Bill's speech at Lakeside High-School 1995

Below is a transcript of his presentation, preceded by Head of School Dr.Terry Macaluso's introduction of Bill.

Terry: It is a somewhat daunting task to try to prepare an introduction for Bill Gates. It seemed as though we would be unable to say anything that had not already been said and then we ran across this document. Now, I need to precede with a caveat and that is that if there are any admissions directors in the room, please know that only once in 75 years or so -- so, 75 years from now we may want to repeat this -- where we break the confidence of a college recommendation, to read it before a cast of thousands, which is what I am about to do now.

This is Bill's letter of recommendation for Harvard, composed in November of 1972 by Vernon Harrington, and I quote:

"As a 10th grader Bill Gates was famous as Lakeside's ranking computer freak. The boy genius who was quicker at mathematical computations than his teachers and who with two friends contracted to set up a payroll tape for a Portland firm. The job was completed. They were paid off in $5,000 worth of computer time -- time which they later used in computerizing the school's schedule. In the last year and a half, Gates has launched out into other areas. He is enormously interested in politics. The result, in part, of a few months as a House page. Gates has a mind which soaks up facts and ideas and all of his information seems to be subject to instant recall. In other words, he is quick and resourceful in debate and thoroughly sophisticated in intellectual areas. This year he has also become involved in drama and is currently playing a lead role in a school production, Shaffer's Black Comedy. Again, his talents are apparent. He plays the role with sensitivity and perception. Gates is one of those who does everything well. If he is not the most able student in the State, he is very close to being so. Beyond that he is vastly concerned with ideas. He reads omnivorously. He is invariably alert, usually enthusiastic, and always willing to tackle a new problem, usually with verve and irrepressibility. In every way he is a stimulating resourceful and ambitious student, one who is determined to succeed."

You know the rest of the story. We are proud to present to you tonight a man who has demonstrated beyond any possible question, the benefit of a Lakeside education.

Bill: Well, after that recommendation, Harvard did admit me to the school. It's a real honor to be here tonight. Lakeside was a very important part of my life. I enjoyed my time here and benefited from it immensely. Dan Aryault was the leader of the school at the time and someone who I got to know very well and had an incredible impact on my life.

What I want to share tonight is a little bit of a glimpse of the future and talk about how the tools of information technology are really going to change a lot of things in the world -- the way we work, the way we play and entertain ourselves, and even the way we are educated. And I think perhaps that last is the most important impact this technology is going to have. I'm very excited about what's going on here. I think it's something that will take quite some time to fully develop. But it's not too early for people to be thinking about how we as a society want to use these capabilities, how we individually want to take advantage of them.

To talk about that I decided to go back in time a little bit and look at the last 25 years and show you the rate of progression that we've been going through. As we're extending that out into the future, you can get a sense that it is moving very quickly.

In a certain sense it all began a little over 25 years ago down in McAllister. There was a mother's club donation. About $3,000 was raised in the rummage sale that year and some of that was given to get a so-called teletype. That's the device that you see there. That's Paul Allen, my co-founder, and myself just entranced by that device. It was so limited in capability. To look back now and think -- geez, how did I get addicted to something like that? It's hard to understand. They didn't get a special phone line. We shared a phone line with the business office.

So, we were always competing to try to get on the thing and staying in late at night. The janitors were always kicking us out. It wasn't a computer. We had to dial out to the computer. Computers were so expensive that you could only time-share the computer. This GE computer we were connected to cost four million dollars. And we were using the basic computer language, writing little programs to play monopoly, play tic-tac-toe and do neat things. And Paul and I were deeply engaged. In fact, we ended up doing a lot of teaching of computer programming and really started to wonder -- where was this thing going? What could it really become?

And we learned about the technology of chip making, taking a lot of transistors and circuits and putting them onto a single very small chip. And the incredible thing about that technology is that every 2 years they can put twice as much capability onto a single chip and yet they do that without increasing the price at all. Now, that kind of exponential improvement just isn't present in any other thing that we deal with. It's unique to this computer technology.

The next milestone about six years later was actually a kit computer called the Altair. By 1975 I had graduated from Lakeside and I was off at Harvard in my dorm room, when I got a magazine that had this kit computer on the cover. It cost $360 and it didn't really do much. In fact, if you bought the kit and put it together all you could do was make the lights on the front of this computer blink at you. But the company who sold it, down in Albuquerque, New Mexico, sold about 3,000 of the kits and Paul and I wrote a letter to them saying that you've got to have some software for this thing. Software can really bring it to life.

