Remarks by Bill Gates

Georgetown University School of Business

March 24, 1999

MR. GATES: Thank you. Well, good morning. It's great to be back here, and it's great to be done with another book. I always feel a little strange speaking to classes like this because, of course, I'm a dropout. Microsoft doesn't hire many dropouts, I want to be clear on that. I'm a big believer in education, including the great value of getting an MBA.

What I want to talk about today are the changes taking place in business. And I think this is a fantastic time to be entering the business world, because business is going to change more in the next 10 years than it has in the last 50. I believe the rules of how a business works are going to be so different that only the companies that seize the opportunity to do things a new way will be the ones that are successful in the years ahead.

Now, this all grows out of the miracle technologies that created the personal computer, the chips that double in power every two years that we've had a million times improvement in the type of computing that somebody can have on their desk. It's when we combine this with the miracle of the Internet, high speed connectivity across the entire globe, and standard ways that that kind of communication network can be used for entertainment, for learning and for business, that's where we get the revolution. That's where we get something that really is unheard of.

We talk about capitalism as the matching of buyers and sellers. Well, before the Internet came along, there was actually a great deal of inefficiency there. If you had a fairly obscure product, there was no way that you could find a distribution channel that would get that product out and get it to be popular. It's been fascinating to see the number of new small businesses that have emerged simply because the Internet lets them find their buyers like they couldn't have before.

There's a little company up in Wisconsin which was actually just a part-time hobby. They refurbish old phones and sell people neat old phones. And the total worldwide market for that may be about five to ten thousand units a year. But as soon as they set up their Web site, instead of just having a few friends that happened to hear about them, they were overwhelmed with business. And so they quit their normal jobs and have started a business that's now thriving. So the breadth of opportunity here, I think, has been greatly underestimated.

Now, as I was writing the book and thinking about the milestone that we have here, I decided that some new terms were important to really capture the new behavior. For consumers, I talk about the Web lifestyle, and this is the idea of people who just take the Internet for granted. They use it many times a day, and it's not anymore notable to them than it is to use the phone or a car or any of the other things that are simply part of everyday life.

And once you're using the Internet this way, no matter what really got you involved in the first place, whether it was staying in touch with friends, or learning new things, or using it to make better purchases, you want to use it very broadly. You want to apply it to everything because it is such a great tool. And so people tend to be almost evangelistic about it. If you way to somebody living the Web lifestyle, you ask them how it could be better, and they'll have dozens of things to say about how they get busy signals, or how the software could be simpler. But if you ask them, would you give it up, would you like to go back to doing things without it, they're vehement that it's critical to them, and they'd like to see everybody get involved in it.

And this kind of evangelism, getting people onto electronic mail, getting them to do personal Web pages, saying to the doctors or businesses you work with, how come I couldn't do the transaction or look at the status of my transaction across the Internet, that's partly what's pulling people in and making this such a transformation, not just a small group of people, but eventually the entire buying population.

Today, the Web lifestyle is just at the beginning. I'd say that in the United States, maybe four or five percent of the population is really doing this in its fullest sense. Of course, students are the leading edge here, it's students that are really driving this into business, and I'm sure many of you will be change agents as you go into your employers and ask, how come we're not using these tools in a more profound way?

I talk to CEOs all the time, and they say, gee, I never learned to type, and when I started my career, it wasn't considered prestigious to be the one who typed your own things and do I really have to do this myself? Do I have to sit down? Can't I have my secretary just kind of print my messages out, and I'll kind of scribble on them, and then have the secretary put them back in? And I say, there's no way that management, and particularly top management, can get into their mind how different this all is, and what it means for their company unless they're personally using it.

We've had some companies really take this to heart. Marriott, for example, made sure that all the top executives got PCs, and they even gave them homework. They said, okay, you have to go buy books, you have to organize some travel, you've got to file your tax return that way, and come back and tell us how that worked, and what it made you think about how things could work in a better way.

So, really getting everybody to see this as the leading edge, and encouraging companies to get out there early, is very important. If you take something like book selling, Amazon.com has become the poster child of electronic commerce, partly because the traditional bookstores simply didn't go out onto the Internet. That gave them over two years to get established as the place to go. Now, Barnes and Noble, and others, are coming on, but it's an uphill battle for them because Amazon is down the learning curve, and it certainly has a broad base of users.

