Remarks by Steve Ballmer and Bill Gates
Microsoft CEO Summit
Redmond, Wash., May 24, 2000

MR. BALLMER: Hello, I'm Steve Ballmer, CEO of Microsoft, and it's my pleasure to welcome you all here today for the summit. I wanted to make just a couple of opening remarks before we turn things over to Bill Gates, and then Bob Herbold from Microsoft, our Chief Operating Officer, who will be the emcee for this program.

This is the fourth year we've done this CEO Summit Event. I will say it's the first year we've tried to host people "start to finish," as they say, right here on our campus. This room doesn't normally look like this. We don't have conference rooms that look quite this cushy, but I hope it is comfortable for the attendees today. We have a very diverse audience. We have folks from over 28 countries, 168 companies minus some last minute calculations, and people from over 37 different industry sectors.

The goal here is a little bit of a lot of things: to stimulate some thought and discussion of technology and its appropriate role in businesses quite broadly, to let people have a chance to share views with peers, which I think attendees over the past several years have found to be some of the most valuable time, and we try to get a few thought-provoking speakers and panels together to just help bootstrap -- both in a business sense, and in a sense of how to use technology -- some new thoughts. We called this year's conference, and I'll read it since I don't remember, Technology Plus Business: Entering an Age of Unprecedented Opportunity. And in some senses I find that to be a silly title in a couple of different ways.

Number one, I think that if you take the right view of your business, there is almost never a time when there's not unprecedented opportunity. And I think that while two or three years ago it may have been less obvious to people that technology was a fundamental part of the evolution of the business, that is certainly not the case today. It is evident to everybody, it's a little bit like saying all business needs marketing and all business needs sales today. It has become a very basic backbone of absolutely ever industry, whether it's in the ways you touch your customers over the Internet, the ways in which your people communicate and understand business internally, those have become very, very important.

We're going to have, I hope, a very interactive dialogue, including the sessions here in this room. So, I encourage you not to be shy. Certainly the last three such conferences that we've done that has not been a problem; people have been pretty engaged, pretty lively. We encourage you to really participate.

I do want to thank some of our partners who were involved in organizing this event, in providing equipment, et cetera, for the event. Thanks to Compaq, Hewlett-Packard, to Dell, to Phillips, to Sony, and to our partners at Price Waterhouse Coopers for all of their work helping get organized.

Please enjoy the conference. We've got a great day-and-a-half, and to kick it off let me introduce my good friend Bill Gates, who will start out with a little bit of a technology session.

My pleasure, thanks.

(Applause.)

MR. GATES: Well, my goal in kicking this off is to talk about how technology is not standing still, and how that will really have a profound effect on how we view this as a revolutionary change agent in business and communication. I've actually given it the title Internet Stage Three, because I think a lot of things are coming together right now that are going to make the Internet qualitatively very different than it has been. Quantitatively, it's easy to track the dramatic growth, the number of new users coming on, the number of messages being sent, the number of transactions that are being done across the Internet. But some of those numbers really hide the importance of the phenomena taking place. The fact that a few purchase orders that used to be sent through the mail are now sent over the Internet, as long as the way that contract is administered, negotiated, doesn't change, the fact that the actual order itself, or even the confirmation is done electronically, that's not a huge change to business. But when you have the entire matching of the buyer and seller, people who never would have found each other because of the frictional cost of traditional means of communications, you have them negotiating an arrangement, and maintaining the entire relationship in terms of surprises, late deliveries, changes, advice, all of those things in a digital manner, which allows those people to be essentially anywhere on the globe. That kind of transaction really is of a different character, and it's the type of thing that is actually reasonably limited on the Internet today, but will be very dramatic in the years ahead.

From a business point of view, we can think of the three phases as being characterized by a little bit different kind of mania. Phase one, the mania was, do you have a Web site, you've got to be up there, you've got to have something neat, just get in the game and people in this phase will talk about hits. You know the notion that how many unique users ere there or what kind of reach you had, things like that, those notions really hadn't been developed. So during that phase, 1995 to 1998, most companies of any size got something going on the Internet. But it was typically a read-only site, where you would just go up, get the information, and that was it.

