BILL GATES: Thank you. Thank you. Well, it’s fun to be here. I was thinking as I came how many people at Microsoft came from Stanford and it was amazing to realize that of the top four people three of them went to Stanford. And as John said, I am the fourth -- (Laughter.) But I didn’t stay. I realized the error of my ways and decided I could make do with a high school diploma. (Laughter.)

Actually, Steve Ballmer, who is now our CEO, also is a dropout but he dropped out from Stanford. He went for a year to the business school and then I kept calling him up and saying boy, I really needed somebody as good as he was to straighten out the fact that we had signed up to develop all this software and we weren’t quite sure how to organize it and turn it into a business and so he’s helped me run the business ever since then.

What I want to chat a little bit about is where the industry is going from where it is today. I mean, the success to date is really quite phenomenal in terms of 50 percent of homes having PCs, PCs essentially having become the most important tool of empowerment and driven a whole wave of productivity and additional communication and research and sharing that’s really fantastic. If the only thing that technology was used for is the advances in medicine that it enables to take place, that alone would justify everything that's been done, but, of course, it’s far more than that and the key point I want to make is we’re really just at the beginning.

The early dream was a machine that was easy to use, very reliable and very powerful of everything that you wanted to do. We even talked back in 1975 about how could we make a machine that all of your reading and note taking would be done on that machine. And, you know, we thought, well, this will be easy. You just get a high resolution screen and do the fonts right, we’ll go figure out about that, and here we are past 2002 and yet many of the key things we do are not enhanced by the PC today.

If you think of a typical day when you’re in meetings, the PC is not helping you that much. Actually, some meetings people are allowed to bring PCs and at Microsoft we even have a rule about whether you’re allowed to type during the meeting or not. A lot of the meetings that people at the front, at the table, are not supposed to use their PCs because they might do e-mail and pay attention to other things, but the people in the back rows are allowed to do e-mail and look at whatever they want to, and so you often want to be in the back row in meetings like those.

But, you know, the PC is not that effective there and when somebody calls you up on the phone when you’re talking over a plan or a budget or a trip or anything like that, you know, your PC is sitting there idle. It might be distracting you with junk mail coming in, but it’s certainly not part of facilitating what you’re doing.

If you go to a meeting and later you want to think, well what happened during that meeting, did somebody talk about this, you want to share a portion of that meeting and send it around to the people who are there, you know, that’s not possible. It’s just not done and yet it should be possible.

By the end of this decade the portion of things that are done through PCs, done on a digital basis, I think will grow dramatically and many of those are dreams that have been talked about over the last few years. You know, 10 years from now, will people think about their music being on a CD? I was walking with my daughter in downtown Seattle the other day and there was a record store and she said to me, "Well, what’s a record?" You know, she’s never seen a record and it’s not clear why it’s called a record store. She knows there are CDs in there but I think 10 years from now we won’t think about CDs because there’s a superior way to do that. Likewise for photos and staying up to date on so many things that we do in a very manual way today.

This, of course, is enabled by the miracle advances in the hardware technology. If there was a key insight in starting Microsoft, other than the importance of software, it was that the industry would structure in a way that people could specialize in what they were best at, and that was a change from when the computer companies really tried to do everything inside one company.

You know, IBM did the chips, the machines, the operating systems and all the systems software, as well as the sales force was all in one company and that machine, if you bought that machine you used only the software on that machine.

Our view was sort of a shortcut view that we only wanted to work on software, and so we hoped that with the advent of the microprocessor there could be this whole structure where we would simply do that piece and do platform software and then enable tens of thousands of other companies to grow up and do other types of software and have other companies do the hardware, and that structure has led to very intense competition. In fact, it’s clear the only real winner, the very big winner in that, is the people who buy these devices and get better and better deals.

There was a paradox in our original founding because we wrote down there would be a computer on every desk and in every home running Microsoft software, but the same week that I wrote that I said I didn’t want to have more than 50 employees. So one or the other must have been wrong. You know, we always said that whenever the company would get bigger, we said, well we might double in size, that would be about it. And finally we’ve come to the point where I’ll tell you that we might double in size and that will be about it, and this time I’m sure that’s got to be the limit of what can happen with a software company.

The improvements in the chips, the LCDs, the wireless network, those things are going to come at pretty incredible pace, and, in fact, I think that means that’s the reason why that software is the most interesting element. After all, if you have all the technology, say in your home you have these different screens and you want to just point to one and get the family’s schedule or pick music you want to listen to, underneath that there is a lot of technology. There is a lot of networking, network management going on, security to make sure that your information isn’t shared with the house next door or anybody that it shouldn’t be, and yet you’re going to want to be able to do that without being an expert in that underlying technology.

And so the challenge we have to dramatically lower the cost for businesses in terms of setting these things up and using them is exactly the challenge we have to meet in getting this technology to be widespread in small businesses and homes as well.

So people often say to us, are we most serious about the enterprise customers, the big customers who from a financial point of view are very attractive to sell to, or are we most serious about consumers, and we say it’s really the combination that drives us to do our best work; the demands for reliability that are best articulated by the big customers, although the consumers expect that as well, and the ease of use that’s best articulated by the consumers, although the enterprise customers want that as well.

