MR. MUNDIE: Good afternoon. Let’s go ahead and begin. My name is Craig Mundie. I apologize at this moment, I have a pretty bad case of laryngitis. I’m going to try to give this talk anyway. I may have to take some help from my friends.

I’m the Senior Vice President at Microsoft of Consumer Strategy, I work with Bill Gates and try to develop and coordinate that strategy for this whole company. This afternoon, I would like to build a little bit on what Bill did in his keynote last night. He talked about the vision that Microsoft has for consumer computing in the future, and what I’m going to try to do today is take one or two of the scenarios that Bill talked about and drill down a little bit and talk about what progress we’ve made technologically, and as a set of industries in the last year, and what I think is going to be required in the coming years in order for us to, collectively, realize this vision.

Clearly, one of the things that’s going to be a requirement for consumer computing to happen and be a factor in people’s lives is that the devices are not only going to have to get smart, they’re going to have to get connected. And we clearly have made a lot of strides building on the work people have done with embedded computing for a long time on getting the devices to be smart. The challenge that is in front of us now is to get them connected, economically, and then to provide enough intelligence through the software to really make these things function as an integrated whole for the consumer.

If you look at the network, as we might even envision it today, the reality is that it’s still appears to the consumer as a technological nightmare. It’s a heterogeneous environment. There’s going to be many types of sources and syncs, many different types of formats in both the audio, video, photo and data areas. And clearly we have to find a way to bring all this together. It’s obvious, certainly, to the people at Microsoft that if we didn’t do any better in approaching the home than we have collectively grown to as we built the corporate networks, we really wouldn’t have a business in the home, because it would be too complicated. So, we that the consumers have to see something that is dramatically simplified. Despite the fact that there is all this heterogeneity, they have to see a transparent network, one that really is no different than they perceive today with their electric power system and their phone system, which is, if you have a device, you find a plug, as long as the plug goes in the hole, it works. And we need to find a way to make that happen with computing, and the attendant networking.

And, I came to this show a year ago and announced the Universal Plug and Play initiative, where Microsoft was trying to take a leadership position in creating an industry initiative that would allow us to move in a direction of realizing this kind of transparent network. And as I go on in this talk, I’ll explain a bit more about the progress we’ve made there.

If you look in the top corner of this slide, you see a stack that shows the different kind of software that are going to be present in the devices, and in the networks. We have the operating system, the traditional TCP/IP networking capabilities. And, of course, people have been proposing different types of middleware solutions, and the traditional applications and services that live on top of that. What we believe is required is something above the primitive networking environment, and below the application platforms, which we call the Universal Plug and Play initiative. It really has two important components, a simple discovery mechanism that allows people to plug in a device and have it automatically be discovered, and to discover the other devices and services that relate to it, and then essentially a description and usage mechanism. And we chose to build both of these on what are evolved to be very important, yet quite primitive parts of the Internet as it goes forward, building on the basic TCP/IP and HTTP protocols, and using XML as the lingua franca to describe this environment in the future, and to make it all machine-readable, if you will.

So, as the year has gone on, we’ve made a lot of progress both organizationally and technically in defining the universal plug and play activities, and what we know will happen is that there’s going to be a set of devices that are intrinsically part of the IP network, and another set of devices that will be connected in non-IP networks. Many of the activities were started, for example, around the hobby initiative on [IEEE] 1394 really got their momentum cranked up before the Internet phenomenon had really taken hold. And, as a result, there’s going to be a big investment, particularly in the consumer electronics area, around devices that are not fundamentally connected in an IP environment. And so, there’s been a lot of dialogue between many of the principals in the hobby world and the UPnP activities. Many of them, in fact the majority of them, have also become steering committee members of participants in the UPnP activity.

