Remarks by Bill Gates
Microsoft Corporation
Empowerment 2001
February 9,1998
Seattle, WA

MR. GATES: Well, good morning, and welcome to Seattle for this conference. I'm very excited about some of the things we're going to be showing you over the next couple of days. I thing we've really got an exciting opportunity to work together to allow government to lead the way into the 21st Century.

Now, the theme is that governments can build a so-called digital nervous system. I'm going to spend some time this morning explaining what I mean about that. The goal of the digital nervous system is to make interactions with citizens far better than ever before, to give them access and enable participation that's not possible today. It also allows businesses to work in a low cost, streamlined fashion with the government, and to create something that really spurs an economy to move into the information age and take full advantage of the opportunities that it represents.

Now, when we say nervous system, obviously we're making an analogy to the way that humans deal with information of all types brought together in an effective way. This era we're entering is the information age, and the tools of the information age are very clear. It's a state-of-the-art personal computer connected up to the Internet.

Many organizations, from governments to businesses, are making large investments in those personal computers and connecting them together. But my belief is that only the governments that step back and really think about how to use these tools will get the full benefit that's possible from them. In fact, I'd argue that governments that simply think about PCs as tools for creating documents aren't getting a tenth of the potential from the systems they're working with. In order to go the rest of the way, the extra expense is actually reasonably modest, and the benefits are very, very strong.

Now, the digital nervous system is not designed to deal with a single application. Part of the beauty of the Digital Nervous System is that brings information of all types together so you can find interrelationships between the information, whether it crosses government departments, whether it covers different regions of the country or different subject areas, and make all of it easily accessible.

You might ask, how is that information being managed today? Well, the answer is that there's massive paperwork, lots of meetings, and it's very difficult for people to get the information they need. In many countries, the government is the largest publisher, yet if you go to a small business and say to them, did you know that this interesting information is available, nine times out of ten, they won't have been aware of it. And so there's lots of information that could be delivering more value to businesses and to citizens.

In dealing with information, there are lots of processes that can be more effective. One thing you'll hear me talk about several times is eliminating paperwork, getting things into digital form as soon as possible. That's the way you can make it searchable. That's the way you can make it inexpensive to deal with. That's the way you can avoid people having to enter the same information many, many times. And so there will finally be a dramatic shift away from paperwork and paper forms into these digital systems.

The system has to deal with basic operations and the kind record keeping that goes on today. It has to deal with all of the typical processes that take place. And it has to be ready to deal with surprises. In each of these areas there are specific applications that need to be managed. Now, when we think about the analogy to a human's basic operations or breathing, blinking, those things that are done subconsciously for a government, basic operations include having a database of business licenses, keeping track of all the taxpayer filings that have been made, any type of registration or government services to citizens. And very typically you have offices where people wait in line for the number of forms to fill out to fulfill these functions. Very often, the records that you keep are stored on paper in massive archival systems that make it reasonably difficult for people who want to go back and find that information.

Planned events for a human being are things that you schedule yourself to do, getting to work, enjoying an entertainment activity, getting to dinner, all of those things organized to make sure that you have the schedule you're interested in.

In the case of a government, planned events would be things like managing legislation, you know, what state is it in, who has approved it, what are the different amendments that people have. And that's a process that benefits greatly by having it be transparent to citizens so that they can see it any time, what the state is. And, in fact, if they're accessing the information, if they've provided the area they live in, you can annotate anything they look up by showing them what their particular representative has said, or what their particular representative has voted on related issues. So you can make democracy more relevant to them. In fact, it should be easy enough to simply click and send a piece of electronic mail to your representative commenting about how you agree with them or disagree with them, so they'll have a lot more citizenry input than they have today.

Every business has to go through a budgeting process, and this can be a very complex process where the real priorities can be lost if you're going through just the process of dealing with all the numbers and the way those numbers are categorized. This is a case where, again, digital tools can let you share information and compress the budgeting process fairly substantially.

At Microsoft, we took a three-month process and turned it into a less than three-week process simply by getting things online. And when you shorten the process, the number of surprises that take place during the process is greatly reduced, and the ability of different organizations to pull together their pieces and see the intermediate steps that other people are taking to make sure what they're doing is consistent with the larger process is greatly improved.

For a government, you also have a lot of instances where you're going out to the private sector to procure services, and making those procurements completely open by putting the information online, by having people even submit their bids in an electronic form, can allow more bidders to get involved in the process and make sure no one misses an opportunity to come forward and get involved.

Even as those projects move forward, managing those projects is very, very complex, certainly as complex as anything that takes place in the private sector. And, again, a digital approach would have benefits.

Well, unplanned events, I think, may be the most important category that I talk about. It's kind of hard to measure how well an organization or government deals with unplanned events. They don't come up very often. They're all very unique in nature. And yet, I think as we look out into the future, there's likely to be more unplanned events than there have been in the past.

For a human being, it's avoiding an accident, hearing something new that's going on, a change in financial status, and humans are very good at gathering all the information and making decisions about the crisis.

For a government, it would be a change in the economy, perhaps the currency exchange rate changing as we've seen in Asia recently. It could be a new set of rules that are coming down from a community organization or a change in the trade status of a company, or perhaps a public safety emergency. Lots of surprises come along, and when they do, you want to make sure that you're getting all the information about what's going on all in one place, and that you make it broadly accessible where.

You also want to get the people who need to make the decision about the unplanned event and make sure that they're all seeing the same thing, and that they can collaborate. In many cases, those people are in very different locations. You'll have cases where governments need to collaborate with each other, and in order to coordinate that kind of activity, doing it through the digital system is far superior to the way it's been done in the past. So there are quite a range of things that this digital system would be used for.

Now, calling up a picture of what this would be like, ideally, is reasonably straightforward. As I said earlier, it is not about investing in a whole new set of hardware. It's not about investing in a new network. The standards of the Internet are clearly going to be used on a global basis by every business and every government for these activities. And the fact that those standards are shared on a worldwide basis is a wonderful thing. It means that all of the equipment suppliers are building hardware and software around those standards, and as you deploy those, the ability to share with anybody you want to is guaranteed because of these protocols.

Microsoft is more involved than any company in sitting down on the standards committees and making sure that the enhancements that are required are being made in those standards. So that's all very, good news. The fundamental plumbing decisions are already clear cut. There's no risk you have to take in choosing to build on top of those things.

Likewise with the PC, you have a building block technology that is getting substantially less expensive at the same time that it's getting far more powerful. Historically, government has had a very hard time buying computers because their procurement process would take a long time to go through, and it had to be a process where you could have many, many companies competing. Before the PC came along, when all these computer systems were incompatible with each other, that meant that the government had a greater mix of hardware systems than any other organization. Businesses could really restrict to make sure that they had a limited number of systems and that they all were designed to work together. The government really wasn't in a position to do that.

Well, when the PC industry came along, for the first time there was a separation of hardware and software. There were literally hundreds of hardware suppliers providing machines with an incredible variety of speeds, portability and service options, all moving at an incredible pace. And decisions to pick individual suppliers or to mix and match machines from different people, those decisions were safe decisions because you knew that everything would be upward compatible, and the systems from different manufacturers would work together. And so you could make those decisions independently of what you were doing on the software side, knowing that you would never have to start over rebuilding software systems.

