COLUMBIA/NYU BUSINESS SCHOOL

March 24, 1999

I always feel a little awkward talking to students because I, myself, am a dropout. And I think education is a great thing. And, in my case it was kind of an usual set of circumstances where I really felt like I could get in on the ground floor of software for personal computers by starting Microsoft when I did. I'm still on leave. When things calm down, I'll go back and get my degree. That won't be anytime soon.

I think this is a very exciting time to be in the world of business. I think all of you are going to have more interesting careers than anyone in your positions had in previous generations. A simple way of capturing this is to say that business is going to change more in the next 10 years than it has in the last 50. There's a revolutionary new way of sharing information that redefines communication. It redefines collaboration, and it even redefines the fundamental mechanism of the marketplace.

The ability to match buyers and sellers will be completely different because of the PC and because of the Internet. People are still, in some ways, underestimating this. There's always this tendency to a little bit overestimate what can happen in the next two years and underestimate what can happen in the next 10. And I think this is a prime example. In fact, it's almost dangerous sometimes because business people read about how they have to adapt to these new changes, and they may wait a year or two and see that, in fact, the predictions that everything is different haven't come true.

Well, any complacency they develop will work against them as in the years to come the devices get cheaper, the ease of use gets better, and you build a critical mass around this completely new approach. Microsoft has a yearly event where we bring in about 100 top chief executive officers and talk to them about technology and their business. How should they be spending money, what kind of value can they get out of it, what are examples of people who are getting ahead competitively. And the level of interest in this has risen very dramatically. You know, five or six years ago, most CEOs would think, gee, that's the IT budget, what a pain. They always ask for more money, and all they do is print out the bills. You know, how can I avoid having that as an incredible overhead.

Today, certainly modern CEOs are saying that IT issues about how information flows in their company, how they empower their knowledge workers, how they reach out to customers, how they collaborate with partners, all of those things are so central to their business that they can't just rely on the CIO to keep track of those issues. In fact, I think it's imperative that they take their management team and have them get hands on experience.

I think all of you have an opportunity to be agents of change, moving into these businesses and really helping them understand the possibilities that are out there. There are really two key milestones that got us to the current situation. The first is the PC itself. This is taking the computer from something that sat at the center of an organization, and was simply in a glasshouse, every expensive. And brought the price down so radically that it could be on an individual's desktop.

The price of computing improves faster than any product known to man. Every two years, you get twice as much computing for no extra cost whatsoever. And so, over a period of 20 years, that means computing is a million times cheaper. Not many industries can deliver that kind of a track record. And so the desktop machines of today are radically more powerful than the most expensive machines of 10 years ago. And the desktop machine, or portable machine 10 years from now, will be so much more powerful that some people are even wondering what are we going to do with all that power. I want to suggest to you that there are some exciting ways to use IT as I get further into my presentation.

I think there are some basic assumptions here. The PC will continue to improve, and the Internet, which is the other big milestone, will also continue to grow more popular. I don't have the slightest doubt that either of those things is true. The Internet came along, and it's hard to say exactly what year it is, but sometime in '93, '94, '95, it moved out of the university environment, out of the computer science department, and into the world at large. And so, what many people have been talking about for a long time as information at your fingertips, it became clear the protocols, the standards, the approach that was going to be used, and that was the standards of the Internet.

We've seen that in the last few years develop into almost a gold rush type environment. The levels of investment have never been so high. The number of new startups, the uncertainty about business models. What are really the long-term assets that are being built here.

If you ask business people today some basic questions, you know, will people use computers for most jobs? I think most people would agree with that. Will paperwork go away and be replaced by a digital equivalent? For example, think of the bills that come into a company, why should they come in on paper? You know, something like a phone bill that has different phone numbers, different charge rates. That should come purely electronically, and those costs should automatically be allocated. There should be software that looks at certain thresholds and tries to decide if somebody needs to be alerted about that.

And, instead of having lots of jobs that are purely clerical in nature that take that data and kind of look at that data in a simple form, you should have that flow in and be available in this digital environment that all the workers will have easy access to. If a worker wants to look at different costs, and how they're growing, what's gone on, or to compare it by department, great. The technology is there to make that very simple. If they want to be alerted to something, then great, they'll have a piece of mail in their mailbox. So, all of those invoices and purchase orders and bills are a very inefficient way of moving the information.