Because I think even if it wasn’t clear a couple of years ago, it’s absolutely clear today that no device, or no set of devices, are going to be able to survive and be effective as an island. So, just hooking up the consumer electronics devices won’t really be that interesting. They’re going to need access to the Internet, and they’re going to need access to the devices that are in the IP network. And so we really have evolved a strategy in Universal Plug and Play that allows us to take both the control point devices, and the devices that are controlled in that environment in a native way using Universal Plug and Play.

And we propose to work with people, for example, in the hobby world as well as other areas, to build Universal Plug and Play bridges that would allow the legacy devices in the non-IP environments to be represented and have interactions with those in the other world. And so, by doing that, we think we become a mechanism for bridging between all of the different heterogeneous networks and device types that will be present in the home.

The partners that have joined the Universal Plug and Play initiative represent all the leaders in the computer industry and the consumer electronics industry for the most part, as well as from other places. And we’re very, very pleased with the progress and commitment that these folks have made this year. If you look at it a year later, since I stood on this stage a year ago and talked about it, there’s quite powerful momentum. There are 65 member companies, 17 of those companies have put people on the steering committee, which is essentially the governance mechanism for Universal Plug and Play. We’ve got six operating working groups producing device control protocols, and six more that have been commissioned and are ready to begin. The universalplugandplay.org Web site is live since the summer. There’s been already 125,000 hits by people coming and looking for information about the protocols or how to participate in the process. And some of the work that’s already been done, both in terms of the underlying architecture and I think some of the initial work that will come out of the working committee is being prepared, and already is in a form of IDP standards drafts.

So, partly what I want to highlight is that it’s so critical to create an environment where the devices come together easily. It’s also clear to us that this has to work in an Internet environment, and it’s also clear to us that it has to cover a great array of industries. And so we’re extremely pleased with the progress that this has made in the last year.

And then finally what’s the application interface is the next set of problems. You could say that’s the next frontier, certainly for Microsoft, but ultimately I think for most companies is that the Internet has got to evolve from the form it’s in today to a new form. Today, you could think of the Internet as being in its second major version. The first version was the development of the TCP/IP network and the most primitive applications like file transfer and electronic mail. The arrival of the browsers and HTTP created the Internet we really all know today, riding just on the back of PCs and the dial-up telephone network.

And so as we go forward, a couple of major changes are going to happen. One, we’re going to move beyond the narrow-band dial-up world into a broadband connected environment. And, two, we’re basically going to put a lot of this capability of being Internet enabled into a lot of other devices, not just the personal computer. When we do that, what we really have to do is create an environment where the Internet becomes not just a mechanism, as it is today, where people click on a link on a Web page and they read a Web page that gets delivered from a server, it’s really people reading Web pages. In the future, if we’re really going to make computing useful to people, we have to create an environment where the computers talk to each other using the Internet plumbing, and together they collaborate, if you will, in service to the people who own them or use them. And for that to happen, we have to have a programmable Web, and that’s not a characteristic of the Web that we know today.

And it’s pretty clear that all the software that people will deliver will have at least a service component. And so every company, whether you’re in the hardware business or the software business is going to have to come to grips with, how do you actually deal with this programmable Web, and how do you build the related service components.

If you look at what’s already happened in the computing world, the Internet world, in the traditional enterprise computing space, you know, it became very clear to us that people were going to build Web sites, and they wanted to build them out of a lot of heterogeneous technology, and there would be many different types of back end applications. You know, a little while ago, before this notion of this programmable Web really emerged, there were proposals, for example, the Sun Java and Jini initiatives, which really contemplated a world of homogeneous, tightly coupled objects, where they all ran the same virtual machine. They basically used the APIs of that environment as a mechanism for them to interoperate.

It’s been our experience that, in fact, it’s not possible, even in the world of personal computing, and the related servers, or just the enterprise, to have that degree of homogeneity. And, in fact, despite even the preponderance of Windows PCs, there’s not enough homogeneity there for that to be a viable strategy in the enterprise. So, it’s pretty obvious to us that it can’t be a viable strategy as you approach a world that has many, many types of devices.