Over the last two or three years, we've been in a phase where there's been a real mania about transactions, about essentially growth revenue over the Internet, where a company will say we have 50 percent of our revenue across the Internet, or our Internet sales grew by this, or even spinning off the portion of the company that has the Internet revenue and trying to get a separate valuation for that piece. And so, gross sales were what people worshiped, even at the expense in many cases of what the long-term profitability model for that might be. I would say that we're just at the beginning of a phase that, in a sense, you can call the rational phase that was due to arrive at some point, although there's still vestiges of the previous phase out there in a pretty big way, and this is a phase where people are saying, "what does it have to do with profit?" And this is where traditional companies can come in and say, "how could we take the things that we've done, do them in a different way?" How does this affect one of our biggest expenses, which are the knowledge workers, and I'll be focusing on that today as one of my themes. I think it's been a constant theme that companies really need to think about their relative excellence in terms of their knowledge workers. Are they giving them the best tools, and really trying to raise people's sights in terms of what they expect out of their knowledge workers with this goal in mind that they can make a huge difference in the things that really count towards profit, great product design, great customer service, efficiency in how things are done, the efficiency with how quickly you respond to changing market conditions. So that's a story that really takes the business angle, and you can even track this a little bit through some of the valuations in the market.

Now, let me take the phase change that we're going through and talk about it from a technology point of view. In some ways, people can say, boy, the Internet, how could it be different, how could it be better. I mean, after all, it lets you connect up to different Web sites, and there's massive amounts of information out there. However, when you're accessing that information, you're basically getting exactly what one company wants to present to you. And so you often find yourself going out finding a Web site, scribbling down some information on a piece of paper, or trying to use the clipboard to move that information around. So, it's one site at a time.

It's information that you read, and if we think about somebody doing planning or forecasting or collaborating, obviously reading is not 100 percent of their job. Their job is to synthesize the information, to take information from multiple places, view it in very rich fashion, add their thoughts, add their observations, and collaborate with other people using that information. And we almost have a dichotomy in phase two where you have the Office productivity tools that have one sort of user interface, the classic Windows applications, Office just being the best example of that, and then the browser. The browser where you're reading, and navigating to these sites, and then the Office tools where you're actually doing your creative type activities. Obviously what we'd like to do is get the best of those working together in a deeper fashion.

So, phase two is reading, phase two is keyboard centric. Today if you want to get to a Web site, you find a keyboard device, and get fairly adept with that. The amount of this that is voice recognition or handwriting, or just automatically done for you because an agent anticipates your interest, that's really not that present in this stage. And so you could say that this phase two is sort of the ultimate library is the way that we're using the Internet.

Phase three, all of those characteristics change. A knowledge worker will generally be looking a set of information from many different sites. So, for example, if they're planning a new product, they'll see inflation forecasts, they'll see demographic forecasts, they'll see cost forecasts, and they'll see that in a view that they create based on the work that they're trying to get done. And that's a huge difference between what you would have to do today where you would literally go out to dozens and dozens of Web sites.

So this information will be customized for you and, depending on the device you're on, the size of the screen, the type of interaction, that information will be presented in the appropriate fashion. Today, if you go between different devices, say you have one of these pocket devices that you carry around, and you've got a PC at work and a PC at home, you're very involved in moving information between those things, manually doing those replications. If you've got a favorites list at home, it's kind of hard to get that to appear on your computer at work. If you want to buy a new computer, transferring the information over is very hard. So it's a very manual activity in terms of dealing with that information. In this next phase three, that information essentially will be magically stored in the Internet itself, and so whatever device you go to, as soon as you prove who you are, whether it's with a password or a smart card, and I'll talk about that because that's undergoing a change as well, the things that you care about, your customized view, your email, your files, your favorites -- those things will appear on the device that you're using, but they'll appear in an appropriate fashion.

So, for example, if you're on a small screen device, you may choose to have only summaries of the messages that are sent to you, or only a certain set of messages that you really care about would come onto that device.

Part of the goal here is to take all the transactions, all the business understanding that today, although some small part of them are doing electronically, and really get those into digital form. So, you could say it's a transition from what is largely an offline economy to a very real-time digital online economy.

Now, as we go forward on this, it's important to recognize places where we've fallen short. You know, the fact that in terms of ease of use, in terms of dealing with system failures, that there's a lot more that can be done with these systems. You know, technology is a very mainstream thing. When you go home you can't pick up a magazine without some story about a great new company, or a company that's having a problem. And you can't go to a cocktail party without somebody saying, gee, I had a problem with my computer. It's good to hear that, it makes us humble, because through the magic of software, those things can be changed fairly dramatically.

So, I would say as well as getting these new capabilities, the ability to have the software always stay up-to-date over the Internet, and the ability for the software to see what's going on wrong and reach out and get the diagnosis, help the user in a much more automatic way, and in the extreme case they really need help, having somebody come up visually on your screen, and being able to take control and help you out, that degree of user satisfaction and simplicity is another thing that will be able to advance pretty dramatically.

