Click Here to Install Silverlight*
United StatesChange|All Microsoft Sites
Microsoft
PressPass - Information for Journalists 

Remarks by Bill Gates
WinHEC 2002
Seattle, Washington
April 18, 2002

(Applause.)

BILL GATES: Well, good morning. I enjoy the yearly opportunity to talk about what’s happened in PC hardware and software and the opportunities the industry has there.

Over this next decade, which we refer to as the digital decade, the PC will be used for a wide range of things that it’s never been used for in the past. Everyone will expect their music, their photos to be there on the PC and that will be the place you’ll organize them, and the PC will make sure that those things show up on any device you want to use them on.

The PC will be where you’ll indicate the information you’re interested in, and so on your smaller screen devices automatically whatever customization you do using that large-screen interface will affect those devices as well.

In the office environment, one of the ways that Microsoft likes to think about software and the value it provides is for every hour of the day that you’re at work or at home, but particularly at work, we’d like to continue making your job simpler. And we look at things like phone calls right now, where the PC is not part of it, and see we can do something revolutionary, having the screen and voice working together there.

We see when you go off to meetings, where you’re generally not taking your computer at all, that with this tablet form factor we can make a big breakthrough there.

We also see that by getting the tablet-size device in your hands, things like really long documents can move onto the PC.

And so there’s an opportunity as we embrace the new data types, as we take advantage of these advances, to have the amount of value you get out of the PC, both at work and at home, be dramatically higher than ever before.

And so we can think of the changes that we’re all involved in here as being both evolutionary and revolutionary. Certainly it’s evolutionary in terms of the basic architecture, in terms of the basic structure of the industry where we have people building specialized components, other people assembling the motherboards, companies specializing in the peripherals, Microsoft building the platforms; so many different areas of specialization all coming together to create the PC ecosystem.

But it’s revolutionary in terms of the breadth of things that you’ll do with it, the way it will connect up to all these other devices. In fact, if you think of communications in the future you’ll be in control, you’ll be able to go to your PC and have a screen that indicates who should be able to interrupt you, whether it’s over your phone or a PDA type device or what e-mail is important to you. You’ll have that central point of control working against all the different devices.

There’s a lot of key software technology around XML, and these XML protocols that fit into this, that make all these pieces go together in a seamless fashion. And we’ve been very excited to see the progress that’s taking place in those standards. You know, it goes back about five years ago when XML was first initiated, but it was only a few years ago that the industry momentum really started to build. We bet our company’s future on this with our .NET strategy that embraces XML, but this year we saw milestones like the Web Services Interop Group, which is IBM, Microsoft and many of the industry leaders coming together to say at the protocol level we’ll have standards that work across all the different devices and platforms in the right way.

So the full screen device will be connecting up to everything, but it will play the central role. People talk about the smaller screen devices and their proliferation and we see that as an important part of this future, but when you think about creating documents or doing rich communication that’s going to be the full screen device. In fact, the mobility of the full screen device both in the portable form factor and tablet form factor will be as great as any other device in terms of connecting up to both Wi-Fi type networks and the wider area 2.5G and 3G generation.

So this whole group of companies, well represented here, that drives forward the PC ecosystem have opportunities in terms of how we specialize things for that office worker, how we do new device form factors in the home and moving up to the server level there’s an important opportunity there.

On the desktop you can say that the PC has a very high market share, that is UNIX workstations versus PCs; even several years ago there was a dramatic shift in favor of the PC as it took leadership in both the price performance and absolute performance, and that’s led to all the new applications being done there and a very strong thing.

That same thing has taken a bit longer to happen at the server level, but with the performance increases that we’re seeing, with the work that Microsoft and others are doing, this idea that it will be common sense from a price performance and an absolute performance to use the architectures of the PC for those server computers over the next two or three years. I think there are things that we as a group can do to accelerate that being common sense. You know, clearly a case where the progress of the last couple of years has been very dramatic, if you look at things like the TPC-C benchmark now seeing the PC server systems dominating the absolute performance there; likewise for the key applications, things like SAP.

