CRAIG MUNDIE: Thanks, Robbie. Good morning, everyone. As Robbie talked about, and indicated by the "Project Natal" demo, there is a huge opportunity to change the way people interact with computers. He demoed how that will happen to change the gaming experience, but, of course, that seeps into the way people might enjoy movies or other forms of entertainment in the home as well. So I want to talk to you a little bit about how broadly we see the opportunity to evolve the human interface in ways that will alter the experience in a positive way and ultimately include a lot more people in terms of their ability to get the benefits of computing.

The evolution of computers has always advanced in waves. And oftentimes the thing that drives us from one generation to the next is some combination of technological advance and the underlying hardware capabilities, and some conceptual advance in terms of how people are going to use the computer— what do they do with it. And those of you that have been at this meeting the last couple of years have heard me talk about how the hardware is going to change dramatically over the next few years— the arrival of high-core-count microprocessors, for example, the advances in display technology. All of these are building up capability. The real question is, what are the killer apps of the next era and, importantly, what will be the user interface that people use to get at them?

And so if you look back over the generations, Microsoft's core business has really evolved when we moved from the text-mode interface of DOS to the graphical-user interface and took those very popular applications at the time— word processing and spreadsheets— and put them into this graphical-user interface model.

You would say the next big change in the user interface was brought about by the Internet— but not about the connectivity. The TCP/IP networks have been around literally for more than 20 years at that time. It was the browser that created a new model of how people would interact, and, in particular, the hyperlink— you know, click on the blue link and you go someplace, you know— was a whole new way for people to expand the domain over which they could get computing to operate. And things like search as another navigation metaphor were important both at a local and certainly at an Internet level.

We've talked a lot this morning already in Steve's remarks and a little bit in Robbie's remarks about our view of how the capability in the cloud— both computational capability and the broadband connectivity— has created the basis of another step forward in the evolution of the total platform— what you can think of now as the cloud plus the client— not as two separate things, the way they've been in sort of the GUI and Internet environment, but where that's really one thing that is going to enable the creation of these new applications. And so even as Robbie demonstrated in the game environment, the combination of Xbox LIVE with the Xbox capability has really created a differentiated gaming experience, even in the generations we've had now. And as the computing capability and these novel capabilities like the "Project Natal" camera are brought forward, we can advance this in other dramatic ways.

Steve Ballmer commented in one slide about the fact that we do see the evolution of this natural user interface. And I want to talk a little bit about how far back you might have to stand to realize how important these changes are going to be.

You know, I just took six inventions on the slide here: the original bicycle, which actually didn't even have pedals, the original radio and telephone, car, early computer, and early robotic devices, and as people looked at those, you know, they said, well, each of these is going to create some big change. But if you look at actually how they evolved today, and what we think of as, you know, a state-of-the-art bicycle, you know, how the phone has evolved to instant messaging and videoconferencing, you know, what we think about today as radio communications, including satellite, you know, how cars look today compared to what they were in their early days. And, of course, the personal computer and robotics have both evolved in important ways. And yet as much as we look at them and say, "Wow, these things are just tremendously powerful capabilities," what we believe is, particularly these last two— the concept of robotics and the evolution of the computer— are really ready to take dramatic steps forward. And I want to explain to you and give you a demo today of how we think that evolution will take place and why it will be so important.

For many years, Bill Gates and I and others have talked about the emergence of this natural user interface as a concept, but up to the present, when I really look back at what's happened, all we've been doing is taking one or two at a time the concepts of how computers can simulate the senses of people— touch, vision, hearing, speech— and use them in the interaction between people and machines. But so far really all we've done is enhance the GUI. The applications, the basic model has still been pretty much the thing you look at on the screen and you want to manipulate and operate a menu that's presented to you there.

And so we've clearly made some steps in this direction, and a lot of the advance we made with the surface and many people now see popularly used in touch-based interfaces on phones and personal computers— each of these still is essentially controlling a model of computer interaction that was largely derived from the basic graphical-user interface model.