Of course, the PC is talked a lot about in the media, and I pulled together a little video clip that just sort of captures the way that all this technology is getting into mainstream culture. So, let's go ahead and take a look at that.

(Video shown.)

MR. GATES: Microsoft's vision for this era is in some ways simply to stick to the exact same thing that we've done from our very beginning, which is to provide software platforms that will let people get at the incredible potential of the hardware and communications advances. Part of the idea here is to move beyond just the individual PC now, to think about solving the problems for the user across all of their devices. So wherever they are, however they're getting access, whether it's a device in the car, the TV set in the living room, or any of the multiple PCs in the variety of form factors that are out there, we want to make sure that their information is available, and that their time is being used in the best way. That the information they care about is what they get, and they're notified about those things exactly the way that is most valuable for them.

Now, that's a very challenging software problem, and one that even with our $4 billion here in R&D will take many years to fully achieve. What we're going to see is that a lot of things that we thought about as software packages, or stand-alone businesses will become Internet services. Things like taking some of these financial management problems and creating a Web site that lets you get access to loans, to stockbrokers, to insurance. There will be a lot of people on the Web who provide that. In the past that would have been software that you installed in your machine locally, and it would have been professionals that you go out and consult with. Now, you'll see that appear as a service on the Internet, where depending on how much you want to pay you'll just get a purely electronic relationship, all the way up to having lots of people involved in personalizing those services.

On the mobile devices you'll see services that go way beyond what we have today. And as I said, it won't just be from one cell phone, it will be across all these things, your calendar, your instant messaging, the voice mail, unified communication so you won't have to think about having multiple email addresses, or think about what email you want to be paged on. It will just be unified so that all the different types of communications, or any sort of information change comes to you when you want it.

Business supply chains that have been very paper oriented will become Internet services that make sure that all those logistics are handled well. And part of this will be because we'll have building blocks for developers that they can go and find on the Internet, get advice on, and just simply get the latest and greatest, and plug it into their software. So the speed with which developers will have access to each other's work, and the latest changes and collaboration also benefit from this new phase three Internet that those services will be built on.

I wanted to spend a few minutes talking about what we're assuming about hardware. Of course, Microsoft itself is not in the hardware business. But, this software vision that we're talking about certainly depends on the kind of miracle work that had been done in many of the hardware sectors continuing to push forward. And many of the things here are guaranteed to happen, the level of investment in broadband and wireless and all the other things I've got listed here are at pretty incredible levels. And just in the next two or three years the improvements will be quite dramatic. When I say broadband, of course, I mean, having a speed of access to the Internet, so that something like video and audio comes to your machine at very high quality, and you don't think twice about updating your software or sharing even a large file or database by doing that across the Internet.

Here, of course, we've got a dichotomy where businesses are overwhelmingly connected up with broadband today, and the prices for that are very competitive, and you're just getting more and more options, whether it's different providers or different wireless technologies that businesses can afford to be in. The big question mark here is how quickly the homes move to broadband, because that revolutionizes the way they think of both entertainment and communications. And I'd say the last year results were a little bit slower than we would have expected.

The rise of DSL, although it's happening, was not as dramatic. The drop in price of DSL was not quite as dramatic. And even cable modems, although they continue to grow in popularity, a lot of the cable systems are still being upgraded, and there have been some capacity problems for some of the areas where the cable modem penetration is very high. I don't think the stumbles of the last year will continue, because the phone companies will recognize that this is absolutely critical to their having a future at all. DSL will get onto the virtuous volume cycle, that the more volume you get the more the price comes down, and the more the price comes down the more volume you get. So that will achieve a certain critical mass.

I think I predicted that again for this year, but I was wrong about that one last year. So in the United States probably, in the best case, even five years from now we'll probably only have 25 to 30 percent of homes connected with broadband. That will be significantly ahead of any other large country, but not an overnight phenomenon. So there's still some dichotomy there between business and home.

The wireless technologies are really going to change how we think about the Internet. The fact that wherever we go, whether it's on a little screen phone device, whether it's on a large portable PC, or a tablet type PC that I'll talk about, the fact that you have that high speed data connectivity will mean that things like looking up restaurants, things like finding the nearest business that can help you with various issues, being on the road with customers and talking through with them, things where you have access back to the Internet through that portable machine, a very, very fundamental change. And of course, we're seeing that in the valuation of the wireless companies, the level of investment that is going on there. This is going to be commonplace. Three years from now everybody will come to this meeting with a device in their pocket that connects up to the Internet, and their PC likewise will have wireless connectivity.

