Spring Internet World 1999

Sean Maloney
Los Angeles, Calif., USA
April 14, 1999

SEAN MALONEY: Well, good morning, ladies and gentlemen, and welcome here to Internet World. I guess you are here because you are born to be wired. Some companies go a bit further than that and they are born on the Web. Those are the companies that have been set up inside of the last five years explicitly to take advantage of the Internet and the World Wide Web. Then there are the rest of us whose businesses were established before the Web that are working to be reborn or reengineered on the Web. And we're going through and we working all of our fundamental processes in our organizations, up to and including our very business model.

But if you think about those companies that were up here in the video earlier on, most of those companies actually were created within the last five years. Those companies, incidentally, Intel is an investor in many of them. And many of them worked on the introduction of the Pentium® III processor.

Five years ago is really a very, very short period of time. And to take your minds back just five years, it's less than five years, it's about four and three-quarter years ago, here in L.A., that the helicopter was going along on top of the freeway where O.J.'s white Bronco was cruising along. That was less than five years ago.

And I remember almost exactly five years ago giving a demonstration of a cable modem. Intel had developed a cable modem five years ago. And we'd really done it to take advantage of what everybody then was pumped up about, which was video on demand. You may remember this.

And the idea was that there would be interactive T.V. that would get video streams. And we wanted to capture this video and try and get it into the PC instead of into the T.V., which we kind of naively thought of as competition. But there wasn't any video on demand to be able to display the cable modem. And so we were looking around for something to demonstrate. And we came across Mosaic. And so we thought, maybe we can display this thing on the World Wide Web going really fast on a cable modem. But it was very difficult to find any content other than academic content and scientific content.

And we actually found a site in Texas that claimed to sell flowers. And so we did a demonstration of this cable modem with Mosaic. And we ordered the flowers. And it was a really good demonstration.

The flowers never arrived. But it was a really good demonstration.

So wind the clock forward five years and, you know, we all have our Internet staggering statistic de jour. In the month where 30 percent of all stock transactions took place on the Web, I was particularly struck by yesterday's Wall Street Journal and the front page article that said there are more than 145,000 pages, just about Pamela Anderson Lee.

145,000 pages.

Content isn't so much the problem, or certainly massive content. Now, that's after America has come on the Web.

Now wind the clock forward three years. You've got this explosion taking place in northern Europe now on the Web. And most conservative projections are that the Web population in western Europe will be bigger than the Web population in North America. And, of course, we've got Asia, which is coming on stream incredibly fast. In Japan alone in the last seven months, Japanese figures are five million Japanese people came onto the Web. That's nearly a million a month.

So this torrent of information and content and opportunity is obviously carrying on.

For companies like Intel and I'm sure for your companies, it represents an incredible, an unavoidable challenge about how you change your business models to react to it and to take advantage of it and not to be the victim of somebody else's URL.

In Intel's case, it's really affected every single part of the company through design, manufacture, and how we service our customers and deal with our customers. It's really impacted every single one of those aspects.

But today what I want to do is to focus on our lessons over the last year in how we relate to our customers, how we service our customers through the Web, very much from a business-to-business context. We only do a small amount of selling to consumers over the Web. The majority of our business is absolutely with our major customers the telecommunications and computer companies. So I want to talk about that experience, and then three fairly simple, obvious conclusions that we have on some of the things that are wrong with the Web, improve information management, improve the infrastructure, and ubiquity of access. And I'm going to try and talk through some options in each one of those three.

We find ourselves today, Intel is a classic manufacturing company, 65,000 employees, 40 something thousand people involved just in manufacturing. We find ourselves today probably as the world's largest E-commerce purely manufacturing company.

We are doing a very substantial amount of our business directly over the Web. And I mean purely over the Web. That means taking orders and not kind of then printing them out and typing them back in again, which some people do, but taking those orders, receiving them 24 by 365 and having them directly fed into our computer systems.

But a year ago, it wasn't like that. Actually, a year ago, we had zero revenue on the Web. All of our business was done in the normal manner, which is kind of typically telex and telephone.