But there is still some work to be done and perception lags reality and so we need to get showcase systems out there and make sure that people understand that.

There are a lot of challenges. It’s not to be underestimated. People’s expectations as to how reliable these systems will be to use, how they’ll be able to get at these new feature capabilities, how they’ll be immune from some of the security problems like the virus problems that were particularly visible over the last year, people have high expectations that will come together to solve those problems.

And one of the ways we looked at this is actually taking the connectivity of the Internet and the intelligence of the software and building a continuous improvement loop where we could see what’s really going on with these systems and help users. I want to talk about the progress we’ve made about that because it’s very exciting now to be getting information about exactly what’s going on with these systems.

The hardware investments are giving us lots of opportunities to tackle new scenarios. You are all probably familiar with the numbers where the CPU performance keeps improving at an incredible rate, Intel leading the way there and we’ll hear from Paul Otellini right after my speech about how he sees that roadmap.

Communications bandwidth, fiber bandwidth is going very well; it’s amazing what can be done. Actually, supply is so great there that it’s a tough business to be in building those long-haul networks.

Connecting up to the consumer’s home, getting broadband there is getting a little tougher and we’re seeing quite a divergence country to country in how particularly that’s being adopted. If there’s one thing in this great, great new hardware that we’re not getting fast enough I’d say it’s that broadband capability, but even that over the next five years I expect to be tackled.

Disk capacity is going up at the same exponential rate, which means that music, videos, you’ve got the storage capacity not only for those but also to do some very clever things with software, making it so you can have checkpoints so people can set their system back to a previous state and so whatever issues they’ve had they immediately go back to what they had before.

One thing that’s not improving, and this is an important dichotomy, even as performance improves latency of networks and of disks and all the various elements in the system, those latencies are very difficult to improve. You know, speed of light, seek times, all those things mean that in order to have rapid performance we need to continue to use the intelligence of the devices for the user. That’s where you avoid the latency.

And so being a lot smarter about bringing things across the network and caching those, using this large storage that we have and all this intelligence, that becomes increasingly important because otherwise people will say my experience isn’t improving as much as all of these raw performance numbers suggest, and the fact is that they’ll be right. Unless we conquer latency with clever local intelligence we won’t be able to pass through those benefits.

At the client side of the PC there are a number of key changes. I already mentioned the tablet. That to me is probably the most important thing to come along for the last four or five years on the client. Reading, note taking; that simplicity is very important. We’ve reached a threshold where the size and weight of that device are very reasonable, but I’m sure that there will be constant improvements there. The first generation devices look very good. They’ll be out from five different OEMs late this year, but I’m sure there will be second generation devices where you get the weight, the size down, you customize the LCDs even more for the kind of reading experience, so a very rich vein of improvements that will come along with that.

You’ll get a sense from all the sessions of WinHEC this year that we’re very keen on audio and the PC, getting it so whether you’re doing a real time meeting or just doing voice recording or voice commands that getting the microphone capability to be a standard element of the PC is a very key goal for us. And so it’s great to see the kind of devices connecting up to USB and Bluetooth that really bring that along.

Another thing you’ll start to see from us is this idea of larger screen area. You know, we’ve looked at knowledge workers and said, boy, the price of LCDs are going go come down enough that they can get much bigger displays than they’ve ever had with LCDs.

And there are some big implications there in terms of the windows itself and the application; how do we manage that screen real estate, how do we make it easy for you to have the information you care about right there whenever you want it. In the years to come we think that the knowledge workers at large will expect to have those much larger display surfaces and so the software will exploit that.

Bluetooth, as I said, is something quite key to us. We think the idea of having all the different peripherals able to connect up that way is a very nice advance. In a sense you can think of it almost as an extension of USB, where everything can connect up that way.

Microsoft has been working with the industry on Bluetooth. We’ll have our development kit out next month. We’ll actually do an update to Windows XP to have support of that later this year. Our Windows CE product already has that.