So we wrote a Basic, actually very similar to the Basic that had run on that GE computer, and made it available for this Altair computer. We were the first software company formed to do software for these tiny little machines. The rest of the computer industry -- DEC, IBM, HP -- they were engaged in making very big machines, mainframes and mini-computers, and didn't really take the time to get involved in these smaller machines.

Well, 6 years later there was another important milestone and that was the move from the original so-called 8-bit technology to the 16-bit machine. And momentously, the very first 16-bit machine was brought out by IBM. Now, the group inside IBM doing this machine was a tiny little group. They had eighteen people. We actually had, because our entire company of 30 people was devoted to this, more people on the project than they did. And they had forecast they'd sell one hundred thousand machines and that's why they turned to a chip company called Intel to get the actual computer processor and turned to Microsoft, which by this time had moved to the Seattle area, for the MS-DOS and the Basic and a great deal of other software.

The power of this machine was good enough that it became a business tool. Things like word processing, keeping databases of customers, doing business analysis. This machine, although it is 1/5,000/th of the speed of the machine you can go out and buy today, was a big breakthrough and in fact went on to sell 10 times what IBM had forecast. In fact, it became the model for everyone else. Instead of making machines that had to have software specially written for them, people built machines to work compatibly with this machine, so-called "clone" machines. And today, 90% of all personal computers made are compatible with this original design, the remaining 10% largely being the Apple Macintosh. And so the impact of that idea and of the standard still is with us today.

The first machine designed around graphical interface, the idea that you can take the screen and put pictures up, was the 1984 Macintosh. Now, these ideas had actually been pioneered in a research laboratory by Xerox, down in Palo Alto in the 70's. And both Apple Computer, who built the Macintosh, and Microsoft hired a lot of the smart people away from Xerox as they failed to commercialize their technology in an effective fashion. This machine was a break-through in ease of use. And it's one that we've gotten behind in a big way to try and write great software to take advantage of it. Graphics came into the standard PC -- it's hard to pick the date -- we had reasonable implementation in 1987, but it wasn't until 1990 that virtually all new PCs had that capability. Even today you'll still find some machines that work in the so-called character mode that came before this.

In the last few years we've had another kind of break-through. The term, multimedia, is a very big term. It broadly refers to the idea of not only using text but also using pictures and sounds and taking electronic content and putting it into the computer, generally using a CD, a compact disk, exactly the same as what you use to listen to high quality music, and then publishing that CD so that people can buy it. Today, the most popular encyclopedia in the world, Microsoft Encarta, comes on a CD. We sell over 5 times as many encyclopedias as World Book does, which is the second most popular encyclopedia. Now, they're sold at considerably less price, so World Book is still substantially the largest encyclopedia company.

The ability to move around in the information, to see it in different ways, is really incredible. I remember when I read the encyclopedia, there's only one way to go through the thing which is alphabetically, and so it's really disjointed if you go through, you know, ancient history and modern history, all in this alphabetic mismatch. Here with an electronic encyclopedia, it can keep track of which articles you've read, so you can go through by topic area. It's far more coherent. And when you get the yearly updates, instead of pasting in little labels on the articles to refer you out to that World Book, you simply take the old CD, get rid of it, and the new CD has all of the information integrated in, totally up to date, in an exciting way.

But this multimedia platform -- it's not just encyclopedias -- it's all of human knowledge. Museums are taking their collections and putting the images along with the stories behind those images, into this media. And you can access it interactively. That means you can go about seeing what you're interested in. You can be quizzed. You can get definitions. And so hobbies of all types -- stamp collecting, sports information, all of that is being moved into this form. And a new form of publishing doesn't come along very often.

And so here we have thousands of companies, not just computer companies, but cable TV channel companies or textbook publishing companies, or companies that print mail-order catalogues, all trying to use the CD to come up with a way that they can work their application using this new breakthrough. It was really this last Christmas that most of the machines had the CD ROM drive in it. And you get this chicken and egg problem of -- before there were machines with CDs, why should anybody publish titles for CDs? And before there was titles out there, why should anybody pay the extra money to buy the machine with the CD?