And we realized that we wanted to completely change that and do 365 by 24, the promise of the Internet. So we kicked off a huge project inside the company, probably about 500 people involved in it. We built our software stack all the way from ground up to do all the obvious stuff like order entry, order rescheduling, checking on status, all those kinds of things. But we also went a stage further, because we were concerned our customers -- remember, this is business-to-business E-commerce -- our customers would be particularly concerned about security. So we built in 128-bit security. And we gave every customer a targeted Web landing zone which was completely secure.

Now, those were the things that we did inside the company to get what we thought would be a really good service.

But we made a discovery, which was that on business-to-business E-commerce, your infrastructural problems are considered to be your own.

Now let me talk through the infrastructure for a minute here.

We have customers all the way around the world. And those customers would then be placing orders on us that would be mission-critical to their businesses. And those -- essentially what we would be doing is we would be taking their local area network and connecting it to our local area network, and between that, we were going out from their LAN to their ISP, which could be anywhere around the world, across the WAN, into our ISP, and then back into our infrastructure.

Now, there are several points on those which are possibly bottlenecks, and turned out all through last year for us to be very, very significant bottlenecks. And, of course, if you have a slow response when you're doing business-to-business E-commerce, you blame your supplier. You don't realize that maybe it's your ISP or it's the WAN infrastructure.

The most extreme example of this we had was in Asia. Now, I don't know if you're aware of it, but there is a huge shift in manufacturing in the worldwide computer industry from the United States into Asia. And Taiwan in particular is now the world's number one site for manufacture of computers and is rapidly becoming the world's number one location for the design of computers.

And so we agreed, with our Taiwanese customers that, we wanted to go really, really fast on this and try and get 100 percent of our business in Taiwan done on E-commerce.

But we had a whole number of fundamental problems. The ISP infrastructure there was inadequate, so we worked with the local ISPs. We started designing our own pinging tools to figure out where traffic was being held up. Because the normal kind of one-second transaction in the U.S. was taking two minutes to complete in Taiwan. And that was completely unacceptable.

We were then -- we then got involved with the Taiwanese government. The Taiwanese government started pushing very hard to get additional backbone infrastructure. And, you know, by the end of last year, it all worked out very well. And we've got about 95 percent satisfaction rate from our Taiwanese customers. 100 percent of our business, multi-billion, billion business, 100 percent done by Web commerce. The whole thing's worked out very well. And none of our customers, I think, would want to go back to the old way of doing business.

We've created a sales force, essentially, that never sleeps, 365 days a year, 24 hours a day. And that's given fundamental benefits to our customers and allowed us to do work on our supply chain and really opening up every day we think of more and more ways we can optimize our supply chain. And as it says here, we are now doing right about a billion dollars a month in 30 different countries.

So this has been a very interesting experience for us and given us a whole number of opportunities and enabled us to do work on our supply chain.

But it's also raised in our mind three fundamental areas of things that need to be improved. And I want to talk through those now. And I want to start with the first one, which is information management, better information management.

That's a very broad term. But it's to do with how we enable the Web experience to be easier, less time-wasting, and prepare ourselves for when we have another order of magnitude, two orders of magnitude of information out on the Web. And this information management problem comes in a number of different facets.

The first one, which I think many of you will recognize, is the kind of incoming problem that you have of messages. We all know the statistic that E-mail overtook first-class post. You know, that was a couple of years ago. As of this month, the latest figures are there's something like 30 times as many E-mails sent every day as there are first-class letters sent. And the letters aren't dying off, but the E-mails are continuing to rise at an incredible rate.

Inside of Intel, like any other modern company, we adopted E-mail in the middle 1980s and ramped it all the way through, with a very sharp ramp over the last five years or so. And I think to a certain extent, when you're connected and you're doing your bills through E-mail, you're almost under siege.

If you take the situation at Intel, we are at something like 35 messages per day. And remember, as I said earlier on, that two-thirds of those employees are essentially on the manufacturing line. It's a very automated, highly advanced, high technology manufacturing line. So they are connected to E-mail. But they don't get so many E-mails. And that means that a lot of the rest of the employees are at -- you know, my personal instance, 100, 150 E-mails a day. I know many of you are in a similar situation. And when you're on the road and traveling around, a huge amount of time is spent dealing with this incredible amount of incoming data.

