Bill Gates: College Tour 2008

Remarks by Bill Gates, Chairman, Microsoft
College Tour 2008, University of Washington
Seattle, April 25, 2008

MARK EMMERT: Good afternoon. (Applause.) Thank you. Good afternoon, and welcome to the University of Washington, I’m Mark Emmert, the president of the university, and I’m delighted to not only welcome you here to the campus, for those of you that are guests, but also to welcome you to the last stop on the Bill Gates Unplugged Tour. Bill has now been to five other universities, and spoken to packed houses, none of them as good looking as this crowd, though. (Applause.) None of them as smart as this crowd either, is that right, Ed? (Cheers and applause.) And we’re really pleased that he has saved the last and the best for the University of Washington in his backyard. This is obviously an audience that we don’t have to spend a lot of time introducing Bill to, but it is important to note that here’s an individual who has transformed, of course, information technology and the PC as we know it, and that in and of itself would have been a pretty remarkable accomplishment for anyone, but now he’s in the midst of transforming philanthropy, and global health, and education, and is likely to have that same kind of profound impact on the world yet again. And the man is younger than I am. So he’s just a child. So there’s much, much more to go, and we’ll see what Bill does for his third trick.

But the fact of the matter is, for the UW, there are few families and few individuals that are more important to us than the Gates family, not just Bill, of course, but his mother and his father, and his brothers  brothers?  his sisters. We’ve got with us today Bill Gates, Sr. Bill, will you please wave to the audience. (Applause.) Bill, as you know, is a regent of the university, as was Mary Gates, Mary for 18 years, I think, Bill, and now we have yet another regent that’s a Gates, and that’s Kristi Blake. Kristi, thank you for being with us. (Applause.) And now, without any further ado, it’s my great pleasure to introduce Bill Gates. Bill. (Cheers and applause.)

BILL GATES: Well, good afternoon. It’s fantastic to be here on the finale of my final Microsoft tour, and I mostly want to talk to you about the great things ahead. But first let me talk about the important relationship that I have with this university. As Mark said, it starts with my parents. They were students and met here. My sister was a student here. I was here a lot, but not as a student. I took an algebra course, that was good, but my main benefit from the university was that when computers were very expensive, hard to find, there was no such thing as a personal computer, the university campus was the best place to wander around and find unused computers. And so Paul Allen and I, particularly during a time when we didn’t have any computer work, and nobody was giving us free computer time, we found a number of machines, and were able to develop our skills during high school. The best one was one that was in the Physics Building, and they ran particles analysis runs most of the day, but they always had three or four hours at the end just in case the things ran long, and we would go up there, as soon as it was done, and grab that PDP-10. So we started stealing computer time, and now I’m giving it back from time to time. (Laughter.)

Also, of course, Microsoft has an amazing relationship with the University of Washington, where the commitment to long-term research at Microsoft has a lot in common with the great work going on at the university, particularly the computer science department, but also now that software is being applied in so many domains, and is really the tool for so many of the sciences, a lot of cutting edge software problems are coming out of the other departments, whether it’s genomics, or the environment, or anything related to biology, we need to solve some very, very interesting problems. So it’s fantastic to have the university here, and a lot of people have joint appointments where they’re involved in teaching, and involved in helping Microsoft do its work. So that has been an absolutely fantastic connection.

Also this university has become a very important partner of the Gates Foundation. The Gates Foundation has made more grants by a fair margin to this university than any other, and the interesting thing about that isn’t so much the size of those grants, but rather the kind of ambition behind them, work related to the AIDS vaccine, work related to malaria, work related to tracking healthcare approaches, and understanding where those are working, and where those are not, because, after all, a lot of things that seem like they’d work when we get out into these developing countries don’t, and making sure we guide that very well is important. So it’s a wide range of things, and I’m very optimistic that there will be big breakthroughs that come out of that work.

Microsoft, in a typical year, has about 100 people we hire from the University of Washington, which makes it the top place that we’ve gotten talent all the time. Ed Lazowska and I were talking today and saying, hey, we want that to be even more. So hopefully some of these dreams about software and the great impact it can have will make people realize that not only is this an interesting field, but really this next decade, the companies that take a long-term approach, are optimistic, hiring smart people, are really going to come up with some things that will be game changing.