And so, we’ve been focused a lot on how you’re going to create a programming environment for this loosely coupled environment. And so the strategy that we’ve got, sort of building up from the same concepts that brought us the Universal Plug and Play initiative, is to believe that all these services are going to get connected and interoperate using XML described messages as basically the primitive mechanism we use to define UPnP and we think that, in fact, that type of message and protocol and schema derived environment will be characteristic of how people, in fact, will build all applications on all devices as the Web really begins to take hold.

So, Microsoft’s offerings continue to expand along with our mission statement change. Bill talked about it last night in his keynote speech. In April of this year, for the first time in 25 years, Microsoft changed its mission statement from putting a PC on every desk and in every home, to basically empowering people through software anytime, anyplace, on any device. And in order to do that, you know, we’ve really had to internalize what it’s going to take for us to be a factor in offering people the tools, platform and service technology in order to allow them to both build those devices, and hook them up. And in an incredibly competitive environment.

So, you know, we have operating system products, quite a broad array now. We’ve been heavily involved and we’ll continue to be heavily involved in trying to develop where necessary and promote where there’s already an effort these middleware standards that give us the basis of hooking all these things up. And a lot of what I wanted to focus on today in this talk is where the industry actually has to have this kind of collaboration in order to achieve this dream picture that we showed at the beginning where people just walk up and plug devices in and they just work.

So, clearly, XML, some of the low level physical networking things, USB, 1394, home phone lines, and a lot of the radio frequency connectivity mechanisms are all areas where we’ve been active. And this will be expanded into things around location based computing, the 911 and GPS mechanisms for example, COM+ essentially will continue to be evolved, and it will fundamentally become a programming model and tool set for Microsoft that will work in a uniform way across all these operating system platforms, and allow people to build these heterogeneous, loosely coupled applications.

And then we will develop some applications — clearly we do some of these already– and we continue to take and develop things like the new micro-browser technology, which is a derivative of Internet Explorer, but made to be portable into very small environments like cell phones. And we also have been working and continue to work on trying to refine our service offerings in order to be able to work with partners, whether they’re telecom operators, or ultimately OEM equipment manufacturers to provide the underlying infrastructure and portal services necessary to allow people to have a uniform experience across this heterogeneous mix of applications and devices.

So, we, at least, will have a pretty complete offering of things that can be powered by Windows. And, in fact, Windows won’t just be the operating system that you’ve known on the personal computer and servers, will not even just be the operating system that you know as Windows CE in some of the early generation of these personal, portable and home electronics devices. We have already delivered now the Smart Card Tool Kit, which is actually an 8-bit version of a tiny Windows environment. And we have a thing that we call the Smart Object Kit, which we are getting close to shipping. All these things live in a uniform way under the Visual Studio application development environment, and all of them basically integrate under this model of Universal Plug and Play, and ultimately it’s a loosely coupled model of system construction. So, we feel pretty good about our progress in being able to offer people a complete range of tools.

Let me take some of these scenarios that Bill talked about, and I’m going to focus today on two because they’re the nearer term ones, and they’re around the devices, certainly, that people use daily in their lives today. Music being the first.

If you look at music consumption today, you know, there’s been a long evolution of the music industry, and in part it was always driven by the adoption of a particular media format, whether it was the vinyl record or the CD or the DAT tape or the cassette tape, each of these things begat another generation of devices, and the way people use the music, whether in a portable or fixed environment.

But if you look at what’s been happening on the Internet, the whole MP3 phenomena, the ability for people to use personal computers, and the storage of those coupled with the computationally intense — and as a result enhanced — CODECS that really reduced the bit rate that’s required for high quality audio to be delivered, we really find that some of the leadership demographics, particular the college students, are very rapidly adopting the personal computer as a place where they store, organization, and ultimately play their music.

And while we wouldn’t posit that that’s the way it will work for everybody all the time, it’s pretty clear to us that it shows a trend that people are very willing to accept an environment where the computer helps them organize and get a more personalized experience out of their music, and also facilitates the ability for them to transport that into different environments.