Another term that I created anew for the book is Web workstyle. And this talks in particular about how employees expect to have information provided to them as part of their job. I've always been amazed at how hard information is to get inside businesses. And just think about the most basic obvious piece of data in a business, which is the sales data. And let's say you have a creative idea where you think the pricing could be better, or you think there should be move investment in a particular product, or a particular geography. Well, if you want to analyze that idea, you ought to be able to take the sales data over a period of years, and have it right there at your fingertips, be able to look at it by geography, compare it to expectations, to look at where different programs were tried, and where they weren't tried, and make those comparisons.

And then, if you see something interesting, you ought to be able to mail that around to someone else, a colleague, and ask, do you agree with this analysis? And doesn't this suggest we ought to try something else?

Well, that isn't happening in 95 percent of businesses today. Instead, the sales figures are printed out at the end of the month on a piece of paper. And because it's on a piece of paper, if you look at the number and say, wow, that number is larger than I expected, or that one is smaller, what do you do? Do you call someone up? You might find out that it's just the way they categorized something, and it's really too much trouble to dig into those. And so it suppresses thinking that ought to go on. It's not empowering to all the knowledge workers that these businesses are investing so much in.

And the incredible thing is that these same businesses that don't do that actually have made the investment to put personal computers on all these desktops. And so, just to run spreadsheets or word processors, or whatever, they've made the investment. They may not run the electronic mail system quite right, so there might be a little more they need to do there. But fundamentally, the infrastructure is in place, and the thing that's missing is for management to think through, what is the information that counts in this business, and how do we get that out to everybody so they can do their job in a better way.

So, I use this term, Web workstyle, to refer to somebody who has got very high expectations for the information that they should be provided, who expects to be able to really think through the business, not just wait for top management to initiate something, but really they're in command of the facts and able to make a much bigger contribution.

So, Web workstyle is also something that is just at the very beginning. In fact, I don't think any company is receiving the full potential of how things can be done on a digital basis. Even the most fundamental thing about getting rid of paper forms, when I talk to executives about that, that's an eye-opener for them. It was about two years ago at Microsoft where I was given one of these forms to fill out, which was about adding new personnel. It was very complicated, and I couldn't understand some of the fields that I was supposed to fill in about transfers and budgets and things. And I thought, gee, we probably have a lot of people who are confused by this form. You know, if I can't understand it, what's going on.

And so, I asked for every paper form in the company to be put in binders and brought to me. And it took four or five binders because there were hundreds of them. And so I did something that really only the CEO can do which is, I said, get rid of them. We don't want these forms because we're filling in redundant information, and having those things online allows you to get feedback. If the form is confusing, for every form you can have the person you send mail to and explain how they can make it better. Every field can be fully explained. And you don't find yourself doing things that are at all redundant.

And so we were able to get rid of most of those forms. In fact, the only ones we have left are the ones that we have to fill out to send to the government. So now we have to work on the government to get them to take a digital approach and be able to let us go all the way.

Even when we buy things today, if you sit in your office and you say you need a new personal computer or you want a marketing plan, that's done electronically. There's no such thing as a paper invoice or a paper purchase order, because those are simply error prone things that create a lot of jobs just to shuffle those things around. Imagine a company getting their phone bill. Well, it's on paper, what you really want is that information to immediately be associated with the cost centers that you're tracking, and to have some logic that says, is this out of line, is this unusual? Should somebody receive a piece of electronic mail suggesting that the pattern of spending is changing here, or should it just go through the system and nobody has to pay much attention to it? And so these are the kind of common sense things that get you going down the path to having an organization that really uses these digital approaches.

I call these organizational advances a "digital nervous system," because the only analogy that I could think of where information is transferred very efficiently, and you really pay attention to the things that count is the human nervous system. And organizations, in their own very inefficient way, have to deal with a lot of inputs. They have to prioritize those inputs, and they're excellence will vary according to how responsive they are to the changes taking place.

So, I believe that businesses ought to put together this digital nervous system, and in the book I try and be as concrete as I can about how you would measure a company and say, are they doing this the right way. I also spend a lot of the book giving examples of companies that I think are providing leadership. Some companies that are really tracking their information now, and have gotten a lot of competitive advantages out of it.