For myself, I’m fairly near a transition point where I’ll move from being full-time at Microsoft, and doing part-time work on the Foundation goals to flipping that around where after the middle of the year I’ll be full-time on Foundation work, and part-time on Microsoft work. And I’ve been writing software as my full-time occupation for ever since I was 17. I had a couple of years where I was back at some other university, but that didn’t slow me down, I was still mostly writing software the whole time then. So it’s going to be an interesting change, you know, not to wake up in the morning and say, okay, let’s primarily focus on software. And some friends of mine were worried about that and thought we’ll maybe they could help me by putting together some thoughts about what my last day is going to be like. So we pulled together a video to help me with the transition. Let’s go ahead and take a look at that.

(Video segment and applause.)

We had a lot of fun making that. Actually, the transition is going great with incredible people like Ray Ozzie and Craig Mundie stepping up to do a lot of the things I’ve done, and free up my time for the new Foundation work that I’m thrilled that I’ll get to focus on. And we’ll have some Microsoft projects that I keep working on as I talk through some of the big opportunities, several of them are of particular interest to me, things like making search very different than it is today, and this whole area of natural interface.

So let’s talk about software and where it’s going. It has come a long ways since the advent of the personal computer. After all, software prior to that was used only by a few organizations with big expensive machines, and if anything they were viewed kind of as the enemy of the individual, they were printing out checks, and keeping databases. And so it’s almost a complete shift that computing has become the best tool for creativity and sharing that humans have ever had. What’s emerged around software and the Internet now is the facility that makes the world a smaller place, and it’s a fundamental element of the pace of innovation, and the efficiency that we have in the world today.

In fact, if you think about growing up when I grew up, it was a paper-based encyclopedia you went to for any question, and you got very limited data, it was always kind of out of date, and if you compare that today to any student who has access and the wealth of material that’s available to them from encyclopedias to discussion groups to models that will let them navigate through and try different things out, it’s a completely different world.

Now, at the center of that have been breakthroughs in both hardware and software. The incredible benefit of having exponential improvement at the chip level has allowed us to be more and more ambitious in software. The number of transistors has mapped to higher clock speeds, larger memories, and so-called “Moore’s Law” has really allowed us to build phenomenal machines.

Now, when Paul and I were young and started Microsoft we had in mind this idea that computing would be free, and so we let our minds wander to any type of computer that might be possible in the future. And even then we thought about some things that are not yet achieved, things like visual recognition, speech understanding, deep gathering of data, and machine learning that would relate to that. So in a sense we can say we’re only part way through achieving that original dream of software empowerment.

Another way that we can say we’re only part way there is just look at the population of the globe. Personal computing, mobile phones benefit less than a third of all people. There’s about a billion PCs, each of those gets used by multiple people. There’s about 2 billion mobile phones that are typically used only by one person. So we’re there in about a third, and two-thirds are not getting any direct benefit. But, the original slogan of Microsoft, going back to its earliest days, was a computer on every desk, and in every home, using powerful software.

So there’s a long way to go in terms of making it lower cost, and more relevant and far more practical. In fact, today we have laboratories, including some work here at our headquarters, and a particular group in India, that focus on really thinking about the poorest 2 billion, how can computing and technology make a difference for them. Often the answers are fairly surprising. The direct application isn’t as important as thinking through what it means for their health activities, or agricultural activities, or learning activities. And we need a lot of new ideas there to push these things forward. The impact of software has gotten broader and broader, but many of our common activities are not yet software-driven.

TV is still a broadcast medium. There’s not much software intermediation. There’s not software that’s picking what segment of the news show might be interesting to you, or letting you indicate, give me more about that story, or skip over this sport that I don’t particularly care about. The ads are not being targeted to things that might be of interest and value to you, which is a win-win, in terms of your interest and value to the advertiser. The game show doesn’t let you interact, the educational show doesn’t let you pause and get information about something that was covered too quickly. The sports show doesn’t let you get more information.