And so what we think is going to happen now is that we will step all the way back— a little bit like they did with the Xbox, where clearly they knew what games historically were with the controllers and other things, but when you step back and say, "Well, what could the concept be if there is no controller?" You are the controller. And in that environment you just want to interact in a completely different way with this. Then you start to think differently about how to develop the application. In the cases that Robbie talked about, that application is gaming or entertainment. And what I want to show is how broadly this adjustment to the natural user interface could take place, and, in particular, if you take these concepts and extrapolate them forward and put them in the business environment instead of the home and entertainment environment, what that might be like.

In this more holistic concept of the natural user interface, I think there are a variety of things that become important. Clearly, gestures are an extension beyond touch. The computer will become more and more environmentally aware— not in the planetary sense, but it will know what's going on around it. Why? Because sensors are becoming really cheap and very, very accurate in terms of their ability to monitor and detect things. There have been substantial advances, even in Windows 7, which we're not really demonstrating today, but where there is a whole model for how developers can really take sensor-based technology and incorporate them into future experiences, even starting with the Windows 7 technology.

The computer will move from being one that is focused more or less on reaction to what you do to trying to anticipate more what you do. It will become more expressive in the sense that you won't just look at text or graphics that are presented, but in fact this is where robotics comes into play. For more than a year now, one of my most popular demos I give has been of a robotic receptionist, and we have been evolving this robotic presentation much in the way we see avatars in games as a way for the computer to have a more humanlike projection and a more humanlike way of interaction.

This will be important, I think, even in the rich world environment, but will be very, very important as we seek to expand our market for computing broadly to the rest of the world's people who don't get benefit from it today. Today, the planet has about six and a half billion people. Only about a billion and a half really get any real benefit from computers, even if you stretch to include smartphones. And so there's five billion more, and in one more lifetime of Microsoft up to the present the planet will probably get up to about nine billion people, and most of those people will come online and just increase the pressure for ways to solve problems around health care, education and improved productivity. And if we're going to do that, I remain convinced that it will only be through the consumerization of advanced computer technology that we'll be able to provide scalable solutions to that many more people. And I think that that will be important for the global economy and the society, and it's one of the things I'm hopeful for in terms of what Microsoft can contribute beyond what we already do in the more-well-developed markets. I think that this contextual awareness will become very important in terms of making the ability to interact with machines a much more natural thing.

If you think about computing as we've done it to the present, it's really been the creation of perhaps the world's most powerful tool. And these tools have gone, you could say, from a more Stone Age sort of hammer and chisel up to the six-axes milling machine, computer controlled. But in each case you have to do your apprenticeship, you have to get the training, you have to master the tool. You can master the tool; you can do some amazing things. But still it depends on you being able to really use the tool.

I think what's going to happen now is that computers will increasingly work more on your behalf. The business of software from the very beginning has been about raising the abstraction level, making it— making higher-level concepts— the way that people interact with machines. In the early days those high-level concepts were just giving programming languages to the programmers that were more robust or higher level. But now we've seen with the advent of modeling, for example, the ability to give people the ability to express and get answers to problems. Excel was essentially one of the early modeling tools where, for the first time, a person with no programming expertise was perhaps able to express and get a solution to a problem. We see this higher-level-modeling-based interaction becoming more and more prevalent.

The last thing before I talk about— give you the demo— is how are the form factors evolving? How will computers be presented to us in our daily life? More and more we will recognize that they just don't— they are not things that we point at and say, "That's the computer," and we sit down and address the computer and work at the computer. And so now that we see it in everything. The company has really come together under this moniker that Robbie talked about and Steve did of three screens and a cloud. In some sense, you could think that there's really like three and a half screens and a cloud, because there's a whole category of devices— in this taxonomy I label them the specialty devices— where computers will also be part of that. Microsoft has really tried to figure out how we are going to engineer experiences for sort of the three big ones, and today we talked about those, as Robbie did with the TV, the PC and the phone.