I think a lot of people are seeing that the business part will change. The part that will take a little bit longer but is perhaps more profound is the behavior in the household. Will most people be turning to the Internet many times a day to get information, and use it for the purchases they make? I think within the next decade in this country, that absolutely will happen. And, the only thing that even makes me say that it will take that long is that there is one element in this equation that doesn't have that Moore's law exponential improvement rate, and that's the access to high speed connections. That's going to be a tough one because that last mile connection that comes into the home, until a major breakthrough in wireless that won't come probably for a little more than 10 years, you're only going to have your cable or your phone line as a way of moving data in and out in a high speed fashion. And although that's gotten going, it's a very small percentage of households, still fairly premium priced, and so I think that will actually be on the critical path for achieving this vision.

But in the home, the thing you call the TV set will be connected to the Internet. It will let you get any audio that you want. It will let you get any video you want. And things like a record or a CD will be a very antiquated concept. You know, today if a kid is growing up and you go by a record store, they might say, what's a record, but they know what a CD is. A kid 10 years from now, I think, even that CD, it will seem kind of archaic that you had to open up the disk and put it in the player, and then when it was done, you had to take it out, very manual. You couldn't organize things the way you want to. And we're seeing the leading edge of people doing this the more efficient way. Right now, there's not a way for people to pay to use things digitally, so this MP3 digital music activity is both convenient, but it's also often done without paying for the music. So, the music industry is actually a little bit panicked about that, which isn't such a bad thing, getting them to think about how do they embrace this, because if they don't it's certainly going to happen without them.

Electronic mail is sort of a starting point in all of this. For anybody who is using electronic mail regularly, you know, if you said to them, well, we're going to take it away from you, I think they'd say no way do they want to go back to working without it.

If you said to them, how could it be better, of course, they'd have a lot of ideas about how the collaboration and the sharing or these big threads where people get into disagreements, couldn't the software help out with those things. And so electronic mail is going to become far richer. We're going to give you back control over your in-box so that you get to decide who can send you electronic mail. If you give out your mail address to a business, even if they pass along, that won't get into your in-box, or if you give it out to them, you can say, okay, once a month I'm willing to take mail from them, but that's about it. And because it's your time that's valuable, you will, through the right software, be able to pick not only how to control your mail in-box, but who can ring your phone, who can send you material, which things are important enough for you to be interrupted when that's actually going on.

Interestingly, we'll actually take advertising and turn it around. If somebody wants to send you something that's purely an ad, you know, why would you read it? Well, there's only one reason, if they offer to pay you. So, you'll be able to describe to the machine, okay, if an ad takes a minute to read, that's worth about $3. And so, anybody who wants to buy my time, great, I'll be glad to read their ads. Effectively, you're engaging in that transaction today. When you watch TV, there's a middleman, which is, of course, the TV company that's collecting the money that your time is worth, and using that to produce the show that you, in turn, are saying is worth watching. Well that, effectively, will be dis-intermediated for a lot of your time. You'll decide what it is that you want to look at from the world of advertising.

Certainly, you know, all of this stuff will be digital. Cameras will be digital, movies will be digital, and nothing will be authored in the world of business without being in digital form. Even the notes you take at a meeting will happen on a digital tablet that's connected up to the wireless network.

In writing the book, I decided I really needed to identify that there was a discontinuity here, a discontinuity in terms of how consumers act, that's the web lifestyle. As profound a change, or more profound than having TV sets or having telephones.

I decided it was important to demark a very big milestone in terms of how knowledge workers think of their job, where they actually expect that all the information is there online, and if there's a sales trend, they can see it. They can mail it around to other people, they can compare it to what they're seeing about the competitor. And if they have a creative idea based on the trends that they see there, they are fully empowered to share that and get the business to consider a new strategy. That's the web workstyle. Again, with that we're very much at the beginning.

And finally, this digital nervous system is really the same concept, but from the point of view of the business. The business saying, how does information flow in this company, what information counts, what information is actionable? Do we record every interaction with a customer and get it up into the file so that anybody can call it up in a few seconds? Do we make it so that I've we've done a similar project in the past, any of our workers within 60 seconds can sit down at their machine and have that document come up and see what lessons they might learn from it. If not, it's almost as if that document doesn't exist. There really is no corporate memory, and that's typical in most businesses today. And so, we're really kicking these things off using this new platform, and the fact that business competition will drive people in this direction.

It's fascinating that for almost any business you can name, there are new companies being started that are taking a 100 percent Internet approach, whether it's selling houses or travel or loans or books there is a little league company doing it. They went public last week, they're worth $300 million today, they'll be worth a billion tomorrow. It's a wild, wild time. And that's because people are saying there are going to be some big winners out of this. I think the only winner that I can say is there for sure are the people who buy products. The goal of business, which is to make better products efficiently, is greatly advanced by having this tool that makes competition all the more perfect.