The solution to that isn't necessarily immediately clear. It probably lies in the area of better client site tools, better filtering, single in-boxes and rules, client side rules and server side rules to filter out what's not really necessary and allow you to see only what's necessary.

That's an incoming information management problem.

But we've also got an outgoing information management problem.

When you go out and you search, it is very, very difficult, particularly if you don't understand Boolean operators, it's very difficult to be able to get that information into any kind of rational, coherent list. You'll get a long, long, long series of hits. By the time you're on your third page of click on the next ten and you've gone down a couple of detours and some frames have popped up, it gets very difficult to put the thing in its overall context.

Bearing in mind what I said earlier on about the European countries coming on, the Asian countries coming on, you're going to be doing searches where you're getting hundreds of hits in languages that you don't even know what they are. So we obviously need some new approaches to try and handle it. And in some ways, the Web is kind of getting too big just for two dimensions.

I want to show you a technology here which has just been introduced by Excite which is an attempt to try and use different techniques to solve that. And it does it by putting the search in a three-dimensional context. And Phu Than here is going to show us this excite extreme. And what you're seeing on the stage here is the result of a search. The search is on the J.P. Getty Museum, which is a museum here. And instead of having a normal two-dimensional list of URLs, the search has returned a series of hits, and they get grouped together into carousels, islands of information, that have a common theme. And those carousels can then be rotated, and instead of seeing the URL, you actually see an extract from the first page, from the home page on the site.

And so you're able to keep much more information on the desk, you're able to see it in some kind of overall context. And then when you find the URL that you're most interested in, you can kick it off and launch it.

Now, this is obviously very computationally intensive, needless to say, with an Intel guy on stage here, we have on stage what you will have at home, which is a Pentium III processor at 500 MHz. (Laughter.)

SEAN MALONEY: There is going to be a lot of them around by the end of the year, I hope.

And this processing power means that you can take that information and you can put it into some rational context. And I think this kind of thing, particularly with the rapid, rapid surge in XML, these kinds of technologies are going to be very, very useful at sorting out that chaos and the overload of information.

Now, that's a way in which -- a way in which you can try and sort through the stack of information that we have, that we as consumers are going out and seeing.

Now, on the other hand, if you are trying to get your message out to people, how do you get your message to rise above all of the noise? I'm going to show you some techniques on that in a minute.

If you are a company out on the Web, you have got -- if you're an older company in particular, you have got large SQL databases storing information. About 15 years ago, the industry went from hierarchical databases to SQL relational databases. And the objective there was to make -- implement a database language that was easy to use.

It turns out now that SQL is actually very, very complex and you need specialist training to use it. But there are thousands and thousands of databases that are out there using SQL.

And so David Sidd here is going to give us an example of a technology that will enable those SQL databases to be accessed very easily over the Web. The technology is called English Wizard.

Hi, David.

DAVID SIDD : Hi, Sean.

Right now I'm connected to that SQL database. And it in this case is a food order tracking system and I actually have no idea how to program in SQL. So I'm going to use English Wizard to make it easy for me to just ask a simple question and have it return the results I want.

Let me give you a quick example. I'm going to say who buys seafood? Whoops. F-o-d.

SEAN MALONEY: There you go. Well, it's -- see that very easy-to-use SQL language at the top there. (Laughter.)

SEAN MALONEY: I guess I could have typed that line in.

Of course this is from the era where you needed incredible precision when you used the computer and you get any one of these kind of parentheses, equal signs, any one of these wrong and you're out of luck. You go back to the beginning. This is a much, much easier way of accessing the database.

DAVID SIDD: Exactly. Let me show you another level of simplicity and show you how easy it is. I'll talk to the computer using Dragon Naturally Speaking and just ask my question directly to the computer.

SEAN MALONEY: Last night when you did this and you shouted at it, it didn't like it.

DAVID SIDD: Yes. It's touchy. I can't hurt its feelings.

SEAN MALONEY: You have to be nice to the computer.

DAVID SIDD: Let me show you real quick. Here's a simple example.

Computer. Computer. (Laughter.)

DAVID SIDD: Who buys seafood but not dairy products?