The first place you'll see this being pervasive is actually in the office itself, in the campus or the office building, you'll be able to take your PC into a meeting and if you have the right modem plugged in you'll still very high speed connectivity. We have this everywhere at Microsoft now, and it makes a huge difference in terms of people being able to take their PCs into meetings, look up information, take notes. I'll talk about that, but the character of how you think of collaboration and finding information is very different once you're un-tethered to either the small screen or the large screen.

Smart cards are going to explode in use. The fact is that the passwords that people are using today -- that they tell you to change, that they tell you not to use names for that are easy to get, but people still do that -- passwords are very much the weak link in the security system. There's all these great scares about software security holes, and viruses and those kinds of things, and there's a lot of work going on in that, but a security system is only as strong as the weakest link. And as long as somebody can call up an administrator and say, hey, I forgot my password could you change it, they can pretend to be an employee that they're not; they get access to all of that information.

So actually having a physical smart card and having in the keyboard something where you swipe that as part of the authentication process, or put it into your cell phone, or put it into your TV set top box, that will become commonplace. And there's a lot of work that needs to be done to bootstrap that. Now, some people would say that these bio-recognition algorithms, fingerprinting, voice printing, will actually be a good complement to that, but right now I don't see any of those getting popular fast enough to really substitute for the role of the smart card.

The PC itself is not going to slow down in any way. The graphics capability we're getting there is really quite phenomenal. It is a device that every year changes very, very dramatically. And that's different from a lot of other devices. For example, video games, even though things like Play Station II or our own Xbox are going to be really phenomenal, those products get frozen for a period of three or four years, because of the economic model, and because game developers want that stable platform. Every year the PC keeps getting better. So it always leads the way, whether it's using video cameras for entertainment, or pushing the realistic resolution, or using all sorts of incredible input devices that sort of bring a virtual reality to the PC. The PC on the desktop will need a lot of power to do the kind of rich viewing that I was talking about. You'll need a lot of power to do the kind of natural interface things that I think will be very mainstream.

When you have a document up on your screen, in fact, your computer will go to work very quickly looking at all the information in that document, trying to find related information that might be interesting to you. So if it sees the name of a person, you just click on that person to see all the data you have about them, or to be able to contact that person. If you see the name of a book that's in a document, all you have to do is click on it and it would offer a method of being able to go out and buy that. So, using that power, that local power of the PC to take the information that you're looking at and make all the different things you might want to do easy, that's how we're going to bring a huge benefit to the increasing process or speed that we have there.

Also, the PC, although we don't think about it today having this, it will have a microphone and a camera as a standard thing. Instant messaging, which probably the people in this room don't use much, is an incredible phenomena amongst teenagers. The idea that you can just fire off a neat little message, see which of your friends are online, that kind of real-time communication actually will be valuable to all age groups, and even in business settings.

Part of that is we'll add in voice and video as well as the type in line capability, which characterizes instant messaging today. In fact, the boundary between what's a phone call, what's a video conference, what's just sending the short little message, or what's something very sophisticated where you're editing a document together, and multiple people are collaborating on that, the boundaries between those things will be completely eliminated, and you'll just sit at that PC screen and initiate any type of communications that you want using that microphone. So, essentially, it takes a phone call and allows you to go to richer and richer levels. That camera will be great for the natural interface as far as telling who is at the PC, are there multiple people there that are trying to work together, so it will automatically increase the font size, make it simple to do that work together. We'll think about phones, these small phones, having very, very good LCD screens, the quality of the displays are improving quite dramatically. And so, even on what would have been a classic phone call where you're calling up to get a schedule or a map, that will appear on the screen on the phone. You won't have to try and talk that through.

There's two new form factors that we're very excited about. In fact, we're putting literally hundreds of millions of R&D into the software to drive these new things, and that's the PC tablet, the PC without a keyboard, and the e-book. And I'd like to have Dick Brass, who is our vice president really in charge of these new form factors, to come up and share some of his enthusiasm and thoughts for these breakthroughs.

Dick.

MR. BRASS: Bill. Good morning, it's been about 65 years since Vannevar Bush, in 1945, proposed the first practical modern e-book, a desk-like machine that had reels of microfilm inside, and displays on top, and it would cost about $250,000 to build. They never did, but it probably would have worked, and it would have held about 1,000 books.

Since then, there have been no less than 100 attempts to build e-books. Why? Because they can store lots of information, it can be delivered instantly, it could save a great deal of money, it could be extremely convenient. There's a million good reasons why people want them. But almost all of the projects have failed. The technology was not quite ready. They were too slow; they were too expensive. In 1991, Sony built a beautiful product called the Data Discman, but it was about the size of a box of tissues. And, in retrospect, it wasn't an e-book yet, although it was a great attempt, and it inspired a lot of work that's followed it.

