SuperComm 2002

Craig Barrett
Atlanta, Ga., USA
June 4, 2002

Download presentation (10.3 MB Zip file)

ANNOUNCER: Ladies and gentlemen, please welcome Walter McCormick, president and CEO of the United States Telecom Association.

(Applause.)

WALTER McCORMICK: Thank you very much. It's my privilege and honor to welcome all of you to SUPERCOMM 2002. This annual gathering is a critical one for our industry. It is the world's premier communications and information technology exhibition, and it is a place where we come together, share our views, share new products and share knowledge.

Over the next few days, we will have the opportunity to glimpse the future in leading edge product exhibitions, to discuss and debate mission critical issues in an extraordinarily diverse array of forums; and we will hear from leading visionaries.

Our keynote speaker this morning will get us off to a great start. The year was 1974. Roberta Flack was on the radio. Chinatown was in the movie theaters. Watergate dominated the national news. And a young associate professor at Stanford University took a job as a technology development manager at a company called Intel.

The rest, as they say, is history. Craig Barrett rose to become Intel Corporation's fourth president, and in 1998 its CEO.

It is due largely to his hands-on leadership that today more than 80 percent of the world's PCs have Intel inside.

As an advocate for the telecommunications industry, as one who spends his time in Washington working with Congress, the Department of Congress, the White House and the FCC, I am especially grateful that Craig Barrett is a leader among the telecommunications industry executives who are so outspoken about the urgent need for a national broadband policy. This is a mission that we all share.

And because of his leadership and that of many other individuals in this room today, this debate is finally beginning to occur.

The telecommunications industry has long been defined by technological innovation. So as we look to the future, we naturally ask the question, "What's next?"

Our speaker this morning will address what many believe will be the new evolution. So please welcome our keynote speaker, CEO of Intel, Craig Barrett.

(Applause.)

(Video plays and concludes.)

ANNOUNCER: Ladies and gentlemen, please welcome Craig Barrett.

(Applause.)

CRAIG BARRETT: I hope you don't mind that little spoof on "Men in Black." We'll see a bit more of them later on. It just seemed to me that the industry needed a little bit of levity, perhaps a new application that's consumer friendly for transporting and to get it out of its current doldrums.

I want to talk about the telecom industry today, and I have a very simple message for you that I think we're seeing a couple of very significant issues facing the industry. Clearly, a little bit of over-investment, a little bit of down economy at the current time, but offsetting that is the convergence at last between computing and communications. And associated with that are two very fundamental issues that we should all be aware of and take advantage of.

One is that convergence is going to lower the cost of operating the telecom space and give operators, equipment manufacturers, an opportunity for profitability. But more important, I think that convergence of the two industries together will give a lot of exciting applications that will hopefully be revenue generators for the industry as a whole.

I want to talk a little bit about perspective of the industry, how we got where we are today, what's happening, what the future looks like, what the network is going to look like as we go forward, what some of the inhibitors to growth are, and then at the end we'll talk a little bit about innovation in the industry and what sort of value-add applications we can bring to really revitalize the industry going forward.

If you look at the industry today, I don't think it's much different than the history of technology over the last several hundred years. Every time a new technological innovation has come forward, there's been the same sort of reception of that technology and the same sort of trend during its growth.

If you go back and look at canal building in Western Europe in the 1700s, if you look at the advent of the steam locomotive and the growth of the railroad industry, if you look at the growth of the steel industry, if you look at the electrification of the United States and the rest of the world, look at the automobile industry, they all have the same trend.

The technology gets introduced, there's a rapid growth and acceptance of the technology, there's something that Alan Greenspan would refer to as irrational exuberance, there's over investment of capital, there's an over inflation of stock price, there's an excess number of competitors in the industry.

There's a period of turbulence as the industry readjusts in terms of its spending its investment, and its capitalization.

Then beyond that period of turbulence there's a period of extended growth as the technology builds out around the world, it's accepted by all of the nations on the earth, all the populous, and we see the best part of the industry going forward.

If I look at the information technology industry today, I think it's in this period of turbulence, this period of overinvestment, overhype, overcapacity and rationalization, as we've talked today, whether we're talking in the telecom space, we're talking the dot-com era that we went through in the late '90s, early part of this decade.

This turbulence is going to pan out. We will have value added competitors going forward with value-add to the consumer and strong business models. The issue is who will be the provider of those applications and who will be the winners.

I'm not going to pick winners today, but what I want to talk about is, in fact, the opportunity going forward that the winners can take advantage of.

So how did we get where we are? Well, we've had an over build-out in terms of overinvestment. Too many competitors in too many fields. That leads to excessive price erosion, leads to profitability erosion, leads to uncertainty in the stock market, leads to market capitalization declines, leads to industry consolidation.