And so at the center of all of this is a belief that with the Internet, there is no turning back. And even though we use this same term, the Internet, to talk about what we have today as to talk about what we had a year ago or two years ago, people should recognize that the Internet is evolving. It's evolving quite rapidly. Today when you go to a website in most cases if you want to buy something, at the last minute if you really want to get some advice and talk to somebody, in fact, you might use the phone if you think you're going to want that.

Well, as we get voice quality across the Internet, as you do that transaction, you'll be able to click, and somebody at that company will see what you're proposing to buy, see your history of purchases, and be able to talk with you to make sure that you've made the right selection. So that will be the best of both worlds. It won't just be the screen versus the phone. It will be the screen to find things, and then talking or even video conferencing to collaborate on is that the right approach.

All the jobs that have to do with moving data into the computer or out of the computer, those jobs won't be necessary, because people will be doing self-service. And so you can think about taking those workers and giving them more information so they can make more powerful decisions. They can be involved in customer service, where people really need it, where they haven't been treated properly and you have to do something to work with them in order to retain them as a customer. If you give those people the right tools they can be involved in that process and add a lot more value. It's a very customer-centric environment that I'm talking about.

Inside companies in the past they've had these mainframe systems that didn't talk to each other. They've had divisional structures where different groups are calling on different customers, and they just didn't have a way of bringing that all together. Now, with the Internet development approaches, where you can create rich views of information, even information that doesn't fit the sort of plastic tabular database, you don't have to worry about what the divisional structure is. As those contacts are made, you can get that all into one logical view that you can look at.

People who think about this as only the website are really missing what's fundamental here. It's not enough just to have rich information for the customer on the website, and let them do transactions. That same system should be used internally, and you should get rid of all the other front ends. The only difference with the internal system that you use to do your analysis is that you might have some additional information that you add about how you think about that customer, what category you put them in. But, whether that customer comes in from the Internet, the phone, comes in face to face, that one customer-centric view is what you should be working off of.

Our last principle here sounds a little bit negative, and that's that bad news should travel fast. I remember when we first got electronic mail, I started getting these messages. I'd get them about once a day. Somebody would send mail and say we won the XYZ account. This is a great thing. And so I immediately sent back a reply and said, congratulations on winning XYZ. And then I'd think to myself, you know, I didn't get any mail on these other accounts. What does that mean, does that mean we've lost every one of those accounts? I doubt that if we did they would have sent an email, because I never got an email saying, "hey, you know, just thought you ought to know, we lost this other account. You know, just reply whatever you want to say to me, here I am." So it's not human nature to send good news around.

And yet, there's a paradox there because what's the most actionable thing for a company? If you get good news, do you change your product plan, do you change your priorities? I doubt it. It's only by taking a pattern of things, where a customer is not quite happy, where a competitor is doing something really good, or economic conditions are a bit different than you had assumed when you did your plan, it's those surprises that have to come together as efficiently as possible and allow you to adopt a business strategy before the bad news becomes a permanent problem for your company. And I believe these digital systems, if they're designed in the right way, can do a fantastic job of spreading that bad news around, making sure the people who design the products know what's unreliable, what customers don't like, and yet do it in a way that's very positive, that everybody shares that news and collaborates together to try and turn the situation around.

Well, how did this all come to pass? There's no doubt that students really led the way in the arrival of the Internet. The PCs were out there. They were a key foundation. The advances in the protocols that people like Tim Barnsley and others made were there and extremely helpful. The cost of optic fiber connections came down radically, making this approach for global networks reasonably priced. But, the place where it was really put into use was on campuses. Today in the United States the situation is very different than outside the United States. College students, and certainly business school students even more so, use the PC as part of their everyday activities, even when they're in their dorms they've got access. You've got schools now that the only way you can apply to the school is online. The only way you can sign up for courses is online. A few places the only way you can get your grades or submit your homework is online. And so that really is about the web lifestyle, getting those people to see it as a fundamental tool.

I was glad to see that Stern and Columbia are both great examples of this, either expecting the laptops for all the students, or really using it to distribute the information so that it just goes without saying, that's how things get done.

For Microsoft, it was students that brought us to our recognition that a real discontinuity, or as Andy Grove would say, an inflection point had come. Actually, it was where we had people who worked at Microsoft going back to their campuses and coming to us and telling us how this is being used. Steve Sonofsky, who worked directly for me, had graduated from Cornell, undergraduate, and when he went back there as part of a hiring trip, he was pretty stunned to see that in the two years since he had been gone, the role of the Internet had changed very fundamentally. And the memo he wrote, and it's sort of a classic Microsoft tradition, sent to hundreds of people and said, boy, if we don't pay attention to this, you know, we could be dead, was very articulate. And it got people's attention. In fact, within a few months of this we had several retreats where we said, okay, here it is, it's the big opportunity. How do we make sure we've designed our products in the right way? And the priority that kept going up higher and higher, until soon after we'd shipped Windows 95 we finally decided, hey, this was the key priority and we articulated that out to the world in late 1995.