Now, there’s a phase that we're going through in the industry of finally moving to do one of the Holy Grails of computer science and that is take all this connectivity we have through the Internet and through the application of just HTML browsing, and say that we want software running on any computer in this network to be able to locate software with capabilities of interest on another computer and even if the people who have written that software have never met, the only thing they have in common is that they’ve seen some standards, specifications or tools, that software will work very well together.

That’s this whole Web services dream and it’s a fascinating thing to be working on, because it really is forcing the technology forward.

The prototypical application for this is, of course, e-commerce. And some people might say, well, didn’t a lot of companies promise that e-commerce would happen overnight a few years ago? And the answer is, yes, they did, but in any meaningful sense it has not happened. It is true that purchase orders are sent electronically and that saves that one paper document, but if you think of two companies where you have the knowledge workers in one company and the systems in that company and then on the other side the same thing in the other company, the inefficiency in terms of the phone calls, the faxes, the e-mails, trying to update the different various systems to have the same view of a complex transaction that the two companies are engaged in, actually there’s more effort spent in just impedance, moving the information around, than there is in the real work itself; that is, the work of designing the products or doing good customer service.

And that’s why, as what’s been a dream in computer science becomes a reality over this next decade, the productivity contribution of the industry as a whole will actually be greater than it was in the 1990s, and that was a very significant thing.

You can get a sense of the breadth of how much we believe in software if you look at products that range from SQL Server, that’s an advanced database, or Visual Studio .NET, which is designed around these XML Web services type developments, all the way down to Xbox. Xbox fits into the strategy perfectly. It’s a lot of fun to be doing a product like that, but it’s also a product that because of the rich capabilities will really kind of surprise people how that fits into the home network. Because it’s a full-blown computer with a disk, we’ll be able to let people have music and photos and TV guide, and as you’re watching TV, if there’s a sports score that’s interesting to you, you have that appear exactly the way you want it, customized to your particular interests.

And so there’s an approach here that says that what we were doing that was focused on individual devices in the past, where your files were on a PC and your preferences were there and then when you went to another PC, it was up to you to move the information around, we have to think about these things as being more centric around people.

A lot of these advances I think will come quite quickly. Later this year we’ve got the first Tablet PC devices coming out, and this is one of the early versions of that. This one is actually made by a company called Acer. In this form it’s just like a normal portable machine, fully capable, runs all the standard software. But if you don’t want to use the keyboard you simply flip the display around and use it this way and it’s a tablet, and there’s a pen here and you take notes. If you like your handwriting, you can leave them as notes and send them around. If you want it to be recognized, the software that does that has gotten to a really amazing level of quality.

Handwriting-recognition software is a perfect example of the kind of software research that requires patience. About five years ago there was a wave, some of you will remember, there was a wave of this type of tablet kind of machines, but basically nothing was ready. The hardware was too big. The battery life wasn’t good. The way it felt to write on the surface was incredibly clunky. You know, we’re all used to pencil on paper, and we’re very demanding about that. And so it all kind of fizzled out, and the people who had worked around that, most of those companies didn’t continue to push forward.

Well, clearly one of the elements that was necessary there was the handwriting recognition work, which is a very tough thing to do, and so we kept pushing on that and pushing on that, and it’s only in the last two years that the big breakthroughs have taken place.

One of the interesting things about this device is, you know, I was saying the idea of reading, doing all your reading on one device, having your notes there, being able to search your notes, as soon as you get reading to be comfortable on a digital device it creates a real mode shift where you think, OK, when I read a magazine and I see an interesting article I’d really rather do it here because I can take the notes and send it off to a friend. And so the idea of just being able to push and select the group of friends you want to share with becomes a very natural thing.

The idea of being able to say, well, I think I read something on that topic and whatever devices you’ve done reading on, of course, know that and can go and find that for you, it’s a very dramatic advance.

So the number of hours that you get empowerment out of this device can be increased very substantially as this form factor matures. There are some that are a bit thinner and lighter because you actually detach the keyboard, so if you’re going to do a lot of reading, say, sit down in bed and read something long, you’d get an even thinner device but it’s a fully capable device.

The one other show-andtell thing I brought is this one, which is basically a PDA phone-type device, and there’s a lot of discussion about whether is this one category or two, and we sort of take the point of view that the two categories will really come together because we can do voice recognition down on this form factor -- the processes have gotten capable enough -- because people want to run little neat applications ranging from games to communications things to just seeing sort of a business alert desktop here, the idea of having a rich color screen makes a lot of sense.

And another part of the vision is that the world of phone calls and the world of screen browsing, which are today very distinct, that those two things definitely should come together. And so if you’re calling somebody up, whether it’s at your desktop or when you’re mobile, the screen is active and so you can sit there and exchange the menu and the map or whatever you’re interested in without setting that up as a different thing.

So the next few years, we’ll see this explosion in devices that work together, wireless networks, both the Wi-Fi and the 3G around that, and a class of applications that are built around these XML standards that really are about taking the dreams of distributed computing, the dreams of object-oriented databases and making those a practical reality while making sure to include all the kind of rich security and reliability requirements that a business infrastructure like this will drive forward.

There are many tough technical problems still to be solved, and one thing I have been a little disappointed in over the last several years is sort of the reduction in commercial research that’s taken place. One of the great strengths that Microsoft has had is that we not only have grown our R&D budget, which is about $5 billion a year, about three times as much as the next biggest software company, which would be IBM, but we’ve also grown the part of that that relates to our different research activities.