Scratch that. Scratch that. Scratch that. Who buys seafood but not dairy products? Execute. (Applause.)

DAVID SIDD: That's just a simple example. Let me do something a little bit more complex.

Computer. Computer. Give me a pie chart showing sales by category. Execute. (Applause.)

SEAN MALONEY: Thanks a lot.

Funny how temperamental these things can be. This one really, really likes it when you talk to it in a very gentle, quiet voice. It can be difficult at times like this.

So those are ways in which we can go and access the information and get it easier to use. But then you also have the problem, as I mentioned earlier on, about how do you as a merchant or supplier get your message to stick through all that noise. And I think there are a number of areas on the Web which have been kind of underutilized. And in some ways, the Web is in its infancy in using some of these techniques.

One of the techniques which has been used in other media for a long, long time is the use of recognizable audio.

Who, for example, could forget the audio that goes with this particular Hollywood product. (Music playing.)

SEAN MALONEY: Thank you. (Laughter.)

SEAN MALONEY: Everybody immediately knows what that is. And that music was inherently a part of that experience. And hopefully, you would all know this high-technology product that's being marketed here. (Music playing.)

SEAN MALONEY: When you hear that noise. And you may find that slightly irritating. (Laughter.)

SEAN MALONEY: Thank you. You may have noticed with a Pentium III, by the way, we've kind of stretched it into three dimensions. But it's very effective at installing a memory.

And also other ways of intriguing people and attracting people are through more clever visual methods. We're all going to be buying an infinite number of objects over the Web. And so we want to be able to look at those objects and be able to rotate them and play with them and look at them. It's a big problem for all of us, because there is no industry standard way that everybody has adopted in describing three-dimensional objects.

But let me just show you one example here. Like all the rest of these things, these are all live Websites. None of these are kind of prototypes. These are live.

Here's an example here from Sharper Image, sharperimage.com that Richard Thalheimer, who's the CEO, he talked on that video earlier on. This is live on Sharper Image's site now. And they use both the 3D and the sound. Let me show you, here's a favorite one here of mine, which is this little gadget you can buy called the Sound Soother. This is the perfect escape from a hectic, noisy world.

And we've hit -- no, we haven't. Excuse me a second.

We have a T1 link here. And I hope we don't have a problem.

Okay. Here we go. Okay. So here's our Sound Soother. And, you know, it's a three-dimensional object. So I can pick the thing up and rotate it and have a good look and look at the light shining off the top of it there.

And, of course, I can turn it on as well, turn up the volume. And what you see here is a series of different options, heartbeats, rain, tropical forest, and so on. And I can actually go and listen to some of those. So this is the sounds from north woods. (Audio.)

SEAN MALONEY: Okay. Had enough of that one. Then the ocean sounds. (Audio.)

SEAN MALONEY: Okay. So those are kind of techniques that, obviously, are getting more and more used in business-to-consumer E-commerce. And Richard's comment was that by directly using those techniques, he had seen a substantial improvement in his hit rate and his purchase rate on Sharper Image.

There are going to be many, many different initiatives over the next year, two years, few years to try and solve these kinds of problems.

You may have seen the announcement yesterday, which was in the Wall Street Journal and other sources, that Intel and Excite have announced a technology partnership. We're working together to bring Excite's technology together with Intel's iCAT E-commerce division to improve the experience of buying and selling products over the Internet to both consumers and to merchants. And we expect to see the results of that technology later on during this year.

So that was the first of the issues that we're looking at, which is improving this information management and getting through that overload and barrage of information.

The second area that we need to improve is in the way of infrastructure. And it's obviously all very well to have information and be able to sift through it. But the key questions are, can we get fast access and is it reliable?

Now, I gave you that example of infrastructure earlier on in Taiwan where we went out and fixed all those various infrastructural problems.

Obviously, we can't keep doing that. Nobody can keep doing that. It's completely impractical. And so the underlying infrastructure has to be significantly improved.

And I think you can break that down into two areas, into the server area and into the networking infrastructure area.

Now, the server area is one, obviously, Intel is very involved in with processors. And I think there's a very interesting challenge there.