We’re showing our commitment to Bluetooth with our own hardware group that’s actually taking the peripherals they do, keyboard and mouse, of course, and doing those through Bluetooth and that involves having a Bluetooth transceiver that’s just a very standard full Bluetooth standard.

This coexists with 802.11. That’s come up at the conference. I think that's very key because these are both critical technologies and we can’t have a situation where it’s a choice between one or the other. They play different purposely complementary roles, where Bluetooth is for connecting up the peripherals over a shorter distance and of course 802.11 is essentially your Ethernet anywhere you go.

We’re working with a lot of partners on 802.11, getting lower power versions of that, making sure that the security installation is very straightforward there, making sure that if you take a device from your corporation and you’re visiting another company that it’s easy to get connected up and gateway through this type approach back through their Bluetooth without a lot of administrative overhead; likewise, working with people who are doing hotspot Bluetooth to make sure that our software makes that kind of connection easy as well. We want to see wireless networking very, very pervasive, and so wherever you spend lots of time with your PC 802.11 will be there.

Well, to show you some of how concretely we see these productivity advances let me ask Mike Van Flandern to come on up and step through some of the things we’re doing to see with this evolving PC client.

MIKE VAN FLANDERN: Thanks, Bill.

All right, our vision for the future is that the PC and the telephone will be tightly integrated and that’s going to enable people to be more productive. One example of that is users will be able to reach people by their name rather than using phone numbers.

Now, I need to call Dave Williams and traditionally what I would have done is look his name up in my rolodex and then enter it in my dial pad. I could, if he were online, send him an instant message using my buddy list.

Instead, what I’m going to do here is use my buddy list in my Windows Messenger to initiate a PC to cell phone call.

So I right-click on Dave’s name and I say Make a Phone Call and Dave’s phone number is right there for me. This is a phone number that Dave’s maintained with his profile. So I could click on that PC but because handsets are a familiar device for using voice telephony or voice communication I’m going to use a USB handset provided by Clarisys for this demo. And I’m just going to press send here on the handset; I could do it on the PC as well, and initiate a call.

And the important thing when integrating a PC and a telephone is that you integrate the signal.

DAVID WILLIAMS: David Williams?

MIKE VAN FLANDERN: Hey, Dave, it’s Mike. I just called to let you know you’re up next.

DAVID WILLIAMS: Yeah, I’ve got everything ready for the demo.

MIKE VAN FLANDERN: All right, thanks. See you in a few.

DAVID WILLIAMS: See you in a minute.

MIKE VAN FLANDERN: Now, the important thing again is that the signaling be integrated and it’s possible for the phone and the PC to coexist. In this particular demo I used a USB phone for the audio input, the Windows Messenger client as my session initiation protocol client or SIP client and Windows Messenger being included in XP there are over 17 million users for the SIP clients. And then lastly we have a phone gateway on the back end. Windows has relationships with several telcos, Windows Messenger has a relationship with several telcos, including Telus, Telstra and Voicestream, with more relationships on the way.

In this particular demo I used the PC as the end point for the voice. If you do that we need to address things like quality of the audio input on the PC, reliability and availability of the PC, which means possibly looking at watchdogs for the client, addressing boot time and recovery or wake on ring.

Now I’m going to address some inbound call scenarios. For this I subscribe to a notification service so that when an inbound call comes I receive an instant message notification on my PC.

Here I can see Dave Williams is trying to reach me, so I’m going to click on his name and it brings up the Outlook call journal or my Outlook log, and this has a list or history of all of my interactions with Dave, and it automatically adds the current call to the log and I can even choose to record it, if I want to.

Now, the key point here is I could have just as easily associated that incoming call notification with other functionalities such as routing or filtering, say I’d like to have all my mother-in-law’s calls go directly to voice mail, for example. (Laughter.)

Now, let me summarize the power of integrating the PC with the phone. Today we believe they’re going to be integrated. Today we’re doing this demonstration by having the phone be the device on the PC. In the next couple of years your desktop phone is going to be tightly integrated with the PC. If you use your work PC as the end point for voice, we need to improve the reliability and the availability of the PC, as well as the audio input system.