Now, part of what goes on there is not just our own research but also reaching out to the leading universities, including a lot of work that’s done with the great people here at Stanford.

That kind of investment has always had a huge payoff certainly for society and sometimes for the companies who do that work. The famous example of that, of course, is Xerox Park, who did a lot of the great work that brought us things like the Macintosh and the Windows PC that progenitor ideas were definitely developed there at Xerox Park. Microsoft is proud to employ many of the people who were there doing that work and may in turn have hired in a new generation of great, young researchers to carry forward the founding work that they really did back in the 1970s.

It’s kind of interesting to think of who are the leaders in American corporate research. Well, Xerox was. That’s certainly come down a lot, as the copier business or expansion to other areas have been tough. AT&T certainly was and as communications has been deregulated, although it’s split into different pieces, the aggregate amount of research there is substantially reduced. And even within the software industry, which you would think would be the most research-oriented, after all when you get a hit product the economics of our business are very, very attractive -- if you don’t get a hit it’s not very attractive because you’ve got your tens of millions of dollars of your fixed cost but if you get a hit product as soon as you recover that tens of millions of dollars of fixed cost, or in our case billions, but that’s okay -- (laughter) -- then whatever additional copies you sell are either there to invest in creating future versions of that product or other products or to be taken as profitability.

So you might think that this would be an industry that would take a very long-term approach and yet actually in the computer business as a whole the amount of research going on is actually quite a bit less today than it has been in many years past. It’s partly because it’s so competitive and it’s partly, I would claim, we’ll look back later and say that that’s a shortsighted thing, that there should be more of that that should go on.

In a sense, you can say that just underlines the critical role that universities play in making sure that the fundamental research continues to be dealt with and that that hopefully will keep the United States in a position of leadership in creating products that are beneficial definitely throughout the world.

I think another thing that’s very important to us is the idea that software is just at the beginning. You know, the way that people update software, the way that people try and secure software and install software, sometimes when we look at it it’s surprising we’ve sold as much as we have. You know, it’s only recently that we put into our Windows product this self-monitoring capability so that if something goes wrong, if the application stops working we see across the network what went wrong. In fact, actually what happens is the user is given an opportunity whether they want to send a report back to Microsoft, and about 70 percent of people say, "Yes." It’s just sort of like a core dump -- that’s kind of an archaic term but it’s just exactly what went on and that allows us to trace was it a device driver that was incompatible with a piece of hardware, was it a third party application and the way that interacted and actually take an ecosystem that is very exciting because of its openness; that is, nobody needs to get permission to write an application or build a peripheral or anything and that you want to preserve that kind of flexibility. You don’t want to have to have something certified and pay some certification fee or have the delay there because not only does that slow things down, the incentives on that whole certification process have a tendency to reduce the number of people who would participate in it and come up with these new things.

So the question is how can you get the best of both worlds? How do you get the closed loop analysis of really understanding what’s going on in that user experience and yet get this openness? And the answer is very straightforward. Now that we have this unbelievable connectivity it’s through software that does that monitoring, sends back information.

Today we're mostly getting information about what you’d say are sort of red light situations where things really don’t work together, but we have enhancements in mind where even if the user is a little frustrated, where they say well it worked but it was a little slow, it worked but it was a little confusing, we want to get the feedback loop working on that as well. And so that’s the information coming back to us and then going forward we want to use the ability to do constant updates to be able to make the improvements so if one person runs into a problem the other people get the benefit of that fix hopefully before they ever run into it.

One particular challenge that comes up there is one of great interest in our industry today and that's the issue of security; a very broad topic, a very tough topic but one that if we don’t’ solve it and make big advances it will actually be something that holds back the realization of all these dreams.

Aspects of this include, will we be able to rely on passwords as a way of identifying users, and the basic answer is no, passwords are terrible. People pick passwords that are easy to guess. They use the same password on very insecure systems. They use it on secure systems. And so we’ll have to go to something more akin to the smart card plus a password in order to have these systems be secure. But that’s just the beginning of the number of elements that need to be put together here.

Another interesting problem that relates to security is the whole privacy issue. One of the features that we’re building into our software is the ability when somebody wants to communicate with you that you can rank according to who it is and what it’s about and what you’re doing, you can rank how important that is to you. It’s a Microsoft Research project called Best Communications. And so something that’s unimportant, junk mail, would never waste your time or something that’s very important might even alert you in the middle of a lecture that, I don’t know, your bathtub was left on or something that you really needed to pay attention to.

In designing that system in such a way that it’s both secure and that your attitudes towards these people who are sending you mail aren’t revealed to them is very tricky. We actually have some designs that work very well but people find out very quickly that you’re not interested in talking to them at all, you’re not interested in reading their e-mail and that’s not very adequate so we’ll have to move beyond that type of design.

So in conclusion now I’d say that this decade is the most interesting one that the software industry has ever had. You know, some of those early years of staying up all night with times sharing, that was fun but now in terms of the breadth of impact and the intellectual depth of these problems, there has never been anything like what we're facing today, and yet I’d say by the end of the decade what we’ll have will be almost completely different, at least dramatically better than what we have today.

Thank you.

(Applause.)