But one way to think about that is really not just those devices and their screens but sort of the screen sizes, which you could think of as large, medium, and small. And what I'll show you is that I think there's a way to think about these things as advancing beyond the idea of just those three devices we historically know.

Clearly the marketplace and Microsoft business have been focused more now on what we think of as the portable, which is the portable desk— aka the laptop or netbook— and the phones. We see the emergence of these other categories, whether they're book readers or your car, which will become an integral part of this environment too. But I think for Microsoft, one of the greatest opportunities going forward is ultimately to realize that there will be a successor to the desktop. It's the room. It's the fixed computing environment. And the question is, what can you do with computing when you have much more of this display capability, a much more robust managing interaction model, and you don't have to fold it in half and move it and run it on a battery? And so what I want to show you today is at least a vision for what that room computer might be like in an office setting in the future.

So what I'm going to do is operate my demonstration here of an office of the future. So, as I might enter my office, it senses my presence and lights up the screen savers, which in this case is sort of some combination of the gadgets that you might historically know on a laptop. I might have the seashore in my artificial window if I don't really have a window. I've got a surface computer as essentially a desktop. And— but as I approach it and I actually want to do some work, it— you know— I address the computer environment and it changes. I don't have to do anything. And so here I have a desk with the papers and things on it that I'm supposed to process. I have both high-resolution capability for display and broader capability, and the whole walls are essentially projection and display surfaces. The office has, both of the traditional type and the type that we demoed in the "Project Natal" and Xbox, so the ability for the office to see motion and recognize gestures.

So, as I come and I want to resume work, you know, I'll touch a button here and convert this display from background to the project I was most recently working on. In this case, I'm an architect and I'm working on some building designs. And so I can reconstitute that by a button push or a gesture. I can go back to my desktop where we essentially have sort of a free-form ability to migrate among the components of my work. So here I want to actually migrate back down to a section of the work that I was doing before. And I'll zoom in on it. And so here I was actually running a model, and I get a control on the desktop that allows me to operate the model. And so I can say I want to rerun some simulation. I'm given a custom user interface that I can essentially alter parameters and look at the output in terms of graphical interface models. And I have a very powerful way of interacting using touch surfaces that adapt the traditional point-and-click type of interface to modeling and graphical presentation in this environment.

The computer can give me reminders, so here's a little note down here that says there's an upcoming meeting. And so if somebody might be coming to visit my office, there's always a question of how do you get back to the context you had before. And so here I might use a gesture, for example, where I'll take the camera gesture, and I'll essentially change the context of my office to prepare for the meeting that's arriving. We were talking about some project planning stuff, so I'll basically reconstitute the notes that we had when we were discussing this environment.

And I actually had some more stuff that we may want to discuss or talk about. And because it's all camera-based, I can essentially have— you know, the system can recognize these images and I might be able to go up here and, say, get an interface, touch it and have it— put that on the wall. So I can add real documents. We expect phones will essentially be an integral part of this. This is the small-screen version, so here I can have a picture, for example, or something on the phone, so I can pick it up with a pinch and essentially move it to the wall. And so the idea that we'll have a very free and natural way of interacting between the devices and computers will be very important.

But I think we've also been exploring the idea of how we can have computer-simulated robotics present themselves as a way to get personal assistance or have the computer help us more. And so here I have a digital assistant. Good morning, DAG, what do you got for me today?

DAG: I have some answers to the question raised at the meeting this morning. Do you want to see them now or should I just post the information to the project site?

CRAIG MUNDIE: Just post it to the site.

DAG: Done. You also have a meeting with Patricia San Juan in a few minutes.

I don't remember the context of that. Can you give me some background on that meeting?

DAG: Of course. Here's some background.

CRAIG MUNDIE: OK, so here we're working on the water project. I see the costs aren't actually what they were the last time we discussed it. Can you give me some background detail on those costs?

