Remarks by Bill Gates, Chairman and Chief Software Architect, Microsoft Corporation
Detroit Economic Club Luncheon
Cobo Conference/Expo Center
Detroit, Michigan
August 28, 2003

BILL GATES: Well, good afternoon. It's a great pleasure to be here. I was last here about six years ago, and a lot of interesting things have happened in the last six years. Six years ago was really the beginning of an incredible wave of activity around the Internet and what computing could mean. And some people would say that that was kind of an excessive period, but what you'll hear from me today is we built the foundation for some exciting things going forward. And even though there were some wild things that came out of that, what it means for the way that work will get done, the way things will happen, is very, very exciting.

I've been very lucky to be there to start Microsoft when the microprocessor was first invented, and had some ideas around the software and what the role of software could be. The original vision of the company, which is now 30 years ago, was about a tool to empower individuals, something that you'd use on your desktop at work and at home. And I would say we've come about halfway towards realizing the vision of what software can do. It's not yet easy, enough. It's not yet reliable enough. And I'll talk about some advances we're making there. But people can glimpse that this is going to make a big difference, a big difference in how people communicate worldwide, how business is done, and we're full speed ahead in driving towards that.

Microsoft is lucky enough that with its success we're able to increase our investments in R&D every year, to build more and more partnerships and pursue the original vision that software should not be expensive, should not be complicated, but it should tackle more and more tasks and really become the best tool that mankind has ever had.

We now can focus on very tough problems, tough problems like speech recognition for the computer - that we're making progress on - and make sure those things get out into the marketplace. We work on a worldwide basis. We have a lot of wonderful partnerships. It's exciting to be here in Michigan because there are a lot of neat things going on here. It's great to see this downtown area, and how in the last six years some great investments have made this quite different. I think a lot of people here should be very proud of that.

Certainly, there are a lot of great people that have come out of this area, Steve Ballmer who is now CEO of Microsoft grew up in the area. We have a close collaboration with institutions like University of Michigan and Michigan State, driving forward computer research to make sure that the students there are contributing, and that they have great opportunities coming out of that.

There's a lot of companies that we work with in this area, particularly the car companies. I'll talk about some of how we're working together on software to drive the automotive business forward.

And nonprofit organizations. In the introduction it was mentioned, but I've also gotten a chance, in addition to my work at Microsoft, to take some of the wealth that that's generated and try and make sure that goes back to society in the most impactful way. And as part of that, I work very closely with United Way. In fact, tonight I'll be lucky enough to be part of a dinner which is a congratulation dinner to the people who are doing leadership giving to the United Way here, and also to encourage other people to step up to that level.

We've also worked with Focus Hope, both on software and making sure they have the resources they need. There's a group that has an office here called NPower that helps nonprofits use software in neat new ways, and we make sure they get the latest technology. And so, it's not just the world of business that sees the magic of software, but also the world of social services and philanthropy can get some of that communications and effectiveness that we're so excited about.

In the world of automotive, there's several ways that software gets used. One is the classic way that all the information those companies are dealing with, in terms of what's selling, what does the manufacturing supply chain look like, what new models ought to be designed, and what might happen there. We provide our Windows and Office software for that kind of activity.

We also have a special thing we're doing with car manufacturers, because we believe that this idea of having access to information even when you're in the car -- that's something that low cost, amazing software can help with. You know, the idea that you get traffic that's relevant to you. You get directions. You can stay in touch with new things that are important to you. Whether it's a voice recognition interaction or whether it's touching a screen, the car is one of the places where, over time, software is going to be very helpful. So, we've got a dedicated group of hundreds of people run by our vice president, Dick Brass, that is particularly focused on that vision, and a lot of excitement we have there, and a lot of investment taking place around that.

Now, let's take a little bit of an historical view of these last six years. Starting in '97 and '98, that's when the industry really started to preach the vision of the Internet, that everybody would be connected up. And, a lot of startups got started around that. A lot of views that things would change overnight, and although some of those views were unrealistic, most of what the vision was all about was really quite valid: the idea that banking could be better, that finding books that you were interested in could be better. The idea that even if you buy things the traditional way, say, before you buy a car, you would go to a Web site like CarPoint to learn about the latest models and how they compare, go to the manufacturer's Web site. That you would stay in touch with people like scheduling appointments with doctors, or being able to ask questions through the rich electronic communication.