I sat down last week with the information organization group from the European Community, and they were talking about how they've moved from having over 20 different varieties of UNIX to now using PC suppliers so they can run Windows-NT on all of their desktop workstations. That's improved their ability to exchange information and deal with some of the challenges they have, including supporting all the different languages that the community has on official basis. In the digital world, translating those documents, making sure that they're easily available in each language, that's been greatly improved.

So the PC has been a very beneficial development. It's really restructured the entire computer industry. Companies used to provide all the pieces. IBM did the chips and the systems and the hardware and the software, and they had their own sales force. So it was completely one organization hooking together things based on only their own standards.

Now we have much more specialization, people like Intel in chips, Microsoft in software, and a variety of organizations for each area of expertise. And based on broadly used standards, these pieces can advance at a very rapid pace, and work together incredibly well. Out of this base investment you'll get not only electronic mail, not only browsing, but also things that haven't been common place yet. Something like video conferencing, that I expect to be standard, and used very commonly, even in the next two to three years. You also get electronic commerce. That's just at the beginning, but the building blocks are coming into place to make it very, very easy to set up an electronic commerce site.

Let's say a department in the government wants to put up a site where people can pay for business licenses. Today, to build that application would be a matter of two or three months, because the standard software is there, and all you have to do is enter in the specifics of how you want to build your site.

Historically, when people talked about computer development projects, they talked about two or three or four years, and there would be cost overruns of 40 and 50 percent. And by the time the system was built, it wasn't appropriate to what it had been built for. Every system was built around a different hardware base. If you have to have a new hardware base for every application, the expense is just too great. What I'm talking about here is using the PC hardware base, and that's a one time expense, and it's used across every single application, even applications that you're not thinking about today can be deployed on that base, without having to change the hardware approach.

So the key technologies here are making this vision easier and easier to get to year by year. The rapid improvement, where we double in speed every two years, you just don't see that in any other parts of the economy. And people aren't used to that exponential improvement. Gordon Moore of Intel predicted it over 20 years ago and the industry has been moving even faster than he predicted. Certainly over the next decade that pace will only accelerate, with many, many companies participating in this, giving both additional power, and lower cost here.

Well, there are some particular challenges for government, in terms of building a digital nervous system. And I want to acknowledge that these challenges have to be overcome. First, is the procurement process. Now, that's improved dramatically with the availability of PCs and standard, low cost software that's run on top of those PCs. A second is that many government systems can't be built unless they're accessible to everyone. There are two ways to deal with that.

One is, to simply keep the paper based system in operation, while having in parallel a PC-based system. The second is to deploy so-called kiosks, which are really just PCs with a simple touch screen interface, available in public spaces, so that everybody can come in and use the system. They either have the PC at home or in their organization, or you make sure that the kiosks are easy to get to.

In a few countries, like some of the Nordic countries, the widespread use of the PC is so advanced that it may be possible to simply expect that the PC is a standard way that people want to interact, without doing many kiosks. In other countries, where the PC penetration is much lower, the kiosk approach will be very, very critical. A third point is one that I've been emphasizing, and that is the idea of sharing this infrastructure across all government departments. You don't want to have different electronic mail systems for all the different departments. If so, that means that when somebody wants to interact with a government, accessing the directory to look up who's who and who has a particular expertise, and where do they go for a particular problem. It's not easy to bring that together, unless the government has a common directory.

And at one department, the department that deals with medical payments, thinks, wow, let's roll out all these PCs to use it. They may find that it's not economical just for that one approach -- but if you rolled it out and it was used for a variety of things, including, say, tax payments as well as medical records as well as access to information, then the savings you can get from getting rid of the paperwork in the old systems means that you can go far more aggressively. And yet, the way that government departments work requires leadership from the top, in order to get people working together and create the plans that will have all those systems come together on a single PC network.

Another tough problem is user authentication. If you're just going to publish information, which many governments are doing today, you don't have to worry about that, because it is public, everybody gets to know those things. But, if you're going to have something like applying for a business license, then you need to make sure that the person who is saying they're applying for that license really is who they say they are. There are several ways to deal with this. You can issue passwords to people. And that's very typical today. It's used for electronic mail security. It works just fine. You can piggyback what the banks have done, ask people to put in a credit card and verify that the credit card is right, or a much stronger approach is to piggy back work going on with smart cards, which uniquely identify the person involved.

Smart cards are advancing rapidly, much more in Europe than in the United States. But I do expect in most countries that most citizens will have a smart card. In some countries the cards will come from the government, and simply be used to make things efficient. Belgium is doing that for its medical records system, where you'll have the card, and whenever you want to look at what's going on, you'll simply go to a kiosk or a PC and insert your smart card.

Today, PCs don't come with a smart card reader, it costs an extra $30 or $40 to have a smart card reader. But, eventually, the standard PC keyboard will have that smart card reader built in, because if you do it that way it's only a few extra dollars. And the industry is working hard at coming up with standards that will work with that. The physical form of the smart card has been agreed on. We need to make sure that the smart cards are very, very inexpensive, that we don't overload them with too much functionality.

A final challenge is privacy. Today, governments have an immense amount of information about their citizens, whether it's about their traffic records, their criminal records, their medical records or their tax records. In the past, when this information was in paper form, you could sometimes not worry about having explicit policies to protect privacy, because just managing the information and searching was so difficult that there wasn't much going on that would force you to think hard about this. Now, as things get into electronic form, you have to be very careful. The technology exists to enforce whatever policies are decided on. But, it's a new thing.

So let's talk about some government examples. Let's make this very, very concrete, what some people are starting to do, and how the ideal system comes together. Information from the government, there shouldn't be anything that's available on paper that isn't available electronically. Electronically it can be updated faster, searched faster, it's easier for people to get at, and that's a simple principle. The government printing office should first put things up electronically. Now, if it wants to make those things free, that's fine, or if it wants to set up and still have some small charge for accessing those things, setting up a site to do that is very, very easy.

You get to the point where the people creating the information are doing it in such a way that the publishing is automatic. So you don't have any intermediate steps, things like government statistics, that can be immediately available on the network. And not like they are on a piece of paper. Whenever you get numbers on a piece of paper, they're very hard to deal with. If you want to get more details, it's hard. If you want to compare numbers with each other and graph them, it's hard.

If you get them on a digital system, where you have tools like spreadsheets, the numbers can come alive. To give an example of that, Microsoft no longer prints out its sales information on paper at all. Because if we print it out, you know, we can't really dive into it. It's simply available electronically.

And so before I visit a country, I quickly call up with a few clicks, exactly what their sales are like. I look at it by product, by customer type, by region, I can see is it on budget, how does it compare to last year, how does it compare to other countries? How does it compare to the growth of PC sales in that country? And I'm just sitting there, clicking and seeing the information at any level I want, even down to the individual transactions.

It's not a system that was built for a few executives. When you build a system that only a few people use, that's a big waste of time, because then you have those people calling up everybody else, who really know what's going on, and citing these figures, that is only in the system that they're looking at. And so you spend a massive amount of time trying to figure out what the difference is. Unless the system is for all the knowledge workers, in the organization, it's not worth building. Here for the sales data we have that one database.

Then you move from publishing information, to things where you're transacting information, like applying for a permit, or filling out tax and insurance forms. When you interact with the government today, you often end up redundantly giving information, again, and again, and again. And let's say you want to change the information. Let's say you have address changes. You're not even sure what forms you have to fill out for the medical world, the tax world, the business license world, for all those things. With an electronic system, if it's properly designed, you just go in one place, change that information and immediately that's done in all the systems, they're completely up to date, because of the sharing that takes place.