One of the basic assumptions is that the Internet will change everything. If people don't believe that, then I'm sure they're not going to find the book interesting at all. You have to take it as a given that this revolutionary communications tool will transform business. And the place it will happen most rapidly is businesses doing business with each other. You tend to read in the press mostly about consumers buying on the Internet, and that is very exciting, that will happen, travel, insurance, everything that you used to go to retail primarily to get. But because it requires people to be comfortable with the computer, and because most of those consumers are still dealing with phone dial-up connections, that's going to be a longer time frame, perhaps as much as 10 years to get the vast majority of people, even in the United States.

Business to business is on a much faster time track, because anything that provides efficiency speaks to what business is all about. And so the competitive impetus will cause people to really move in this direction, I think, in the next two or three years, the majority of transactions will move this way. When you work with your bank, when you want to look at your portfolio or your cash collection, you'll expect to do that electronically, there won't be any paper involved in those processes.

And other key principle that I think is important is that every worker is a knowledge worker. That is, every worker should be given a full set of information tools. The kinds of jobs that were typical in the economy in the past, where you're simply taking paperwork and entering it into the computer, or customers calling up and asking about the status of their account, those jobs won't be necessary. Those jobs can be transformed into higher value added. So that instead of just simply telling a customer what's going on, you can actually have enough freedom, and the data that allows you to really help the customer with a complex problem.

If they have a simple inquiry, they'll just do it through self-service. You know, they'll have a little screen on their cell phone to call up the information, or they'll have their tablet PC that they carry around. It will be very easy for them to get the information.

I also think a central principle is to put customers at the center. A company should be able to call up every interaction with a customer in just a few seconds. Any employee can type in the name of that customer and see all the things going on with them, and not just for one division or one set of products, not just their billing status, but any phone call, any meeting, any insight that somebody in the company has about what that customer is thinking or what they might need should be online and easily accessible. And it's only through that that you can provide the best customer service, only through that can you think about what additional business should we be doing with this customer. And it's fairly stunning how few companies really have that customer-centered view and have been able to create it.

A lot of companies come to me and say, well, we're going to do this on our Web site. We're going to go from having a Web site that's just a lot of text to letting customers come in and see all their information, and it's it nice, we're doing that on our Web site. Well, I say, then the Web site is your business, you shouldn't have any other ways of looking at the information. When the customer calls you on the phone, they want you to be looking at the same screens, the same information that they're looking at. When they come and see you face to face, you should be dealing with that same information. So, all the front ends that you've had in the past, those are obsolete. The Web view is the one that should define how you're dealing with this information. And all you need to add to it is a little bit of additional data that you might want to keep internal, but otherwise the whole navigation and the integration should be used in all cases where somebody is thinking about that customer.

A final principle, and one that often surprises people, is one I think that I think this nervous system should be designed so that bad news travels fast. When we first got our electronic mail system, I started getting messages where somebody would say, this is fantastic, we just won the XYZ account. And of course, I'd send them back a nice piece of mail saying, hey, good job, it's great we won XYZ. But, after I'd send that mail I'd often say to myself, I didn't get any other emails today about accounts that we won. Did we lose every other account? You know, people weren't sending messages that we were losing accounts, because it's just not human nature to send that kind of thing along. You know, if they sent that mail what kind of response did they want, you know? Great job. Try not to do that again.

But, in fact, what's really actionable? If you're going to change your priorities, if you're going to reallocate your resources, if you're going to rethink your pricing, what should cause you to do that? Well, not good news. Good news just confirms that the plan you're on is a good plan, and you ought to go full speed ahead. But, rather if you can get early warning signs that your most demanding customers, your leading edge customers, they expect some additional features, or that a competitor is doing something that's particularly clever, if you can get that early, then the organization instead of responding in a way that it's somebody's fault, or somebody should feel bad, the organization can have that as a common piece of knowledge and really reconsider their plan.

Sometimes the response, to be effective, needs to come within 24 hours. Sometimes it requires taking lots of those data points, going off on retreats, thinking it through, a period of months before you reformulate a new plan. But, whatever it is, the sooner you get that bad news and really pattern match it with all the other things going on, the better off you are. And there are ways to design this digital information flow that makes that natural and guarantees that that's going to happen.