As Bill showed, we have the ability to take all these devices that are connected, whether it’s through these remote tuners, or intelligent stereos, the personal computer itself, on any device essentially that has digital storage capability and a reasonable CPU becomes a potential playback environment for this common set of bits.

So, to some extent, the users, you know, the tyranny of the medium, the thing that has prevented them from collecting the songs in a form they want, in a compilation that they like, and being able to take it sort of anywhere they want is going to be broken by essentially the fungibility of these different stores and playback environments that are all network connected. We think that that’s going to be a really big change in the music industry. It’s already begun to change the way people are distributing and pricing music, and we think it will be ultimately a big change in the way that people buy and deploy these devices.

So, before I drill down into that, I wanted to give you a little demo. And to save my voice, I’ve asked Randy Smith from the media group to come up and just go through this demo with you. It’s a little bit of just how people are doing this on personal computers today and, for example, how we can move it down into these personal portable devices.

Randy.

MR. SMITH: All right. What I’d like to show just briefly here is a product made by one of our partners built on Windows Media Technology that really shows how you can use your PC as a storage for all your music and turn your PC into the central music server for playback on your stereo, for portable devices, and so forth, in a very easy fashion.

This product by MusicMatch essentially keeps all your songs that are stored on your hard disk in a library that can easily sort, put into play lists, by simply clicking along the situation, artist, situation, so you can organize it by either party music, barbecue music, and then just by dragging and dropping into your play list, organize a play list, and hit play. It’s very easy to organize, you can save those and move those on.

Now to get some music onto your computer, there are really two ways to do so. The Internet, download the music in Windows Media Audio. And actually one of the advantages of Windows Media Audio is because it’s half the file size of MP3, that’s half the download time. It’s also secure, so four or five labels are supporting Windows Media Audio technology because of its secure format. You can also, more simply, take the music off your own CDs you already own using this same product, for example, you can load a CD into your CD-ROM drive, record right off of it with a simple push of one button, and you’re good to go.

And once you organize your play list in your PC, the next step is portability, taking it out on the road. Several products are on the market now which are for Windows Media Audio. The first one here, I have the Windows CE product line. We have the Casio player here, which will play back your music as well. Keep in mind, that’s all digital, of course, so there are no moving parts, nothing skips, nothing jumps. Also, another device is, besides the Windows CE, the portable devices such as the RCA Lyra here which will support Windows Media Audio as well. This will give you, with Windows Media Audio, double the capacity over MP3 in the same device. So this one device with a 64-meg Compaq Flash will give you two hours of music. That’s near CD quality music on here.

Our format is being supported by a number of products, all the top players, including Creative, Diamond, Rio, RCA, Sony, the new Music Clip will be supporting it as well, Samsung, and a number of other companies as well.

MR. MUNDIE: So, music is an example where, again, many industries are going to be brought together in order to make these kind of things happen. We not only have the devices, but we, again, have an incredibly interesting array of connections between these devices. And this would not be ultimately an exotic environment for people to have in their home. And if you look at the number of CD players and stereo systems that exist in people’s households and vehicles today, and that they carry around with them, it’s not unusual to think that each of these is going to ultimately evolve to be connected and intelligent in some way. And so, it’s another example of not only a requirement to have these things interoperate in the network in a compatible way, but the need to continue to have flexibility in the evolution of the formats.

So, when you realize that you can take a device like this, I mean, this Casio, that’s a 340-megabyte Compaq Flash card. So, you know, that’s enough for 10 hours of CD quality audio. And when I buy a standard PC off the shelf today, I mean, for probably $1,000 I get one that probably has nine or ten gigabytes of hard drive on it. And at that point, you realize you’ve gone over the threshold at these compression levels, where a single disk on a single PC probably can hold more of your music collection than you actually like. And so, it really does become possible to carry these around with you.