Now, part of this is about business. It's about people buying and selling things. The way this gets measured is, I think, a bit misleading. It's very easy for a company that's done business with another company to take the paper forms and move to doing things electronically. That's good, it's important, but it drives the numbers up almost in an artificial way. Microsoft does $10 billion a year with our distributors. About a year ago we shifted that all to be electronic. Was that a big change? Yes, it saved us some money. But, it's very different than when you have buyers and sellers that never would have found each other, using the Internet to do that.

Consider a product whose natural marketplace is about 5000 units a year. There is no distribution system that's efficient for that product. You can't afford to put it into a store structure. People can't afford to know about it, you can't afford to run ads for it. And so you could have lots of small companies doing neat things like that. But in today's environment they wouldn't thrive. Well, already on the Internet we're starting to see that companies like that can find their customers and do business.

There's a little company up in Wisconsin that does a truly obscure thing. They take these old phones, and they refurbish them, the beautiful old phones, and actually make them work with today's phone networks. They'll either take one that you own, or they'll sell you one that works that way. And so they've put up a website, this is just a hobby for them, they have to have normal jobs to have food to eat. But, then when they put up their website they got customers from all over the country, all over the world, and were eventually able to quit and form a business around it. So that market is now being mediated in a very different way. And you have a business there that you wouldn't have.

The poster children for the Internet and e-commerce are pretty well-known. Amazon got out in front in the book space, and got about two years to really establish its brand before the traditional booksellers came in. Dell whose business model was very well adapted to the Internet from the beginning. And their experience has really been phenomenal. We've been their technological partner, putting the systems together, and so as quickly as Dell can they've been going around the world rolling out these systems. And so our people were with them in France a few months ago when they turned the system on. You know, with localized French data and the right tax calculations and good shipment and all that.

They turned the system on and they said, geez, we haven't run a single ad to tell people we're here in France. I wonder how long it's going to be before somebody buys a computer. Well, 15 minutes after that site was turned on they sold a computer. And so it really makes you wonder, was there some guy waiting there for years, just typing in, Dell.com.fr? No, not there. Let me try again 15 minutes later. And it wasn't really a fluke, because we went to Germany and, you know, got the system running there a few weeks later and there again, it was about 15 minutes and somebody had bought a computer. So that's an example of where the Internet is profound. Where it's allowing people to have better customer service, to pick options in a new way. And to consider more options than they ever would have in the past. Here's the actual website that we use to do that.

Well, it's very easy when you're involved in this as all of us are here, and we're using this technology and we almost take it for granted. But I think it's eye opening to recognize how little adoption we've really got so far. And if you take Canada and the U.S., and you take the most optimistic number, which is people who claim that they use online and Internet at all, it's about 87 million people. Now, only a quarter of those are really using the Internet in a deep way. Only less than a quarter, I would say, are living the web lifestyle. Most of them are using very occasional electronic mail. If you adjust for that, you'd say less than 10 percent here in the United States and Canada, which is by far the furthest along. In Europe, any comparable number you pick is going to be about a fifth or a sixth as big. And yet, if you take those economies as a whole, in Western Europe, or take the population, they're actually larger in both dimensions than the United States. That's a fair bit of lag time there, and I think it is one several reasons why our economy has been so strong, particularly our exporting companies that are using technology to get this flexibility are doing extremely well. Once you get outside of Europe then you very quickly get into numbers where less than 1 percent of people are connected up.

Now, PCs are much better than this. In a country like China, over 4 million PCs a year are being sold. So it's actually the fifth biggest PC market in the world. They do need to get the communications infrastructure in there, but that's going to happen. It's just a matter of time, and the time isn't that great. For the first time you really have politicians paying attention to this. You know, the chancellor in Germany did an event with me where he talked about, hey, Germany needs to catch up and all the government policies he was excited about. Well, two or three years ago you wouldn't have -- a German chancellor wouldn't have uttered the term Internet and known what that meant whatsoever. And so because of all the issues that come out here, because of all the opportunities, even the political realm is coming in and looking at what can be done here, and how can a country move into this in an early stage.

Part of the way that we keep ourselves grounded in terms of how people are thinking about this is we go out and talk to users on a regular basis. In fact, we got a consultant to go out on the street and do that for us. And I've got a video of that, which shows we still have a long ways to go.

Let's take a look at that.

(Video shown.)