You look at the telephone network, and the telephone network was designed from the bottom up to handle very, very big spikes in load. And that's obviously what we need to have in our server infrastructure. But the spikes, I think, are going to be even higher. It's going to be even more bursty than the telephone network.

I'll give you a for instance.

If you go talk to online banking companies, one of the big fears right now is what happens three minutes after midnight after the millennium. Everybody who has a bank account is going to be inclined to think, mm, I wonder if the year 2K bug just wiped out my bank balance. Compare the burst in traffic five minutes after the millennium with five minutes before, when everyone is having a toast and champagne.

And I think you're going to see over the next few years that more and more and more, you're going to see incredible bursts and swings in access. And it means the server infrastructure has to be engineered to handle huge swings either way.

Now, from Intel's role in this is relatively straightforward. We're trying to reengineer the server architecture to make it able to handle that, particularly in the last year, the Pentium II, Pentium III, Xeon™ processor, and we're trying to concentrate on breaking records from performance, particularly world performance. And here you have some of them this month.

Going forward from here, we have some other things coming up. You're going to start seeing a large number of affordable eight-way servers coming on the market very, very soon. And then behind that, later this year, we have the introduction of the next-generation Intel Architecture called IA-64. IA-64 will be cable with the existing architecture, but will -- has something called explicit -- it's explicitly parallel. And it will be designed so you can bolt together multiple of these things to handle big spikes on the Net.

Now, the infrastructure alone, the server infrastructure alone obviously isn't enough. You've also got to improve the IO. Coming soon is something called NGIO, next generation IO. That will be an IO which would enable the intersection of the telecommunications space and the data space. With the convergence of telecommunications equipment and data equipment, the convergence around IP, we have the need for a new high-speed IO, independent IO system. And that is NGIO.

Okay. So these are some of the technologies that are going to be kind of inside the walls of the enterprise of the ISP, servers that can handle high burst, much higher-speed IO. Outside of it, we are very, very interested in the possibility for virtual private networks. And in our case, we are so interested, we bought a company called Shiva a few months ago that is a leader in virtual private networks.

What I want to do now is to show you an example of what a virtual private network can do for you. And I think VPN is going to be a major, major phenomenon in the next few years.

Phu Than is back on stage again. And what we're going to do is show you a connection from here in L.A. to our Boston design site. And the connection is going to be over VPN. Now, that means that we're just going to dial from the stage here into a local ISP and then we use the Internet backbone to carry the traffic.

So, Phu, maybe you can show you what you've got here.

PHU THAN: Hi, Sean.

What I have here is a laptop. And on the laptop. To save us some time, I have already dialed into the Internet. And so right now I'm on the Net.

And what I will do next is I will use the Internet to dial back into two sites. One site is in Santa Clara and the other site's in Boston. I'm going to access those site over a very secure, encrypted link.

SEAN MALONEY: Okay. So we have one telephone line from this laptop computer. That one telephone line, using VPN, can access the Web and simultaneously access two different sites at the same time. So that's something that you couldn't possibly do on a normal point-to-point telephone call.

PHU THAN: Right. So what I brought up here is a VPN client screen. So on the top, you see that I'm connected to both sides. Bedford and corporate. That's in Boston and then corporate demo is in Santa Clara.

So I'm now connected.

What I did is I typed in the log -- the password and log on to save us time. So now I'm connected.

So what I will do next is access the network, select the Boston site and access the network in Boston.

So now, as you can see, it came up with a list of computers in the Boston site. So now I can -- may choose access, whatever file I need from the Boston site. So I will do that. I will go to the computer that we want to select is Marley. So now I'm going to go to Marley. And those are my files.

SEAN MALONEY: Okay. Thanks very much.

So that's a very simple demonstration. But what it shows is that we can kind of take advantage of the Internet infrastructure to significantly save money and reduce complexity. From a single telephone call to an ISP, we can access three different sources or more rather than just one particular source. It means we have a local connection charge rather than having to dial all the way across the U.S. and get a slower response. And it's a much more secure transaction. That's actually using 196-bit encryption, which, you know, so far, if you've got a Pentium III, a smart hacker could probably bust through 48-bit encryption. 128-bit encryption I don't think anyone's broken through. That's 196-bit encryption. So that's really a very, very secure link.