Well, what are some examples of leadership here? I want to be clear, there is no government today that can't have a digital nervous system. There are some great pioneering efforts going on. And one of the goals of this conference is to take examples of pioneering work, and talk about them to all of you. So that you can benefit by the work that's been done already. And we want this to be an ongoing process of really fertilizing the best ideas.

I think a few governments really have a clear dedication to doing this right, doing it across the departments, using the right equipment. And so in three or four years, perhaps some country will have a full-blown digital nervous system, and we'll really be able to see the massive benefits that flow from that. During this conference, we're trying to be very concrete with the examples. In the U.S., actually the federal level has been a little slow to jump into this. The states, partly because they're smaller, partly because a few of them want to take leadership, have done well. Many of them are publishing information, and some are doing transactions over the network. So there has been an elimination of paperwork, an improvement in procurement efficiency, things like jobs, data banks, things like government requests. Some of those are online. Many things are not online yet, you know, applying for a trademark is not online. The patent office is still very paperwork intensive, although there is a plan to get that online.

In Germany, one thing that's been done well is having these archive records be available in digital form. There is a general principle in most countries that citizens can get at information whenever they'd like to. And you'd like to make it so they don't have to pay large fees to do that, but you don't want a big expense for that as well. By having these archival systems be electronic, the ability to find the information you care about is dramatically improved. And Germany has been a leader there.

In Asia there has been a lot of work on the medical area. And when you think of the medical system, over a third of the money spent on it is information management. It's nothing to do with medicine or patient care. It's a system that's ripe for dramatic improvement, particularly in countries where the government plays such a central role in medicine. It can mandate standards in electronic exchange of information, and really the government is the only organization that's in a position to get this going, because any of the individual hospitals or doctors simply can't do it, unless everybody they're interacting with is on an electronic system.

In the United Kingdom, there is a great example of an area, Hampshire, that's using the Web to try and attract new businesses. They are making it easy for people who do business in that area to know everything that's going on, seeing the business parks, the spaces available to rent, the information about jobs, and the kind of people that are there. Certainly, in letting businesses take advantage of local resources, that's a model example.

A lot of the examples we're going to talk about represent government interacting with businesses. But many examples will be government interacting with citizens. Businesses, of course, will move to have PCs and use the Internet as a common tool more rapidly than all of the citizens will. But, both of these are very important target audiences.

For a citizen, the type of thing they ought to be able to easily do, I'm showing here, you know, look at police records, and understand the crime statistics. That, today, is actually public information, but you're not going to take the time to go through and look at that. Simple things like renewing things, being reminded in advance, in many cases not having to actually go somewhere to do it, and even if you do, make it easy to schedule, so you're not waiting in line when you get there.

Initiatives like recycling, data about property transactions, information about assessed tax rates, and property transactions, again, it's a matter of public record, but you have to pay very expensive intermediaries now to go search that information for you. Why shouldn't it be on line, why shouldn't it be very, very simple to call that up and see even a map that shows all of it? Well, the software to do that is easily available.

We're going to be talking a lot about schools. It's a particular theme of this conference, because we think in the long run, getting all the students to be excited about these systems and comfortable with the technology is a great investment in the future. And we've -- we're very interested in helping out with those school-based systems and so we'll be talking about that.

One of my favorite examples is up in Ireland. And you will note in this conference, a little bit, that some of the smaller countries, or some of the, you know, states or regions, have been the real innovators. That's not surprising. It's a little bit easier for them to get consensus and get going. And we should all look at that very openly and take advantage of the fact that they're out there, charging ahead.

In Ireland, what they've done is they've built a PC-based kiosk, and put it into all of the post office sites, which is over 1,000 locations. Now, here is a case where the concept of the post office is a very broad one. There are an amazing number of things people go to the post office to do. Here you can pay your bills, you can apply for a passport, any type of license, look at your pension status, medical records, even buy stamps from these systems. And this has only been rolled out in the last couple of years, and it's a big success any way that you measure it.

New applications are being put on the system. It was put together very rapidly. The user satisfaction is high. Over half the population is using the system weekly, and so the user interface, which is easy to tune, was done in a way that people found very accessible.

This concept is being embraced in a lot of other locations; Singapore, London, and I expect it to be a very typical kind of application where you do a broad set of things through one infrastructure. The State of Victoria in Australia has been a real pioneer in driving this by looking at when people interact with the government, and sitting down and talking to them about their experiences -- was it easy for you to find what you needed to do? How many offices did you have to call? Where were you confused?

So, instead of forcing this citizen to think in terms of the various departments, they have defined an interface where you just talk about what's the life event that's brought you in that you want to do something. For example, turning voting age -- and it has all the different things you do at that point. Changing address, getting married, starting a company, so it can start with your needs and navigate you through all of the different government organizations.

Part of the beauty of doing it this way is, you never enter the same information more than one time. And, the information is filed with the different levels of government without you having to think about the state level and the national level, and whether there's differing requirements that take place there. Often, if you're employing people, the forms are very complex. I know here in the United States, a lot of small organizations or households that employ one or two people find it quite difficult to fill out all of those different forms. And here electronically that could be made to be a very straightforward thing. When you hire a new employee, you simply fill out electronic forms, and it steps you through the system.

You might ask, what's missing? What's holding this back? The answer you're going to hear over these couple of days is that basically there's nothing holding these systems back. There are some new things that need to be done for the more far-out applications, but today the Internet standards are there, the PC standards are there, the proper building blocks are there. The only thing that holds people back is just seeing the examples of the pioneers so they can charge forward.

However, if you want to do things like video conferencing and have very, very high bandwidth, that is a tough challenge. At the technology level, this idea of using optic fiber and putting unbelievable amounts of information on the fiber, that is moving ahead at the same exponential rate that all these other technologies are. But, actually running that fiber, having the encouragement of organizations to do that, and then digging the ditches and really putting it in, that's going to take a long time.

In most cases, businesses will be connected up at very high speeds in the foreseeable future. But people in their houses, in most cases, will have to continue to rely on the phone network to dial up. Now, the speeds you get across the phone network is adequate for pages of text or even pictures. It's not adequate for video type interaction. Now, in most applications you don't need video interaction. Distance learning, you do want video interaction, certain service functions, it would be nice to have the video interaction. So there's a lot of investment that needs to be made, and we'll be talking about how we see that rolling out.

There are very clever schemes of using infrastructure that are already in place, things like DSL that uses a telephone infrastructure, things like PC cable modems that use the cable infrastructure, and there are also revolutionary approaches where you use new, very high end spectrum for wireless kinds of approaches.

The most radical and most exciting development is the so-called low earth orbit satellite. There are a number of companies that are pursuing this, and these systems will probably be launched in three to five years. There's three or four, I'm an investor in one of them. But whichever one gets going will bring something new to the equation, which is the ability to connect up at high speeds every point on the globe. If you think of your rural areas where you're not going to be able to ever run fiber, once these satellite systems are in place, they will be able to connect up with the same type of high speed and high reliability as citizens who are in the urban area. So, investing in fiber for the high-density urban areas is perfectly complementary to being willing to take advantage of these satellite systems that will deal with the low-density areas.