We're seeing that industry consolidation across the board, whether you're a local exchange carrier, whether you're a long haul carrier, whether you're a wireless carrier, whether you're an Internet service provider. Whoever you are, we're seeing that consolidation in the form of bankruptcies, mergers, acquisitions, or people just plain going out of business.

A simple way to look at this is we're seeing 10 years of consolidation under normal conditions take place in about 12 months. It creates a great dislocation in the industry, big challenge for any of the surviving players in the industry.

Now, how do you look at that change that takes place? One way to look at it is just to look at the equipment suppliers who supply the basic infrastructure, the backbone of the telecom industry. And we've listed a few of these infrastructure providers here, and basically looked at two aspects of them as indicative of the change taking place: head count and R&D spending.

You can see in the space of about 12 or 18 months that dramatic change has taken place. Roughly, across the board, if you add up these seven companies, you get something like a 50 percent reduction in headcount, and about a 50 percent reduction in R&D spending.

This is indicative of the transformation that's taking place in the industry, as the industry goes from a really proprietary industry where every company has their own set of standards to more modular building block, open interface, standard building block, more horizontal industry, somewhat like the computer industry is today.

This is a transformation that is about 10 years of change, 10 years of consolidation sandwiched into about a 12-month period.

The tremendous revenue decline for the companies shown on this slide has been mirrored up and down the food chain. If you go to the service providers, they're seeing the same sort of impact. If you go to the component suppliers who supply this industry, people like the semiconductor industry, you see the same sort of 30 or 40 percent reduction in revenue over a 12-month period.

Phenomenal dislocation in the industry, but it is going to provide an industry that moves faster, gives greater opportunity going forward.

So we've kind of created an interesting mess, if you will, in the industry. I don't have a better word to talk about it than "mess." We have people who are competing for the same customers. Long haul providers want the local access, local access wants long haul or long distance, cable wants both, wireless wants both. So everyone is competing for the same customer. It's almost a zero sum game.

The real problem here is we haven't seen new applications that provide new revenue streams that provide growth opportunities for the industry as a whole. This is part of the consolidation we have ongoing.

I don't want to spend a lot of time today talking about this consolidation. The over expenditure of wireless spectrum, the hundreds of billions of dollars that have been spent on 3G spectrum around the world without a single consumer desirable application associated with those expenditures, the slow rollout of that technology, the lack of real broadband here in the United States and in other places around the world. And by "real broadband," when I talk about broadband today, I'll be talking not standard DSL capability but I'll be talking let's have a minimum of 5 or 10 meg, and more desirably, a hundred meg to every home and small business around the world. That capability.

The fact that we're stuck with really flat text, a representation of the Internet, rather than rich media. Non-interoperable devices. We've had a period of great competition between handset developers, PC developers, without the realization that we need those devices to be interoperable with one Internet going forward.

We've really had a lack of exciting customer applications, things that people are willing to pay for. And that is, if I could name it, the one big issue that faces the industry today: getting value-added applications and services that people are willing to pay for. And that's what we'll come back to a little bit later on.

So we have a Catch 22 situation. We've made a lot of investments. Over a trillion dollars invested around the world in infrastructure in the last few years. Unfortunately, it's not particularly profitable.

So what happens is we look for opportunity for growth, and every opportunity we have for growth takes a new investment. But unfortunately, we've already made the investments. We're worried about servicing the debt, the huge debt that saddles this industry. So we're not willing to make new investments to generate that growth, so we stop investing. And this is kind of where the industry is today.

If you look at one thing that the industry is interested in today, it's servicing its debt or return on invested capital, whatever financial term you want. This is, I think, the biggest challenge we have.

If I have a simple message for you today in the industry as a whole, it's really we have to find a way to get greater capability out of the system with modest investment for a bigger return. We're not going to get a bigger return for zero investment, but there has to be a way to make the industry more efficient, more economic, and bring greater value to the end user. That's the challenge we all face. That's what we need to look at going forward.

So this concept of the way we've done business in the past I think is dead. We're going to see consolidations, we're going to see bankruptcies, we're going to see rationalization of the investment. We're going to see some survivors out of that, but the new structure of the industry has to be different than the old.

It has to be more cost effective, it has to be able to introduce capability and technology faster than the old industry. It has to be interoperable. There's only one Internet. It has to be accessible from handheld devices, laptop PCs, desktop PCs. And we have to be able to port applications and technology from device to device in a scalable fashion in the easiest possible way.

This is the challenge that we have.

I think the opportunity going forward is phenomenal. Perhaps brighter than it's ever been. But you have to look out one, two or three years to see that opportunity.

The reason I think it's bright is if you go around the world and talk to government leaders or business leaders, and each year I get to visit about 30 countries and talk to those leaders, and I ask them what is the future of their economy, what are they looking forward to next year and the year after, and they all say exactly the same thing. "We're not going to have an economy based on natural resources. We're not going to have an economy based on low labor rates. We're going to have an economy based on knowledge. People's capability. A knowledge-based industry. A knowledge-based economy."