MODERATOR: Well, thank you. That was wonderful. So I thought you'd be in an interesting position to address the question, which I think has bothered me and certainly which I hear very often from our students, which is how do you see this battle between content providers, who seem to be so focused on intellectual property protection, and technology companies, which would like to get new media and videos and music out to a much wider audience in a new way? How do see that battle resolving itself?

BILL GATES: Well, that battle is actually pretty intense right now. There is a move in the Congress, pushed by the content people, that these PCs should be required to basically have sniffing software that looks and if you're playing a piece of music it looks and says, hey, that looks like one of these popular pieces of music that I know about and hmm I don't see that you have the special rights envelope for that. You know, I'm going to basically report you because of that.

MODERATOR: Microsoft doesn't do this, right? (Laughter.)

BILL GATES: Oh, no. This is Intel and Microsoft versus I'm afraid it's actually Disney who's on the other side of this one. And we consider that an extreme position.

Now, the idea that if something has a rights envelope around it that there should be a mechanism in the operating system to honor that rights envelope we actually believe in that because that comes up in terms of not just music but say business information or medical information. If you want to have a piece of e-mail about some medical condition that you're sending out to three or four people, the system ought to be able to guarantee you do that that mail is not forward to other people, that it's not printed out, that you're not cutting and pasting it into something else.

And so we need a rights management structure, and this actually has an impact all the way down to the microprocessor level of the microprocessor verifying that you're actually running an operating system that you want to run, that it hasn't been spoofed and a whole hierarchy of trust going on.

So we're doing a lot of research that would say to the content industry, okay, if people buy things with envelopes we won't make our devices always be this hacker's tool, but then again if people get the bits we don't think it's appropriate for PC manufacturers to be forced to have software that does that.

Now, what the content companies say to that is oh great you don't mind because you get to sell to businesses and so you can fund most of your R&D selling to businesses and then you can have special educational prices or make things very attractive at the consumer level so it's in some ways less of an issue for you, and that's a fair criticism. I mean they are facing an interesting problem and at the end of the day I don't think there are many people -- you know, maybe the free software foundation or something -- who doesn't think that there ought to be a mechanism for authors of various types to basically if they want to be paid for their material, whether it's a book or music of software, some mechanism that works pretty well and isn't too inconvenient for the listener to make that work.

MODERATOR: I think that's probably right and the challenge is to find something that's not too inconvenient and that reflects some reasonable cost, and then I people would probably use it.

BILL GATES: Yeah, it raises some really interesting questions about the nature of the rights you get. You know, if the rights are to you personally it raises the question of well do you log on to every music player that you have, you know, have that authentication take place. And so I think this debate will rage for the next five or six years as the technology matures.

MODERATOR: You know, one of the things I've observed is my PC has gotten faster and faster and faster and when I now use the Internet all the time it's great when I'm here at the office but then I go home and I have this tiny little pipe and now so much of what I do is on the Internet and the performance stinks. It's terrible. A lot of the things you do here you can't do at home. So it doesn't seem that that situation is improving very quickly. Do you see any hope for solving that so-called last mile problem in the home?

BILL GATES: Well, you'll have to move on campus. (Laughter.)

MODERATOR: I did. (Laughter.) It solved my problem. (Laughter, applause.)

BILL GATES: Yeah, it's interesting that that dichotomy of essentially the business experience versus the home experience, and, you know, a lot of students are going to be spoiled by having broadband and then having to give that up.

Yeah, I was saying that the hardware people are performing most of the miracles we want them to perform. Really the one footnote to that is this consumer broadband question. And it's not totally that they're just being sticks in the mud. The economics of that last mile are particularly difficult.

Now, there is one country, Korea, where through sort of fierce competition and good density and good policies broadband costs are down at about $20 a month and so over half the households have broadband and it's had a super dramatic effect in terms of that's how people get video and online games are more popular in Korea than anywhere. So it's almost a microcosm of seeing what's going to happen as that works in other locations.

I think at $40, $50 a month you're never going to get beyond, say, 15, 20 percent and that really holds back the breakthrough applications that you'd like to see.

Now, if we think in the five to 10-year time frame there are ways of using the airwaves and getting more information across that basically can get us around this last mile problem. And anybody who has an idea to get around this last mile problem our treasury has a little spare cash; we'd love to help finance them because it's the one thing that is a big disappointing and is going to slow things down versus what we'd like to see.

MODERATOR: Is this something the government ought to take an active role in trying to promote or make happen? I know you're not big friends with them right now. (Laughter, applause.) With the states' attorneys general, I mean.

BILL GATES: That's right. Well, you know, there are two ways to look at that. One is to say we have this business where we sit up there and write software and all we do is send people bits and they send us money. And why does that work? Well, the government deserves some credit for that, so overall it's a good system, and every once in a while it doesn't work perfectly -- (laughter) -- but in terms of this broadband thing it's hard to think -- the government, of course, is in this and that's partly why it's such a messy thing is they encouraged these CLECs to come in and sublease infrastructure and then they didn't do it well enough for that to work and so most of the CLECs have basically disappeared and now you have this funny thing where the phone companies say, "Well, we're not really incented to do it right because somebody can sublease it from us and so we don't want to, so give us our sole permission to do this." And it's really tough; once you get down that regulatory path then figuring out how to back out of that is tough.

I don't think the magic is in the government. I don't think the government should subsidize these things. There are plenty of other priorities for the government.