Well, that kind of boot-strap, getting that going, you really have to have a lot of what our industry calls "evangelism" -- going out and telling people what the potential is, getting them the tools, and a few people who just believe, who are willing to take the risk. And this took a long time to happen. It started in 1986 when people were first publishing these things. But now it has achieved critical mass. That means that you don't have to have a faith that there will be a lot of volume with these machines. People are making a lot of money coming out with titles. And so now competition will drive the quality of those up, drive the price down, and even drive the variety up at the same time. The competitive nature of this industry is illustrated by this timeline showing what a typical machine looks like. Now the way this is often phrased is people say -- computers are getting cheaper. Well, in a certain weird sense that is true. If you wanted to buy that 1990 machine, that 286 2-meg machine, you could probably get it for $500-600. But in fact, almost no machines of that type sell. The dominant phenomenon here is that the typical machine is getting dramatically more powerful. So, while there was a decline in the average PC price from $3,000 to $2,000, the memory size is 8 times greater. The speed is 15 times greater. And there's new features, like the sound and the CD ROM, which are part of what goes into that system. And that will continue out into the future.

So, it won't just be some qualitative thing that we can grab onto and say, yes, computers will eventually cost one dollar. It won't be quantitative. It will be new features that change how it can be used and what you do with that computer. We didn't take the computer industry which was large mainframes being used by businesses and simply drive the cost down. We made it a tool of the individual to go about doing their work. And that phase is basically complete.

There are, today -- over 140 million people have this as an individual tool. The next phase is now beginning. And that is to turn the PC into a communications device, into the vehicle we use for sharing information. They say this is the information age. But what do they mean when they say that? Well, what it should mean is that we'll have a revolution in how we go about creating and finding information. And, in fact, this future PC will be the tool on which that occurs.

Going through time, let's look at how the PC has been used for communication. In the 80's all you would do is create your word processing document, print it out and bring it to a meeting. It was very expensive to connect the machines together. If you wanted to use a phone line to dial up, it was very slow and not much information was available that way. Today, there is some sharing of information through the CD, but that's high volume publishing. It's a broadcast media. Dial-up services like America Online or Prodigy or CompuServe, or the popularity of the thing called the Internet, are starting to show us though that by taking computers even that are very apart, we can start to get information sharing.

The phone line, we're able to use it at a higher speed today, so-called 9600 baud, but that still isn't good enough. You can't really send pictures very well. You certainly can't send sound or movies across that type of connection. But already, with things like bulletin boards or electronic mail, we're beginning to glimpse the potential. Around election time there's these bulletin boards where people can put up the latest information and share their ideas. And there were literally millions of people around election time going up and seeing what the results were and sharing their thoughts about what was going on. So, when you get a strong community of interest, even today's technology brings them together in almost a virtual community where they are sharing different ideas.

Later this year we'll start to see some new things -- an improvement in the Windows product that brought graphics to the PC and we'll start to see a higher speed of communications that, like all things in our industry, has a kind of a funny acronym. It's called ISDN. It's actually been around a long time but the phone companies built it and there was no real idea of how it would get used, no application. And now with some new features built into the PC we can do video conferencing. You still have to add a few thousand dollars to each machine to do it, but you can share documents with each other and have meetings at a distance.

And that's just the beginning. If we go out further into the future we'll see a period where not only will ISDN be cheap and easily available, but a technology even beyond that, so called ATM, asynchronous transfer mode, which is so high-speed they call it broad band, will be prevalent. And it lets you share high quality motion video between any two points on the planet. So, whether it's watching a movie or seeing a video conference, you'll be able to do that at very low cost.

So, a number of factors, again, have to come together. We've got to get a network built out that can do this. We have to have cheap hardware that can connect up to it. We have to have applications that make it all make sense. And all of the hype, all of the discussion that's going on right now is -- are companies really positioning themselves for this next big opportunity?

The PC will be at the center of how this happens. Today about 35% of US homes have the PC. Over the next three years that will go up to almost 60% and the PC will start to have this higher speed communication and high quality motion video capability. And so in the den or on your desk in business you'll have a PC.

But there will also be two other foreign factors that will be important. As you walk around you won't want to carry an entire PC. You'll have what we call the wallet PC, something you can put into your pocket. We call it the wallet PC because it's essentially a replacement for your wallet. Everything you have in your wallet you'll be able to do on this. You'll be able to look at maps, gets messages, get news. Instead of having money, this computer will have the equivalent of money, digital money. Instead of carrying around pictures of your children you can have hundreds of pictures of your children stored on this device, even little movies and you'll have your credit cards. You'll have digital keys to let you get into places. Instead of having to pay for an airplane ticket, as you walk through the boarding area, using wireless communications, the airline will see that you have a ticket, tell you what seat to go sit in, and record that that ticket is being used. So, no more paperwork, no more manual intervention. When you leave a hotel you'll just say, "Check me out." You'll see the bill on your little wallet PC, accept it, and be on your way. And so this will be a very common device and it will tie in to the same information but it will appear different than the PC does.