MR. GATES: One of the reasons I think that people still underestimate what can be done here is that the innovations that are still to come really will make the device so much more natural than it is today. For example, take reading. What do you read on paper, and what do you read on the screen? There are some documents that have moved to be primarily screen based. Microsoft makes an encyclopedia that's on a CD that outsells the best selling print encyclopedia by a factor of 10. And it's also about a 10th as expensive and that's really become the standard way, because of the video, the audio, the cost and being up-to-date.

In fact, my friend Warren Buffett actually owns the World Book, and so he and I joke around a lot about, well, whatever happened to the World Book? He thought that that nice leather smell would protect them. And I told him, I didn't think that was going to be good enough.

In any case, there's this transition. For long documents, most people don't read it off the screen, they tend to print it out, even though magazines are up there on the Internet, there's not many magazines where they've seen the subscriber base fall off because people are saying, hey, why should I get this paper, I can get it there on the Internet. But as we get the better screens, particularly the device you can carry around with you, so you can sit with it like you would with a book, which implies either downloading the material, or using the wireless connection. Then you'll have a drastic shift towards reading off the screen. And, you know, I think note taking and everything will go that way.

So, having the wireless infrastructure and putting voice and video on the Internet so that video conferencing is as simple as a few clicks will change behavior very radically. You won't have that big phone device sitting there on your desk. You'll have something that's a screen that you can pick up and carry around. That one will be smaller, and so depending on where you're going, you'll either pick the small screen or the big screen that you want to have with you.

I think the software that lets people collaborate, lets them work together on decisions and help with the decision process, that will be far better than anything we've got right now. This is a hot area of research for us, because it's all software driven.

I also think speech recognition, although we've been talking about it for decades, is finally going to get to the point where it really is quite usable. That could be anywhere from, say, two to six years away, and the extra power, the software that understands the grammar, will help get us over the threshold where this is commonplace.

I've also talked about this tablet PC, and that's the thing I see coming soonest, because handwriting recognition software has improved so dramatically, this I see only a couple of years away, drawing on the great advances in portables that we saw over the last year.

We'll also see the smaller form factors in a variety of places. The device that is just an electronic book, so it's like that tablet PC, but instead of being able to create documents and edit and annotate them, it will be less expensive because all you can do is read.

Your TV set, as it's connected up to cable or some other infrastructure will have digital capabilities including graphics that are way beyond anything you would expect today. When you walk up and want to see a show, you'll interact with your TV, it will help you through that guide that's becoming more and more complex. It will be able to record things onto a hard disk that will be very large, and it will deliver a computer experience, and we're doing a lot of pioneering work there as well.

And so, all of these things combined. The fact that you get more content on the Internet, the fact that people want you to be there to communicate, the fact that it's an imperative for businesses to do things effectively, and so people whose job involves this web workstyle will be going home and saying, well, as long as I'm used to this, I'd like to use it for other things as well. And so that's why, in a sense, this gold rush approach is valid. The numbers are going to continue to skyrocket in the years ahead.

What that means for some products is that they won't even have physical forms. Books will be e-books, music will be digital music, software will be delivered over the network. And so imagine what the value chain looks like between author and reader, and how different that might be than today's world where things like physical distribution, printing capability, all of those skills have really created barriers to entry to anybody short of getting in and becoming a publisher themselves.

We're seeing real pressure on pricing models. You know, relationship type selling, where you have some products that you lose money on, some you make money on. That comes under attack on the Internet because the product that you make money on, somebody just does an Internet focused business to go after that special sales stream, and then you have to rethink what you do.

You also, in that case, will get adverse selection, where they'll go after the customers who are particularly profitable. So, take a travel agent, the people who will go tot he web are the people who have simple transactions. The ones who will be left will be the ones who take a lot of time. And so, at some point, that service benefit will have to be recognized explicitly as part of what people pay for. So a radical transformation, not only in information flow, but the business model that comes into play here.

And so, inventing these new ideas, driving new companies or old companies to take these approaches, that's the opportunity that's in front of you. There's no doubt in my mind that the successful companies of the next decade are the ones that take these digital tools and reinvent the way they do work.

So, I'm certainly excited about this, and I hope you are, too. Thank you.

(Applause.)

MODERATIOR : Ladies and gentlemen, there are four microphones in the house. Mr. Gates will take questions from each of the microphones in order, one, two, three and four. If you will proceed to the microphones in the aisle, you can start your questions there. Thank you.

MR. GATES: Yes. We've got plenty of time for Q&A, and whoever goes to a microphone, go ahead right there.

QUESTION: I think --

MR. GATES: I'm still here.

QUESTION: I think most people would describe you as influential. I was wondering how you define power and does it ever scare you to have so much?