I think that it will take some patience and some technology breakthroughs and then we just should make sure the regulatory framework doesn't do anything to hold these things back.

There is a particular issue we're interested in right now, which is saying that if you give somebody a cable modem or DSL that you can't restrict them from using that to create a neighborhood 802.11 Wi-Fi network. You know, some of the broadband providers are trying to put that restriction into the contract so that you can't do that and that, we sort of hope that doesn't happen because there is this way of using Wi-Fi to sort of do the equivalent of the party line and get more people sharing the connection.

MODERATOR: If we were sitting here 30 years ago and I asked you what do you see coming that's really going to change the face of computing you might have said the microprocessor. Twenty years ago you might have said the personal computer. Ten years ago you might have said the Internet. What do you think is really going to provide that kind of monumental change now? Do you see a technology on the horizon that's going to change things as dramatically as those technologies did?

BILL GATES: Well, I actually believe that when you change the form factor and when you get speech and handwriting to work that is just very dramatic and it's, of course, because of the way that builds on the Internet connectivity and all those things.

But people say, you know, why aren't certain Internet activities mainstream? Well, I think when you have to go up to a chair and have a CRT and kind of turn the thing on there's just tons of stuff you don't do.

Take photos. We, when we had the early tablet did this field study. We sent out a hundred tablets and had people use it. And we had people ranked, we had monitoring to see what they did on it and we compared that to what they'd done on their PC before. One of the things they did over twice as much was photo browsing, but we did nothing to do with photo browsing, but it's just that when you have it in your hand you can say, "Here's my family photos; let's look at this one, look at this one, okay let's look at that one." It's very different than you saying well come over to my CRT where there's only one chair sitting there and it's a bit of a different experience.

So I actually think this stuff that's coming out over these next two or three years that in a way we'll look back and say it's common sense. Of course I didn't use the Internet for quite as much shopping when it wasn't in my hand. Of course I didn't use it for news and organizing my things.

And this idea of getting the phone experience unified in so that you just have this little tablet you carry around and when somebody calls you up you can see all your previous e-mail with them and documents and things like that, I think that's such a nice evolutionary step that we might almost miss it as something quite dramatic.

So to me that's probably the biggest thing remaining is to get the natural interface, handwriting, speech and it's in my hands like a tablet and solve those and not much compared to that.

MODERATOR: It would be great if voice recognition was good enough that it would be better to talk to my computer than to type, and that would be wonderful if we could get to that day, I think.

BILL GATES: Well, Butler Lampsen now does all this stuff with voice now and he thinks it's good enough, so the place this really should catch on is in China and Japan where the keyboard is relatively disadvantaged and we think we are starting to see that.

We had a contest at Tsinghua University where the best Chinese typist got up and competed with our speech recognition people and the speech recognition was a lot better. But, of course, typing Chinese is a lot harder than typing English.

MODERATOR: Yeah, that's probably true.

You mentioned the issue of computer security but I think one that lots of people are just as concerned about is the whole privacy issue. You know, they go up to a Web site and who knows what that Web site is snarfing about them, or they run an application and you mentioned the case of the recording companies wanting to send the information back to find out whether you're a valid user.

So I think people are also concerned about what data is being captured about them as they run these applications, whether it's downloaded from a Web site or whether it's native in the application.

How do you see this whole concern resolving itself so that people don't become afraid of what information their computer is sending back to the mother ship about them?

BILL GATES: Well, we see this phenomenon in a pretty big way because we have this Hotmail system. And a lot of the people who sign up for those e-mail accounts just want an e-mail account that is unknown to everyone basically. You know, it's not your work e-mail account. Your employer is not seeing what's going in and out of that. It may be an e-mail name that can't be traced back to you because you want to just be anonymous in terms of the things you're doing, and it's just a huge phenomenon.

MODERATOR: What exactly are you doing that you want to remain -- (Laughter.)

BILL GATES: You're resisting an oppressive government. (Laughter, applause.) In some other country.

MODERATOR: A right guaranteed by the Constitution, right? (Laughter.)

BILL GATES: I don't know what they're doing. (Laughter.) That's why they like Hotmail.

No, we're going to have to have the right balance there, but people, you know, today when you're using a wireless phone there is a record basically of where you've been and actually there's been quite a few court cases where people proved that somebody really was somebody by using that kind of tracking information.

And society is going to have to decide where information like that is available and not available. I think in general it will not be available, but you will have situations where you get serious terrorist type events where maybe the information in special circumstances that can be accessed.

The unusual thing today is that people are reluctant to give their credit card number, whereas that's actually probably the least damaging thing in the sense that if things show up on your credit card it's a reversible situation. It's really up to the credit card people to deal with the fact that you say, "Hey, I wasn't there, I didn't do those online transactions."

Partly people have got to get more used to this system. Partly we have to have levels of privacy where you go to a Web site, you can actually recognize what's going on. There's a thing called P3P that we've been involved with that's a step in the right direction. I won't claim it's as good as say the movie ratings system that's just kind of common sense at this point, but that's really got to be done, and then people have to realize that there's a lot of information about them out there, that it's only as they're getting involved in using computers that they realize that they ought to really think about what the rules should be for the use of that information.

