. The cost of care is a hundred thousand times more than the cost of delivering a drug that's a cure. So for purposes of humanity, for purposes of economy, the interest is providing cures, not just care.

OM: Whom do you listen to?
LJE: I listen to my best friend Steve Jobs. I think Steve is one of the few visionaries in the world.

OM: What other leaders do you admire?

LJE: Certainly Winston Churchill was one of the greatest people in the twentieth century. He saved Western civilization when everyone else was ready to give in to the fascist dictators.

In technology areas, I greatly admire General Electric (GE). GE is a magnificently run company. They can be tough to do business with, because they want so much from you and they're tough negotiators. But the people are very rational. Very demanding but very fair. GE has an incredible corporate culture. They keep constantly asking the questions "How can I do more with less?" "How can I make better products while spending less money?" It's a remarkable company that Jack Welch created. And it spreads beyond Jack. There are a lot of talented people running that company.

I'll tell you, my favorite visionary of all time was Galileo. Conventional wisdom at the time of Galileo said the earth was the center of the universe, and the sun revolved around the earth. But Galileo said this guy Copernicus was right: The earth goes around the sun. He got into a lot of trouble for saying that.

OM: With your track record as a visionary in technology and business, you've certainly gone against conventional wisdom yourself several times over the years. What's it like to be in that position?
LJE: When you're the first person whose beliefs are different from what everyone else believes, you're basically saying, "I'm right, and everyone else is wrong." That's a very unpleasant position to be in. It's at once exhilarating and at the same time an invitation to be attacked.

There are really four phases. In phase one, everyone tells you you're crazy and it's the stupidest thing they ever heard. In phase two, they say, "There is some merit to the argument. It's still crazy, but there's some merit to it." Phase three is, "Well, we've done it better than they have." And phase four is, "What are you talking about? It was our idea in the first place."

It's fascinating as we continue to innovate and lead the way in both the application space and the database space. In the very beginning, people said you couldn't make relational databases fast enough to be commercially viable. I thought we could, and we were the first to do it. But we took tremendous abuse until IBM said, "Oh yeah, this stuff is good."

We were the first company that said all the applications had to be on the internet and not client/server. Everyone said that was a bad idea. That was 1995. Now everyone has moved all their applications to the internet.

And now we're saying you have to have a suite—that this best-of-breed approach is crazy. You can't sell parts that were never designed to fit together. They're still saying we're crazy about that. But it's interesting, SAP and PeopleSoft are now advertising they have suites. Everyone has started using the "suite" word.

And so the four phases repeat over and over again. As long as we continue to innovate, I don't think that's going to change. When you innovate, you've got to be prepared for everyone telling you you're nuts.

Julie Gibbs is vice president of Corporate Marketing at Oracle Corporation. She has been with Oracle for 13 years, serving as editor of Oracle Magazine for many of them.