Now, one of the incredible things about all rapid changes is that there's a tendency for there to be resistance. There's a tendency for people to want to do things the old way. And we're definitely seeing that here with really the most rapid change that's ever taken place. If you think about previous advances, you know, the use of the phone, it took several generations, there was a generation where they were kind of hearing about it, a generation where it was an expensive thing, and then a generation that took it for granted. Well, this transformation will happen in less than the span of one generation. And so the need for people to really get people to think more open-mindedly is far more important here.

There's no doubt that the United States leads in these changes. Our companies, as little of the potential as they've realized, are actually doing more than companies in any other country. And 100 percent of the credit for that goes to the students that are moving into these companies and really driving the change. The Internet itself grew out of a set of work funded by the government to have a university network. In fact, when I was a student we were using what was then called the ARPANET, but all the protocols are exactly what became the foundation for the Internet. And if you look at the use of Internet in U.S. business schools and universities, it's really quite impressive, the access to PCs is almost universal. The regular use is almost universal.

It is moving deeper and deeper into the fabric of how things are done. You've got some schools now where if you want to sign up for courses, you have to do it electronically, that I think is a good thing. You've got some schools where if you want to apply, you do that purely electronically. You have some schools where for each student they not only set up an email address, but a home page where you have your information. A lot of schools are actually requiring laptops, so you can -- as you assign course work you can assume people have that as a tool to go out and get information. And more and more the assignments involve research which is collaborative over the Internet. This is a great thing. And the faster we can move this forward the better.

Here at Georgetown, now they've inaugurated the ability to take presentations, which I think this is one of the first, and put them up on the Internet, so that people can come back and look at it later, people who didn't have a chance to attend or whatever. And that kind of accessibility really just gives students more flexibility, and it's a great application of the Internet.

Microsoft's awareness that something very dramatic was going on around the Internet really came from employees that we had that went back to some of the universities that were on the cutting edge of this. Steve Sinofsky, who worked for me had graduated from Cornell, back in '91. But, when he went back for a recruiting trip he saw that the Internet really was being used not just in the computer science department, but very broadly. And so he became a change agent at Microsoft. He came back and, in classic Microsoft style, wrote a memo that said, unless you pay attention to what I'm saying, we'll go out of business. And people looked at that, they looked at the other memos they had saying similar things, and we said, boy, this is profound. We did lots of retreats in '94 and '95, we kept raising the priority of what had to be done in this area. And by late '95 we had made it the top priority, and were defining everything around it.

And even though you can say, well, why didn't you magically know about it at the beginning. The fact is, if we hadn't had electronic mail, and the type of culture that that creates, it would have taken another year or two, and there's no doubt that our leadership would have been eclipsed if we had taken that long to really get it, and drive Internet standards into our products. So for us this digital approach has made the difference between being on top of things or actually falling behind.

Now, one of the ways people measure this revolution - certainly the articles you'll see in the press all the time talk about electronic commerce, and the amount of business that's being done over the Internet. That measurement I think in a way doesn't really capture what's going on of importance. I mean, after all, if Microsoft takes $6 billion a year that we were doing with our distributors, using paper systems, and we simply move that to be electronic, that boosts the numbers up by 6 billion. And it's a good thing. We should do it. But, it's not the big change. The big change is when you're matching buyers and sellers who never would have found each other before.

Amazon is an example of that kind of change. E-Bay is an example of that kind of change. In most markets today the percentage of business being done on the Internet is still quite small. Stock trading is where you get the highest percentage, even there it's only about 30 or 35 percent. Books, although it's been such a visible category, it's still only about 4 percent of book sales. Now, those percentages will get to be very, very high. In fact, in some categories people really wonder, where is the equilibrium? how many stores will there be of that type? Because as you take the business away, you're obviously making a decision to have less physical distribution of those products. And that's one of those great open questions. If you look at the valuations that are being created for these companies, there are certainly some people who believe that business will overwhelmingly be through the Internet, and they believe that will be fairly high margin business. And both of those are major questions that in every product area are yet to be answered.