MODERATOR: You know, in some ways it's a mixed blessing because if there's some information about you stored there you come to a Web site it makes life a lot easier, right, Passport and things like this really can make your life easier for doing transactions on the Web and then you just worry about are you secure and is the site trustworthy and that's the hard problem.

BILL GATES: Yeah, you have to get into that tradeoff because otherwise you sit down and you say, well, I'd like infinite privacy. You know, why not, give me all the privacy in the world. Well, is it really true when you go to a store the second time that you expect them to pretend they never met you or anything? Well, if you come with one girlfriend and you were there with another girlfriend before, maybe so. (Laughter.) I don't know.

MODERATOR: Absolutely.

Well, why don’t we throw it open to -- (Laughter, applause.) Why don’t we throw it open to the audience for some questions. Let’s take some questions from the audience. We have two microphones, one in each aisle there. Over here.

QUESTION: Microsoft curator Michael Kline has talked about the focus being on collecting contemporary art from living artists, which is very exciting and I know you yourself have collected a lot of American art. And I was wondering if you could speak about the importance of preserving our cultural heritage and being stewards of fragile artistic treasures, and as a follow-up to that, if you see any of the artwork that you’ve either collected or been intrigued by, any of the complex concepts that are embedded in those works as intersecting with any of these issues and problems that you are going to tackle in the next decade that you talked about today? (Laughter.)

BILL GATES: OK. I’m going to have to look harder at my pictures. (Laughter.)

You know, art is something that I think can be made more approachable through technology, learning about the artist and the period that they were involved in, you know, being able to see reproductions of the art online, you know, not having to deal with all these slides and things. So I think in a museum some of them are actually using technology in a very nice way to make walking around and picking the things you care about, learning about those is fascinating and being able to follow-up on those things after you leave the museum.

You know, I enjoy the art I have. I can’t think directly how it’s influenced my work in technology. The main thing I have is actually a library where I’m collecting, like I have the drafts of Feinman’s books and I have -- anyway I have some --

MODERATOR: New artistics.

BILL GATES: Yeah, very. (Laughter.) Uh, yeah.

MODERATOR: How about over here?

QUESTION: Hello, Bill. My question is the following: What applications do you foresee in the future for Microsoft Xbox?

BILL GATES: Well, first and foremost the way that Xbox is making a name for itself as the best game machine ever, and -- (laughter) -- and --

MODERATOR: That’s an unbiased opinion.

BILL GATES: Well, it’s got to be. (Laughter.) At least that’s what the Xbox people have been told. (Laughter.)

And there are a lot of neat things the Xbox can do beyond gaming, but one thing that was clear when we went out and talked to consumers, if you first brought it out and said it slices and dices, that the message that it was the best slicer ever made just wasn’t going to get across and anyway because there isn’t that much broadband and things or not that many people have Wi-Fi in their home yet, many of those things are more interesting as you get further down the road anyway.

Now, the breadth of software we’ll have on the machine, because we have a disk, things like kids’ education software or encyclopedia type things, things you wouldn’t think of on the videogames that don’t have a disk, you will see us broaden out quite a bit. But we do have to be the best at the classic genre of game things and this Christmas will be the really interesting Christmas because last Christmas we were hardware constrained. We sold every box we could make. This Christmas, both ourselves and Sony, who’s our primary competitor, will be able to make as many as people want, and there will be a ton of games. So the fact that they numerically have more games won’t matter; you’ll be able to just see we have plenty of games, they have plenty, which ones are really the best. And so we think it’s going to be an interesting Christmas. (Laughter.)

MODERATOR: Let’s go back over to this side, yes.

QUESTION: Yes, I was wondering, it seems that as your vision for connectivity between applications and devices becomes more of a reality there also seems to be an ever-increasing possibility for some sort of security threat to produce some sort of digital catastrophe. I was wondering how you go about balancing innovation with security and do you ever think there should be some sort of limit on how connected things are?

BILL GATES: Well, there’s definitely society’s reliance on certain infrastructure today; electricity, you know, we’re very reliant on that working very well. Not many of us know how to plant seeds and grow food and we’d better have this.

We’re already pretty far down the path of being reliant on these digital systems. You know, if the Internet, say, shut down for a week, and I have no reason to think that will happen but there’s no guarantee that it won’t, that would have a huge effect on business productivity.

We see it with our e-mail systems, that if e-mail is not working at Microsoft, people go out into the halls, like, geez, let’s go home. (Laughter.)

MODERATOR: Do you lead them out the door, Bill?

BILL GATES: No, I head down to the Exchange Group to find out what’s going on.

So there are cycles of going back and focusing on these basic issues and we’re definitely in a cycle right now where the next big innovation of Windows is what we call Longhorn, and we’ll be about seven months longer doing that because of this priority we’ve put on doing all of these security reviews, going back and saying, how do you have less security vulnerabilities, how do you fix those vulnerabilities very quickly and if somebody ever has a problem, how do you make recovery into something very trivial versus what it is today, where you kind of have no visibility of what’s going on and you can’t just say OK, I want this system to be like it was two weeks ago when there wasn’t a problem. I just want to push a button and I want it to be like it was two weeks ago.

And so there’s a huge investment in all of those things. The nice thing is that it’s not a dilemma for us. The marketplace is demanding these things and some are the very avant-garde users and we use the demands of those users in order to get down into the learning curve, because we know the other users will want that as well. Some of the government users are the most demanding in these different areas.