So we’re on the verge, in the same way that listening to music, or organizing photos has changed for a lot of people, the way TV is delivered is in the process of changing, but just at the beginning. There’s a million households in the United States today, primarily customers of AT&T, who are getting TV over the Internet. And as you get that scale, which has to get to critical mass, then the content creators are saying to themselves, okay, let’s do the extra work to take that show experience and make it far better, and take it genre by genre, and come up with new ways that the personalization and interactivity can work.

Today the world of video is bifurcated. The tail videos, the ones that are very popular, say, videos about your young kid’s sports activities, you’re not going to find those in the broadcast media. And so you go to your PC screen, navigate on the Internet to find those, and then that’s not part o the TV Guide that’s part of that living room 10-foot experience. But, we will have a complete synthesis of those things, where what you’re interested in will show up on all your devices, whether it’s your cell phone, your PC screen, or in the living room. So the tail video, the course that you wanted to watch and see the lectures that you happened to miss, those things will show up right there along with those mainstream things.

Today we don’t have robotics, it’s kind of a speculative field. Part of what’s missing there is a software revolution, so that the software  so that the robot can gather information about its environment, whether it’s visual or audio, that planning type module, standard so all these different sensors can connect up, and we can experiment with robotics hardware without having to change the software base every time. So that’s an area where Microsoft and lots of startups and universities are investing. And I think that people are really underestimating over the course of a period like a decade how much things can change.

With technology we’ve always got that people tend to overestimate what can change in a year or two, and they underestimate the cumulative effect of change that can take place in a 10 or 15-year period. We’re also subject to cycles of over-optimism and pessimism. Certainly the late ’90s were kind of an insane period, where every startup was going to replace your bank, and your retail store, and people forgot that there are some benefits to experiences working those other ways, and the economic proposition that’s brought there. And in any medium where the barrier to entry is very low, the ability to build up an asset is all the more difficult. So only a few of those companies managed to get to the critical mass and do something interesting.

That was a fantastic thing, there was some crazy investment. It was like the gold rush. Some people did lose money, but that’s what capitalism is good at, taking lots of wild ideas and continuing to back the ones that work. So it was a period of, in the final analysis, quite a bit of innovation.

Then when that bubble burst some people went to the other extreme thinking that these changes were not really valid, that it had all been over-hyped. But, it was only over-hyped in the sense of the time frame. Some of the things were not thought through. Some of the technical foundations were not there yet.

Take something that I’ve been a big believer in, and gotten Microsoft to invest in, and I still totally believe in, things like the Tablet computer, where you can take notes and do your reading off of the screen, that takes a certain level of hardware and software, usability, price, size, battery life, and what that magic threshold is that makes that a mainstream thing, where business people going to meetings take it with them, students going to classes, say, of course, I take notes this way, share things this way, we’re not quite there yet. We’re there in certain verticals, the medical market with doctors, insurance market with claim writers. So that’s getting us down that learning curve, making the software and hardware better all the time.

Now, when that comes to fruition, when it really is at critical mass, it starts to have a big effect on reading, where you’ll be reading things more online. We’re seeing Amazon with the Kindle, Sony with the Ebook, Microsoft with software that runs on the portable machine, doing early stages of that. And even though in the next two years you can’t say it will change, but certainly in this 5 to 10-year period that will be very, very different. And we can think about students working that way.

My daughter goes to a school that all the kids use tablet computers, and it’s fascinating to see how the curriculum has changed, but when you really think, not just about putting the normal curriculum there, but when you think about taking advantage of that tool, how fantastic it is. It also facilitates sharing information. The teacher can mail out the test results to the parents, so you go home at night, and whether your daughter wants you to know or not, you know exactly how she did on those fractions. So you volunteer to help no matter what.

So there are some big, big changes that are about to come about. I think one of the most important ones, and this is an area of a lot of great work going on at Microsoft, and at the University of Washington, is the way we interface with these devices. If you look at this last 30 years or so there have been very few changes. We’ve gone from a keyboard, to a keyboard and a mouse fundamentally, 90 percent of the interaction with the machine is driven that way.