Okay. So that is the second of the points, additional things we need from the infrastructure in terms of speed, security, and robustness.

The third of my points is ubiquitous access.

The Internet is a challenge to the computer industry, but also it's a wonderful, wonderful thing.

We had a dream in the past that you'd sort of have one computer for everybody. Now you can make an argument for having an Internet connection, a computer, whether it's a PC or an appliance, you can make an argument for having it almost anywhere you are. You certainly want to have it in your office. You certainly want to have it when you're mobile. You probably want to have it in your car. And, of course, in the home, you can make an argument, if it's easy enough to use, for having Internet in almost every single room in your home, up to and including the kitchen.

So this year is a very big year for the industry, because we've spent several years talking about home LANs. And we've had all kinds of arguments about having servers in the basement next to the boiler room, all this kind of thing.

None of that's going to happen. What's going to happen is simple, easy-to-use peer-to-peer networks. And there are going to be a whole series of them rolled out over the next few months.

Intel announced, we announced our own technology last week called AnyPoint™, which appropriately enough is sold over the Web. And the idea on this technology, which is very, very simple, easy to use, I'm going to show you in a second, is that you use redundant bandwidth in the telephone line inside your house to be able to send data around. So you can share an Internet connection going out and you also have the LAN within the home just on the existing telephone line, using a different frequency. It's using a technology called TUT technology.

So I'm going to show you now an example of how you would -- you would use this inside your home. And we have two computers here. And you may know, if you're a parent, next week is Bring Your Child To Work Week.

Well, I'm actually being rather cheeky, and I'm bringing it in a week and I've brought my kids down to have a look at the show. And they're actually here.

So this is George and Rachel. And you're going to see -- (Applause.)

SEAN MALONEY: They're going to make sure I pay them for this. Don't worry. (Laughter.)

SEAN MALONEY: I'm sure in your house, you have arguments over who's on the phone. Obviously, like everybody else now, we have arguments about who's on the Web.

So we have got two very normal Pentium III 500 MHz PCs here. (Laughter.)

SEAN MALONEY: And this simple little home LAN that's running on the telephone line. We've actually got a telephone line here and we've plugged it on this AnyPoint thing. We've got two people accessing, sharing the Internet at a single time on a single line to the ISP.

Hi, George. What do you normally do on the net?

GEORGE MALONEY: Well, when I come home from school, I like to check my homework, just in case I forgot to write it down. 'Cause my -- my school has a address on the Internet. So I can just look at that.

SEAN MALONEY: Right. And I can also look at it as well, right? So I can look at your homework wherever I am?

GEORGE MALONEY: Yeah.

SEAN MALONEY: Unless you mess up my browser. (Laughter.)

SEAN MALONEY: Okay. So in addition to that, what kind of other stuff do you play with or look at.

GEORGE MALONEY: Well, in my spare time, I like to look at this one certain site where you can look at specific places from a 360-degree angle and you can just look at the whole place and get a good view of it.

SEAN MALONEY: All right. That's beautiful.

And what about what else do you do? I know you've got a big hobby, which is soccer; right?

GEORGE MALONEY: Yeah. I like to follow my favorite soccer team. It's called Manchester United. It's in England and it's hard to sort of -- hard to speak to them. So -- Well, I can look on the Internet here and I can see, like, the big games, which is usually them, 'cause they're the biggest team. And I can see, like, the scores. Because my team played your team this time.

SEAN MALONEY: That's right. And normally you win.

GEORGE MALONEY: Yeah. 'Cause mine is first; yours is twelfth. And, surprisingly, I was upset about this, it was a 2-2 draw. And the good thing about this, I can just see video clips of the goals that were scored.

SEAN MALONEY: Right.

Okay. That's great. And, again, a big change over five years, as you can probably tell from my hybrid accent, we are British and sort of have been traveling around the world for years. Of course, now you can keep track with any sport any time. That game was only played a couple days ago. It's very real time.

Rachel. So what do you do mainly on the Net?

RACHEL MALONEY: I check my E-mail and listen to CDs before I buy them and look at clothes and stuff.

SEAN MALONEY: And how many of your friends are on E-mail?