And so, a vision of some things, that were very powerful. You know, getting rid of paperwork, making it easy to stay in touch with other people. And so a lot happened, building out these systems in those several years, and there was a spirit of optimism that drove things forward.

Now, starting a few years ago, say early 2001, people sort of came to look at it as a glass half-full. After all, the business models of the startups, 95 percent of them really didn't make sense. Some incredible companies are among the 5 percent that did: Amazon, eBay, a number of companies that will be here for a long time making very important contributions, and continue to grow because they got it right. People saw that systems weren't connecting together as well as they expected, a lot of software was hard to learn, hard to customize, for the particular needs that people had. The security crises, which has been a big deal recently, where people get e-mail that if they open it they can do damage to systems. Those are problems that came along with the pervasive use of e-mail.

And so now, we need to make it so that all those drawbacks aren't there, to make it so that it's a glass half full. And fortunately, by driving the software forward, we can do that. Making sure the systems don't crash. Making sure that you don't have as much spam in you inbox, mail that's really unwanted. That comes because it costs about a 10,000th of a cent to send a piece of e-mail. And so if 1 out of 1,000 people click on one of those things, and they get paid a few dollars for that, then it's an economic business. What we'll do with the software is we'll make it so you don't get those in most cases, and we'll drive the cost up for the people sending unwanted e-mail by about a factor of 1,000. And so we'll actually reach an equilibrium that's even better than paper-based mail is today, where you are back in control, you'll decide exactly the information you want to receive.

So, despite this view that came with the tough economic times and some of the over-promises that came out of that Internet period, I still say that this is the decade, between now and 2009, that digital approaches really will become mainstream.

We're continuing to have amazing hardware advances. There's nothing like the speed of improvement that we have in chips. These chips get twice as powerful every year and a half, and that's referred to as Moore's Law, it's exponential improvement. You know, there's the old joke about how much mileage a car would get if it was based on this kind of improvement. There's also the punch line of that joke which is how cars would crash several times a day.

So what we want to do is we want to give you that mileage and take away the crashing element that exists there. And actually, by monitoring these systems and getting feedback from users, we're getting now a perfect view of exactly when the pieces don't come together. And some of you may have seen this thing where it says, 'Do you want to send this error report?' Well, we get lots of those, and those are driving our priorities, and that's why I can say even within a few years these devices will be dramatically more reliable, and that will make people more open-minded to do more ambitious things. To do more family scheduling, and more collaboration on the devices.

The hardware improvements are not slowing down. You might find that surprising with the economy being tough, with so many startups going out of business, but the fact is the people who build the disks and make them twice as large, so you can now store videos and photos and music. People like Intel who build the processors, they're not slowing down in that rapid improvement. And so, on your desktop, typically you've had a 15-inch CRT. In a few years, at a price actually lower than that, you'll be able to have a flat screen, an LCD screen, that has about three times the area, a 22-inch diagonal. And so, a huge change in terms of the information you can look at there.

You'll have wireless networks, a thing called Wi-Fi that you'll have in homes and businesses, where you can take portable machines around and always be connected up to your information. You'll have a device in your pocket, the phone-PDA type device will be a lot more powerful, and it will have GPS location, camera capability and we'll make the software so it's a lot easier to use.

We need to put the user at the center, so you're not moving the information between these devices. Rather that automatically happens as you move from the car, to the phone-pocket device, to the desktop device, to the PC at home, to the TV -- all of that comes together. So the hardware pieces are coming; our role is to make sure the software breakthroughs happen, that you can do electronic commerce without paperwork. That never happened, as yet: the benefit to the efficiency of the economy of getting that right is very, very dramatic. Today, finding the best seller with the best price, staying in touch -- are they going to be able to deliver, what changes are taking place -- it is way too complex, and it means that people overpay, they buy from the wrong source. Software magic with the Internet we've got, using an approach called Web services will solve that problem -- not overnight, but over the course of this decade, that will get solved.

A good example of that is people get a phone bill today, or a company does. That's all this paper that they have to retype in, and they don't really know if something is unusual and out of line on that. That won't come as paper. That will come as just an electronic piece of information that will automatically be analyzed. The way that you look at sales and profitability, look at quality control, there will be a rich understanding of that, because you can dive down into the information and see it even at that detailed level.