The users won't even have to know which kind of network they are hooked up to. It will be completely transparent, whether it's the satellite system or the fiber, and everyone will be able to connect together. And so, those satellite providers will come in and make it very low cost to connect up schools, hospitals, libraries, and even the residences out in the more remote parts of the country. And that's very important politically in many countries where you want to have equal access between rural and urban.

There are some legal issues that the broad digital nervous system brings up, things about encryption, things about allowing transactions to be done electronically. Most countries are moving to really clear the barriers away, whether it's the U.S. or Europe or Asia, there's a recognition that this is great to have this happen, and the rules are coming into place. The one area that unfortunately is still controversial is encryption because if you allow information to be protected for everyone, criminals can also protect information. And so we see a debate going on about what information should be insecure, and is there a way to deal with it?

Again, I don't think that will block things, but that's the one area where there is some controversy. In terms of running infrastructure there are some countries that have gotten out in front. And the United States is not among them, although the U.S. is doing good things with cable modems and DSL, in terms of running fiber to homes, which is sort of the ultimate, the fiber gives you so much bandwidth that even today we can't imagine how we'd need more that. Here we have parts of Australia, we have Stockholm, we have several places that want to run a lot of fiber.

The furthest along is Singapore. They have a lot of advantages. It's a high-income country. It's very high-density. And they've really taken on as a political priority to make sure that they lead in the information age. Half its citizens have PCs today. And the information is really already in place. They have many companies coming down to take advantage of this and do neat applications. Microsoft has created a special group in Singapore just to focus on this. Actually a law says new housing that's put up, that it's required just like electricity and water that high bandwidth access be built in there. So, when you put up an application, you know everybody has access.

A lot of exciting things are going on. Our applications we're doing with partners in Singapore. This government has even gone so far as to say they want judicial records and judicial processes to all be done electronically. And so it's the most comprehensive approach to saying, let's get everything into electronic form.

While I'm talking about using the computers in such a fundamental way, the natural question that comes up is, does this create a gap, a gap between people who have access, and people who don't? And that's got to be a serious concern. If you look at where the new jobs are going to be and how important this system is, you really want everybody to have access. Now, the way to do that is in the long-run, by getting it into the educational process, even using this as a learning tool where kids can reach out and pursue their curiosity in new ways.

It's very similar to what happened with books. Books were originally available only into an elite group, and they were controlling things and had all the information, but eventually it was decided everybody should have books. And in different countries, different approaches were used to make sure that a library system was rolled out that was free of charge.

Here we really need to do the same thing, not overnight, but certainly in the next decade. Schools should all have these systems. Libraries should all have these systems. And that's going to require real political will, and lots of private companies coming in to get involved.

In England, there's an example where the schools actually have taken what they've done, their electronic mail and their publishing and the students' enthusiasm for helping them make all this work and they've become a hub not just for education, but for a lot of things going on in that community. It's amazing how students learn these systems and can help drive them. Sometimes they're the ones who are teaching even the teachers or the people in the community what's possible, and certainly that's been the case here.

By having this system, they've really connected the school into the community in a new way. Parents can see what their kids are studying. They can see what's going on. They can even exchange electronic mail with the teacher. So, in-between parent-teacher conferences, they're kind of up-to-date with what's going on, and it's a much better process.

Another example is one from the United States in Oklahoma. This is a case where the schools were very aggressive in putting in very strong infrastructure, hooking up all the elementary schools, and making sure the PCs were there. And with this infrastructure, they are now set up to do distance learning. The video conferencing works extremely well. And so they're able to do virtual field trips where they go and have somebody with a camera go around and look at new and interesting things and put that up to make it acceptable to all of those students.

And so you can really build something that is as engaging or more engaging than TV is. And, of course, that's what these students are used to. So when you've just be using the chalkboard to put up information, it just doesn't draw them in because they're used to the production of video and the ability to really see things. Well, with this system, there's some wonderful ways that it's being used for students.

Another set of examples that we'll be talking about is where you go the full distance and give each student a portable computer. Now, that is an expensive approach. You've got to make sure you can fund it. But it really is dramatically different than just having a few PCs in a classroom. As long as there are only a few, you have to line up for them, you get limited time, and you end up some of the students feel comfortable and better at it, and others feel like, well, maybe I'm not good at this. When you give them all their own PCs that they can take home at night, there's a sense of ownership, a sense of involvement. If they feel like they're falling behind or confused about something at night, they have the PC, their friends can sit around with them and try different things out. And the most dramatic results, the biggest benefits of bringing the PC into the curriculum are when you have everybody with a portable machine.

Now, the price of those machines is coming down. Today, around the world there are probably three or four hundred schools that have done that. It started in Australia, but it's spread now to more and more schools. And that's something I am very enthusiastic about. You can't do it unless you get the parents involved, but in many cases they are so excited to be drawn in and get involved they're at the ones who really help make it happen.

One key point I want to make about all this is that for governments that think about jobs, getting these information systems in can make a very big difference. Whether it's simply having educational courses out on the network, or a jobs database out on the network, or training people specifically in information technology, there's very few businesses that you can say on a worldwide basis there's a shortage of workers, but in information technology that's absolutely the case. When we go out to our partners and talk to them and ask what's their top priority, what's the thing that's holding their business back, it's the ability to hire qualified people. And that is true in every country where we do business. There are literally hundreds of thousands of jobs available. These are high paying jobs. They're non-polluting jobs. They're fun jobs. They're really kind of ideal jobs.

We've had cooperation in Europe, with governments that have made employment a priority, to take some funds that we've come up with and use those to train out of work people in information technology. And we've been able to get thousands of people into high-tech jobs. There's a lot more that needs to be done. We've got a couple of initiatives, but we're very interested in public-private partnerships around the world to help this gap be filled, and get people into these jobs. Part of it is working together with the universities or the community colleges to get the curriculum to be great there, and so that the people are ready for those jobs.

Well, what's Microsoft's role in all this? We're convening this conference to be a facilitator. Our specific role is simply providing the very low cost, high-volume software building blocks that make it easy to build these systems. And so that means things like Windows, Office, BackOffice. We need to keep improving those building blocks. We need to make the PC easier to work with. We need to bring down the administration costs of having large networks of PCs. And we've got some breakthroughs there that are coming out. We're able to do that because we get a lot of feedback from users of PCs, and that's really what drives what we're up to.

We have been very enthused about the role we built in working on some of the pioneering projects and we're willing to put some of our most skilled people into helping pull those together. Because we're not a consulting company, our whole emphasis is on skills transfer, making sure that these projects can be done by the governments themselves without outside help as much as possible as similar projects come up in the future.

A key element of what I've been talking about is the personal computer. It is the device that has really allowed the Internet to flourish, having so many personal computers that can easily connect up. Governments are spending quite a bit on these, and we feel very good about the improvements in the next year that will reduce the total cost of ownership of having these PCs.

One thing to understand is that the rate of improvement is not just a numbers thing. It's not just saying it's 200 megahertz, 400 megahertz, 800 megahertz. The PC will change in a qualitative way. The PC 10 years from now will be a tablet like device you can carry around wherever you go and will be connected to wireless keyboards to enter information. You'll be able to talk to the computer, it will be able to talk to you. You'll be able to take notes with handwriting, or working directly on the surface of the display. So at a meeting like this, virtually everybody would have that tablet type device.