Also, in the home we won't just have the PC. We'll continue to have something that looks like the TV set. That is, whenever you sit close to a device we call that a PC. Whenever you sit far away from it, we call that a TV. Inside that TV will be a lot of electronics, almost identical to what's in the PC. But multiple people can sit in that living room, use a little infrared control, watch movies, see your grandchildren at a distance, play games together, get medical advice, whatever it is that you want to do that relates to information you'll be able to do very easily with these three different devices -- wallet PC, the future desktop PC, and this TV PC.

So those are the things that people are working on today. What it means is that a lot of industries that have been separate -- communications companies, TV broadcast companies, PC makers, video game makers -- they see that there's a convergence with all of those businesses coming together to build these very important tools. The cable companies see that the telephone companies will be offering video. And the cable companies want to offer telephone services as well as all of the new things. And that's a great thing because it engenders a rush and it engenders people wanting to get out ahead and help all of this happen.

One of the applications I mentioned is the idea of having a map. In fact, that little wallet PC will even have what's called a global positioning system, GPS, in it. And so the map will be updated at all times. You can say, "Where the heck am I? Where's the nearest McDonald's?" And the device will show you. Or, "What's the route to get from where I am to someplace that I want to go". This area of mapping products has gotten to be really popular. And so I thought I'd give just a real quick demo of one that we have called AutoMap.

(Demo of Automap)

Well, this term, "Information Super Highway," it's hard to get away from it. Almost any magazine you pick up has some crazy article about the information super highway. It used to be that if you worked in the computer field, when you were at work you'd read about computers and when you'd go home you didn't have to read about computers. Now, that's simply not possible. And a lot of the talk about the information super highway is pretty wild. You know, people saying it will happen over night, people saying we ought to be unbelievably afraid of it, or when it happens we'll never go outside of our homes ever again. And it's really hard, before the technology is really there, for people to get a sense of it. I thought it would be interesting to go back and look at other big advances that have taken place in the past and really compare how the information super highway fits in with these other advances. And so we made a little video that can help us have the information highway in perspective.

(Video)

The key is trying to figure out what are we going to do with all this technology. A lot of talk has centered on the idea of movies on demand, the ability to see the reviews, have the system recommend things you might like based on what you've watched in the past, and then have the movie be right there. But that one application can only fund a small percentage of the cost of building these new systems. We need far more than that. We'll be able to take all of these CDs that we've been building, and instead of asking people to go out and buy it and then every time they want to use it, putting it into their machine, it will simply be stored at the center of the network. And you'll have your own personal bookshelf where you pick the things that you're interested in and every time you turn on your machine they're right there. You click on it and then you're interacting with that information. In fact, this approach eliminates any size constraints or the difficulty of linking information between various titles.

And so we'll be able to draw on the phenomenon that we've had over the last few years of this CD publishing. Certainly "Yellow Pages," "Classified," "White Pages," being able to look up real estate, make travel plans … all of those things will be there. And it will be very easy to get somebody to come on the screen and give you advice. Even doctors will be available to talk to you about a condition. They can send you a little video clip or refer you to text information. And of course they electronically will have your information. Games could be a big part of this, even gambling from your own living room. And it won't just be people getting information sent to them. This will be a publishing media that everyone can participate in. If you have thoughts about a bill in the legislature, if you have a group of people that you want to bring together to do something, you'll be able to put up a bulletin board and send articles to that, filter in the things that you want, in a very straightforward fashion. So, there are a lot of consumer applications. This is what is going to justify upgrading these cable and phone systems, so that you have the new optic fiber going out to every home.

In order to help people visualize what this might be like, we've put together an example of the typical day about 10 years from now, in November 2004. And what we have is a mother and a son getting up in the morning and doing the typical things they do, just using these new tools as part of their everyday life. Let's take a look at this family and see what it's like.