That’s not going to go away, for creating the documents, sitting in a solo basis, and reading things, navigating things, there’s a certain utility and efficiency of that that will probably always be justified. But, for the first time we’re starting to see, whether it’s the Nintendo Wii with the 3D controller, or the iPhone with its touch capability, or the Microsoft Surface, where it sees what’s going on, objects, and any sort of interaction that you want to have take place, or vision-type capabilities in a broader sense, that this idea of the computer knowing what’s going on, knowing more than just how you move the mouse, or hit the keyboard, that that really is bringing computing into new experiences.

It was just a few weeks ago that we rolled out Surface into retail stores. They were actually AT&T phones stores where people could come in, put their current phone down, put a phone they were considering buying, and see a comparison, what things were different, what things were better. They could look at the different plans, and try those things out. They could look on the phone and see the different places they go to, it would show them the different coverage capabilities that they had for those things. And so you’re starting to see what that will be like.

We actually think it’s time to amend our slogan of a computer on every desk, because with this kind of technology we want to put a computer in every desk. We want the desktop, or tabletop, we want the whiteboard to be something that’s completely intelligent. As the price of the hardware comes down, even something like a mirror will change to be essentially a screen with a camera. Sure, some of the time you want to see what you really look like, but some of the time you’d like to see what you’d look like wearing something different, or have it maybe point out something that you missed, or show you extra information. So you can create a very pervasive sense of computing.

Part of this will be having screens anywhere. Even the mobile phone itself will have this ability to project onto a large surface area. So if you want to read lots of information, yes, the mobile phone can connect through Bluetooth, or some other means, to another computing device, but it can also simply have projection capability, these laser displays, and some of these different ways the screen hardware works, it not just going to bring us high resolution, it’s going to bring us screens that are on all the walls, the ceilings, different places. So when a kid thinks about their bedroom it will be customized the way they want it, until the parent walks in, when immediately it will be customized a different way, so that everybody stays happy.

Now, part of the things that drives computer science forward is doing risky things, having a long-term time horizon. And when Microsoft was a small company we benefited form the fact that AT&T and Xerox, and other companies had made investments in research, that they’d reached out and worked with universities on a broad basis. So we felt very lucky that as we got to a certain level of success we could start our own research group, and have that research group be in partnership with top universities all over the world, doing advanced programs, things that range from the work on speech recognition, all the way up to quantum computing. We have a fairly serious program that has a high risk of failure, which makes it fun and interesting, on quantum computing. Even in the best case it’s probably a decade away before that makes a difference, but that’s the kind of thing that when it worked can have incredibly high pay-off.

It’s surprising to me how low the levels of real research investment by businesses are. We’re certainly always trying to get the message out that more businesses should do these things. It’s unfortunate that AT&T did it almost because of their regulated nature, and a lot of it wasn’t directly related to the business. So they’re not, in a certain sense, a role model, it’s kind of a unique thing that happened, and those unique circumstances are gone.

Likewise, Xerox, anyone who has been in the computer industry for a while thinks of that as an interesting example, where they for that period of time had a greater concentration of IQ in one place than anyone did, and came up with ethernet, laser printing, graphics user interface, the early stages of these things that were later refined. But, the basic ideas came together there. Yet, Xerox managed to not get any benefit out of it. In fact, they lost a lot of money.

There’s a lot of people who have tried to analyze that. There’s a book called Fumbling the Future, and Microsoft soon after PARC  the research center that did that work kind of fell apart, we had more Xerox PARC people working for us than Xerox did. So we owe them a particular debt of gratitude, not only in terms of their work, but amazing people from Butler Lampson, Chuck Thacker, anyway, just an incredible group who not only do amazing work, but also have nurtured the next generation of brilliant people coming in, and taking on very ambitious challenges.

I mentioned that there’s a lot of good collaboration between Microsoft and UW, the Center for Collaboration Technologies is a place we’re very excited about the work, and we’ve been backing that in a number of ways. Analyzing photos, that we call Photosynth, the University of Washington did some key work on that. I think people will be amazed that you will be able to take just simple 2D photos from, say, a cell phone or a camera, and software will be able to figure out where that is, and connect it up with other photos. And so you’ll be able to synthetically, essentially get a sense of all the things going on in the world in different places, and you can even create full 3-D models out of these fairly low-resolution 2-D photos.