You won't have to worry about managing your PC information, whether you put information on one PC or another PC, it will all be brought together automatically for you. And so there's -- the breakthroughs will continue, in the same way you look back on the PC of 10 years ago and saying, what a limited machine that was, you'll also look back on what we have today the same way. Despite that, we'll have compatibility every step of the way. Applications you build today, those applications will run on the PC of tomorrow.

Now, connected up to this network, although the PC will be the most important device, there will actually be quite a variety of form factors. Whether you call them PCs or not, it's important to talk about each of these. When I say the full-blown PC, I mean you have a screen that let's you do word processing, do a spreadsheet, and sit close to that device. In the smaller categories, once you move down from a portable PC, you get the handheld PC, which has a small keyboard and there are a lot of these in the marketplace today. Now they're getting color displays, the ability to easily synchronize information, and connect to wireless networks. So that's an explosive category. And we have a lot of great partners involved in that, from Phillips, Compaq, Sharp, HP, Casio, companies around the world building those devices.

A new form factor for us was introduced just about a month ago, and these machines will ship in the next couple of months, and this is even smaller, it's called the Palm PC. I'm holding one here that actually comes from Casio, although once again, all of those manufacturers will be building these. This is one where you give up the keyboard, and you have the touch sensitive screen, that you can hand write on, and call up information very easily. Now, this is for your messages, your schedule, your name and address lists, and whenever you come near your normal PC, there is an infrared connector here that makes sure that information is brought totally up to date. Anything you enter here goes up to the PC and vice versa.

And so that's a nice little tool. These cost something around $300 or $400. The displays will keep getting better. The price will come down. So that is a very nice form factor, but complementary to the larger machine.

Another area that computers will be present is actually in cars. We showed just a month ago, the first model of what we call the Auto PC. Here you have a nice color display, but for the first time, we use speech as a way of controlling the device. So you can simply say what traffic information you want, what radio stations you want. And immediately it uses voice synthesis in that display to quickly give you the information. Eventually every car will have one of those. And it's a huge opportunity to manage traffic in a better way, and always be able to stay in touch, even when you're in your car.

Finally, another new form factor, that I think is incredibly important, is taking some of these same chips and electronics that we've put in the PC and putting those in the TV set, whether it's built in or it's in a box like a set top box, or Web TV. The beauty of using the TV display is twofold. First, you don't have to buy another display. Which, as the PC has been coming down in price, from $1000, to $800 next year, to $600, the cost of the display is a substantial part of that. So by using the TV we eliminate that. We build a simple device that costs about $200. It's not as powerful as the PC, but it does let you do electronic mail and it does let you get out onto the Internet.

Today, this Web TV device connects up through the phone lines, you dial out and connect up. The second big advantage of using the TV set is that you can integrate with the TV watching experience. So any show you're watching, you can get digital information, any ads, you can say you'd like to know more about them. You can see the TV guide in a nice way. And with the latest Web TV, which is called Web TV plus, we've really shown that integration in a powerful way.

Now, outside of the United States, we're going to work with partners on this, with cable companies and phone companies in particular, to get this into the market. Again, we do not do the hardware. It's consumer electronics companies, like Sony and Phillips, who actually do the hardware. We simply do the software, and then let any hardware manufacturer come in.

As you move from phone connections to cable connections the speed and the ease of getting at the information is much greater. You don't have to wait for it to dial out. It's just there all the time. In fact, we have an indicator light that shows you whether there's mail waiting for you, or some new information.

Now, the software is going to have to improve very rapidly. The software that we make available today, certainly won't be the software people want in the future. We're spending a lot to make sure we're the company that provides for future needs. And we've got a lot of research that we do, some that's here, some in Cambridge, some in California. And we'll be expanding the number of sites we do research in, in the years to come. R&D investment has gone up very dramatically, and really the only thing that holds us back from making that go up even faster is simply the hiring of the great people.

The goals here include the very ambitious goals, of computers that see and can tell who's using them. The computers now, these cameras that are used for video conferencing are very low cost, and that ability to know who's there and interpret their gestures will be important. The ability to listen, the ability to learn, so that you don't have to work with a computer on a low level. We are making great progress on those things. So it's going to be a much more natural interaction. And it's going to be a technology that is not as complex or as intimidating as machines are today. I think anybody can use today's machines, but we'll make that even easier.

So what are the opportunities? the opportunities are to have the government spur the businesses in your country to be leaders in this area. To me, embracing the digital nervous system will make the difference between what companies are competitive and not competitive. This is particularly true in businesses like financial services, where managing information is everything you do. But, even in other areas -- take the car manufacturers -- the ability to do the designs quickly, and coordinate the factories and get the customer feedback, the information systems will separate out the winners from the losers. And governments can help, particularly in setting a model for small and medium sized organizations that are the backbone of basic -- most economies around the world.

We can make government more accessible. We can make government get more feedback, easily, from its citizens, through single point of contact. When you design those kiosks, it's easy to interview people, and help understand what they need, and make government really something they feel like is listening to their requirements. Getting the investments in the telecommunications infrastructure, that's a tough problem. We'll be talking about that more.

Getting the technology into education, to both improve education, and have people that come out of education really help make sure these systems are used well is very important. And it's important to remember that the way we think about these systems today is nothing compared to what they'll be in the future.

The opportunity for continued invention and innovation is very, very clear. This will be the engine of growth in all economies. And Microsoft is very pleased to have the opportunity to work with all of you to make this a reality in your country. Thank you.

Well, now I'm going to ask Daniel Franklin to come on out. He's the editorial director of the Economist intelligence unit. And he'll take some time to ask some of the questions that you've submitted.

MR. FRANKLIN: I'm going to be incorporating questions that a lot of you submitted by fax and email, before the conference. So thank you very much for that. That's been a great help.

Bill, you gave a very compelling vision of the future, with some very practical examples of how that's already being put in place. And I think a few of us here would question that that is a vision that, more or less, is going to come about. I do, however, have a question over the timing, and how fast it will actually come about. The pace, the speed, exponential, as you say, of develops in the private sector has been quite staggering. And that happens, because risk takers, such as yourself, are prepared to take risks. A lot of failures happen on the way. There's a lot of experimentation happening all the time. But, in the state sector, with governments, it's different. It's not a risk taking business, on the whole.

So my question is, how do you create a risk-taking environment for governments? How do you create the government entrepreneurs, if you like? And how do you overcome the resistance to change that I'm sure a lot of people in this room are actually experiencing in their daily environments?

MR. GATES: Well, a key point is that the risk involved is very low. Can you eliminate paperwork, and make things more efficient? Absolutely, there is no doubt about that. The real risk is to step forward and promote that idea, to say that you want to do things in a different way. People often resist change. And here you'll be changing a lot of jobs. There are a lot of people who process that paperwork, and the right way to re-deploy them is to make sure the quality of the service in the new system is very good. So some of those people, including the unions associated with those people, may want to hold back. You do need to dive in and do a system that people are going to have to believe that you can get improved ease of use. Prototypes can be very important for that.

We want to help those risk takers by arming them with examples from other parts of the world, so they can say that the costs are pretty clear, the ability to put it together quickly is very clear, and the reaction of the citizens to what it's possible to build is very clear. So in the government environment, we need to find some people who do think about new things, and who are sharing this vision of what goes on. I'm sure the people who came here, you know, part of the reason they came is to have that opportunity.