Well, with that, Dave should be ready so I’m going to turn things over to him.

DAVID WILLIAMS: Thanks, Mike. (Applause.)

What’s the interesting thing is a real call from the PC out over the Internet live to a phone back stage; now they can track you down anywhere.

So what I’d like to talk about in productivity advances is actually some research that Microsoft or the Microsoft Research Group has done around multi-monitor and productivity gains we can get around that. It’s also an attempt on our part to see how much screen real estate we can actually display at WinHEC.

Now, maybe by a round of applause how many people have what you could consider a traditional multi-monitor on your desk, a couple monitors sitting around? (Applause.) Okay. How many of you would like to have a couple monitors on your desk to improve productivity? (Applause.) So I’ll give you a little hint here; as you go back to your boss and talk about productivity gains, our research shows that productivity gains for standard tasks, window management, office, switching between documents, you can get productivity gains with multi-monitor anywhere from 20 to 50 percent.

And what users find is they know how to manage their windows better, they’re more aware of which is the live task and when they have background tasks that they’re still up and available, they remember they’re there, and we’ve found in our research that people find multitasking to become more natural.

But there’s a lot more that we need to do and the opportunities in the hardware space are tremendous, not only around form factors for how do you build the monitors, right, but the size of the screen real estate that you use, the type of features and functionality you put in the monitor, multi-monitor head cards for the IHVs and ISVs especially and there’s a lot of work that Microsoft is doing in the operating system to really understand the deep, deep details we need to make this very natural and very immersive for the users.

And what we’ve got here is where we’ve just taken this standard multi-monitor scenario and we’ve stitched two screens physically together and put a very narrow band between the two. And our research has found that this improves the user’s perception and ability to multitask as opposed to having two physically separate monitors. You still have a bar between the two and it’s the type of situation where you might be having a focus over here on the left and your toolbars for PowerPoint are up for formatting. Now, those are the same toolbars for, say, Excel and when you go to the right screen and click on Excel your toolbars show up appropriately in the right place and they’re the same toolbars you’re using over on the other screen. And this really helps users understand the transition between applications and things like that.

Now, our research goes on to tell us that we can even improve on this. Users, human beings in general are very aware of what’s going on in their peripheral vision. It’s a talent you pick up from driving. If you have small children, like I do, your peripheral vision tends to get pretty good, keeping track of those guys. And what you want to do is have a screen where you have a focal point, which is your primary screen, and then you have the screens around for the periphery. And what we have built here is what’s called the "inverted T." That’s a very clever name and you can see where here you would have your primary work surface that you’re working on and of course above you have your toolbars, your notifications and of course you listen to music while you’re playing so one of your media players is playing up on top. And I have my e- mail on the left, because right now I’m not doing e-mail as my primary work, and I’m doing a lot of research on the right for the presentation I’m doing.

This becomes very natural and it begins to become an immersive experience for the user in what they’re doing, the task they’re trying to accomplish. If you think about the multiple tasks that you’re juggling, our researchers show just tremendous productivity gains in the ability to balance and juggle and drive these forward.

So, for example, if you go back to productivity gains, the quick ROI on just what a two- monitor is from 20 to 50 percent, if your company says all right, let’s just be real conservative and say we get $25,000 a year value on your productivity, a 20 percent productivity gain, a second monitor would pay for itself in a matter of weeks.

And when you really start getting into these more advanced scenarios where obviously day traders or CAD/CAM type operators where you have a very intense usage but you need to have a lot of toolbars, even these expensive inverted Ts start to pay for themselves in a matter of months.

So you can all go back and tell your boss, hey we did a quick ROI on me and you can buy a second monitor for me and my productivity gain will pay for itself in a matter of weeks.

Now finally what we looked at is how do you push multi-monitor into a large single display surface. Now, this is a research product, a research station and it’s built to simply show the concepts of what you’d have if you had one continuous desktop surface and there’s no bars or anything and it’s on a curved screen. And what our research has found is this actually provides the maximum productivity gain. It’s more immersive. It gives the users an area where they can have a focal point but they can also change their focal point as they see fit.