MR. GATES: I think that people overestimate the freedom of action that somebody in business has. You know, we get up every day saying that our -- we have to obviously lead our products, we have to drive those forward. And so between sitting down with customers, getting involved with hiring the great people, meeting with those engineering teams, there's not one element of my day that has any arbitrariness to it. I think of somebody who is truly powerful saying, maybe I'll do this today, or I don't like this, maybe I'll go after that and change that. And certainly when you're driven, you're in a business that's fast changing, that kind of arbitrariness just doesn't exist. You have to fine-tune everything, including the culture of the company to really go after building the great software products, and that's all you're really focused on.

Companies in our position do have -- are in a position of responsibility. So when it comes to tough issues like privacy and security, the policies about encryption software, whether you can export that or not, you know, we've got to take a position and put a lot of time into those things. But, I think, you know, maybe some of the companies in our industry coming together can make a very big difference in these things. So certainly, I don't think about it in terms of being powerful. I think about it in terms of what is our responsibility and delivering on that.

Over here.

QUESTION: Hi. My name is Zach, and my phone number -- no, just kidding. I guess I'm wondering what is your personal opinion about the sort of tradeoff between the need for standards and enabling adoption to take place by having standards, and the other side of needing innovation and creativity. I know it's a big issue. And I'd just like to know what you think about that, in relation to the fact that when you started with DOS there weren't a ton of operating systems, and you came in at a great time and looking to the future. If there are other areas that have a need for standards, what the approach would be.

MR. GATES: Well, the computer business has always been very driven by standards. There are a lot of things that are going down in the plumbing, like how characters are represented. We've gone from ASCII to a thing called Unicode, which represents all of those characters. The way these computers connect up with each other, whether it's electrical standards like Ethernet, or 1394, or higher-level protocols, TCP/IP, HTML, the number of acronyms could fill a book. There are some new standards coming that are very, very exciting around XML. And that is taking so-called business objects and being able to exchange them across businesses, so that you're not just working at bits and bytes level, you're really working at the semantic level.

So we're taking things like EDI, electronic data interchange, which was not Internet based, moving it into XML on the Internet, so that you have interactive systems and a lot more flexibility there. So standards are more important than ever. You know, we've actually got more people on standards committees around the Internet than any other company, because when it comes to security or quality of service on the network, we've got to do that.

Now, that doesn't take away from innovative work. The marketplace has plenty of room for both. As we're doing the investments, which are literally billions of dollars, to create the computer that can see, that can listen, that can learn, that can recognize handwriting, those implementations, even if they're done in a standards framework, those are major pieces of work that particularly using our high volume, low cost approach, we can get out there and build a great business model around. So the two things tend to help each other. The PC standard was great for Windows. Windows supported all the standards so that we could interoperate with the other computers that were out there. So I don't see any problem being very enthused about both.

QUESTION: Hi, my name is Emily McCue from Columbia Business School. And first I'd like to thank you for talking to us today and I have two questions for you. The first one I think there's probably no one in this room who doesn't want to ask you this question. And pretty much that is, how do you respond to people asking you outright for money, and would you mind giving any of us maybe say a million dollars just off the bat? The second question, if I may.

MR. GATES: That's the second -- okay.

QUESTION: Okay. Answer then, if you'd like.

MR. GATES: Okay. Well, the -- the success I've had in terms of the -- you know, the value of my stock in Microsoft, that creates a real responsibility to say, you know, how are those resources put back to very good use in society broadly. I've been very clear that I don't think it's helpful for society broadly or to my children to take that wealth and pass a substantial portion of it onto them. I just think that would be a huge burden. They may disagree, but my views are fairly well set. And so instead of just randomly giving out a million dollars to people that I run into, you know, to be nice, I've decided to take something I believe in a lot, which is given my optimism about the advances in medical technology, and information tools that are very empowering, you know, I just love the fact that we're going to solve many of the tough diseases and we're going to get these tools out and use more human potential. I'd like my philanthropy to help make sure those advances are available to a broad group of people, and not just a very small group.

And so in thinking about that, and talking with various people, actually the priority that's become the top for me is world health. You know, I have a daughter who is almost three years old, and she gets great vaccines, great medicine, and most children around the world don't get those. And so a higher priority than getting computers out there is getting these vaccines that already exist and are used in developed countries right now and giving those out to people very broadly. And so, you know, it's been fantastic to work with those doctors and learn about that. Those are the things that I think someone in my position, or at least what I have chosen to do with the wealth that I've gotten from Microsoft.

(Applause.)

MR. GATES: Let's go to number three, and then we'll come back.

QUESTION: Yes, Bill, my name is Mike Little, I'm with the Columbia executive MBA program. And I'm aware of your, I guess, meager position in investment in Lernout & Hauspie, the voice recognition company. And I was curious as to what your take is, or what your calendar is for when voice recognition is going to really take hold, because it seems to be dragging along and not really making an entrance as I would have expected a year ago.