And what do you need for that? You need an information technology infrastructure -- computing, telecommunications infrastructure.

So if I look at the build-out of this technology and capability around the world in the developing countries, the opportunity is phenomenal.

Then if I come back and look at the developed economies like the United States, Western Europe and Japan and see what they have to do to be competitive with these developing economies, they have to invest even more in their IT infrastructure to stay one step ahead in the value-add game.

So whether you're in a developed economy or in a developing economy, I see exactly the same thing taking place. Knowledge-based industry is going to be the driving force, and to be successful, to be competitive, to be efficient, you need to invest in the IT infrastructure, and that's investing in communications and computing, which increasingly merge together as one.

So this is where we're going. That's why I think the opportunity is phenomenal. The world's economies are dependent on this. And connecting anyone, everyone, anyplace, anytime is key to that, whether you're connecting through a two-inch screen, a wireless laptop computer, a desktop computer. However you're connecting to the world's economy, this is key.

This is a technology that we collectively provide. This is the opportunity that we have. This is a capability we bring forward.

But what does that bring? Well, if you look at the economies of the world and you say they are going to be knowledge-based, and kind of another way of saying knowledge-based is service-based, and you look at the infrastructure build-out in terms of those economies that have a high fraction of their GDP as service-based today - the United States, Western Europe kind of lead the way in that area, and then you look at the infrastructure build-out that's necessary to be efficient, because basically services are interaction-based transactions, and interaction-based transactions are more competitive, more efficient, more productive if they're automated. The capability there says, well, the United States, which has the highest GDP that is service-based today, has the highest number of PCs per thousand people, has the highest number of Internet users, one of the highest phone concentrations around.

Everyone else is going to have to move in that direction. If you look at the opportunity in countries like China, India, the rest of Asia, Latin America or Eastern Europe, you see exactly that same sort of build-out happening today. That's an opportunity for all of us to participate in that build-out of that capability, whether it's infrastructure or services.

But the fact that the world is going to go to more of a service-based economy and the infrastructure build-out has to accommodate that I think is key to all of us moving forward.

Now, what's unique about this IT build-out, this capability? The Internet is really it, if you will. And the Internet does four things for us. It allows us to do commerce. There's perhaps about $600 billion worth of commerce done on the Internet done today. In a few years, that will rise to over $5 or $6 trillion worth of commerce. Why do you use the Internet to do commerce? It's efficient, it's 24 by 7, it's productive, it's competitive.

If you look at communications, the Internet is really it for communications. Whether you're talking about voice over IP, whether you're talking about rich digital media going forward, the Internet is it for communication.

If you look at information, information access, the Internet is it. Today, for the first time in our history we can access all the information in the world just sitting at a terminal, keyboard, and a seat belt connection. Never before in history has that been available. That's here today. And the Internet is it for sharing rich digital media.

Each one of these as a standalone is kind of interesting in its own right, but if you look at them combined, and that's exactly what they are, is the combined capability of commerce, information access, communication, and rich digital media, that's the trend for the industry as a whole. That's the trend for the Internet.

You want to have these not as separate standalone capabilities. But as integrated capabilities so that the half a billion Internet users today, going to well over a billion users in a few years, the half a billion or so devices able to access the Internet today going to a few billion in a few years, you can use those devices to carry out these four functions in an efficient, effective way.

That's the future of the industry as a whole. That's why I'm optimistic about it, because it is going to grow around the world, and we're going to not stay stagnant with one set of capabilities. We're going to integrate these four separate capabilities, make us all productive, more efficient, bring more value to us all as users or consumers of the technology.

Now, despite the gloom in my industry, the semiconductor industry, and the gloom in the telecom industry, if you look at the Internet and the growth of the Internet, it hasn't recognized the slowdown at all. And perhaps the best way to look at that is just the movement of data along the backbone of the Internet over time.

This shows just over the last couple of years with a forecast for 2002 showing the strong growth in data movement, which is really the backbone of the Internet. And that's nice to look at that. The issue is will that continue to grow going forward. I think there's every anticipation that it will.

It's really this massive movement of information, massive movement of data that is the exciting part of our industry. This is the growth part, and this is what we need to attach ourselves to.

So if you just fast forward from where we are today to what we look like over the next three or four years, I think you're going to continue to see this exponential growth.

It's driven by two factors. One is there's a ton of glass fiber in the world, various estimates of how much of it is lit, 5 or 10 percent, but it's there. To light it, we need a lot of electronic infrastructure. And we also get more clever at how we send photons down those glass fibers every day, so we can kind of double the number of photons or double the bandwidth of a fiber every 10 months.

We know we can double the compute power every 18 months or so, the stuff you attach to the end of that. We know that the electronics capabilities are coming along to continue to decrease the cost of lighting that fiber and then manipulating, massaging the data that goes from one end of the fiber to the other.