So, for example, if you’re running Excel and normally you’d have Excel here and maybe you want it over on the left or in the middle or something like that, but if you’ve seen like in your accounting department, your payroll department people are real power users and they want to get a view of the data all at once and maybe have a graph, it really allows you to become immersed in your data and human beings are able to absorb and look and view at this type of data in this fashion.

So typical today this is what you look at if you’re doing a word processing and you want to play out in a document and you’re sitting there typing, but really you want to get a view for what the document looks like as a whole: How does it flow, how does it look, how do the charts interact, how does the document flow. And, of course, you can imagine we could increase the height of this a little bit if you want a little wider aspect ratio, and truly immerse yourself in the product and document you’re working on.

And, of course, the most natural one that we want to show is if you’re a graphic artist or a photograph, so this is one of those panoramic views but now you can actually sit down and take a look at the whole view and this truly becomes immersive.

Now, this technology we’re showing is nothing that we’re saying this is the technology that needs to be applied, but really our vision is in the next few years with organic LEDs and other alternative technologies for displays a very thin, flat, curved screen will be very affordable and in the normal productivity an office user, even though like we talked about, these are going to be more expensive, so productivity gain for this type will be very cost effective.

And there’s a huge opportunity here for hardware vendors to innovate and innovation in the form factor and how people immerse themselves in technology and we need a lot of help in the software sphere and of course the operating system. The work we’re learning from this we’re going to apply directly into our operating system to bring this to market.

So with that I’d like to thank you and bring Bill back up. Thank you.

(Applause.)

BILL GATES: Okay, now let’s take a look at the server level and talk about the advances there. As I say, there’s a key threshold that we’re in the process of crossing, which is getting people to understand that the kind of performance and reliability they expect for the most demanding applications can be done with PC server architectures. The performance improvements are very key to that, software things, particularly the management level software that we’re putting on top, some of the neat new storage advances where you can control your volume can actually do your backup disk to disk, not having to worry so much about tape and you can aggregate all your different storage services together; so a lot of innovation that many of you are doing in these servers that can drive things forward.

One of the biggest investments the industry is making as a whole is this move up to 64-bit, moving up to in particular the IA64 architecture that Intel has put forward. For them, of course, that’s a billion-dollar investment; for us it’s been the compilers, the optimization, a lot of work there to start to build this as the new foundation for very, very high-performance servers.

We’ve made very good progress with that, taking the early chips and getting those out to software developers and now lots of tuning going on, lots of application development going on and we’re starting to see some very powerful server designs build around it, and so that will be part of this transition to getting people to see PC architecture as being mainstream for this.

To give you an example of how this works with the latest McKinley I’d like to ask Valerie See to come up and show us how this looks. Welcome, Valerie.

VALERIE SEE: Hi there. How are you this morning?

What I’d like to show you is a really cool end-to-end use of 64-bit computing power and scalability. We thought it would be nice to do an original piece of art for our WinHEC attendees, so we found one of those rare clear blue days that we have here and found a nice picture of the downtown Seattle skyline. And what we wanted to do was use a technique developed by Rod Silvers, who’s famous for his photo mosaic rendering technologies, to turn this into a piece of art.

So let’s show you what we wound up with. When you take a look at this closely what’s happened is an image analysis was performed on this and you can see that it’s composed of thousands and thousands of tiny little pictures from the Corbis stock library that they were kind enough to lend us for this demonstration.

We’re displaying this using an HP workstation, a McKinley powered workstation, with six display surfaces. And just to prove the point that we do, in fact, have six of them let me identify them for you. You’ll notice we’ve got it turned sideways since we’re presenting a piece of portrait artwork and we put the thing together the wrong way the first time.

But what did we do to make this happen? If we take a look, we do have a server architecture slide, so we’ll take a look at how we got this built and then I’ll show you some of the programs behind it.