MR. FRANKLIN: And, as you say, providing concrete examples can be a great help, and this sort of occasion is an example of that. But, you're monitoring innovation around the world. Is there a clearinghouse, where people can keep up to speed with the very latest examples of best practice, or is there a need to create more information on the information revolution?

MR. GATES: Well, Microsoft has made of several vehicles, to help with this. First, is our Web site, www.microsoft.com. We have sections of our Web site today that are devoted to government. We're going to be beefing those up, not only with the examples you're going to see here, but new examples, as they come in. So every month, our goal is to put up at least five new examples of something that's being done well. And that site is very searchable, you know, so if you want to look at education, or medical records, or whatever, you'll be able to call those things up. I think that sometimes having an award makes sense.

In Europe, Commissioner Bingaman had an award process that he went through to get some of the city level governments to think about what they can do. That was valuable. Now, we need to have some more awards, recognition for people who are taking things to the next level. So Microsoft will do some of that. And I hope governmental entities, as well, will recognize the risk takers who went in and helped these systems come together.

There is a little bit of a sense of competition here. I went and spoke six months ago at a get together of all the governors of the United States. And before we went down, we wanted to make sure we were really picking the states who were doing it well, because we knew the governors whose states we didn't pick would think, oh boy, you know, are we behind. And so we highlighted, and it was only three or four states really doing things the right way. And, you know, that was wonderful, because the states we highlighted want to go to the next level. The ones that weren't doing anything want to get involved.

Politicians are more interested in information technology than ever before. When I spent last week in Europe, virtually every country I went to, I was meeting with Cabinet ministers, or in many cases the prime minister, talking about what should they do, how do they seize this opportunity? So the atmosphere is incredibly ripe. People are paying attention. You know, just to take France as an example, we had, except with very few groups, information technology wasn't on their radar screen. Last week, the prime minister actually went to a computer show and gave the keynote speech. So there is a dramatic change and that's pretty typical.

You mentioned competition. There was a topical example from one of the delegates here, in fact, from Argentina, who gives the example that they've organized the national olympics in Argentina, for students to come up with good ideas for Web sites. And there is a proposal to do an international olympic games among students for ideas. And the best ideas get posted up. So maybe that's the way of the future.

You mentioned right at the beginning of your talk doing away with paperwork, and I remember reading in one of your speeches that you actually did that in Microsoft. That you took all the forms, and you basically put everything online. Can you tell us how you did that and whether that experience is directly transferable to the government?

MR. GATES: We got rid of paper forms and things where you fill out fields. Those are ripe for elimination because those forms eventually are going to have to be entered into a computer anyway. If you fill them in electronically to start out with, you're going to have a lot more guidance in terms of how you fill them out. You can check to make sure what somebody's filling out makes sense, and that it's very consistent. You can go get information from other forms they've filled out and automatically fill in those things. But it's just a much simpler experience. And that's where having kiosks or letting them just use their PCs come in.

At Microsoft I got all of the paper forms in one place, and I was just stunned that Microsoft had over 1,000 paper forms. And I said to people, okay, let's get rid of every single one of these. Well, we did. We got rid of all the ones that were used internally. Now, we still have some that we have to file with governments, and so we've got an incentive to teach those governments to take that information electronically so that we can get to total purity of not having paper forms.

Now, getting rid of paper forms is not to say to get rid of all paper. You know, when you have a long document where you have lots and lots to read, the computer screen is still not good enough if you're going to read a very, very long document there. I read the Economist off of paper still, although I might go up and search the Web site to see what the latest articles are or, if I'm very anxious, I can get it there more rapidly than I get it through the mail. But when it comes down to sitting down and just going through everything, I read that off of paper. So long text documents, you can distribute them electronically, and you'll still be printing them out. The screen breakthroughs that will have us reading everything off the screen are out there in the future. But for forms, where you're filling in data, the time has already come where there is just no reason those should be left in --

MR. FRANKLIN: Was it easy to do? I mean, did you find resistance even within a company such as Microsoft where everybody is wired to doing that, or it's going to be harder for governments?

MR. GATES: Very little resistance. Take the human resources group -- if you wanted to change one of your payroll options, you had to call them up and get the form, and there were a lot of different forms, and when the government changed the rules, the forms would kind of get out of date. I mean, it's just a lot of trouble. Instead now, anything related to human resources, you just click on that on the home page. You see all the options they have there. You get guidance on all the pages. If you're unclear about what's going on, you can click and send electronic mail to somebody who is going to be very responsive about it. But because we have PCs everywhere in our company and because people are comfortable with using them, it happened very quickly.

A key thing for a government is to think about electronic mail. What I always say to businesses is that if you care at all about the information age, if you want your employees in any way to get ready for it, you should use electronic mail. But you have to get everyone to use it, or else it's really useless. If you don't -- if you send mail to somebody, you have to know that they're going to read it. And you have to have a directory that makes it easy to find everybody and have group names defined so that you can send out to lots of people. Governments are just starting to get into really using electronic mail. And that is a key step, a very big step. Because then, once you're using electronic mail, you realize why not send forms, information around that way. Why not receive it that way?

MR. FRANKLIN: But don't you have a problem whereby not everybody is on electronic mail? It's fine for Microsoft where the entire company is on email, but what about a country where the literacy rate may be low, or where only a very small percentage of people actually have access to a computer?

MR. GATES: Well, literacy, it's tough to get a lot of paper forms there. Eventually, we'll have speech recognition within the public kiosk, and so we'll be able to interact in that fashion. But, the lack of widespread use of computers can hold these things back. For business-to-business, I really don't think that is a huge problem in developed countries today. But the reason we talk about these kiosks, and the reason Ireland did it the way they did it, is to have universal access. And deciding where you need to have that will be very important.

Eventually, you know, everywhere where there's a pay phone, there eventually will be a screen type phone where you'll be able to do these things, but governments often have to step in and fund these kiosks.

MR. FRANKLIN: Now, since we're looking into the future, I came across in Cyber Times the following statement: Who knows, maybe one day Microsoft will decide to move to India to avoid the U.S. Justice Department. Now, I don't want you to speculate on a move to India, but I think this audience would be interested to hear your perspective on where things stand with that affair?

MR. GATES: Okay. Well, the key issue here is simply about Microsoft driving forward, putting more capabilities into the product, and as operating systems get richer and richer over time, you know, those R&D costs are there to make sure that everything people want to see in the computer is there. For example, building in support of the Internet. So that's a very standard evolution that people expect.

Our issue with the U.S. government is not going to be a major thing. In fact, it's been covered in some incredible way. I was surprised, when I was travelling around, how much it came up. There is an issue where they're saying they want us to ship not only fully featured products, but also crippled products. Now, they're not telling us that we can't ship the fully featured product, and certainly customers want the fully featured product. It could be a real drain to us and hold us back in our innovation to also have to ship the products with less features. And so we are pursing this case very seriously. But it's not something that's going to hold back Windows 98. It's not going to hold back Windows NT 5 or any of the things we're doing.

And so people can expect Microsoft, in particular, but perhaps more importantly the software industry to continue to add features that make software both more powerful and easier to use.

MR. FRANKLIN: And just in terms of your own time, one reads about it, as you say, an awful lot. It must be occupying a little bit of time and time that you could be spending, as you would like, more creatively elsewhere.