One of the interesting companies that's really been a leader in this is Dell Computer. Their direct model lends itself quite handily to the idea of using the Internet. And so we as their technological partner have been helping them set up these sites. For them it's more than just the order process. The fact that they can do customer service this way, answer questions, the fact that people can find out the different configurations, the fact they have these custom pages for their large customers that come in and see what price discount they get, what the recommended machine is in that company, and even the approval process for buying a machine is understood so that you can just order, and it fires off electronic mail to the people inside your company who need to make that approval.

We've been working with Dell to take their site around the world and it's been kind of interesting, you see here the screens of their French and German sites. We took their site in France, and we brought it up late at night, and there was a question, because this site hadn't been advertised at all, how long would it take before people found it and started buying computers. And, in fact, 15 minutes after the site went up somebody came in and bought a computer. It really makes you wonder, was that person sitting there for years just typing in Dell.com.fr every 15 minutes just to see if it was there, and then boom, they were ready? It wasn't a coincidence, though, because we brought the site up in Germany a few weeks later and sure enough, 15 minutes later somebody bought a computer. And so that's an example of the real change that's brought about here, because those are buyers who never would have been doing business with Dell if they hadn't been up on the Internet.

Well, we look at this statistically, we can see the huge disparity between online use in the United States and in the rest of the world. If you take Europe as a whole, the population, the size of the economy, they're actually somewhat larger than the United States. And so they're having 5 percent of their population involved online, that's about a factor of six less than what we've got here. So a huge difference and one that a lot of people in those countries are worried about. And I'm always saying to them, you know, number one is getting the students exposed, number two is getting your communications market to be competitive so people have options in terms of how they connect up. And having the government itself be a leader. That's one area where it's actually some of the smaller countries, New Zealand, the Nordic countries, Singapore, that have done more than even any of the government entities in the United States. Some of those governments really are getting rid of the paperwork and going online, and that's a best practice that even the United States can learn from.

Now, the number I'm using here the percentage online actually tends to overstate how many people are deeply involved, because over half of these people are just doing simple electronic mail. And so it's a subset of those that I would say have gone all the way to that stage I would call the Web lifestyle. One of the ways we track how people are thinking about this is we often will go out and talk to users. In fact, we hired a consultant to even just go out in the street and ask people questions and see what they think about the web lifestyle. And I brought the video of that along. So let's go ahead and take a look at that.

(Video shown.)

MR. GATES: So we've still got a long ways to go before everybody understands the Web lifestyle. Part of the reason I'm so optimistic about this is that the rate of innovation is actually faster today than it's ever been. Every element of the system, not just the processor, but the size of the storage, the quality of the screen, the speeds of these connections, getting the software so it's simpler to work with, all of those things, the level of investments are at record levels, whether it's Microsoft growing its R&D budget to over $3 billion, whether it's the record level of venture capital investments in new companies, all of this is driving this forward at a pace that I don't think is widely recognized.

Some of the key upcoming advances will really transform the power of these systems. First of all, we'll have wireless networks. In the business environment, instead of having the phone that's just fixed to your desk, and the PC that's just there on your desktop, you'll have a wireless environment that lets you take a small device, one that fits in your pocket, take that with you, or take a tablet sized device and take that with you into a meeting. And that will be connected up, you'll get your messages, you can browse information whenever you want. Well, also take the fact that the Internet will support voice and video in rich ways, and things like video conferencing will become mainstream. We already have built into the PC this ability to share screens across the Internet, so two people can work on a spreadsheet together, even if they're continents apart. But, also including in that the video and audio is very important.

Now, that's had limited use today, because as long as you can't prioritize the network traffic, network administrators are worried that that traffic will crowd out the important business transaction traffic. So we need to have a prioritization scheme that's part of the work that we and Cisco are building into all this equipment. We'll also build in advanced collaboration. This digital approach will redefine how people are able to get to problems very quickly and work together on those problems, how they share information. It's a very tough software problem, but one in which we think we've made some big breakthroughs.

Speech recognition will become a standard part of the machine. That's not going to happen overnight. You have a small minority of users that are happy with the quality levels that are achieved today. Those will improve because of the research that's being done, the faster processors, greater memory, better microphones, and if you bring all those elements to bear, some time in the next five years dictation will be something that even a typical user will be doing. Even before that things that are limited vocabulary, like calling up and getting stock quotes or the weather, or travel status, those kinds of things you will find will be very popular. But, broad dictation has proven to be a tough problem, but the right work has been done.