Since then, the technology and the software has improved. We have much better LCD screens, faster processors, more memory, and the devices are getting to the point where we can deliver a slate-like appliance, thin, light, with enough display and enough power to last for hours, eventually for days. I'm told that many of you have been given, or all of you, I hope, have been given one of these devices, which we built with our friends at Hewlett-Packard, it contains a great little processor. Yes, it's a PDA, but because of the power of the device, and Microsoft's ClearType screen technology, which roughly doubles the apparent resolution of the LCD, it makes a pretty good e-book. And there are several titles included, including Michael Crichton's Timeline, which yesterday we announced we were going to provide. It's not yet a book, but it's getting very close.

And in the next few years, we're going to continuously improve the technology until it reaches the point where it's about as good to read as paper, and certainly more convenient and faster. The devices will get larger. We're working on what we call a tablet PC, and that's basically a big e-book that's not only great for reading, but you can also write on it. It runs productivity software. It's the sort of thing that might be a reasonable replacement at meetings for a PC, and one day when the technology is solid, perhaps even an evolution of the laptop itself.

We figure, by 2005 with e-books and tablet PCs, there will be at least a billion dollars in sales in periodicals and e-book content. Now, that estimate was developed last year. Since then, Stephen King, as you know, sold 500,000 copies of his novella in 48 hours after it was released, actually swamping and crashing the servers that were designed to provide it. We think we may be underestimating the potential for the delivery of this kind of information and the interest in it. We think if you provide great reading capability and terrific technology, great authors, reasonable prices, there could be a genuine revolution in the distribution of content far faster and far more profound than you might have expected five or ten years ago.

By 2006, I suspect that on airplanes you'll not be reading electronic magazines, but the e-books that we've developed, and the tablet PCs, but unfortunately they'll still be the same boring magazines that you currently get on the airplanes. I think that by 2008, e-books may over take t-books in certain categories in certain sales. I think most magazines by 2010 will be delivered in that format. By 2020, I think the revolution will probably be complete, and the definition of a book will be changed from paper bound by leather to a substantial piece of writing delivered on a computer display.

Now, there are a lot of things that have to take place for this revolution to occur, the technology has to improve, there has to be great partnerships between publishers, technology companies. We need to have standards, but we also need to have copy protection and reasonable rules to enforce it. Michael Eisner at Disney has done some terrific evangelizing, pointing out that if an author creates a work, you have a choice to buy it or not, but stealing it is not a viable third choice. And it will be very difficult to create this market if the public attitude is overwhelmingly toward stealing content. We have to make the Web a safe place not only to surf, but a great place to buy information and to distribute content and entertainment.

If we do our jobs right, if we create really great software, if we create devices that are, in many ways, better than paper, I suspect that in 2013, you'll see a commercial like the one we're going to roll. Now, I should say that there have been predictions in the past of paperless offices, and they haven't come true quite as we expected. Part of it is that we created the devices to manufacture documents before we created the devices to consume and read them. But with apologies to our good friend Steve Rogel at Weyerhaeuser, let me suggest that if we're on target, by 2013 you'll see this commercial.

(Video shown.)

MR. BRASS: Thank you, Bill.

MR. GATES: One of the things that's fascinating about these breakthroughs to come is that they won't happen, their popularity won't arrive gradually. That is, when you're able to create a system that, say, the handwriting recognition is good enough; it will be good enough, essentially, for the entire market. If the recognition rate is, say, 80-85 percent, there's no market. If the recognition rate passes over some magic threshold that we're getting very close to, then you have a very broad market. In the case of book reading, we've had to do a lot of research about what really allows you to have that immersive experience, where you just sit for hours and you don't think about the book, you just simply are concentrating on the great writing that's going on there. And there are many, many elements that we have to pull together before that can be done on an LCD screen. Now, it's partly about resolution, it's partly about the size of the device, the fact that you can shift the angle, move it around. It's a very demanding set of requirements. And yet, we feel quite clear that our partners in the hardware business will actually be able to make those within the next couple of years. In fact, this fall we'll have a number of prototypes that we'll be using here at Microsoft that, although the expense to build them is impractical, we actually think the quality is good enough that when that comes down, it will be a mainstream phenomena.

So, I'm looking forward to saying, okay, when does handwriting, when does e-book reading, when do those get into the mainstream. It's a very tough prediction, it's like saying, when was the Internet or online information exchange going to reach critical mass. We went for a decade where certainly all of us at Microsoft were disappointed that online activity didn't seem to be catching on. And then, around 1995, we were actually stunned at the non-linear rise in the popularity. So, we have four or five things here, speech, screen phone, e-book reading, tablet-type PCs, which I would say I feel very confident that in the next five years we'll reach that magic transition point for all of those things.