RACHEL MALONEY: About 75 percent.

SEAN MALONEY: About 75 percent. Right.

RACHEL MALONEY: Yeah.

SEAN MALONEY: So this is Delia's now. If you have teenaged children or if you're a teenager yourself, you'll know Delia's is a really, really big deal.

RACHEL MALONEY: Yeah.

SEAN MALONEY: It's targeted very much at the youth market.

So what do you have here?

RACHEL MALONEY: A dress that you're going to buy me.

SEAN MALONEY: Uh-huh. (Laughter.)

SEAN MALONEY: Uh-huh. And don't tell me you have your credit card; right?

RACHEL MALONEY: I have the number.

SEAN MALONEY: You have the number. I'm sure you do. (Laughter.)

SEAN MALONEY: Uh-huh. Okay.

How much is that dress?

RACHEL MALONEY: Fifty bucks.

SEAN MALONEY: Fifty bucks. Okay.

RACHEL MALONEY: Plus tax and shipping.

SEAN MALONEY: Plus tax. (Laughter.)

RACHEL MALONEY: You know. And, of course, I'd want to straightaway the next day, so --

SEAN MALONEY: Yeah. Okay. Well, we'll see. I daresay you're going to win this particular argument.

Okay. What else do you have?

RACHEL MALONEY: I want to listen -- I'm going to buy a new CD by Garbage. And I can listen to a song by them.

SEAN MALONEY: Okay. This is garbage's home page; right? Like every other band now they've got a home page.

RACHEL MALONEY: Yeah. (Music playing.)

SEAN MALONEY: Great. Well, thanks very much.

So are you still there? (Applause.)

SEAN MALONEY: What you saw there is an end to the normal woes of, please, will you get off the Web, I've got to get on, I've got to do my homework. No, you're lying, you're just going to get on and play, all those kinds of arguments.

So that's sharing the Internet. And I think you're going to see huge numbers of simple home LANs installed in the next year or two.

One of the fastest growing categories on the Web is games played across the Web. Obviously once you have the LAN in, you can do that. But you can also play games inside the house. And you've probably all seen these arcade games where you have cars all lined up and they're, actually, basically normally PCs on a LAN.

We can do the same thing here. And so what we have here is we're going to have a race. And we're going to have a race between Rachel on the right-hand side, and this is Need For Speed III, and George on the left-hand side. And this is a real grudge match. These two are linked together over a LAN. And you can send little notes to each other. They're already sending notes to each other. They can do chat.

Why don't you get going and have a race.

COMPUTER: One, go.

SEAN MALONEY: Okay. So Rachel is in the lead at the moment, which is pretty good, because she's only a few months away from getting her driving license.

And George is a little bit further ahead.

RACHEL MALONEY: I'm going to go --

SEAN MALONEY: Okay. She's following the airplane. She's letting her free spirits fly.

Okay. Thank you very much. It's a draw. It's a draw. Hey, thanks a bunch. (Applause.)

SEAN MALONEY: Good. Thank you.

Okay. So that's home LANs, a problem that's been around for quite a while. And I think you'll be very impressed by this kind of product and numerous other products in the next few months. I think we're actually making some progress on ease of use on home LANs.

But, of course, if you really, really, really want to be hard-core about the Net, you're going to have it in your kitchen; right? And, of course, we all have access to the Web in our kitchen -- not!

But it would be actually really, really useful if you are a chef or even if you're partly interested in cooking. You've got these bookshelves full of recipes and dozens and dozens of books, and you have a great people you cook one night and you forget which book it's in. Where is it? It's written down on a scrap of paper somewhere.

Now, if you could actually just have access to that information on the Web right in front of you, the recipe and your notes on, that would be incredibly useful. So let's give it a try.

So I'm going to become chef, and I'm going to put on my little fancy Intel apron here. (Laughter.)

SEAN MALONEY: And I'm going to go the whole way, with the hat. And I'm going to bring you into my kitchen.

And, of course, as with any other kitchen, right here I have my chopping block, my cleaver, all these various little tools here, and, of course, the keyboard that's always very, very useful in the kitchen. (Laughter.)