MR. GATES: Yes, I spend a lot of time thinking about this, because it's fairly central to how we design our products. There have been demonstrations of voice recognition where somebody would come on and do the demo, even 20 years ago universities were funded by DARPA, the same people who funded the Internet, which was called the DARPANET at the time, they had people making very good progress. So we got the, depending on how you measure it, we got the quality rate up to about 93 or 94 percent, and it looked like, hey, this must be an easy problem, we'll just drive that number up to about 98 or 99 percent and people will want to use it. Well, that's right. The threshold is somewhere in the 98-99 percent range before people will want to use speech recognition. And it's proven to be incredibly difficult to get here, because it's not just a question of what the waveform is. If you just have the waveform it's very ambiguous.

In fact, our speech group calls themselves the wreck a nice beach, the wreck a nice beach group, because when you look at that waveform it looks the same as recognize speech, unless you're very careful. And of course, if you're in the right context you would never mistake those two things. And it just shows your mind subconsciously is doing very deep pattern matching and probability, and even sometimes you have to go a few sentences down before you disambiguate the things that came before. So it's a tough, tough problem.

The speech products have been sold for many years. Three or four years ago we surveyed the people who bought them and about 0 percent would be using it 90 days after they bought the product. Today with speech products about 5 percent are using it on a regular basis 90 days after they buy the product. So that's an improvement. We must be getting close. But, those extra percentages of accuracy are very tough, and people get very frustrated when the computer does it wrong.

This is a case where handwriting is better than speech, because with handwriting you can actually look back and say, okay, that e looked like a c. So next time I do an e I'll loop the heck out of the thing. It's a conscious process that you can get better at. Whereas, speech is very tough. You know, you come in some day with a cold and your computer just doesn't work at all. And so this is one where I am still optimistic. We're going to get the speaker hardware better, we're going to get the software better. Microsoft on its own is one of the leaders in this, but I really can't give a solid date. I think it might be more towards the end of five or six years before we use this for dictation. And I distinguish that from command and control, where it's already okay in cases where your hands are tied up, like our AutoPC when you're in the car, if you want to pick a radio channel or call somebody or get directions, we do that through voice recognition.

Number one?

QUESTION: I wanted to follow up with you a little bit more about information devices. First of all, to understand a little bit better where you see information devices going, to what degree you think they will replace the role of the PC or be an add on, whether you see the PC being the center of a network of information devices, and also to understand better what Microsoft's strategy will be. Do you see Windows CE becoming the Windows of information devices, how are you going to prevent Java and other forms from taking that role?

MR. GATES: When you think about information devices, the key differentiating factor is the screen size and what applications you run. And so the tablet PC that I talked about, that device even though you've got new things like note taking and the wireless connection, it is a clear descendent of the PC, and it will run Microsoft Office, it will run the tens of thousands of applications that have been developed for the PC. You'll be able to take those with you and we'll get people to do enhancements so they work in a more natural way with the annotation pen type approach. That will be, in my view, the most popular full screen device, a device that you sit close to, and create and edit documents.

There will be less expensive devices that you can't create and edit on, that you just read off of. Those we'll offer Windows CE for. The amount of systems software revenue from a device like that will be pretty small. Actually, it's much more our breakthroughs like clear type that we'll get license fees for than anything else down in that space, which will be a fairly inexpensive space. In the -- where you get a much smaller screen, that's where we have chosen to go after Windows CE. And that's a companion device. Part of the beauty of what we do is we take all the documents, and contact lists and schedules that you create in Microsoft Office on the larger device, and because we match the display and the semantics, we can move the information back and forth automatically, including whatever customized views you've done. So that's one of the things we'll do on the small devices, I think, I hope, better than other people who do those things.

There are markets today for the smaller screen sizes are actually very small. It's three or four million units at, you know, $300 or $400. When you compare that to the PC market, 100 million units at still an average of about $2500 per unit up there. And the full screen device will always be a bigger market in terms of software opportunities, but it's very strategic for us to contribute to the small device, as well, because we think about this knowledge worker, and how they go to meetings, and how they go out and see customers, and our software is aimed at that total scenario, working across all those different devices. And so there is some deep integration, whether it's security or UI or customization that we provide across all of those things. That's the knowledge worker.

The TV set is even a little more complicated. There people certainly want to run a lot of software, they want to run games, they want to run educational software, and you're not going to run that remotely across the server. The game you want the interactivity, you want the speech recognition, you want all that richness right there. And so that game API, which is called DirectX, is a key asset as we move into the living room and create this device, which is like a PC, but the user interface, the robustness may make people think of it in a very new way.