I already mentioned this tablet PC. Part of that is having a screen whose resolution is good enough that you would be willing to read very long documents there. Today if documents get very long, people have a tendency to print them out, because the screen doesn't have the resolution, and because as long as it's on your desk you can't move around and change your angle of view like you do with a book. It's kind of interesting, you know, where is the tradeoff nowadays between the screen and paper? There are things like encyclopedias that have moved from paper to the screen. The encyclopedia we offer outsells the most popular print encyclopedia by a factor of 10, because having the video, the audio, the searching, the lower price, all those things have come together to make that the standard way of doing delivery. Likewise for these paper forms and sales data, that will move off of paper onto the screen. But, for longer documents, you've got to make very big improvements in the readability.

And this first generation of electronic books takes a step in that direction, but it doesn't really achieve something most people would think of as equivalent to reading off of paper. We feel that with some software work we're doing, and the collaboration of these screen companies, we can solve that. And we can solve it very rapidly, I believe in the next few years. We'll also be able to include handwriting recognition. This is an area that there's been work going on for a long time. Again, users are very demanding about the accuracy level, but we think we'll achieve that even before the speech recognition.

And finally there will be devices of different sizes. The full screen device that you have at your desk or take with you, that will be the PC. It will be an evolution of the PC, it will run all those applications. You'll be able to see the spreadsheet, create, edit, do all those rich things that are at the center of knowledge worker activity. There will also be the device in your pocket, which will connect through the wide area wireless networks, so that not only will you be able to get voice, you'll be able to get updates to your contact list or your schedule, or any email messages that you want to get. And here there has to be a partnership, not only with the hardware people to make it small and cheap, and long battery life, but also with the people who have these wireless networks, to get the new protocols built in there.

We'll also be putting computing into the car, so that with this limited vocabulary speech recognition you'll be able to say what radio station you want to listen to, or who you want to call, or get the latest traffic conditions very easily. And we will be building these capabilities into the TV set, either through a set top box, or actually in the set itself, to where you have a high speed two-way connection, you'll have the power of the Internet connected up to the TV. The work we've done around Web TV has been a real pioneer there, and that field is going to grow pretty rapidly.

We're even going to see a lot of products that have been physical products like books or software or music go into purely digital form. Music is the one that's probably the furthest along here. And on a lot of campuses people have servers with lots of MP3 music. In some cases they haven't paid for that music, but the studios need to make it easier to do that, I guess. In fact, the studios are very worried about this becoming a sort of piracy haven, and they're working with the software industry on what sort of rights management might be appropriate there. But, the flexibility you get in digital form is really quite incredible. You can sequence it the way you want, you can take it with you. You don't have to be putting the CD in and out, that kind of thing. So I have no doubt that with the right framework in place, that will come very rapidly.

Likewise with photos, with which you've always had to go through a developing process, and they are fairly hard to get duplicated and sent around, these digital cameras are getting good enough that that will be the standard way, you can take photos, organize them, send them off to relatives or people who are interested. Another frontier there is motion video, these new digital motion video cameras are starting to do a good job, and if we combine that with our software and these big disks, you'll even be able to do film editing, a little bit of sequencing and even special effects on your PC, the equivalent of what would have cost $100,000 on a special workstation just a few years ago.

So those are radical changes to those businesses, and if you think of the value chain between the author and the reader and who's in between, the fact that you move to digital allows that to be done in a very, very different way. I think the bottom line here is that there's incredible opportunity for all of you. You're not entering into a static environment. You're entering into a world where you can make a huge difference, and help the companies you join be the successful companies of the next decade, that take these digital tools and really reinvent the way they work. I hope I've been able to make it clear how excited I am about these possibilities.

Thank you.

(Applause.)

MODERATOR: Ladies and gentlemen, there will be a short Q&A. There is a gentleman with a microphone in the house. If you'd raise your hand, he'll come to you.

Thank you.

QUESTION: During the course of the presentation, you mentioned job reduction a number of times. While, as business students, we can all appreciate what that means for the bottom line, have you put any thought into what it means for society as a whole?