I remember five years ago people thought that was absolutely impossible, and yet there’s been breakthroughs in terms of these algorithms involved, to not only make it possible, but to make it practical. The breakthroughs in efficiency, along with brute force computing, are combining in a pretty magic way.

We’re getting immense amounts of data, and one of the projects I think is very cool that Professor Lazowska and someone in his lab, Keith Grochow, are working on is this Trident, which is a lot of marine biology undersea information. We really don’t know what we’re going to learn from that, but it is state of the art, in terms of what sort of visualization, data collection, pattern analysis that needs to get done. And so it’s fun for software people to have things like that, where it’s interesting, practical data, and yet we have to make advances, interaction techniques, and how we look at data in a rich way.

I have one good example of that, that I want to show you. And this is kind of a new thing. I’ve never demoed it before. So we’ll see how this goes. Let’s see. So this  this is the Worldwide Telescope, and this is Curtis Wong at Microsoft Research, along with some people working with him came up with this idea of taking astronomy data and making it widely available.

It actually started with Jim Gray who was in our research lab, who thought that all those databases were not pulled together. And he worked with some astronomers who needed data from different databases to try out different ideas, to test their model of the universe. And so the idea of having a schema, and a taxonomy, and taking things done at different times, different wavelengths, and pulling that together was done for the scientist to advance their work, and what’s been done here is to actually take that data and say, okay, how do we take that and connect it up in a way that can make astronomy interesting for anybody who wants to learn about it, to make it absolutely approachable in a very rich way.

So you start out, you’ve got a vision of the sky here, and this is a pretty neat application. That’s the sky. First let’s go to something that’s familiar for you, just go in and we’ll find Jupiter out in the sky. So there’s Jupiter. We can zoom in, zoom out, and that. You actually see these dots here, those are the moons of Jupiter that when this image was taken, that’s exactly where they were, and what it’s doing is, it’s showing Jupiter as it is in the sky right now. So we can go anywhere in the sky that we want. Let’s pick Cygnus, which will let us illustrate some very interesting things. So this is data that’s been collected from many, many, many different telescopes, the Hubble, the Sky Survey, literally hundreds of different things have been used to pull this together. So here we are at Cygnus. So I can go in and look at that. Okay, so let’s look at this in different ways. To start out with, we’re in the visible wavelength. Now we’re seeing hydrogen alpha, we can see dust map. One that’s particularly interesting is to look at this in the X-ray, because if you look in X-rays you can see where there were supernovas. There’s always an echo left over after that supernova. So we see this big thing down here. We can go in and zoom in on that a little bit, and say, okay, that’s interesting. Well, what does that look like if we just look at it in the visual wavelength. We can see, okay, there’s what the dust cloud looks like from a supernova that this is actually five thousand years ago, but if I go back here, I can say, okay, it shows up in the X-ray, but not quite the same way in that different wavelength. So we can try out different things to see what patterns different types of phenomena are going to create there.

Let’s go do something else. Let’s see. Let’s go over to Crab Nebula. You can see it actually navigates the sky, kind of gives you a sense of where you’re going. You can go straight there, if you want, but I always think it’s interesting to see what the relative positions of these things are. So that’s the Crab Nebula in the visual range. And so, here again it is in the X-ray, there’s in the IR, and these things are registered on top of each other. You see we’re not moving at all because these databases have been correlated in a way that it’s all in that same location.

So far what I’ve shown you is how I’m guiding things around myself. Another thing you can do is go out and look into databases about this information. So I can just right click something, and it will show me all the different things on the Web that might be interesting about that information, anything that I want to point to here. We also have what are called guided tours where instead of going around yourself, somebody can record the things that they’ve done, and they can have voiceovers. We make it very easy for them to build those things. This is one that Alyssa Goodman did. She’s at Harvard.

(Video segment.)

I can pause at any time, and look at the images in different ways, and go back into the tour, or, again, I can go right click and see this is a spiral galaxy, and again go out and get any data that’s out there on the Web. To give you a sense of how easy it is to make these guided tours, let’s go in and see one that was created literally by a six-year-old about the Ring Nebula, where he’s able to share his fascinating to using this Worldwide Telescope, and share that with other students.