Basically what we’ve got behind me here on the rack we’ve got this HP McKinley workstation, which is our display technology, we’ve got a two-way HP McKinley server and then a four-way McKinley server down at the bottom. The two-way server is what we’re doing our rendering on, and, in fact, if I go back to the screen where we showed our original photograph I’ll take a look at the rendering application. This is Rod Silver’s application and we’ll go ahead and do a render in real time while we’re talking about this.

Rod’s company, Runaway Technologies, ported this to the 64-bit programming platform to get that additional scalability capability. It allows us to pull in the original piece of artwork, say a portrait of President Kennedy, we have a tile database full of lots of pictures of President Kennedy, we could set some parameters as to how wide we want to make it, how many repetitions as it does the image analysis, how many processors we want to use -- we decided to use them all so it would go really quickly -- and as we’ve been sitting here talking it’s completed and you can see the resulting mosaic.

And we did a pretty coarse one here because we wanted it to go quickly so you could see the technology, but you could see how fast that went and the power of that 64-bit Itanium family computing platform, and that’s running on a preliminary build of Windows .NET Server for the 64-bit.

Now, on the back end I have previously mentioned we’ve got some 80,000 photos from Corbis; it’s a lot of data. So we’re using 64-bit Microsoft SQL Server 2000 to manage that data. But for the average professional that’s not necessarily the most friendly way for them to deal with their data. So MediaBin Technologies has an engine and it turns out this is a 32-bit engine that actually front-ends the 64-bit SQL Server database and it is running on that four-way HP McKinley server in the Windows on Windows 64 technology, so it’s running in emulation mode. We put it on and it just works, so we’ve got great 32-bit and 64-bit interoperability on that 64- bit platform.

So if we were to go and browse our image data what’s happening is that engine is doing queries on that 64-bit SQL database for us and if we were to look at some active lifestyles in the Northwest, you can look at the sailboats and stuff like that, you can see how friendly this is for an imaging professional.

So the power of this programming environment and the scalability of this hardware gives us a rendering environment for content creation that just boggles the mind, a power that we never had in this way before.

So that’s what we have, end-to-end 64-bit computing, an original piece of artwork. Thank you, Bill.

BILL GATES: Thanks, Valerie. It looks great. Super.

VALERIE SEE: Thank you, Bill.

(Applause.)

BILL GATES: So there’s a lot of progress there around McKinley and one of the key points there was that only the performance-sensitive applications need to be ported to the 64- bit and the other applications can run with the compatibility capabilities that you saw there for that front end of the photo application. The operating system, the SQL Server are all there and running now with great performance on the latest McKinley hardware.

Well, now I’m switching venues down to the consumer side to talk about how all this is going to look in the home. PCs in the home traditionally have just been sitting in the den, you know, you’ve got the one screen, you sit down to use that screen and it’s used for pretty much the same things you would do at work.

But we want to extend that. We want to change that. We want to connect multiple PCs together, connect them to all the different things in the home and make that power of the PC available wherever you go. So the PC is now going to be pervasive as well as embracing these new media experiences, and we have a number of initiatives that are rolling out this year and next to get us to this full vision of what we sometimes think of as the electronic home.

One of the things we need to make sure we think about in this is how do we distribute the information around the home, both the PC screen display, how do we get that to be available on any display surface, including TVs in the home, including the idea that you want your audio and your video there and, of course, audio you want not where there are screens, wherever there’s any type of speaker.

There are two key ways that the industry is coming together to do this. One is the wired approach, taking an IP Ethernet type approach and being able to send synchronous audio/video over that network. There have been some breakthroughs that deal with the kind of skewing issues you would have dealt with in the past, the noise sensitivity, interference and things of that nature, and so we see that leveraging good ol’ Ethernet in a very rich way as a very key infrastructure for this.

The other of equal importance is the idea of using the 802.11, advancing 802.11 in terms of the quality of service that’s done there and eventually moving up to the A format, which provides for quite a bit of additional bandwidth, which for multiple high quality video transmissions turns out to be very important.

And so these pieces really have to come together, making it easy to install so that a consumer can think, "Boy, if I bring a new display surface into my home it’s automatically discovered, any of the PC things that I can do can be brought onto that, and it just is part of the overall system without me having to think like an IT administrator to get those things pulled together."