MR. GATES: Well, part of the lesson of economics is that there are infinite demands for jobs out there, as long as you want class sizes to be smaller, or entertainment services to be better, there's not a lump of labor where there's a finite demand for a certain number of jobs. And so, as efficiency changes, such as in food production, the jobs shifted to manufacturing. As efficiencies were gained there, those jobs moved into services. In fact, there's no shortage of things that can be done. So, it's not like we're going to run out of jobs here.

We do have to deal with the issue of making sure people get the appropriate training, and that we can take somebody who has been in a clerical job and have them, ideally, stay at the same company, but doing something that's value added. So I do see a lot of this concern about the rate of change that's here, but it's important to see that the U.S. economy, which has been embracing these changes more rapidly than any other, is the one that's generated the most jobs. In fact, we're really the only economy that's done a great job with that.

And it's paradoxical, if you just look at the headlines, you'll see that big companies on average are reducing their size. The Internet makes it so a lot of things you would have done inside your company you'll now do through outside partners. The average size of a company will be going down. But at the same time you've had all those headlines of how big companies are reducing staff, the total number of jobs in the economy has been going up pretty rapidly, to the point where there's even concern about the labor shortage and the inflation that kicks in.

That's an example of the marketplace at work in a very positive way. And we will see countries that decide not to embrace this change, and simply say no, you can never let any workers go, you can never change their job assignments, and those are two very radically different approaches to dealing with the changes that are coming.

QUESTION: A lot of people today are nervous about privacy. So, what do you see with the advance of the digital nervous system, what do you think the future of the digital immune system will be?

MR. GATES: That's a very important question. I think to some degree it's only as people have been using personal computers and seeing their ability to search information and bring information together are they recognizing how much information about their activities actually is being captured. And it's not just the Internet. I mean, whenever you use your digital phone, not only do they know that you made that call and who you called, they actually know where you are. And that's come up in some court cases recently, where they proved that people are in different places. And you've got your credit cards, you've got all this information, taxes, medical records, has been stored in large computers for many, many years. And I think it's a very healthy thing that people are saying, wow, that data is out there. There should be rules about that data.

Now, some of them are fairly complex. For example, what should an employer, a potential employer, be able to see about somebody who is applying for a job? If somebody wants to be a teacher at a school, should they be able to see that person's criminal record? And if somebody is granting insurance, should they be able to ask you any questions about your medical history? These are important political questions that the rules need to be laid down for.

Now, the technology allows us to encrypt the information, it allows us to enforce policies on how that should be used. And so, Microsoft is one of several companies that are in this dialogue about industry self-policing, and also with the government about what those rules should look like. If we don't handle that well, it's really going to hold back people being willing to jump in and take advantage of this environment. And, so, there's a lot of work still to be done.

QUESTION: My question has to do with the question of connectivity. You're talking about how many of the businesses in the U.S. are becoming smaller, and especially a lot of them are being started from people's homes. And I noticed the best way for me to get my product or service is to go through the Internet, but if I don't have that connection, you know, if I don't have a T-1 line or an ISDN line, what do I do, and how far along are we to have that connectivity everywhere?

MR. GATES: Yes, that's a very good question, because I think connectivity will be on the critical path for the things I talked about, particularly connectivity out of the home. If you're in a business district, you'll have lots of companies coming to you and offering to provide you high speed Internet connections. Not only the phone company, but all these new sorts of companies that have emerged. And so the price for business will be very reasonable.

Now, if you're in your home or doing business out of your home, that's a tougher problem. The only companies that are really providing the high speed connections today are the cable companies, and even they, as a percentage of homes, is a pretty small number. We also hope to get the phone companies using a technology called DSL to be more aggressive in terms of pricing high speed Internet connections so we can really start to get lots of homes connected up at high speed.

For a small business, there's two ways you can deal with that in the near-term, before high speed bandwidth comes to you. You could either have a set-up in a business district and connect up there, or what you can do is, you can have your business Web site hosted by someone else. We actually have a service where you can go onto the Microsoft Network, say you're a small business, and describe what products you're offering, and we'll set it up immediately so you're hosted on our site. And then you can just use your dial-in capability to go back and check what messages are coming to that site, what orders are coming to that site.

And, so the fact that you just have dial-up is not a drawback to having lots of customers come in and do lots of transactions because the two systems are coordinated with each other. So, I think for businesses, there are some great solutions.