DAG: Let me check on that. Here. If we zoom in on this part, we can see that the pricing of some of the upstream dependencies has changed. Would you like me to get Patricia San Juan on the line?

CRAIG MUNDIE: Yes, please. Initiate the call.

DAG: Is there anything else?

CRAIG MUNDIE: No, that's all for now. Thank you.

PATRICIA SAN JUAN: Oh, hi, it's good to see you again.

CRAIG MUNDIE: Good morning, Patricia. I'd like to keep talking about our water project.

PATRICIA SAN JUAN: Now, my team has refined three of the designs since our last conversation. Here's what we came up with. Now, given what we know from you and the customer, these are the three we think would work the best. And number three is the one the customer liked the most, but we've come across a small problem. I'm hoping we can talk through some of our ideas on how to make this one work.

CRAIG MUNDIE: All right.

PATRICIA SAN JUAN: So let's set aside the other two for a second.

CRAIG MUNDIE: Good.

PATRICIA SAN JUAN: OK, now here's the problem: the water system the customer was so excited about is in a great spot and the model shows that it really enhances the existing day-lighting system. But the model also indicates that we may come up against some of the regulatory changes that are starting to take hold in some other cities. So our solution is to use a different water capture system here and here. OK, now watch the other variables and see if there's anything you're concerned about.

CRAIG MUNDIE: They look OK.

PATRICIA SAN JUAN: Plus we get a sourcing incentive since we can get most of the materials right here in town.

CRAIG MUNDIE: The locals will like it.

PATRICIA SAN JUAN: So this way we keep the aesthetics the customer likes, we're forward- compliant with the regs and we end up saving a bit of money. What do you think?

CRAIG MUNDIE: What's not to like?

PATRICIA SAN JUAN: I think it's going to be great. I can't wait to start construction.

CRAIG MUNDIE: OK, thanks. Have a good day.

So, more and more we see a way of integrating the cyberworld with the physical world and using the computational capability to build understanding from that.

So the last thing I'll show you is the idea of running models— 3-D computer models— that would help me, in this case as an architect, understand what would this water project really be like if situated on that roof and in that city.

So, DAG, can you get the simulation for the water project and run it here? Thank you.

DAG: Right away.

CRAIG MUNDIE: So here I've got a simulation computer model of that architectural project. The background is actually the cityscape that it's in, but that also is a 3-D model, just like we do in our Virtual Earth capabilities today on the Web. But here the "Project Natal"-type technology cameras are actually modeling where I am and my interactions and placing me in that physical model from a computational point of view. So if I actually move around, the view will actually change because my position on the roof changes. So if I walk over here, it actually recomputes in real time what that is. And so here the question might have been, if I had the client here, If I stand in this water feature on the top of the roof, would I be able to see the water past these buildings? And you could answer the question by saying, yup, you can stand right here and see that you could see that water.

And so here I'm not playing the ricochet game, but I'm using exactly the same technologies to bring these things together and to create a way where I have a very natural human interaction model between the computer environment, the real world environment, and the task that I want to work on. So this is our dream, but it's really not that far away.

And if you think about the extrapolation from what Robbie showed in the Xbox utilization of this camera technology, what I've shown here, we see a pretty direct path to make this happen. We've got all the technologies necessary to do this in our research labs, and, you know, our goal right now as a management team is to really create an accelerated pipeline where the great work we do in the MSR labs is able to go into these products on a very rapid basis. Robbie talked about the speed with which "Project Natal" was completed, and part of that was enabled because we, in fact, had spent quite a bit of time over the years developing different aspects of machine learning, machine vision, and many of the underlying algorithms for that, and being able to do that work in conjunction with Robbie's development team to produce this "Project Natal" product at a time we think other people will have a tough time copying it.

So that's the vision we have of perhaps work of the future, and at this point I thank you for your attention and ask Robbie and Bill to come back and join us for question and answer.

Due to the varying sound quality and subject matter of tapes, the information in this transcript may contain inaccuracies.