So all of that efficiency and capability continues to come forward, and that's going to allow all of those billions of users of the Internet to be more productive, to be in communication with each other, and allow every company around the world to use this technology and its commerce will be more efficient and more capable.

So I think this exponential growth is going to continue with this backbone. And then all the services and devices that attach to this backbone are our opportunity to move forward, to bring greater value to the consumer, and to be successful in the marketplace.

Now, the success here is also associated with the merging of computing and communications. I've been in the industry about 30 years from the computing side, and I've heard discussions during that entire period about how we're going to merge computing and communications.

If you went back and looked in about 20 years ago, NEC had the corporate motto of the merging of computing and communications.

People have been talking about it. "Next year," for about 15 years, was the year of ISDN, digital transmission of information over the telephone system. Took a little longer for ISDN to come around than most of us anticipated. It took a little bit longer for these technologies to merge. But we're starting to see it today. We're starting to see deep integration, both technologies, both capabilities, and we're starting to see the industries come together, both from the infrastructure standpoint and from a capabilities standpoint.

Both industries need each other to move forward. Both industries need each other to bring more value to the consumer, which is what the game is all about.

So how is this going to happen? I think it's going to happen by modular building blocks, open interfaces, and people on both sides of both industries being able to innovate on the system as a whole. And that's what really has to happen. You have to have open interfaces to allow the whole industry to innovate on an infrastructure and not let individual companies try to do it themselves.

If you want to think of a simple model of this, it's the personal computer industry with its open interfaces and allows anybody to write applications on a Microsoft interface, is a way that we bring the end users more value on a daily basis.

But that open interface concept is what this industry as a whole needs such that the service providers can have the entire industry innovating to bring new capabilities, new value-add capabilities, to their consumers.

I could talk about this merging of computing and communications. Perhaps the best way to show you that is to give you a simple example of what applications look like on a merged platform. So we'll do some communication applications on a computer platform for you.

I'd like to invite James Mi who is the vice president of marketing and business development for Itelco to come out. We're going to do just a simple little demonstration of some of their technology showing this deep integration of computing and communication.

JAMES: Hi, Craig.

CRAIG BARRETT: Welcome.

(Applause.)

JAMES: Thank you very much. What I'm demonstrating here today is carrier grade conferencing and collaboration product. And it's designed specifically for service providers to offer an integrated service for audio conferencing, video conferencing, as well as data conferencing all together in a single IP network. So it has a lot of unique features. I'm just going to walk through some of the highlights.

CRAIG BARRETT: Great.

JAMES: The screen to my right is showing the Web interface for the chairperson of the conference. And if you turn to the screen to my left, it's showing a slightly different interface for the participants of the conference.

To join the audio portion of the conference, people can just use a regular telephone or can even use a PC. And also, the chairperson of the conference can say in the middle of the conference, "We'd like to invite some people on the fly," just by pulling some numbers from online address book, and the system will place calls starting out to the invited people. I'll show you. I'm being invited here as well.

TODD: Hello, this is Todd.

JAMES: Hi, Todd. He is somewhere in the audience. I really can't see him here. But it's just as simple as that. If you have 10 people you need to invite, it's the same single mouse click.

To join the video portion of the conference, all you need is a very simple PC-based video camera like this one. And the different video images can be controlled by the chairperson or the participant as we are toggling through different images. This is a live video conference. We also have people joining from the booths, at Intel booths. We also have a mechanism to automatically detect the speaker. We're seeing David here.

If people are talking on the conference, the system also can detect the speaker and automatically switch the video image to that person who is speaking.

CRAIG BARRETT: I hope you have the capability to silence people you don't want speaking at any time.

JAMES: That's a very handy feature to have. We do have mute capability, put people on hold, you can digitally record it, all different things you can do.

Another important feature for the product is data conferencing aspects. What I mean is let's say we have a conference to talk about the network infrastructure for a customer. And David would like to show some of the network diagram for that project, so he is going to start an application, in this case, it's a Visio application on his desktop. And over here, as a participant of the conference, I'm sitting here at the other end of the network. Remotely, I can see the same application, and also I can request for control and make necessary modifications remotely over the Internet.

So this is really leveraging the true capability enabled by the convergence of the communication and the computing.

Traditionally, the audio conferencing and video conferencing are all done using separate DSP-based hardware systems. But what's unique about this Itelco product is everything is done in software. It's running on standard-based compact PCI system, like Intel NetStructure product. And this give us tremendous advantage to add new features, for example, supporting wireless cell phones and PDA devices. We can bring these new capabilities very quickly to the customers with a very simple software upgrade, and also at a much lower cost as well.

CRAIG BARRETT: So standard building blocks, open interfaces, software upgrades, software applications built on standard building blocks.

JAMES: What a tremendous advantage to have.