(Video)

In this scene we saw that Jackson is doing his last minute homework over in the kitchen. Meanwhile his Mom is using the TV set to look at a few shows. Before she walked over to the TV, she just simply made a selection on a wall panel. It said good morning and she selected that she wanted the lighting, the security, the temperature, all of the systems in the house set for her to be over at the TV set. When she got there it simply had the rest of the show that she stopped watching last night. You don't have to keep watching a show. It can be called up at anytime. So, she saw the last little bit of The Late Show. And then it went to her menu, the things that she likes to see all the time. There were little icons, pictures there, for every one of the shows. And the ones where there was a new episode were lit up. And so she knew that there was a new Oprah show, looked at that a little bit, and then went over into the morning news. And the morning news is a lot like the normal TV show. You can just sit and watch it, but it also had interactivity. On any of the stories they were talking about you could just point with your little infrared pointer, click on one of those stories, and select that to come up, and go into some depth on it. Take a topic like the weather. Some people love to see the weather forecast. I think it's repetitious, so I would just skip by it. And so it's very personalized how you can use the system.

Now, when she was over there, Jackson had a device like a PC, a future PC, that he was starting his homework assignment on. And he's been asked to make a presentation on pre-Columbian art. And so he went out to the library at the University of Washington, saw what articles they had, saw a cross-reference to a museum down in Mexico, and saw some pictures that were stored on the computer system down at that museum. Now, one of those really caught his eye, this one in the middle here. And so he asked the system to do a search and find if there was anyplace near to him where he could go and buy a replica of that particular artifact. And in fact it gave him a map. It said it was down in Pioneer Square and gave him directions for him to go and get that artifact and use it in his presentation. And so this is just a very natural thing for him to be working on his assignment. Later we're going to see him come to school and see how he really does.

Well, that brings us to -- how will this change teaching? Teaching, of course, hasn't changed a great deal over a long period of time. You know, somebody could say the best teacher ever was, you know, 30 or 40 years ago, and you wouldn't have any kind of a way of determining if that was a correct statement or incorrect. In fact, it's been very difficult to share best practices, to really take presentations that other people do and custom tailor them. Well, as things move into this digital form you'll have people posting their ideas, what they did that worked, out on bulletin boards. Other people coming in and looking at those. And awards being given for good ideas. And all of it easily available to be shaped for any teacher who wants it. Likewise students can learn interactively whether it's during the school day or at home. All of this wealth of material will be there. And so they can proceed at their pace. They can see what their level of skill is. And it will be very engaging because it will be done in a full high quality multimedia fashion. Students can collaborate together at great distances, share ideas. And even people who are outside the educational system will be able to tap in to this rich material. And so in this little movie, Jackson is of course a student at Lakeside High School. He's got a class up in the Allen Gates building. So, let's take a look and see how he does with his assignment.

(Video)

In the same way that PC word processors gave an individual the ability to make a document as high quality as somebody with their own typesetting machine, now we're taking what special effects, what animators, what people with very large budgets are able to do in movies, and making those easily accessible to anyone who has a computer. It certainly is a big difference from the time I went to school here. This is the rogue's gallery at the time, where really all the PC was used for was trying to figure out how to do the schedules, how to get all these different classes to meet in different places. I can still remember how difficult it was and being worried that by the time the semester started we wouldn't have the schedule all put together. Well, instead of that, I think, that the kids will grow up very familiar with this technology. It's quite different when you just take these things for granted and they're around.

Jackson here, we saw him actually doing some of the animation specification on the sort of portable computer that he carries around here, with kind of a cool little keyboard. We saw some of the pictures that he brought up. He simply down-loaded those from the Internet. They were out there and the museum had marked them as available, at least for educational use, without any fees being paid. And so that was very straightforward. And he had all sorts of different effects that he was able to put together there.

Now, business applications will be the first to emerge. That's because the ease of running the networks is greatest in city centers and the cost justification is strongest for the business applications. So, even in the next year or two these will emerge in a big way, to be followed a few years later by the other applications. The first stage here is for people to use electronic mail. Companies really ought to make the effort to make E-mail a key communications technique, not just within their company but also to their suppliers and their customers. I think the time will come in the not too distant future where, if you're working with your lawyer, you'll expect to be able to set-up an appointment electronically, to be able to send in a question electronically, have them send you back advice, and work through that medium. Likewise a doctor, an accountant, or a travel agent. It's so nice to be able to just send in a query, have them get around to it in their time, mail back to you not just a message, but your whole travel schedule, a spreadsheet with some analysis, some pictures of some places that you might go to. And all of that can be done with today's mail systems.

