Don't be disappointed. Read this excerpt from 1995 interview and you'll see why.
160.111.7.240
Excerpts from an Oral History Interview with Lawrence Ellison, President
and CEO Oracle Corporation.
Recipient of the 1994 Science Applications International Corporation
(SAIC) Information Technology Leadership Award for Global Integration
Interviewer: Daniel Morrow, Executive Director, The Computerworld
Smithsonian Awards Program
Date of Interview: 24 October 1995
Transcript Editor: Thomas J. Campanella, Computerworld Smithsonian
Awards
The Quest for Video-on-Demand
DM: Now, ideas and questions drive visions. Another story I have heard
is a story about a British Telecom request regarding software that would
run a video server, for video on demand. Tell me that story.
LE: Sure. Actually, we had signed a contract with U.S. West, where we
were the supplier of enabling technology for U.S. West's Information
Highway Initiative. And in that contract, we signed up to deliver video on demand in about 5 years. And, to tell you the truth, we did not know if we could do it at all. We really had not spent much time thinking about it, but 5 years is a long time and we saw technology moving a lot in 5 years and, who knows, maybe we could even keep our commitment. We had some ideas, but we really had not come up with a solution. But we thought we could solve it generally in 5 years. We felt that was safe.
And that was what U.S. West expected; they thought it was unreasonable
to try to get it very much sooner. They did not think this was our area of expertise anyway. Then British Telecom had the reverse smuch sooner.
They did not think this was our area of expertise anyway. Then British
Telecom had the reverse set of priorities. They wanted video on demand
immediately, and were willing to wait for some of the other services
that our software offered. A guy just a few offices down from mine
received the British Telecom request for proposal, and I was walking by his office and he said, "I've got the BT request for proposal and
they're asking for video on demand almost immediately; they need it in
about 6 months." I said, "video on demand I'm not sure I e
And it was very, very clear that the reason this is such a difficult
problem is that you have to move a tremendous amount of information;
every movie is 1.5 megabits every second. And then there are lots of
movies. You multiple that by a factor of 100 or 1,000; you have lots of
movies being played and stopped and started and individually controlled.
So the problem was moving information through a computer at much higher
rates than any computer could handle at the time.
That is, any computer but a massively parallel computer. It turns out,
all the work we had done in understanding massively parallel technology
and working with massively parallel technology could be re purposed to
handle the video-on-demand problem, which suddenly became trivial.
Because a massively parallel machine's ability to move information was
10 hundred times greater than conventional machines, that the problem
was not much of a problem. And we generally don't admit this.
I then called in one of our senior programmers--not a manager, one of
our senior programmers, actually a fairly young mathematician from
Harvard, Bill Bailey. I called him into my office and I suggested that
we could use the N-cubed massively parallel machine to do video on
demand. And I hope Bill remembers this. He told me I was nuts. And,
obviously, people here just are not afraid of me enough.
And I said, "Well, just hear me out a half an hour." In a half an hour
he said, "There's no question we can do this. In fact, I can do this.
And I think I can do this pretty fast." And Bill Bailey and Mark Porter
and a couple of other people got our video server working within months
of the idea. Now, again, we were standing on the shoulders of all the
work that was done by others in the MPP group, the massively parallel
group. But we were able to get our video server working almost
overnight--I mean, overnight is an exaggeration; even the overnight
successes take a long time. We had to build this foundation of 5 years
worth of massively parallel computers. But we were able to demonstrate
within months--and I will never forget, they called me up on a Saturday
night, they called me up at home at 12:30 am. And I was actually on a
date. I had a date. And there was just kind of screaming into the phone,
"It's working, it's working!" And we both, you know, went into the lab
and, sure enough, we had video streams, rock videos, coming out of this computer in our laboratories. And it's a pretty big computer. Well, it is not that big a computer. But here's the lab with all these stereo speakers and screens. It was not the normal Oracle computer lab. And these rock videos blasting all over the place. It was pretty exciting.
But a very small team of people had done this work. Now, I have the
utmost respect for Microsoft, but I always hear Bill Gates saying he is spending $100 million a year on doing this. Again, software has never been a capital-intensive undertaking. If it was a capital-intensive undertaking, you know, I suppose we would have a tremendous number of huge software firms in Saudi Arabia; they certainly have a lot of cash. Again, in a very short period of time, a very small number of people did remarkable things by combining the right technologies and finding the right solution to the problems.
British Telecom opened the first on-ramp to the information highway.
Again, the last thing I want to do is slam Microsoft, because I have the utmost respect for them--but it was interesting that CNN identified that Microsoft's information highway demos were precisely that. They were fake. You know, they were not real demos. They were movies of what it might look like. And then Bill said, "Everyone's stuff is fake at this stage." And someone in the audience stood up and said, "Well, then you should come to Leeds, where it is up and running. You can go to some people's houses and this stuff actually works and it is turned on."
So there is an N cube video server delivering across regular telephone
wire, twisted pair, not fiber, not coax, but across twisted pair, movies to the homes of people in the United Kingdom. It is interesting that the UK and British Telecom beat Bell Atlantic now. The first North American project turns on, I believe, in mid-July; Bell Atlantic, in the area of Alexandria, Virginia. What a perfect spot to turn on the first digital library in North America.
|