CRAIG BARRETT: Where can I buy some of it today?

JAMES: You can.

(Laughter.)

JAMES: We also show you the product at the Intel booths. Come visit us, and just bring the P.O. as well.

(Laughter.)

CRAIG BARRETT: People are rolling this out today?

JAMES: Yes. Actually, NTT Worldwide Communications, and also KDDI America, they all have rolled out the commercial service based on this product recently, and they have seen very strong revenue generation based on the new services.

CRAIG BARRETT: Great. Thanks, James.

JAMES: Thank you very much.

(Applause.)

CRAIG BARRETT: I think that's a good example of the merging of the two. That's the standard compute building blocks and software applications on that applied to the communication space. I think that's what you're going to see increasingly going forward.

There are lots of other applications you can look at if you're looking either from the computing side or the communication side.

Looking from the computing side, obviously the next wave of laptops are all going to be built out with wireless capability, whether it's a wide-area network type of 3G capability or a local area network 802.11 type of capability, but built in. So the communication is built into the computing device as you go forward.

If you look at voice and data handsets, increasingly they are computers with wireless capability. The same sort of compute paradigm as you see in computers today, but built into standard communication devices.

Software programmable switches, media gateways, and increasingly as we go forward, with the transistor budget that we have, increasingly you're going to be able to put radio capability or interface capability for any communication protocol you want on essentially every microprocessor that's produced. Whether you want DSL capability, whether you want cable capability, whether you want wireless capability, whether you want gigabit Ethernet capability. The transistors budget to put the MAC and the PHY for those four separate technologies that I just mentioned occupy today about 10 percent of the transistor budget of a state of the art microprocessor.

As the manufacturing technology continues to increase, that transistor budget becomes even a lower and lower fraction of the total, so that you're almost going to get that communication capability built into all standard semiconductor, microprocessor type devices for almost free going forward.

So that deep convergence of computing and communication I think is a critical feature we all have to be aware of and have to take advantage of going forward.

Well, this obviously gives you opportunity across the entirety of the telecommunications system. Whether you're talking about long haul, the fiber that's in the ground and lighting that fiber, building it out with electronic capability, electronic optical interface, whether you're looking at the metro build-out which is increasingly going to be based on optical capabilities, it has to be more cost competitive to get its build-out.

As we start to worry about getting fiber to every business or, in fact, fiber to every home, it's the cost per connection that gets to be exceedingly important. That cost continues to come down with electronic integration and integration at the electro-optical interface capability.

DSL build-out, digital messaging sent over twisted copper is increasingly important as a build-out, not here just in the United States but around the world. And wireless capability, whether we're talking about wide area network wireless capability, whether it's street 3G or GPRS or local area network capability with 802.11a, b, g. You pick whichever one of those you want. But that capability becomes increasingly important and becomes increasingly economic, cost effective, with the level of integration we can provide today.

So this cost competitiveness that starts to come into this space allows us to bring greater capability to the end user, and that greater capability translates into more exciting applications the end user is willing to pay for.

So it's a bright future. Unfortunately, there are a bunch of clouds in the sky at this point in time.

The economy is down in the United States and Western Europe. The economy is down in Japan. The economy is growing and vibrant in most of the emerging places, but still the developed economies around the world constitute the bulk of the demand for our products.

But going forward, those established economies have to continue to invest in this capability to be competitive, to be efficient, and to go forward. That's why I'm excited about the future.

But in the meantime, what happens? Well, in the meantime we all worry about return on invested capital. We all worry about profit after tax, or if we don't worry about profit after tax, we worry about EBIDA, we worry about cash flow.

And that's the great dichotomy, is what do you do to worry about ROIC and also make investments for the future? How do you bridge that gap? How do you make investments to get new revenue, new growth opportunities at the point in time when the market is hammering on you to get the maximum return out of your invested capital today?

The only way I know to do that, frankly, is to focus on the customer. I mean, you can look at the various layers of the industry, whether you're a service provider, equipment manufacturer, maker of software, if you assume you're going to communicate over the Internet with IP, but then if you look at the hardware, the protocols that you're going to use, the software and the usage model, what the industry has to do is focus on what does the customer want.

Too long, our industry has been focusing on blind capability as kind of the Field of Dreams: Give them bandwidth, and they will come. Give them 3G, they will come. Give them digital capability over a handset and they will come.

There are clearly models that are very efficient, models where the consumer leaps at the new capability.

All of you probably have short messaging capability in your handset and in your Blackberry type device you carry around with you, whether it's a Blackberry Palm, whatever it is. That's an immediate application that we all have demand for, we're all willing to pay for. That's revenue generation, profit generation, return on invested capital.

I think if the industry has to do anything going forward, it's focusing on the consumer, focusing on the consumer's usage models, what the consumer wants. What value-added services can we bring there. Too much, we've been focusing just on the infrastructure.