So, Microsoft has been very active in trying to not only develop the underlying CODEC technology, but to basically evangelize everything from the Web mechanisms for distribution, dealt directly with the rights management issue. So, if you look at this music pipeline and think about, wow, you know, look at all the places where people have to come and work together, and you start at the head end and say, I have to have the music, however it got recorded, I put it in a server, so I have to have a transaction mechanism, you know, to find the stuff, and then buy it. You have to have a way of putting it on the server, so you need the CODEC to basically encode the data. Then you need tools that actually allow people to set up the streaming of it so that people who want to do Internet streaming media, or Internet radio, they use basically the same mechanisms.

Then you have to have the search mechanisms, play list analysis, like was demonstrated in this MusicMatch environment. You have to deal with the digital rights problem because now, in fact, the problem as far as the copyright owners are concerned just gets more and more exacerbated. The whole notion of what was the stereo copy management systems, for example, are just out the window because those things were all tied to specific media. And, again, a preordained arrangement that each new media type would somehow carry forward. In this environment where the player is just a piece of software, and it can run in an interchangeable way on any of a variety of computing devices, you don’t have the ability to create and enforce those kinds of a priori arrangements, so you need an architecture for describing these thing and enforcing them that can, in fact, be standardized at a different level.

Clearly, you have to have the decoders that then get put into the client devices. You have to figure out how you’re going to store it. So any of these storage systems, as long as they have the right properties in terms of real-time access or stream-ability give us a way of putting the bits in any of these devices. So he showed us Compaq Flash, Casio holds Compaq Flash, but they now have Compaq Flash size disk drives. You know, basically, there’s just going to be an incredible array of mediums. Sony has the memory stick, for example, as another one. And then, finally, the client devices, how do you get these things to come together, how do you have enough uniformity, how do you ensure that when they all get networked together that there’s not just a cacophony in terms of what the user perceives of how they innovate with them.

And finally what the application interface is going to be is both a blessing and a curse as we begin to put these things together, and provide graphical interfaces. We can either get rid of the historical problem of the six remote controls, where each device has their own completely different type of remote, or in fact we could create an unlimited number of remote controls in the form of new graphical interfaces.

So a lot of thinking has to be done in order to get that to congeal in a useful way. But, ultimately we think we are moving to an environment where people will get music pretty much anywhere any time, on any device, essentially any device that has a minimum capability of computing power and storage. So today we have a lot of people making players, and new companies building services, juke boxes, and ultimately stereo components in the traditional sense that would integrate into this environment.

We’ll move on and talk a little bit about video in the home today. We have really two places where there’s some video being done, the television, and of course the related video capture equipment that people have, which has really gotten to be relative to the traditional display mechanisms, a very high quality, easy to use environment. But, the reality is it’s not very personalized. Television as a broadcasting system hasn’t been interactive or personal. And certainly the problems associated with video, personal video and the management of it, whether you capture it on VCR off your broadcasting environment, or whether you are taking your own video images, there really hasn’t been a good place to put those, to organize them, store them, index them, edit them, and ultimately enjoy them. So many people take a lot of movies, but frankly they don’t watch a lot of it. And we know that that’s similarly true with photography because its’ really such a pain to get access to the stuff that you’ve collected.

The personal computer obviously is a place where people are accessing the Internet. We’ve started to see the arrival of low bit rate video there. The fact that the connections have been so slow, the computers have been so fast, has really been the mother of inventing these low bit rate video compression technologies, and so people are beginning to enjoy low bit rate video through the Internet. The interactivity is great, but again there hasn’t been a lot of personalization of how that’s presented. But, in fact, we are in an era where not only is the traditional television system becoming a digital delivery platform, but clearly the Internet as we change the pipes into the house from narrowband to broadband, that gives us the capability to do this too. And this map actually shows the locations where companies like iBeam and Innerview already are operating broadband media servers in what we call a jumpstart program, to allow people to have broadband, low bit rate video services, and deliver them on demand to people over the Internet.