Now part of it has to do with the fact that more and more of the things in the world will be digital. We're seeing that with photos today. There's finally a crossover where people are starting to see digital cameras as a very mainstream thing. The resolutions have come up, the ease of use still could be better, and we're part of the need to improve that, but the richness and flexibility, the fact you can mail them around to people, the act that you can edit them, and put them in an album in different ways, that's getting those to critical mass. And so it's a mainstream thing.

Digital music, certainly for the younger generation, is very mainstream. It is raising a lot of controversy about the issue Dick mentioned about people paying for the music in the appropriate way. And everybody who has intellectual property, including Microsoft with software, and all the entertainment companies, have got to be concerned about that. Video will be the next thing, in fact you'll see both Windows PCs and Macintoshes promoting the idea that the latest hardware lets you take home videos and do your own editing. You'll also have a lot of these devices, so-called digital video recorders, that let you record shows and play them back whenever you want. And although those have been fairly expensive, there will be a generation of set-top box from the cable industry that just includes that as a standard capability, watch any show exactly when you'd want to.

The note taking, that's the tablet PC, and we see that as a huge benefit because today you're really split, the notes you take in meetings, the things you've got on your PC screen, you're spending a lot of time just moving things back and forth across that boundary.

So, digital media in all these different forms. Digital business in every respect, even the records so that things are filed away electronically and those are the easiest way to call up the information.

For knowledge workers, there's these processes. How did they learn about a new product, how do they get together to decide about something, how do they do deep analysis. And that is largely not digital today, and we think through the right software even in the next five years, that will change quite a bit.

So the experience of using this third generation Internet will be significantly different. There will be the natural interface, which particularly for the small devices having speech will be very important. All the typical information you want you'll be able to call in and get. I mentioned the idea of the user being in control. People say, well, won't this overload me, won't I get more junk mail, won't I have to go out and poll different sites to see what things have changed? The fact is that through the magic of software, we should be able to tell what you're doing and, therefore, be very appropriate in terms of how you get notified about things that change. Many pieces of mail you may never want to pay attention to. And so, those simply shouldn't take any of your time.

I talked about having access to your information anywhere. So you won't think of the devices themselves as holding information. For practical reasons, because of bandwidth limitations actually behind the scenes, we'll actually be loading the information down onto the device, but you won't have to think about that at all. It will just all be out there, and it will be the best of that browser world that's reading, and the best of the Web, and the Office capability where you can create and edit the document.

I wanted to spend a few minutes talking about how does a business get to this point of having all this digital activity. Of course, people aren't going to go and change the back-end systems they've got. People who bought SAP or I2 or JD Edwards, or all the great enterprise software packages that are out there, they put so much into those that they'd like to leave those in place, and yet put around it the software that lets them operate in this new way. And so we have to standardize the communications so that the different software packages can talk to each other. And, as you go across a corporate boundary, you can talk to each other about what is a purchase order, or what is an agreement, or what is a delivery.

And the buzz word that's associated with this, I tried to think about how I could give this speech without using this buzz word, but I don't think there's any way, the buzz word is XML. And that is a format, an Internet format, that allows you to encode sort of arbitrary information in standard ways, and the information essentially describes what kind of information it is so that it can be passed around from system to system. And so it's really about connecting things.

There's a generation of software being built today, and the first wave of that comes out late this year where XML is sort of built into these things. The development tools, the databases, even the Office productivity software that lets you go out and browse and see things that are going on. So, it's a very big change. Every industry actually has to take XML and look at how are health records going to be done, how are purchase orders going to be done.

You could say, well, isn't that just what we did with EDI? And the fact is that EDI took a small part of this problem, simply some pieces of the supply chain, and did come up with standard formats. And so matching that over to XML was very easy. But, EDI never dealt with real time interchange, where you say, okay, I've got a certain number of packages that were broken, didn't you announce a sale, so shouldn't I get a discount on some of these cartons, all that complex interaction was outside of the EDI protocol. So we need to go way beyond what even is present there. And that is happening right now in the different industry groups.

I wanted to give you a sense that setting these things up is not going to be like switching to a new piece of back end software. In fact, this glue software, these tools around XML will make this pretty straightforward. So what I did is I asked some of our guys, using these new tools, to take an example company that just arbitrarily we said, okay, let's say they have a JD Edwards order processing, I2 supply chain, McCue warehousing, and a special picking system in their warehouse. So let's leave that software as given and see if two people in two weeks can take and use these XML tools to create essentially an Internet presence and a very digital enterprise out of that. And so I want to quickly demonstrate exactly what they were able to create in this time frame. What I'm saying here is that with these advances, for budgets that are more on the order of a few million dollars, getting these things to work together, and getting the information to the knowledge workers in the right form should absolutely be possible.