Now, what are some of the opportunities we have here?

Broadband. Great capability. You've all read the papers. You know the things you can do with distance learning, rich media transmission. We know how many users Napster had before it was essentially put out of business. We know how many users the follow-ons to Napster have.

There is a huge customer demand for broadband transmission in the delivery of rich digital content. This is a capability that we not only need for consumers; this is a capability we need in the enterprise as increasingly companies like my own do all of their business over the Internet. Not just with our major customers like Dell, IBM, Hewlett Packard, et cetera, but with 60,000 or 70,000 small distributors of our technology around the world. We don't get in front of those people with a human-to-human interaction. We get in front of them with an Internet transaction.

That Internet transaction needs to be rich in terms of content and capability. You do that with broadband.

So broadband build-out is tremendously important. The barriers to it are obvious. The U.S. government hasn't responded with a broadband initiative for the country as a whole, for its competitive future, its economic vitality of the future. I think that has to happen, should happen going forward.

We need more than probably just a mandate by the government that this is important. What we need are regulatory reforms to allow people who invest in this capability to get an adequate return on their investment.

We need to have the people who own content to have business models which can deliver that content in a rational fashion to the end user, assuming that most people are willing to pay for that content, be law abiding citizens and not try to rip it off the Internet for free.

And it needs to be real broadband content. I get to travel around the world and see what real broadband content looks like compared to some of the things we try to pass off for broadband in the United States. 300Kbps or 400Kbps is not real broadband. When you get to 5 or 10 meg, going to a hundred, that's real broadband, and the capabilities you can provide the user benefit, the value to the end user gets to be phenomenal.

What I want to do again is not just talk about broadband, but let's give a demonstration of what some relatively local area broadband capability looks like, but we'll do it from a wireless standpoint.

What I'm going to do is have someone come out from a company that provides this capability, Navini Networks, and Sai Subramainian is going to come out and give a little demonstration of what we can do in delivering broadband capability over a several mile distance for a very customer friendly, ease of use capability. So Sai, why don't you come out and see what we can do with this little demo. Welcome.

SAI: Hi, Craig.

(Applause.)

SAI: Thank you. Now, Craig, what I'd like to show you is a whole new way to deliver wireless broadband.

You know, everybody wants broadband, but it's really difficult to get real broadband. Many service providers find that the delivery systems for broadband access are complicated, expensive, and difficult to install and maintain.

However, there's a new way, that's the Navini way. And with our system, installing broadband is a snap. In fact, the Navini solution provides nomadic anytime, anywhere access to broadband that's really easy to install.

There are no truck rolls, no lines to provision, no antennas mounted on the side of your house or business. Just plug it in and go.

CRAIG BARRETT: Can't believe it. Can you get this to my ranch in Montana?

SAI: Absolutely, absolutely.

CRAIG BARRETT: My ranch in Montana is out there some distance, now.

SAI: We can do that. Service works through windows and walls.

CRAIG BARRETT: How about pine trees?

SAI: Pine trees, through the cows you might have on the ranch. It doesn't really matter. You know, the transmitting tower could be as much as five miles away, and it would work just fine. The range is similar to a PCS type network, so all you would have to do is install a tower and it would work.

CRAIG BARRETT: As long as the tower looks like a pine tree, you've got a deal.

SAI: Exactly. There you go.

CRAIG BARRETT: All right. Let's see what this can do.

SAI: David here is going to show our zero install technology in action.

DAVID: Thanks, Sai. So to connect to the network is as easy as connecting this small Navini antenna over a USB or Ethernet cable to your PC or notebook, and you have instant access to the Internet, and of course instant access to rich media data types out there.

So what I'm going to do is show you a quick example of that, and if we take a look at the screen here, we're going to take a look at a Shock Wave 3D. It's very photo realistic, very interactive.

What we did was we used a digital still camera and a suite of 3-D applications, tools from Real Biz, and we went on a trip with you to Taipei and created this.

So this is what is done very quickly and very easily. You can see all this rich media easily, and deployed over broadband, of course, and there will be tons of content available.

CRAIG BARRETT: David, this is a real live demo. You have an antenna someplace that is beaming down here, right?

DAVID: Correct. The antenna is out there in the ether somewhere beaming the stuff over here. And what I'm going to do is actually download one of my other favorite examples of rich media. And of course we had a little fun. We took a couple of pictures of you while we were there in Taipei to create this wonderful little example that we're going to stream over this little broadband connection that we have. So the highlight of that is it's all being streamed wirelessly to my notebook right here.

CRAIG BARRETT: Just to give you an idea how you create that, show you what digital content creation, all they did was take several photographs of my face, and then they superimposed that in different directions on that little wire frame body, and I can do just about anything, right?

DAVID: Anything you want. But that was almost real. I do that in real life, right?

CRAIG BARRETT: Right. What else can you show us with this?