(Video segment.)

Anyway, it gives you a quick glance at this Worldwide Telescope. I think it illustrates a lot of things. We’ve got a lot of data out there, and with the right software we can make it approachable and understandable. And, in fact, in the next month or so, this will go up for free at and just be a piece of software that any teacher or student can use, and we’ll be collecting the really good tours that people create, and have those just show up in the catalogue so that over time people get better at making that a subject that’s fun and approachable using the magic of software and the rich data that’s available.

Now making sure that this innovation is going ahead at full speed, and we want to enable people of all ages, we want students to have access, we want them to have the latest and greatest tools, many of the breakthroughs have been made by people that are very young. Microsoft has a lot of things, one is called DreamSpark, which is where we made all our development tools free to students now where they can try things out. And we’ve also created a lot of online essentially TV channels, but of course very interactive, one we call Channel 8 that has all the information, programming courses, advice, sample code, and really putting a big investment in that to make sure that anyone who can get to the Internet can polish up their skills, see what good code looks like, and really pursue their ideas, whether it’s about software itself or rather they have an idea about another field, one of the sciences, or any idea, and actually you’re using software merely as a tool to understand those things, and encode the algorithm that they have.

Now when we look at this very optimistic view, which I think is a totally valid view, one lens that it is important that we continue to apply to it is to say, okay, with all these advances, the breakthroughs in software, and the tools that make TV more fun, and mobile phones more fun, so you’re talking to them, and you can share with your friends where you are, how do we take these breakthroughs of all the sciences, particularly including medicine, and make them available to everyone?

After all, there’s sort of a natural thing that can prevent that from happening if we’re not careful, which is the interest and needs of the richest have the most powerful voice in the marketplace. And so you’ll see disparities that take place where medical conditions like malaria or an AIDS vaccine receive substantially less focus than something like baldness or erectile dysfunction. In fact, you have a pretty significant ratio in those things, and it’s not that the company’s involved are doing something wrong, it’s that there isn’t a mechanism that outside sends that message that those things are worth working on.

The same thing is true specifically in computing. How should we get computers out to people on a very broad basis? There was a joint project between the Foundation and Microsoft to get computers into libraries. That was done in the United States about 60,000 computers were put into 18,000 libraries. And it was a huge success. It was particularly transformative in rural areas where people didn’t have broadband access, and they weren’t as likely to be able to afford the computer on their own. First the librarians were a little worried, would kids still come and get books, would the machines work well, and there was a lot of learning that took place. But in the end, it was phenomenally successful. In fact, the number of books being checked out went out as more traffic was coming to the library. And kids didn’t just play games, they did things that were interesting and valuable. And a lot of adults came in to learn new skills, to stay in touch with relatives, to understand about health conditions, and it’s been phenomenal to track that. That’s now been taken to a number of countries, Mexico, Chile, Botswana, Lithuania, there’s another six countries that we’re in the process of doing the pilots, and doing the rollouts to take this idea that in the same sense that libraries were deemed to be something everybody should get to, because literacy should be available to all, now this access to computing, and software, and the Internet, and the things it makes available, we should push on for that same type of accessibility.

In fact, we looked across many areas of activity, banking, the ability to have a savings account. The poor need that in particular. The ability to get loans. And it is interesting to think how we could draw in some of the large successful companies into helping out with these things. It’s an area that I’ve used the phrase creative capitalism to describe that and, in fact, if you just took the best practices of the leading companies, and had the average company doing as well, I think that this gap between how quickly and how well we map technologies into the needs of the most needy, of the poorest, would change very, very dramatically. And so that’s certainly a topic that I’ll be putting a lot of time into, and sitting down with companies, talking to them, creating awards, creating positive feedback for the ones that do it, maybe somebody will figure out how to get negative feedback for the ones who don’t do it, and creating hopefully some real momentum behind that.

So, you know, I’m very optimistic about software. I can’t imagine why software is not the most overcrowded field in the world. What could be more interesting than working on these tough problems, and being able to have this kind of impact to build magic new devices? And, in fact, what I and my generation got to do these last 30 years really pales in comparison to what you’ll be able to do in the next 30 years ahead.

Thank you. (Applause.)