MR. GATES: Well, for me, I don't -- I'm not involved in the legal issues whatsoever. We have lawyers. This country is blessed with lots of lawyers.

(Laughter.)

MR. GATES: Which I don't necessarily recommend for other countries.

(Laughter.)

MR. GATES: So that, you know, they're off doing a good job on that. In recent months do spend maybe a quarter of my time really explaining what we're doing, because at first, we didn't spend enough time on this. A lot of people are thinking this suit relates to a lot of things that it has nothing to do with. And, you know, I think we need to get smarter about getting that message out there.

In terms of the people at Microsoft who really build the products and develop on this stuff, they're full speed ahead. I mean, they come into work every day, thinking about building great software. We know we have a lot that we need to do. And if we don't do it, there are plenty of people who are ready to take our place. That's the fun thing about our business. Some people don't like it, but no one has a guaranteed future. It's not like many other businesses where you know who the leader is going to be in 10 years. In our business, we could be out of business in 10 years, even though our current products are popular, we have all these great people, we've got to really drive forward based on what customers are telling us to maintain any type of position. And so, government should wish that every business was as competitive as the PC hardware and software business.

MR. FRANKLIN: Okay. Now, on the publicist side of what governments can do, I'm going to read you a question from a delegate. What are the most important steps governments can take to facilitate the operations of a company like Microsoft?

MR. GATES: Well, I'd say that we'd really highlight three things. One thing to understand about Microsoft is, our goal when we go into a country is primarily to make our partner successful. That means the software developers in that country, the integrators, the consultants, the training companies in that country, because we have decided not to be in those businesses, we can work with them on a very pure basis. That's different than some other computer companies that are really building up a large profit center around their services activities. And that's perfectly valid for them to do that, but it makes it harder for them to work with local partners. In our case, we come in and sign people up for partnership programs, and really work with them. That determines our local success.

So, when we go to the government, we say, what is the government doing in education? What is the government doing to make sure the communications networks are going to be powerful and reliable? And what is the government doing itself to be a model of how information technology can be very positive.

I have to say, around the world, we've had a great reception. Governments in general are very interested in any insights we can bring into this. They get groups of the local industry leaders and some of our people together to advise them on what some of the next steps should be. So, we've had a very, very positive relationship with all the governments around the world.

MR. FRANKLIN: I guess this is a related question. It also comes from a delegate. What are the most important criteria that are taken into account by Microsoft in deciding locations of future projects? Now, you've recently decided to locate a research facility in my country, in Cambridge, what in practice were the decisions that made you choose Cambridge of all places, you have the whole world. It's a bit cold there as well.

MR. GATES: Sure. We do most of our development at our headquarters location, and that's so that all the product groups can be working together with each other. So, even in the United States we are very, very centralized. Now, we've gotten to a size where we feel we have to diversify some of that. The majority of our work will always be in the one location. So there's no fundamental change to that.

But we are building up a number of development centers around the world. We're in Israel. We're in China. We're starting to build some things in India. The research group wanted to make sure that they were drawing on all the great research going on around the world. And so, we knew we wanted to locate in Europe, and not just to be in a single location there, but also to have an advisory board from some of the top universities to really tell us, where is the great government fount of research, where is the great entrepreneurial work going on? Contribute to some venture capital funds in each of those countries, so it's clear that we are rooting for the local entrepreneurs because, as they build on top of the PC platform, that's only good for Microsoft.

Cambridge was particularly attractive because the university wanted to work with us. And, they have a history of being involved not just with theoretical science, but also with practical science. And so we were able to put a building there that was outside the university, completely funded by us, but yet close enough that the doctoral, post-doctorate, the doctoral candidates could come in and do some of their projects working together with us. We also surveyed around Europe and found it was a location that would work for us to be able to draw people in from other countries.

And we will be doing other things in Europe, but the hub of our activities and the basis we're going to use to learn about other things is going to be out of Cambridge then. We don't yet have a research facility in Asia. At some point, we definitely need to do that. But we're just at the early stages of exploring that.

So, our development is going to be in many locations. The key is the quality of the university systems, and turning out lots of IT graduates that can do very strong work. Unfortunately, there is no country with a surplus. Even in India now, where they've done very well and they are exporting the work they do in connecting up over the Internet, even there a lot of people have discovered that and they're jumping in. So any country that gets their universities going, those people are going to have great jobs.

MR. FRANKLIN: Well, let's stick with education because that's obviously a core area where the digital nervous system is both affected by, in terms of investment decisions, and can be effective in delivering results to the education system. I want to read you a paragraph from a book by a gentleman called Dr. Fred Bennett, who is a great fan of computer-based solutions for education. But he points out that getting the system installed is only one step, can be a gimmick, even. If you think that just putting the computer in there is the answer, you're wrong. It's how you use it.

And he writes: Alexander the Great benefited by having a king, Phillip of Macedon as his father. The king was astute and had the power and foresight to recruit the brightest person in the known world, Aristotle, as the private tutor of his son. Not everyone can have a king as a father, but each student can have a private tutor with even more knowledge than Aristotle. This tutor is a computer. Of all the advantages that computers will deliver to education, foremost will be the capacity to act as an individual tutors.

Now, do you share that view?

MR. GATES: I definitely share that vision. Today, the computer is not -- because there's more software to be written, more material to be connected up, it's going a little bit too far to say that the computer is a tutor. But the computer lets you access all of the things you're curious about, and even find other students who are out there. And in different subject areas, that idea of it being a computer, that is happening. Skill software is selling very, very well.

In the United States, where we're going to pass over half the homes with PCs this year, in homes with children, it's over 70 percent. And so, there's a real correlation of parents saying, boy, if I want my kids to be able to reach out and pursue new things, getting a PC is a good step in being able to do that.

The number of pieces that you've got to put together to get the full educational benefit are quite strong. You know, just buying hardware is one step, but you've got to get those computers connected up, you've got to get the teachers feeling enthusiastic about it, and embracing it, which in many cases the opposite happens. I believe that getting the teachers a portable computer so, over the summer months, or however they want they can play around with it and get comfortable with it, that is a well-spent investment.

And then you have to get the curriculum to change to get the textbooks to talk about going out to the Internet and try things out. And you have to think about structuring the school day so that if kids have found things that confuse them, or excite them, that they bring those in and, as a group, the class can pull it up if they have a screen they can all look at, and talk about what's been found there.

So the ability of the machine to see what you're confused about, to adapt to your learning style, to give you immediate feedback about what's right and what's wrong, and let you have the same access that the most privileged student in the world has, because it's all on the Internet, that possibility is there. But there are quite a few pieces necessary to make it happen.

MR. FRANKLIN: I mean, there are presumably a lot of very exciting pilot projects going on around the world, certainly around this country, but are we doing enough to share those experiences, to get the best practice adopted and to show what really works?

MR. GATES: No, I don't think so. I think that there hasn't been a place you could turn to see the best educational activities, even inside the United States, it's very hard to do that. And this country is not doing nearly as many experiments as it should. Our public educational systems are a little rigid in terms of being able to try new things. And so it's another case where, despite the US's lead in technology and purchasing technology, I don't think that we're that unique, in terms of having ideas about how this can be used in education.

So in this conference, we're going to be sharing examples, again, the Web site, there will be a special part of the Web site that is related to education. But, I think there have to be awards. I think there has to be ongoing conferences. I think there is a lot more that has to be done. The beauty of the world having so many countries is that we can learn from each other. And when things move quickly, we have to really put some of our investment into spreading the word.