So let's start out, let's say, in this company. Here I am on my PC. I'll go and log in here. I'm being at this point a customer. Now, this is a company that had no Web site at all. They just took orders manually. So what we did was we said, okay, let's take Commerce Server, which is an XML software product, and build them a Web site. So we put some sample products on here. I'll just for simplicity's sake say I want to buy one of these. And I see the price, I see the information, and I'll go through the transaction here where I'll say let's do the checkout, and I select exactly where I want it to be sent, and how I want to pay for the shipping. And I confirm the price, and I can choose exactly how I want to pay.

And so all the modules, the tax calculation, the different shipping options, those are standard modules that are using this XML approach to communicate. So we went out to the Internet and got those modules, so that would all be done properly. So now I've got -- I've made my order and I've got my order number. So that Web site interaction was one of the things that was built during the next weeks. Now, when that order goes in, it goes to the company. And say in the past the company did the typical thing, that is they would print out all these orders, and they'd have a lot of employees that would rip those things off the printer, and kind of look at them, and then run around and try and pull together the order.

Now, that's been replaced. We used the XML glue to send the information not on paper, but just across the Internet to the warehousing system. So here we can see that the employee in the warehouse has one of these little PDA devices, and he's being told to go in and get one of these particular products that we ordered. So he'll go in and say that he's got that, and say that he's finished with this order. So now that warehouse employee is far more productive.

On this order I said I wanted them to call me. If I wasn't on my PC when the order was shipped go ahead instead of sending me email to go ahead and call me on my cell phone, because I really want to know that this thing was shipped. So all that software was connected up to a voice synthesis system. So there I was notified of that, and that it's on the way.

And the final element I want to show about this is how an employee in this company who historically would have gotten the information by seeing some sort of sales print out at the end of the month, either in too detailed a form or too summary a form, with what they've got here they'll be able to go in and actually look at the information, in a different, rich way. Okay. Here's one view of taking that information and doing rich sales analysis. The idea is that in the spreadsheet interface, not just in a print out or in a Web-type page where it's frozen in one view, now you can select the things you care about, look into the trends, look into the changes, and that didn't require any change into that actual back end software. So with a two week process we were able to take a business and get most of the benefits of doing things in an e-commerce type way. So the complexity of building these systems will be dramatically lower.

It's actually something that we've surprised ourselves on, because we said the first problem we set out to solve with this XML stuff was to say, okay, you've got to go across that company boundary between buyers and sellers. And only after we'd worked hard on that and come up with some really great stuff for that did we realize that the information barriers inside a company between these different software packages, exactly the same approaches would be very radical, in terms of simplifying getting the information to flow there using XML as well. So it's a very dramatic thing that I'd say the software industry as a whole is embracing pretty dramatically.

The theme here about empowering workers with these digital forms is one that we continue to push on, because of the potential of doing this, I'd say, even American business as a whole is probably doing the best on this, has captured less than 10 percent of this. That is, the manual processes, the paper forms that are still out there are the rule rather than the exception. Just to get some figures from Microsoft, we've gone from a purchase order being a $60, 20-minute event for an employee, down to it being a $5, 3-minute event. And the number of errors, that is where something was coded in the wrong way, or tracked the wrong way, or tracked the wrong way, that is dramatically down, because now we have that person's email address, and anything they're filling in on the form we give them immediate feedback about whether it makes sense, or if there's something they haven't filled in, which you wouldn't have without the digital process.

So that continues to be quite dramatic. Also, this idea of having more business intelligence, letting employees see what's going on, make more suggestions, because they have a sense of the activities in the business, and they can see that real-time analysis, that is something that can't be overstated in terms of its importance.

One that we talked about before, that I think has been particularly dramatic for us in the last year is the idea of the digital feedback loop, taking what would have been yearly paper form surveys, and now really on an ad-hoc basis -- every couple of weeks, every couple of months -- being able to survey people about how are you feeling about your project goals, how do you feel about your management. And as we do business reviews now, we're taking these measures, and it's really part of the discussion, what percentage of people are excited about the goals of that group. And very often in the discussion we'll come up with a different way that we need to look at it if the numbers are low. Look at where that problem is, come up with different answers. And this is very actionable stuff, serving employees and knowing exactly what they're thinking about things.