Now, making sure that computers are super secure is a tough problem, we call it Trustworthy Computing. Today, when you use a computer, you enter a password. Well, that's a fairly weak system, because you often pick passwords that other people can guess, or you use the same password on different systems. That's a weak link, it's only one of the weak links, but it's one of the things that will have to change. So in the future, if a system has information that's important to you -- your electronic mail, health records -- you'll either use something like a finger print or a smart card to verify that you really are you, and that you should have access to that information.

Likewise, we need to build into our software the ability to stay up-to-date all the time, so that we don't have viruses coming along. We need to build the spam elimination into the mail software itself. And so that's our responsibility, we can't just say, 'It's all great, just put up with that.' We have to get rid of those things. And so, it's a whole new mind-set for us to think about building these things in, and advancing the technology so that this infrastructure is extremely reliable. I used to say as reliable as the electric power system. Now I guess I need to say even more reliable than the electric power system. So we're not the only ones that have work to do on our critical infrastructure. But, we definitely have the tools to solve that problem.

So, if you look at what kind of devices will there be, there will be multiple form factors -- different sizes of devices just because of the natural ways that you want to use things. When you're sitting anywhere, you should be able to glance at your watch and see not only the time, but also a sports score that you care about, or a short message, or check and see, 'OK, what appointments do I have today?' All of that should show up on your wrist watch, it should just be there.

The device in your pocket, you shouldn't have to have a bunch of those, one that is good for maps, and one that is good for photos, one that's good as a phone, one that's good for note-taking. We should have a single device that can be fairly inexpensive, and connect up to both the wide area network, the phone type network that's everywhere, and whenever you're within the range of one of these Wi-Fi connections, often called the hot-spot type connection. If you're at a Starbucks or a hotel or a convention center, they will have that capability. Or if you're visiting another company, you can gateway through their network out to the Internet, you should have the higher speed connection without paying any per minute charge, whether you're data browsing, or voice browsing. That should be something that that pocket device does for you.

Another form factor, of course, is in the car, where voice recognition comes in to avoid driver distraction. The TV set. Right now the TV Guide is becoming a form of information overload, where finding the shows you care about on the different channels is too difficult. With things like what we call Media Center, you'll be able to say what shows you're interested in, have those recorded and advanced to you, and not have to have multiple remote controls: just a single remote control for your music, your photos all of those different things. There's no doubt that things like film will go away, things like having to have your music on an actual physical CD, that's very inconvenient in terms of organizing it the way you want, taking it where you want to go. By the end of the decade, that will be gone.

My daughter and I were walking down the street and I said, 'Let's go to the record store,' and she said, 'What's a record?' She's never seen a record, she's seen a CD, so I said, 'Well, it's confusing, but a record store is where they sell CDs.' But in the future, they won't. It's a changing environment, but it's because we're making those things a lot better.

Another form factor that we believe in a lot is called the Tablet-sized, a thing that you take to meetings, and you just use a pen to take your notes. And so, if you see an article that you're interested in, you put a little note on it, and say what friend you want to send it off to. If you see a budget you want to dive down and see more numbers for, you just point and get that information, and then share that with another person. This is far better than printing out slides, or just having the information that's there in the meeting, because as the meeting goes in different directions, you can get all the information that's important. The desktop computer I mentioned will have the bigger screen, and then we'll have wall-sized devices. All these things connected up to one network, so software will be working across all the different devices.

There are a lot of scenarios that will be very different. When you call somebody on the phone, it will be very typical that you're not just talking to them by voice, but your screen and their screen will automatically be connected up, and you can take a budget or a travel plan and sit and edit it, and talk about it while you're on the phone. That will be common sense that your screen should be connected. Scheduling meetings. Most people don't use electronic scheduling, but we've made it rich enough in this latest version of Office, we think that will become very mainstream.

Reading magazines. We've got that Tablet screen that's very high resolution, and you can hold in your hands. More and more things like reading books, reading magazines, will be done off of that screen. So, a big change that is coming, that younger people are probably driving towards faster than older people, but it's going to happen in a very, very pervasive way. And it does affect lots of businesses, and how they reach out to their customers, and it does affect how you can collaborate across corporate boundaries and do better partnerships.

We think a lot about the worker and business, and how this affects them, what kind of cycle of information they go through. That is so inefficient today, and yet people aren't bothered by it, they call somebody and get a busy signal -- well, what do they do? They try and call them back later. They try and set up a conference call, but it's too difficult. They have a meeting that somebody at a distance they want to involve, but the software doesn't make that easy. Somebody wasn't there for the meeting, they want to have a digital recording of it so that person can go back and see a piece of the meeting. Well, you can't do that right now. So many things that should be common sense are not. And yet, people aren't pushing for it, because they haven't seen it.