So these things are clearly moving rapidly toward each other. A few years ago when I talked at NAB and CES, it was at a time when there was a lot of confusion of what would be the roll out mechanism for digital television, whether it would be focused on high definition television, or whether it would be focused on digital data services. We’ve clearly been proponents that say digital television has got to be a vehicle for delivering digital data, in addition to traditional high quality video.

And I think as the market has evolved, certainly outside the United States, and increasingly inside the United States, both companies seem to be coming around to the recognition that these enhanced forms of programming, in fact, are potentially more lucrative perhaps than just a transition to a traditional broadcasting form of high definition television. And the technology to do that is clearly moving quite rapidly.

If you look at what we think video will look like in the home tomorrow, we’ll see TVs with set top boxes in a variety of forms. And we’ll see the Internet connected to those televisions, providing a basis for interactivity in the TV itself, and also providing an alternative mechanism for the delivery of natural video bit streams. And, of course, we’ll see the personal computer and lots of other devices being married up. With the arrival of 1394 support on the personal computer, for example, all those devices you used to buy and hook up to your TV to display, suddenly become peripherals or digital video sources for your personal computer.

And with this evolution, as I pointed out in the audio world where the disk has already become big enough to virtually store any reasonable entire music collection, this rapid evolution of the disk drive hasn’t stopped, and we expect over the next few years that it will continue to grow. And it’s highly likely that we’re going to near quickly the point where it’s big enough to hold most people’s video collection, too.

And so these disk farms, whether they’re purchased inside the computer or purchased as a separate device in the home, will become an environment where people will catalogue photos, music, and video. And of course, at that point you’re going to need a way of organizing and delivering that. And so these techniques that we’ve seen first applied to music are likely now to be applied to video.

So let me, for those of you who haven’t really focused much on this, stop and give just a quick demo here of what’s happening in this world of low bit video compression. Here I have an application program that’s the Windows Media Player, and it has clips and the ability to play these clips at different rates. The thing of course that necessitated this was the fact that people had dial up modems, initially 28.8. So here we’ve taken partial frame rate video, and if I load it to full screen you can see when you blow it up that it’s very grainy and low quality. But, this of course is video over a phone line on a 28.8 modem.

So if you hook up at 56 it starts to get a little bit better. And of course when you get up to 100, change right here to 100 kilobits a second, you still see a little

—

some blocky art effects, but again I’ve now made this to a screen resolution that’s much greater than people’s traditional television. And if you get to 300 kilobits a second, which is certainly readily achieved even on what’s going to be known as the G-dot Life environment, you’re starting to get the audio quality up to almost CD quality, and the images certainly blown to that size are watchable without question. At 700 kilobits a second you’re probably bordering on images that maybe better than what people have enjoyed on the traditional television with VHS tape. Of course, we know how many hours people have spent doing that.

And we can keep going, but clearly by the time you get to 2 megabits and beyond the video quality becomes very, very high, and of course you have very high audio quality. So all these things become

—

it becomes clear that the traditional notions that people have had of what kind of bit rate is required to deliver high quality video services is really dramatically different than what we’ve engineered traditional television systems for. The satellite systems more recently the digital cable systems have begun to recognize this and are using the more sophisticated compression technologies to get their channel counts up. But, ultimately as the broadband wires come to the house, or even through wireless mechanisms, you’re just going to have a lot more competition for how video and audio are going to be delivered in this environment.

So just moving on, if you look similarly at the video pipeline, we have many of the same problems we had in the audio world, and then we have some new ones. One of the biggest issues, of course, is going to be in the home where people want to do this, the client side applications are going to require that we have digital video editing capability in the home. And while this has become popular in the last five years in the professional environment, it’s pretty clear to us that this has now become something that’s viable for people to do in the home. And as Bill Gates indicated last night the next version of Windows for the home will actually include an integral nonlinear video editing system, and so people will be able to build up their own video collections and watch them, much as we saw the assembly of audio play list and the playback of that.