So I don’t think there’s any absolute limit we should set but I think there is a little bit of a mea culpa industry wide saying okay we haven’t been as focused on this as we should. You know, basic things like the current mail protocol you don’t know if the mail you get really came from that person, and you can create mass confusion by sending mail that appears to come from the IT department or the president of the university or your friend because it’s just an insecure infrastructure; so a whole bunch of things like that that need to be improved.

QUESTION: While I recognize that you cannot comment on any pending litigation, I’m curious as a former judge what impact all of this litigation has had on you personally? And looking back, do you wish you had gone to law school? (Laughter.)

MODERATOR: I don’t. (Laughter.)

BILL GATES: Well, my dad is a lawyer and actually when I was quite young, I thought maybe I would be a lawyer. I certainly have gotten a chance to learn a lot about the law. (Laughter.) And, you know, you always learn as you go through these things. There are some things we were naïve about in terms of reaching out politically, being involved politically. We went too many years saying, you know, boy isn’t it amazing we’ve never had any office in Washington, DC and all that, which up to a certain point really is a great credit to this country. I mean, I had to go and lobby in the capitals of dozens of countries before I ever had to cover my back in our capital.

And, you know, I don’t think there are many companies that have had a phenomenal level of success without running into having to learn a lot about the legal system and having lawyers work for them.

I do think that it isn’t going to be a fundamental impediment in terms of the industry or Microsoft or anything like that, and we have done, I think we’re very proud of the fact that actually during these last three, four years, which were very unusual years because we had to make sure the lawsuit wasn’t a distraction for us, we had to make sure that a little bit of the short-term hype, which made things a little bit more difficult in terms of hiring and getting people to focus in, you know, my friend is worth 5 million and here I am working on handwriting recognition. Yes, that’s great. You’ve got to make sure. Well, then 5 million disappeared, so it’s okay. (Laughter.)

But staying focused on the things that we’re really good at and not letting ourselves slow down is something that I think we’ve done quite well.

MODERATOR: We had originally thought we’d end at six but we have an extra 10 minutes or so, so we have another question here.

QUESTION: You were mentioning that people can use technology to resist oppressive governments, and I was wondering by the same token technology can be used to repress people, for instance, Cisco provides the routers and the packet system technology for China. And I was wondering what do you see the role of the technology company being in the process of freeing people as opposed to trapping them? And is it a legislative solution or a technological solution?

BILL GATES: Well, fundamentally technology makes it hard to control information, and I don’t see any dilemma that we’re facing. I mean, nobody is coming to us and saying, you know, put backdoors into software reports or anything like that. There’s nothing like that at all. And there’s almost a perfect correlation between the use of personal computer technology and how democratic a society is and, in fact, I’m very optimistic about China because right now you’re seeing this huge rise in the use of the personal computer. And there may be some substantial lag time on the order of a decade but the kind of developments that are taking place there and the kind of empowerment that comes out of that will lead over time to more political choices probably in a smooth evolutionary way.

So I think everybody in the technology industry can feel good about that and there’s nothing pushing back the other way. There’s nothing where somebody is saying build this thing in a way that prevents information from being shared.

MODERATOR: It’s been fascinating to watch China where, early on, there seemed to be a lot of filters and parts of the Internet that were blocked and people couldn’t get to and that seems to have completely vanished now.

BILL GATES: Well, it’s tough. The scale of the thing just gets to be a real problem. And so I think Singapore was an early interesting case, because Singapore has a lot of fascinating things about it, and people who run governments differently, we should always be open minded to learning from them, they always had this view of a little bit of a paternalistic point of view of magazines we don’t like just won’t be in the country. But as the Internet came along they had a choice. They had to say OK, are we going to be the high-tech front-office place for that part of Asia and really embrace technology, or is this idea of being able to control information consumption in our country the thing that’s the highest priority? And they knew there was a choice and they explicitly made that choice.

I think in China today that, certainly we have a research group in China, and we’re exchanging information with them massively, you know, videoconferencing, exchanging things. There’s no firewall sitting there blocking the bit flow between what we’re doing between our research group in Beijing and ourselves. I mean, that would be a very visible thing the way the technology works. So I think the whole thing about how much filtering is taking place is in some ways more discussed than reality.

QUESTION: So it seems like one key ingredient of making software more powerful and yet more easy to use is making it more intelligent, and I know Microsoft has a really incredible group of researchers doing artificial intelligence work, but it seems like not very much of that has really has really made it into prime time. And so I was just wondering what you think of as kind of the criteria for evaluating when that type of software is really ready for prime time?

BILL GATES: Yeah, I’m one of the great last AI optimists that I think it’s strange how little is being put into it. It’s another case where there’s less being spent on AI today than, say, 20 years ago or 10 years ago and I guess DARPA is under attack in terms of what scale they’re going to be able to do things at.

We have quite a few people who work on various different types of decision systems, BASEAN decision systems and doing what might be called learning, using those modeling systems. And we feel like we’re getting very good results out of that. For example, this thing I talked about of being able to prioritize your communications, you would never go to the trouble to explicitly write a program that had the/if statements that sort of did that. It’s watching you in terms of which e-mails you pick, and then if it misprioritizes you point to something and say, that’s not quite as important and it’s up to it to figure out, hmm, I wonder why; was it who it was from, was it what it’s about, it’s various things like that, and so putting a rich BASEAN decision system underneath that is what will make this information agent type thing a reality.