DAVID: Well, the solution is great for, like, the home and office. But what's truly amazing is their Nomadic solution. And what they can do is with a battery operated antenna, they can attach this to a laptop and go anywhere and have access to the Internet.

They can even attach it to a PDA, and they have another solution that fits into the sleeve of your handheld device, and you can have access to any media that you want. Is there any preference that you want to see, Craig? Anything you want to see?

CRAIG BARRETT: Well, I'll trust you to your own devices now, I think, after that last one.

DAVID: Since you liked that video so much before, we'll going to just stream that over and play that again here. There it is on iPAQ. Just thought everybody might like to see that again. So what you saw is that we have wireless broadband, so we have access to rich media data types from anywhere and anytime.

SAI: That's cool. Thanks, David. As you pointed out, the future is supposed to be broadband and wireless. Well, with the Navini solution, we've combined the two, and it is ready to go today.

CRAIG BARRETT: Where can I buy it?

SAI: Oh, our entire sales team is here. They'll be happy to take your P.O. We're actually in the early commercial deployment and in trials with a number of carriers, including larger ILECs and IXEs such as Sprint.

CRAIG BARRETT: All right. Just remember Montana, okay?

SAI: Absolutely.

CRAIG BARRETT: All right. Thanks very much.

SAI: Thank you.

(Applause.)

CRAIG BARRETT: I think if you really believe the whole issue of broadband is exciting, and there's a huge opportunity not only for application delivery but for industrial competitiveness, and I really think that the United States needs to get its act together in this space, as all the other developed countries around the world need to get their act together, with national priorities and national infrastructure build-out. Not necessarily by financing it, but by at least getting the regulatory issues resolved so that people who do make that investment can get a return on their investment.

If you look at digital media, which is one of the unique applications for this capability, I think you see some excitement there as well. There's a lot of discussion in the industry today about digital content management and about protection of copyright laws. Napster probably led this discussion. It's now being taken up by the Motion Picture Association and other people in Hollywood.

I think there are four very clear fundamentals that have to happen here. One is we do need technical solutions to protect copyrighted material as it's transmitted around the world.

Obviously if you make one digital content, one digital copy, and put that in the clear, you can make an unlimited number of pure copies from that without any loss of resolution capability. So we need technical solutions to help protect digital content.

We also need to protect the consumer in their fair use rights that they've grown to expect with tape media or CD media in the past.

So articulating and documenting what the fair use rights are for the consumer at the same time you put technical protection in to the system, I think are appropriate.

Third thing we need are usage models or commercial models for the distribution much digital content, digital media, which meet the consumers' needs at the prices consumers are willing to pay for.

So what we need to do is have business models from the content owners who will then provide their content over the Internet in the way that consumers want them.

For example, consumers may want to buy one song at a time and not an entire CD at a time. But having that capability I think is important.

Then you need the government, obviously, to protect against massive fraud or misuse of copyrighted material.

Fundamentally, copyrights should be protected. The artists should be paid for their content. But you need to have business models that are satisfactory to the consumers as well.

So this four principle model I think is important to getting digital media on the Internet, and combine that with broadband for exciting new applications and new business models that will benefit not only the consumer but all of us who put infrastructure in place.

Cost effective infrastructure. The three layers of the distribution tier I think are important, whether you're looking at the enterprise or home or we're moving information around at the local area, local network, or, in fact, we get to the access or the edge that local area network starts to interface with the wide area network, or then we look at the metro or long haul capability.

All three of these areas can benefit from the capability that the integration of computing and communications is bringing forward. More cost effective capability, faster time to market, easier proliferation around the world.

And this is where the whole issue of standards comes into play, modular building blocks, open interfaces. That's the most cost effective way to put this infrastructure in place, the most cost effective way to be able to interface between one protocol and another protocol.

That's what the transistor budgets of today's microelectronics capabilities are going to allow us to do very simply, very easily going forward.

So what we end up, then, is, in fact, a value added capability where we bring the end user more capability, bring the end user more applications, bring the end user more value and give the service provider and the infrastructure provider more return on their invested infrastructure.

Now, what does this look like going forward? What's the structure of the industry going to be in this period of turbulence that we have today?

Well, it's going to go from a vertical model to more of a horizontal model. Horizontal model is indicative of standard building blocks and open interfaces.

If you look at the computer industry 20 years ago, it looked very much like the communication industry looks today. Vertical. Vertical means proprietary. Proprietary means proprietary hardware, proprietary software, proprietary interfaces.

The computer industry transformed itself to a horizontal industry. If we had counted on vertical mainframe industry, mainframe computer industry, proliferating to the home or the small business, it would have never happened. It would have only happened if I dumped terminals where you didn't give value to the end user.

What we want to do is provide the end user value, so we want to give the end user capability to put the applications in play they want to put in play, and have the compute and communication capability that they need and want at their disposal. And that happens if you take the industry from vertical to horizontal, open interfaces, open capability, so the whole industry can innovate around those open interfaces.