And so, with pilot tests, with pioneering customers, we want to get these things out there so that the way you think of meetings and scheduling and collaboration -- the way you think of diving in and seeing business information -- is far, far better. And what that does is it just unleashes creativity. It takes the job categories that have been fairly mundane and makes them different. We don't have secretaries -- most companies don't -- whose primary job is to type in information. Now they're administrative assistants who are doing scheduling and organization things and looking over budgets, adding value in a much more interesting and more important way than just purely transcribing the information. And that up-leveling of jobs -- whether it's a customer service job, where you no longer need somebody to tell you what your bank account is -- you see that. You only need them to help you with advice about what you should with financial instruments. So the productivity and opportunity across all the different activities of a worker is very, very substantial, and will be realized in this decade.

The productivity improvements this decade will be far better than the productivity improvements of the '90s, and yet they won't be as hyped. In some ways that's a good thing, because people see the benefit as it comes, rather than having a heightened expectation that happened quite a bit during that last decade. So in parallel, we fix the things that make it hard, and we drive these productivity advances.

Now, that optimism that we have we're backing up by increasing the amount we spend on R&D. We're just over US$6 billion now, and we're really kind of out of step with the rest of the industry, where everyone else is cutting back. But we're pushing forward. That's because we believe that the natural interface that we can create, where the computer can recognize your handwriting, or it can recognize your speech. Those things are just around the corner, and we are now building those into these products, and so it will become common sense that things can be done this way.

A good example of this: we try to make ourselves a showcase -- we call it eating our own dog food -- sometimes it's a good thing, sometimes it's not. We're learned a lot just by always putting ourselves on the cutting edge. This is an example of when you come in in the morning you see an employee Web site that's customized to their particular interests, and it has corporate things. It's just a simple portal, and it's easy to search things, see all the different new things going on. If you pick a project that you're interested in, then you'll navigate in and see something like this, where the project information is presented in this very rich way, and you can see what's changed since the last time you looked.

Very often, you'll look at one of these and send off mail, and it will stick this view, this very rich view, into the e-mail, so that person doesn't have to do anything. They can just take that view and dive in and see more detail, understand what it is you were surprised about or you think should change. So, what you're seeing on the Web site, the way that the e-mail works and the way you collaborate are very closely tied together. A very simple scenario, very straightforward, once people see it they have a sense of how they would take advantage of that. So, software is kind of a magic ingredient. Not nearly complete in terms of what's possible here.

If you look at IT budgets, the spending on developing software, supporting these systems, all of these things can be substantially reduced. Certainly, the hardware costs will be going down because of the magic of Moore's Law, and so overall, even in a time of flat IT budgets, there's room in there, because of these efficiencies, to let people do new things.

Coming back to the automotive industry as an example of this, all of these pieces can really come together. How do the car companies talk to their dealers? How do the dealers monitor events that have taken place inside the car? How do you then take that information and use it as feedback in terms of design ideas, or approaches you should have in terms of pricing your vehicles, and driving your supply chain? All of those things really come together in this digital realm.

And so, we're able to do some very pioneering things with our customers here in Detroit, where we're getting out on the cutting edge and saying, let's supply information technology to help revolutionize your business, and make sure that you get to put more of your resources on things that really count for the customer, not just the mundane things. That's true of many, many industries: the finance industry, the retail industry. A lot of these industries will be in a period of dramatic change because of the digital approaches that are being used behind the scenes, and actually in the product itself.

So, I would say it's an exciting time to be involved in software. You know, lots of people are telling us the glass is half empty and pointing out the new thing that we need to do next. That's very important for us to get that feedback, and to make sure we're prioritizing that with the R&D things that we're driving forward. And it's very surprising to me to be able to say that the same kind of excitement I had when I started Microsoft at age 19, I have that today. Many people ask me why I don't retire and just focus on philanthropy, which I love doing a great deal, and the answer is that I want to take the half of the dream that we've realized today and make it in the rest of my career, which I have a little more than 10 years; make that full dream come true, and make all of these scenarios a reality for everyone here, and everyone around the world.

So, it's going to be very exciting, and I think that it will make a huge difference.

Thank you.

(Applause.)