One of the things that’s been clear is that when we move video beyond the world of just passive linear video, we had to have mechanisms that were going to allow this to be done. And in a way it actually predates the work that was done in Universal Plug and Play, but I think it’s become much more prevalent in the last year. Microsoft worked with 18 other companies a couple of years ago, started a thing called ATVEF , which was a mechanism for creating essentially de facto standards, where people would be able to design content, put it on multiple playback environments, and deliver it on a lot of different environments.

And so while there’s been a lot of talk in different quarters, both in the United States and outside the United States about how do you build an intelligent receiver, what’s the programming model, the reality is you can take these same techniques that have been used for people to design Web pages, or do the kind of things we’ve just been talking about here with low bit rate video and audio on the Internet, and we basically are able to use that as the fundamental mechanism embellished only with some primitives, for example, for synchronization and triggering, to create a very rich environment for enhancing digital video. And I’m pleased to say that much, as has been the case with the universal plug and play initiative, this ATVEF initiative has similarly taken hold.

So to put that in perspective a little bit, let me just play one more demonstration here, this is a pilot of a new TV show that’s being put together. And here people are starting with a traditional television production, but in the environment of WebTV, for example, they can have this kind of ATVEF rendering today. And then all the personal computers that come out in the coming years will similarly have this kind of ATVEF rendering capability. And so now what you’ve got is

for people to have the ability to optionally click on that, just as they do on today’s WebTV product, but the programming is going to get richer and richer. So here, you know, you have the ability to interact with this, you know, at will, to do many, many different things. You can chat, you can shop directly as you watch the television programming.

And what’s interesting about this is all of these kind of enhancements, and all this kind of rich interactivity doesn’t require that you do anything about creating application programs that are shipped with the digital video. This is just a lot more like Web pages, where you use mechanisms like HTML, a markup language, basically embed that along with the triggers to synchronize action and the changing display environment, either interacting with a mouse, for example, or remote control for your television. And I think many people have been surprised by how rich an environment you can create using these markup languages and trigger mechanisms of ATVEF. And so this community is going to continue to take that forward, and I think that you’ll see this not only adopted and broadened within the United States market, but increasingly accepted outside the United States. That’s probably the way in which people will really get started producing this type of enhanced and interactive television content.

Let’s conclude here. Much as has been the case with the work that Windows media group has done around audio CODECs and the Universal Plug and Play group has done, all of these things require the coming together of an incredible number of companies in order to create the de facto standards that allow the industry to really move forward. It’s absolutely clear, again, as Bill said in his talk last night, that this world of convergence it’s not about having one company do it all, or one standard, it’s pretty clear there can’t be just one network type, there can’t be one computer type, there can’t be one language type, there can’t be one television standard. And as a result it really takes a lot of people to come together to make any one of these things approachable. And, again, we’re very happy to have seen the really rapid growth of the number of companies that have supported the ATVEF initiative, both inside and outside the United States, and expect to take that forward, as well.

So my view, and I’m quite enthused about this now, and we try to reflect this in our booth here at the show, is in fact that we’re going to see a tremendous array of smart connected devices, the personal computer among them. You know, it’s our hope that the personal computer, perhaps being the most capable of those, will add value when it’s present in this environment. But, clearly, we have to create an environment where all of these heterogeneous devices are able to discover each other and their attendant services, and operate without anybody being master to all of them. And so it’s really our view that what we’re about to do is begin the next phase of the evolution of the Internet. And despite the tremendous impact it’s had on society and the economy already in a relatively short number of years, I think that will ultimately be dwarfed by what will happen when, in fact, this form of fully connected and intelligent appliances are part of all aspects of our daily lives. And so my view is that the personal computer, as it evolves, and the consumer electronics world, as it undertakes this incredible revolution together will, in fact, define what becomes the next generation Internet.

Thank you very much, I’d be happy to answer some questions.

(Applause.)