MR. FRANKLIN: The delegates have a lot of questions about how Microsoft in particular is spreading the word on education. What does Microsoft consider to be its core products, for education, in the light of the global impact of the Internet?

MR. GATES: Well, we believe that the products that students should use should be the same products they're going to encounter when they get out into the business world. And so you don't want to really limit, or take things in a different direction. It's the same Internet, the same Windows, it's the same productivity software. And so really your hardcore products used in education are Windows, and Office. And then the server-based software that can connect all those networks together.

We do some special things, for example, in many languages we have an encyclopedia called Encarta. We do a teacher's training guide, of how can we bring that into the classroom, and help bring subjects to life. So there are some add on elements that are there. We are one of many companies that are doing teaching software, to teach young kids math and reading, through the computer. We've only done that in English, so far. But, as we make that successful we'll be spreading that out into other languages as well.

At the core of though, we're able to take the work that we do for the commercial sector, and price it in a dramatically lower way, and do some add-ons to it, and make that available to schools.

MR. FRANKLIN: I don't know whether this is a pointed question, or someone who is, I believe, technically still on very long leave from Harvard, but can Bill define how Microsoft will promote lifelong learning opportunities?

MR. GATES: Well, I certainly believe people should go to school, and graduate from school. Ninety-nine percent of the people Microsoft hires are college graduates. In my case, I just -- you know, I felt there was an urgent need to be the first, and probably it was a good decision. There are two things that facilitate lifelong learning. One is that when you're young, if you feel like your curiosity is reinforced, there is a way to go out and learn new things. And if you want to try out your ideas, you can get feedback, and find other people to do these things with you.

If you have optimism about that, then you're going to retain your role as a student for a long, long time. And the Internet is at the center of this. Almost any hobby, or interest you have, there is an amazing number of things, including bulletin boards, and chat areas, on the Internet today. You know, if a friend of yours is working in a certain area, going to a new place, somebody you know is sick, and you want to learn about that disease, and what new is being done there, the Internet is the place to turn to.

And so what we can do is tell adults that even though they're kind of reluctant to get on the computer that if they'll put in three or four hours and relate it to the things they care about, soon they'll be telling their friends to get involved as well.

Older people are some of the most active users of these systems. And they get incredible benefit, particularly if they can't get around or can't go to see their grand kids, being able to send mail, exchange pictures, contribute their knowledge on the forums out there is great for them. Having older people brought into this is definitely something we need to do better on.

MR. FRANKLIN: You mentioned the shortage of IT people, IT skilled workers, perhaps this relates to that. What is your advice, asks one delegate, for young people to make up their mind what to learn and what to study, to handle the requirements of a job in the 21st Century?

MR. GATES: Well, I think no matter what area you're working in, the computer will be an important tool. That's one thing that the U.S. universities did well, is they didn't confine the computer access and the Internet to the engineering or science departments of the universities. Even if you work in the humanities, the only way to sign up for a class is to get onto the Internet. The only way to get your grade is interacted through the Internet. So they -- the universities were pioneers in eliminating paperwork, because they have the density there. And I don't think universities outside the U.S. have been as aggressive on this. I really think they should.

In terms of what kids should learn, basically learning the things that are interesting to you, so you're going to enjoy your job for a long time. Understanding how the computer works is valuable, so you can know its limitations. There are a lot of people who, they'll take a few computer courses and may think, well these computers are magical. They can do everything. Well, they cannot do everything. They are just a tool. They can't think and reason. They're simply an extension for our minds, the same way that a tractor or a shovel is an extension for our muscle power. So it's up to us how we apply them.

And having some common sense about what they do well, they don't do well, I think that will be worthwhile, no matter what profession people are going into. So a little bit of exposure to computers, and then going after the subject areas that you really think you're going to want to spend your life working on.

MR. FRANKLIN: You talked about competitiveness, and the fact that countries that are most advanced are going to be the ones that are fastest with the information revolution. One of the delegates asked, can you give examples of initiatives by governments, which you believe have contributed significantly to the country's competitiveness in the information age? You mentioned Singapore, of course, as one example. But, are there others?

MR. GATES: Well, I think there are examples of almost any country you can think of. You know, why is India creating these jobs? Well, partly because they have high IT and some other educational institutions that are world class institutions. And that -- you know, that's a very helpful thing. I think every government is starting to put computers into schools. And so we're starting down that path. The real standouts are mostly the community level activities, or the smaller governments.

MR. FRANKLIN: Is small beautiful in this game?

MR. GATES: In the early days, where you can get things going. If you have a large country, then you have a hard time, you know, the project is so big, and people are so worried about what it means, and if you just pick one part of the country, people are worried, are you discriminating in favor of that part of the country. And so, yes, the smaller countries, I think, are going to stay at the forefront for some time. And I love these bottom-up initiatives, where you start at the city level, or the regional level, because a lot of the opportunities are there.

The European Commission has funded a project that we're involved in, working with a number of companies in Europe, where six different cities around Europe have been picked to bring up Web sites, and use them for neat new things. And there will be a lot of visibility of how those pilot projects go. So I think that is a good framework, that the EC is now funding nice activities. And they have made it a very visible thing. So I would commend them for what's being done in that area.

Some countries are just beginning to look at this. The deregulation in communications, that's almost a global phenomenon, that is very, very important. If you have very bad phone systems, then it's tough even for businesses to come in and participate in this. You have to have very reliable phone systems with lots of new options. And the competition is bringing that.

MR. FRANKLIN: We're almost out of time. So I think this will have to be the last question. But, it relates, I think, to a very important issue going forward. You mentioned what I believe is referred to as Moore's Law that the computing power doubles at least every two years.

One of the delegates points to recent research, which suggests that maybe this won't hold in the future, going forward, that there are physical constraints on that exponential growth continuing to apply. And did I detect in your presentation that you were shortening from 20 to 10 years, the period over which you think Moore's Law is going to continue to operate?

MR. GATES: No, I would say that it's extremely likely Moore's Law will continue to operate for the next 20 years. It is true that, if we just keep doing brute force, shrinkage with the current fabrication techniques, you do run into some problems in 8 to 10 years. And that there is some cleverness required. But, if you go back to any point in the history of the electronics industry, you would have heard the same thing, that okay, we sort of knew how to extend for 8 to 10 years, but new techniques would have to come along.

There are at least five or six breakthrough techniques, new materials, new ways of doing the printing, with electronic gating, and some things that rely on very quantum effects to do the computing. There are plenty of good ideas out there. And the economic awards for the people who make those breakthroughs are so gigantic that I can say there is 100 percent probability for the next 10 years there is really high probability that we'll be able to continue to drive these things forward.

And the power that we're talking about there, you know, there's no doubt that machine is good enough for speech recognition, visual recognition, learning. I mean, even the machines six to eight years out, I feel very comfortable will be doing all of those things.

MR. FRANKLIN: I'll take 96 percent. I think what's striking about this whole area is the optimism that pervades it. What we're really talking about is astonishingly powerful techniques for improving the human condition.

There were many more questions, unfortunately, we're out of time, but there will be time this evening to speak to you at the reception tonight. Bill Gates, thank you very much.

MR. GATES: Thank you.

(End of presentation.)

NOTE: Links on these pages point to servers that are not under the control of Microsoft Corporation.