SEAN MALONEY: We all know how useful the keyboard is; right?

Oh, I spilled something on it.

Of course, having a keyboard in a kitchen is a particularly dumb idea.

So I think we'll dispense with the keyboard.

Hey, we just stick with the mouse, right. Because point and click, that's real easy to use. But as I'm a gifted pastry chef, guess what, the little roller is going to clog up and become useless. So guess what, I think we better get rid of the mouse, too.

That leaves me with a quandary, because I've got to access the net and no way I can use a PC with all this cooking stuff around. We have to use something else. And the thing, of course, that chefs do really well is shout at people and bark orders out. So I'm going to do the same thing. And I'm going to go back to our tried and trusted speech technology. And in this case, what we're going to do here is we have Internet Explorer, and on top of Internet Explorer, we have the IBM ViaVoice technology.

And I'm going to use it to demonstrate how you may be using the Web in your kitchen in a while.

Wake up.

Wake up. Take me to weather for Los Angeles. Go to sleep.

So what I've done here is I've zapped off to have a look at a local weather page and find out what the weather's like. That's all very well and good. But what I actually want to do is get involved in cooking.

So wake up.

Take me to Epicurious. Take me to my recipes for drinks. Go to sleep.

Okay. So this one's actually a little more obedient than the other one. And it's taken me off to my recipes for drinks.

Now, obviously, you're going to have customized sites just in the same way in Yahoo! and Excite and the other portals, you know, you have your own areas.

In this instance, where we have a very, very well-known, Epicurious, very well-known cooking site, I have my own area where I can go and install my own recipes.

And we have a choice. This is now audience participation time. And you have to vote here. Do I cook the incredibly healthy Californian-style banana orange smoothy, or do I go straight to the bloody Mary? (Laughter.)

SEAN MALONEY: Okay. Now, you can tell, hands up if I go straight to the bloody Mary. Yeah, I'm going straight to the bloody Mary.

Okay. So here I am connected to the Web. And I want to cook this bloody Mary. And we're going to do it now in one sequence.

Now, pay attention, guys, because this is going to move fairly fast here.

Wake up. Take me to bloody Mary.

Select all.

Begin reading.

COMPUTER BABY: Select drinks, general drinks, bloody Mary, 46 ounces, tomato juice, four ounces, lemon juice, two tablespoons, Worcestershire sauce, salt, freshly ground pepper, lemon, two and a half ounces vodka per serving, 20 ice cubes, ten thin slices lemon.

To prepare a bloody Mary, combine the tomato juice, lemon juice, Worcestershire sauce. (Laughter.)

COMPUTER BABY: Salt, and pepper. Stir and mix slow. Pour into container and refrigerate until thoroughly chilled. Put two ice cubes in each eight-ounce glass. Add (inaudible) two and a half ounces vodka, five ounces tomato juice mixture. Garnish with a sliced --

SEAN MALONEY: Thank you. Actually, you missed -- There's our bloody Mary.

Unfortunately, you may have noticed, there was an error in that demonstration, and we had a very slick little talking head there. And the talking head disappeared off somewhere across the Internet and left its voice.

But I think that speech recognition and speech synthesis and so on are particularly -- are going to get particularly useful over the next couple of years. When you're accessing Internet on the go and you're retrieving your E-mail messages when you're in your car or via some appliance, all of those kind of technologies are going to get more and more useful, more and more generally applicable.

Okay. So let me then just conclude with the three different areas that we think are worthy of special attention over the next few years.

Firstly, information management, developing tools and products to enable us to be prepared for this ever-increasing onslaught of information, making sense of it from a consumer point of view, and also from the point of view of the merchant trying to get their message across.

Secondly, on the infrastructure, building an infrastructure capable of handling extremely high burst information flows and access flows in a secure manner.

And thirdly, the ubiquity of access, getting access to the Internet in many, many different areas so that it can truly become the significant, number-one medium for us getting information and doing business.

I'd like to thank you very much for coming along and listening to us this morning. I think that we are all born to be wired. And on behalf of Intel and the organizers, I'm looking forward to us working together over the next few years in this incredible experience.

Thank you very much. (Applause.)

* Other names and brands may be claimed as the property of others.