One set of programs that has emerged this year in the United States has actually been fairly popular outside the United States, even before this, is the idea of companies providing an option for employees to get PCs at home. In some cases they do it totally for free, in some cases there's a monthly charge. In many of these cases they have taken it really to quite an extreme, to say look, even what you might think of as blue collar jobs, we want them to be open minded about these new digital approaches to doing things. So we're going to make sure they're out there, using the Internet, seeing what this new world is like. So I've listed Intel, Avon, Raytheon, I think about seven or eight companies are now doing this, and there's a lot of companies emerging to make this simple. I don't think everyone will do it, but it is an interesting thing to super charge the move to getting a company to think about there is a new way of doing things.

One of the things that I mentioned a year ago that we were just starting on was this idea of digital meetings. So in the last year virtually every meeting at Microsoft that's of broad interest, we actually not only recorded digitally, so that anybody from their PC who should have access can go and look at what went on in that meeting, but we also broadcast it as it's taking place. And that's had a huge impact, way beyond what we expected. We'd have these product review meetings, where you'd have 50 or 60 people in the meeting. Now, what we have is the core group of about 15 people actually go to the meeting, the ones who want to converse and engage, and the rest of the people either just watch the meeting from their PC, or they can look at the progress of the meeting, and what's coming up, and then they go into that meeting at exactly the point where their involvement is worthwhile.

Likewise, everything that went on in that meeting, because it's digitally recorded, you can take and just email around to somebody a snippet of the meeting. A very typical email might read, your product came up during this meeting you weren't in, here's the two minute clip, you ought to hear what Steve is saying about your product, or what people -- how this group is dependent on you. So you get a level of communication where people can see that information. And in many cases, because of that, they don't have to go sit through a meeting that only 10 percent of it might be relevant to them, just to make sure that their product didn't come up, or somebody's not worried about the dependency that's going on with their thing, they can have that involvement on a much more efficient basis.

To really do this well you have to have your networks enabled for reasonably high-speed data, video and audio capabilities. Most networks the amount that it costs to get them up to this level is actually quite modest to do that. Often it's helpful to have the wireless network as well, so that when you go into the meeting you're connected up as well. And you'll find that during these meetings people are looking information up, showing things to the group, checking on things in a way that makes the meeting itself a lot more effective, and taking the notes, and letting people know about things more effective, as well.

This is just an example of -- and we have an events page, where they list out all the things of general interest. And sitting at your PC you would just click on something, this is kind of a technical session here, but you'll see what it's like sitting at the PC.

(Video shown.)

MR. GATES: So the quality of that is very good. And, in fact, the way we prioritize traffic on our network is set up so the video traffic can never slow down the business critical traffic, the email traffic, or the SAP system traffic. So even if you get very heavy loads of this stuff, in no way is it going to interfere with the other things that are going on.

So I'd say that today the opportunities to get out in front in using these tools include putting in the wireless infrastructure, it includes this idea of digital meetings. And I don't think any company should -- I think every company should take those meetings and just get them up online, so people can access them through the PC, 90 percent of the expense of doing that is buying the PCs and having the network that you already have. So the increment is fairly small. As this catches on, you'll find that it becomes universally popular and expected, and the opportunity for learning things, sharing things, is pretty big there.

Digital supply chain, I think most people have initiatives along those lines. But, now that term is being defined to include every interaction, including the matching of buyers and sellers. And I think this idea of worker involvement; last year we used the term digital dashboard. And that really caught on, some companies took that and did some really fantastic things with it, to say, what is the information that ought to be in front of the individual workers. One of the great things about that initiative is you can actually track what people are really using. Are they diving into the quality statistics, or the customer service statistics, are they going down and looking at the account information. So as you do these things you have a very clear guide of what's being used, what's not being used.

Tomorrow Craig Mundie is actually going to show us some examples of interacting in this digital meeting environment, which will be pretty interesting. So the leadership opportunities today sort of take us beyond the basics. The basics are the pervasive use of email, a real metric inside a company to get rid of paper forms, constant evaluation of people who visit your Web site, serving some reasonable percentage of them to say, when they came to the Web site did they find what they want, using that to kind of track the attitudes of your customers, and what they're thinking about various products. Those I consider the basics. These things here are the things that the top 20 or 30 percent of companies in the United States will be doing in the year ahead.

So a lot of change, a lot of opportunity. We're not just talking about taking the advances of the past and suffusing them out into 100 percent of companies. We're talking about new waves, and new ways of thinking about the Internet, and that's going to keep all of our jobs very, very exciting.

Thank you.

(Applause.)

Transcript by Federal News Service, Inc. 1-800-211-4020