MODERATOR: Two more questions.

MR. GATES: Okay, number two.

QUESTION: Alan Greenspan, the Chairman of the Federal Reserve, came out publicly stated that he felt that the philosophy behind the antitrust laws of the United States were based upon laws of economic fallacy. And one of the people that influenced him was an author Ayn Rand that wrote the novel in 1957, Atlas Shrugged, which deals with antitrust laws in a way that seems very similar to what's going on today. And I was curious if you've read that novel?

MR. GATES: Yes. I did read that. I think that having people like Greenspan speak out about these things is quite impressive. You don't even have to go as far as Greenspan, or certainly not as far as Rand to think that allowing businesses to keep innovating in their products, including supporting the Internet in their products, and providing those features without extra cost to users, you don't have to go very far at all to think that's a good thing. To me, you have to go about as far as common sense. And so, that summarize my view on that topic.

I think we have time for one more. I guess number three.

QUESTION: Again, I don't know if this is really the best way to end it.

MR. GATES: Okay, number four.

(Laughter.)

MR. GATES: Okay, go ahead. I'm just kidding.

QUESTION: And I certainly don't want to paint myself as anti-technology, because I certainly have embraced it, and I think it's been helpful. But when it comes to everything that's gone on and what's going to be available to my kids and your kids, the question is, do you think there's a lost art to be had in future generations because technology is going so fast, and kids aren't going to understand the card catalogue from anything. It's just a click away from getting anything they want. And their research skills could be compromised.

MR. GATES: I think it's a very interesting question. And there's a very valid point of view where people say, hey, this rate of change, it's unprecedented. And, you know, it creates almost a little bit of instability in terms of which businesses are going to do well, and which job skills really matter. And I think in different countries, you may have some who say, gee, let's try and turn away from this, let's try and hold back what's going on instead of trying to mitigate whatever the consequences are that people don't view as positive. Fortunately, in the U.S., I think people are saying, by and large, hey, this is overwhelmingly a good thing, and let's go into this with our eyes open, let's think about privacy and security, and all of those things.

I've certainly been part of these debates where people talked about, "what are we losing as we move forward?" There are a few things that come out in those debates. If somebody thinks that things have gotten worse over the last 20 years, or the last 50 years, if they think books were a bad thing, and TVs were a bad thing, and phones were a bad thing, then I guarantee you this is a bad thing. Because this lets you communicate with people and be part of many communities, it is a dramatic extension of what happened before.

One of the claims that people make that I think has some validity is, if you belong to these communities digitally, you know, what about your physical community, doesn't it make it a little bit easier to be more detached from the people around you know, your neighbors, the people who live around you? I think that's a valid concern. And certainly if you look at Japan, and how they help out their neighbors and they think more in terms of that physical community, there are some benefits that come out of that.

I also see that some of the things people are nostalgic about, I'm not that nostalgic about. You know, I've used a lot of card catalogues, I wouldn't mind, honestly, if there was never another card catalogue in the world. And somebody should collect them. Great, there can be, you know, some museum of card catalogues. So, I think, we all have our different views about what might be lost as we move forward.

Things that people want to have preserved, you know, take plays, live plays, those thrive today. You know, they're as popular as ever because people love that in person presence. You know, I didn't send a video tape or a video conference here, I came here to see you to talk to you about the things I'm enthused about, so you can really see how I feel about this stuff. When people want that, they're going to get it. These tools, they will shape the way they're used, and whatever the outcome they want, that's hat they'll get from them.

Thank you.

MODERATOR: Bill, thank you very much for a great presentation. I want to than the audience, too, for doing a great job. We'd like to welcome you to New York. We want to make you feel comfortable here. You can see that Columbia teachers are in presence out there. We didn't want to be different, and so we've brought a tasteful T-shirt as well.

Just one other thing, I don't know quite how to say this, but and maybe you didn't notice, you've been very busy, but, Bill, you're not wearing a tie. I don't know if you noticed that. And I know you're going to the Council of Foreign Relations next. And, Bill, this is New York, this is not Seattle. And as it turns out, the Faragamo family made for us a tie at our school. I always carry one with me, you never know when I'm going to run into a rich person, you know. And you'd do us a great honor and maybe feel a little more comfortable over there if you'd wear this as well.

MR. GATES: Okay.

MODERATOR: Well, you've seen the shirts and we have a hat to add to the shirts, and we really do thank you for the time that you gave us this afternoon. I think everybody benefited extraordinarily from your presence. And you're right, being here does make a difference. It's not the same as a videotape.

Thank you very much.

MR. GATES: Thank you. It's great.

(End of event.)