So I don’t know if you will use the term AI because that’s become so wildly defined. It’s sort of like is it human, which certainly none of the stuff we’re doing in the next decade fit that path, but the intelligence of these devices in terms of knowing what you’re trying to do, being easier to work with, where eventually you shouldn’t have to say which application you want to run; you should utter the thing you want to do and it should bring in the application that deal with that. So I think there will be a lot of progress that will surprise people in even the next two years.

QUESTION: Palo Alto has a cyber-ring that is used currently by businesses and we’re considering extending that to the home, inexpensively we hope. For how long do you think this could be a sustainable municipal utility, like gas, like water, like electricity? Or put another way, do you think that airwave technology will quickly supercede the cyber infrastructure? And if you think it will work, do you mind helping fund it? (Laughter.)

MODERATOR: You can skip the last part, if you want.

BILL GATES: There are lots of different priorities to picking philanthropy and right now the foundation is primarily focused on global health and learning-related things.

(Applause.)

In terms of the broadband stuff, I don’t see anything in the next five years that’s really going to change the idea that it’s some form of cable modem or DSL will be the primary broadband. You know, I hope somebody can change that, but generally, say this UWB stuff and how that might be used, five years is very optimistic for all the pieces that need to fall into place, although it looks like it has some real promise. So I don’t think it would get obsolete very quickly.

The whole issue of whether municipalities should essentially get into businesses like that and compete with the commercial providers, that’s a political question that can be decided in different localities.

I think Portland did something like that, and I don’t know if it ended up being a success for them or not, because I don’t really track that, but it won’t get obsolete in any rapid fashion.

QUESTION: I just wondered what is your biggest challenge in Asia and how are you dealing with that? I mean, they have a very different value and different legislature and everything like that.

BILL GATES: Well, very early on, Microsoft decided we had better be in Japan and China and India because in the long run those were going to be very big markets. And so there was a very steep learning curve for us in terms of figuring out how to really hire people in those markets and how to relate to the companies in those markets. There’s a very steep learning curve in just getting our software to work with the languages there. That now seems really easy in retrospect but there was a lot of very hard work that went into that.

Amazingly, the interest in basic productivity things, like forecasting, sales analysis, e-mail are not that dramatically different in most of the Asian markets. In Japan e-mail hasn’t caught on quite as much and it’s a special form of e-mail that we’re actually customizing our e-mail for Japanese workflow because they’re very used to this group stamp type system, so you get some nuances like that that are very interesting.

In Japan the chart types that Americans are used to understanding are very basic: bar chart, pie chart; that's all they understand. (Laughter.) In Japan they love all these double donut, separate connected things, and so we put all those chart types into Excel for anybody in the world to use that they want, but they’re still fundamentally used in Japan. (Laughter.)

But the software we like to do is the software that’s fairly universal in nature. Now, we have exceptions to that, things like our Encarta encyclopedia or the Microsoft Money package, but then you’ve got to watch the economics because that means you’ve got to rebuild it for each of these different countries.

Parochially, I’d say that our greatest problem in Asia is software piracy. In the long run that will be okay, but our dollars per PC in a developed country, revenue per PC sold, is on the order of $100, and in China it’s less than $10.

Now, OK, they’re using our software -- (laughter) -- so over time I’m sure there will be a revenue source that comes out of that, but we’ve really wanted to be in those markets early on and our business in China took off a lot faster than we expected.

India is another special case for us because there are about a million software developers there that are doing software services export, so it’s very important for us to reach out to them and get them our software tools so they can use the Microsoft platforms in the work that they’re doing.

And I’d say one of the most fun things about my job has been going to these different countries and learning about these countries. We’re starting to do -- we do quite a bit of research in Beijing and now we’re doing more and more R&D in India and so we’ll get a lot of chances to go over there and build up the investments we’re making in those countries.

MODERATOR: Well, let me ask you to close with a personal question. If you were alive in the time of John D. Rockefeller or Andrew Carnegie, they were recognized as great business leaders. Now 50 or 100 years later, they’re primarily recognized for their philanthropy. What do you want to be recognized for 50 or 100 years from now?

BILL GATES: Well, I don’t think optimizing for being recognized is a good approach, because I think you could really get confused. I mean, I enjoy going into work and working with very smart people on problems that are very complex but if you solve them well that they can have a big impact. And as I’ve gotten involved in the foundation that same sort of problem solving thing, you know, getting these great medical specialists, getting the chance to meet with them, who’ve dedicated their lives to schistosomiasis, which nobody in the U.S. has ever seen, but hundreds of thousands of people are suffering from, and, you know, listening to that and learning about that, that’s great, it’s exciting, and being able to bring some insights about intellectual property or political popularity and how these issues can get on the agenda in these countries, that’s a really fun thing.

If I had to say what is the thing that I feel best about, it’s being involved in this whole software revolution and what comes out of that, because you can go all over the world and go into schools and see these computers being used and go into hospitals and see them being used, and see how they’re tools for sharing information that hopefully leads to more peaceful conditions and just the great research advances that come out of that.

So from a professional point of view, the software is the thing that, you know, I think I find most gratifying.

MODERATOR: Please join me in thanking our guest.

(Applause.)