And if the whole industry can innovate, then what do you is you move technology more rapidly into the industry and give the end user more value and provide more value back to the service provider.

So value at all layers: hardware, operating systems, applications, value-added services on top of that. That's the direction that the industry is going to move through this period of turbulence. It's a much more cost effective capability and a way to bring the end user value more rapidly.

How do people like Intel play into this game? Well, we play into it by, in fact, some of the exciting new technology we bring into the space. Ninety nanometer may not mean anything to you. It's really the size of the next generation of transistors we produce, but the transistor budget you get when you go to that level of technology allows you to integrate all sorts of communication capability into standard electronic semiconductor components.

If you start to integrate microelectronic module type capabilities, put antennas, passive devices, integrate those with active transistors or electronic devices, then you can start to build whole radios, integrate radios into your standard semiconductor devices.

So those capabilities are happening.

Integrating memory or Flash memory capability with logic capability. If you look at a cell phone today, probably the two most expensive components are the compute component and the memory component. Why not integrate those together to get higher performance, lower power capability, more value to the end user? That's happening.

And if you look at the proliferation of optical devices, optical transponders, the termination between fiber and electronic signals has been very expensive in the past.

Today, as we integrate that capability more and more into one or two components, shrink it down and decrease the cost of connection, this is going to proliferate fiber to the small business and event actually to the home in a cost effective, economic fashion.

This is what silicon technology brings to the game, and this is why I'm excited about the future of both of our industries. I think we're bringing value to the game, and you can take that value and provide that to the end user and consumer.

This also allows you to, in fact, worry about standards. Standards, whether they be hardware platform standards, whether they be communication protocol standards, whether they be standard application packages for handheld devices. A whole array of standards, much similar to the compute industry as we know it today, but bringing that capability to the handset industry in the communication space, to the back planes of the communication space where we have standard form factors and standard interfaces attached together. Lowers the cost of operation, brings greater capability to the end user.

I want to show you what you can do with standards in this space by giving you a very simple example of how we can bring rich media content over standard platforms to the end user. And the example I want to do is, in fact, show you how you bring high definition TV over an Ethernet connection to an end user.

Diana is going to come out and give a simple example. And perhaps I'm dressed appropriately for this because one of the things we want to show you is a trailer for the new movie "Men in Black II" streaming over the Internet.

DIANA: Thanks, Craig. I'd like to show you an example of the exciting video-on-demand applications that are now available with robust broadband and a powerful PC. We have here a carrier class Xeon(r) server. It's going to be acting as our content provider and streaming high definition video over gigabit Ethernet and then down to 100 megabit for the home.

So representing the home, we have a 3 GHz Pentium(r) 4 processor-based PC and it's running software from Access TV that allows us to code this high definition video in software.

So I actually have a clip here from a film featuring Craig's alter ego, so let's take a look at it and see how it looks.

(Playing clip from "Men in Black II.")

DIANA: So like I said, that's a high definition video stream, and it's being streamed on this PC all in software, no hardware acceleration needed.

But I know you're not satisfied with just a great demo, so we decided to go one more. You can see here we actually have two high definition video streams coming to this Pentium 4 processor-based PC, being decoded and displayed completely in software.

CRAIG BARRETT: Great. I think the other one is the movie Spiderman, right?

DIANA: Yes, Spiderman.

CRAIG BARRETT: I decided not to come dressed as Spiderman this morning for obvious reasons.

(Laughter.)

DIANA: So as you can see, this is a very powerful solution. And I should mention it's very easy to develop using standards-based building blocks.

CRAIG BARRETT: Thanks, Diana. That application takes about 20 meg per bit of streaming media for each one of those clips. I guess we could have done three or four for you. But if you bring real high capability broadband to the end user, that's the sort of features you can get.

And you can apply that not just to this sort of entertainment streaming media, but apply it medical technology streaming over the Internet, education streaming over the Internet, what have you.

How do we like to interface with this industry going forward? Well, it's all about bringing value to the consumer. It's about bringing standard building blocks, which we're about, interfacing with people like yourselves who bring systems for the infrastructure or solutions for the end user.

So the interface we see is at various levels in the food chain, but ultimately, it's bringing more capability to the end user and more business for each of us.

So working with industry partners, as I've shown here, we like to work obviously at the silicon level and at various other levels within the value-add food chain, but ultimately it's bringing service capability, value to the end user.

Let me just summarize very simply. The industry is in transition. It's a difficult time. We're going to see consolidation.

The real issue, though, is what investment brings really the reinvention of the industry, the redefinition of the industry. And I think that that redefinition is going to be standard building blocks, open interfaces, and more value to the consumer. That's what we all have to work toward, that's what we're interested in cooperating with you in the next generation.

Thank you for your time this morning.

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