QUESTION: Given these positive changes will come to pass in the near future, it's easy to see how both organizations and individuals could potentially have to be dealing with massive, massive amounts of information, huge amounts of data. What's your vision on how organizations and individuals will deal with the massive amounts of information, how will they be able to separate the relevant from irrelevant?

MR. GATES: Well, people often say to me, gee, you have electronic mail and anybody in the company can send you electronic mail, isn't that a bad thing? You know, aren't you overwhelmed with electronic mail? And the fact is, I'm not. If there's a group where the morale is low, somebody thinks things are being done wrong, it's great that somebody sends me electronic mail, and I can look into that and think about what changes we need to make. If somebody has a new product idea, it's great they send me electronic mail. And, it is a serious part of my job to read that mail and go through it, but I wouldn't have it any other way.

In terms of the information overload, inside a company it's fairly clear what's important, the sales data, the project data, the customer data, and management can really work with people to make sure they're spending time on the information that counts. We used to have to spend a lot of time on an expense account approval, another one of those paper processes, but we said, no, we'll just set some thresholds, and we'll trust people. We won't waste time dealing with that unless it goes over a defined level.

Once you get outside your company, where you really control the information and how it gets organized, there you will rely on trusted names because out on the Internet there will be a lot of information that's right, there will be a lot of information that's wrong. And so, that's where the brand of a newspaper or a consultant, or a partner who you've come to trust will come into play, and that's the information that you'll look to.

It all comes down to the knowledge worker having something they want to get done, and going out to see the information that's relevant there. The search engines will improve in terms of their ability to automatically find things for you, but also the number of people who are out there sort of creating edited web pages. Say you want to find out about a certain disease and the advances being made there. You'll find trusted sources who will author the links to all the projects and thinking that might be worthwhile there. So, you'll go through that page and follow those links to get the information that you can trust.

MODERATOR: One last question, please.

QUESTION: Sometimes successful companies are the victims of their own success, serving their current installation base, sometimes blinds them to developing new technologies. I was just curious how Microsoft strategically balances serving its current installation base and developing interesting new technologies you showed us, the hand-held devices, et cetera?

MR. GATES: Well, that's a very important question for our business, because the history of the computer business is the history of giants that have fallen. IBM focused on mainframes and didn't do a great job on minicomputers. Digital focused on minicomputers, and sort of pooh-poohed the personal computer. Wang dominated word processing and yet didn't make the transition into this new era very gracefully. They went through Chapter 11, and now they've created a good company, but it's a very different company.

And so anybody in this business, including myself, comes to work every day saying, what are we missing? What is the revolutionary thing that's going to come along and change the framework? And by working with universities, by hiring top researchers, by having this open electronic mail environment, by thinking about everybody as a potential customer. I think that, at least so far, we've been able to avoid having something like that come along.

It's partly a cultural thing to say that the person who is saying that something new ought to be done is a voice that ought to be heard. And Andy Grove writes in his book about these inflection points where incredible things happen in businesses. In our business that happens every three or four years. In typical businesses, it's only every 10 years or so.

I claim the Internet is a strategic inflection point for 90 percent of all businesses. I mean, there might be one or two that aren't affected. You know, if you're good at finding oil, the Internet doesn't change the fact that you just go ahead and find oil. The price is a little low, but it's still probably a business to be had there.

So, I think it's important for companies to acknowledge that just because they've had past success that won't lead to success in the future, and not to develop the kind of insularity that we've seen in those cases. That's partly what makes my job interesting. I know that I've got to stay on my toes, we've got to keep hiring great people, because the opportunity to have new products, and have us be the one to obsolete our products, that's the only way we'll stay ahead.

MODERATOR: Okay. That's the last question. Mr. Gates, I did want to tell you that while you were speaking, I checked the faculty handbook, and if you'll come back and talk to us about your third book, I'm pretty sure I can get you tenure in the business school.

MR. GATES: Okay.

MODERATOR: We'll work on that. Until that time, on behalf of Georgetown and the McDonough School, I'd like to thank you very much for your time, and sharing your wisdom with us this morning, and thank everyone for coming. It's a great session thank you very much.

MR. GATES: Thank you.

(End of event.)