Now, a lot of companies set up these mail systems and because not everybody uses it regularly they get no value from it. Unless you can assume that everybody really reads their mail, then you still have to communicate with them the old way. And it's simply not effective at all. Now, mail will expand so that it's not just sending a message. You can dial someone in and see their picture at high quality. It should reduce business travel. It should let people take what they're good at and let them reach out to a broader market.

Even the idea of matching buyers and sellers, the fundamental mechanism of capitalism, will be mediated far more efficiently by these electronic networks. And so companies may find that instead of having a lot of capabilities inside the company, they can reach out and get consultants, get people to come in, check their references over the network, know they're available, and then collaborate with them that way. Some people may choose to work out of their home more, either people who today can't be part of the workforce, or people who just a few days of the week find that attractive. And so job advertisements will talk about what portion of the hours in that job need to be done at the business location versus how many can be done flexibly. And technology will make that very straightforward. The secondary affects of this are hard to predict, but they certainly could include the idea that people clustering together in urban centers is not nearly as important as it has been. You can go off and live anywhere you want and still exercise your skills for your business activities.

The last little video clip that I'm going to show actually starts out with a little bit of a sad development. Our friend, Jackson, has been hit by a car, not too badly, we don't think. And he's in an ambulance. And so what we're going to see here is how they're able to take care of Jackson's injury in a new way, using this technology. Let's see how Jackson is doing.

(Video)

Well, there were a lot of things going on there. First we had the wireless connection from the ambulance to the dispatching nurse. And she was using her large screen there to call up medical information. As soon as she heard it was a shoulder problem she got the diagram and was told what questions to ask and was putting in those answers. And based on that the computer decided what kind of specialist needed to be called up for consultation. And so the doctor that came on the screen there could have been anywhere in the world as long as she had access to a computer screen. And she was also being relayed all the information, including the scan that was done right there in the ambulance. As soon as the phone number was entered, not only did they find out Jackson's name and how to call his mother but also his medical records were called in. And so they could see if there were any problems there, and put in the new information, and even decide where the ambulance should go, and let Jackson know that it wasn't a serious injury. And if you compare the quality of service and simply the overhead of being able to do things this way versus the way it's done now, it's quite a dramatic change.

Now, we don't have a video clip to show what happened but, of course, in the future you'll be able to select your own endings. And so you could certainly choose for Jackson to recover just fine there.

This will all take some time. This year what we're seeing is this explosion in CD publishing, a lot of on-line usage, about four million people in the US. That's still a small percentage of the people who use PCs -- only about 20% today. And this thing called the Internet that schools and others are now connecting into, all of those will come together. As the speeds go up, all of the good ideas will be brought into one seamless on-line system. In '96 there will be some trials where not only is the very high speed optic fiber put into businesses but also into some homes. And, in fact, here in King County we're working very closely with a couple of the cable companies, Viacom and TCI, to have a trial in 1996 and try out these different services. Do people like to do their home banking using this system? Are they willing to talk to each other? Do they like to call up movies? What about travel? What about medicine and education? So, we'll be trying things out.

And we'll be tuning the user interface. These things will have to be very easy to use, say 100 times easier to use than today's VCR is. And that's very easy to do because when you have the screen that can interact with you, it can see if you pause. It can remember what your skill set is and prompt you along. So an interface that goes a step beyond today's graphical interface we call the social interface, will be a big part of this. In '96 businesses will be sharing information with this ISDN. And some of the build out will start because there is such a rush to get going. It looks like the United States will be in the vanguard of this revolution if we can get the right deregulation bill passed in Congress in the next year or two. That will unleash the kind of spending that needs to go on. And it's rather amazing that most of the key elements that define this "highway" whether it's the networking technology or the hardware or the software technology, high tech companies here in the United States are the leaders in virtually all of these areas. Not only is it an opportunity to get things to work more efficiently here, there will be significant exports and a lot of new companies that are very successful built around this opportunity.

It's very hard to predict where we'll be by the year 2000. I think it's fairly conservative to say that by that time about one third of the households that have cable today -- which in the US is about 63 million -- about a third of those will be connected up and able to have these services. I think it's a very exciting opportunity, one that really goes back to the vision that convinced me to drop out of school and start Microsoft 20 years ago now. And that is this idea of the computer as a great tool for everyone. I think not only is it an exciting time, it's one that we'll all benefit from.

Thank you.