To give you a sense of where we’re making progress on this and what some of the new scenarios are, let me ask Steven Guggenheimer to come on out and how us the latest in the digital decade consumer experience.

STEVEN GUGGENHEIMER: Thank you. Good morning, everyone.

We’ve shown a little bit on business productivity, a little bit on enterprise. I want to take just a few minutes and talk about directions for the home, directions for consumers.

I think there are two key messages for everyone this year coming out of WinHEC. One is building on top of Windows XP and the work we’ve done there with Freestyle and Mira and other advancements currently that we’re making to extend access to the content throughout the home. We have an opportunity in terms of the hardware design to do a better job of making the PC, and really the peripherals, more of an integral part of the home. Instead of a gray box in the corner there’s a lot we can do with industrial design.

I think you’ve seen some prototypes of Freestyle over the course of the week. You’ve heard about Mira, which is the ability to take a flat-screen monitor and then make it accessible and access your PC from anywhere in the home with it. I wanted to show you one other thing, an early concept from Philips in terms of what a Mira monitor and in this case keyboard combination can look like, okay, so it’s a very nice stylistic approach to a flat-screen monitor plus keyboard, which you can put back easily when you’re not using it.

And when you think about it, when I want to have now a pad that I carry around the house to access my PC for browsing the Web or listening to music, it’s this type of form factor and design that fits very nicely in the home with some of the other devices and traditional consumer electronics that we have in the home today, and even TVs. So I think there’s a lot of opportunity in terms of differentiation in the Freestyle technologies with Mira just with industrial design.

Now, the second area that I want to focus on is really the notion of experiences. One of the things again that we can do building onto the new capabilities is really begin to think about not the PC as this gray box in the corner, but how the PC, the TV, the phone and the stereo begin to relate together.

Some OEMs have already started down this path; the combination of a camera plus the PC plus a digital phone frame or a printer create an experience. Some people are doing it around media.

What I want to do is give you a scenario, moving out into the future, of where I think Mira and Freestyle can take us.

So in the future I think it’s going to make a lot of sense to not just invest in Mira’s technologies, CE .NET and 802.11 into a computer monitor but in the flat screen TVs that are shipping. It’s a very logical extension. That way, if I bought a new TV for the home, a flat screen TV or any other TV for that matter but in particular the flat screen, I would automatically have access to my Windows XP room in another room, so there’s no real additional cost to the user, go ahead and purchase a TV and now I can immediately access the content and media from my PC in the living room, even if the PC is in another room; no wires, very simple.

The second thing is if that PC is running, the Freestyle technologies or future versions of the Freestyle technology then not only can I access the PC but it’s easy and logical to get at the content. So, for example, if I want to look at my pictures on the TV in the living room and my PC is in another room I sit back with one of the new remote controls they talked about earlier this week and now I have access to my photos.

And if you think about it, when I was a kid and my dad wanted to show his home slides he’d get out the carousel, put the slides the carousel, hang up the screen and that’s how you look at home photos. In the future you’ll have your television, your PC will be in another room and you’ll look at your photos or your home movies, so it’s a very logical evolution of the marriage of sort of PC technology and other screen form factors within the home, in particular the TV.

The other thing that I think will happen over time is the ability not just to have access to the content on multiple screens in the home but ultimately to coordinate amongst them. So here you’ll notice if an instant message pops up I could take it on the TV but I might not want to interrupt all of you watching TV, so if I had one of my Mira devices around, another screen, access the same PC you’ll notice I get a pop-up here as well. So instead of interrupting you I’ll go ahead and take the message here. It’s asking for a video chat. I’ll go ahead and do that. Hello?

WOMAN: Hi, Steve.

STEVEN GUGGENHEIMER: Hello. How are you? It’s my babysitter.

WOMAN: Hey, am I interrupting anything?

STEVEN GUGGENHEIMER: Just briefly, go ahead, real quick.

WOMAN: Okay, well I have to show you what the kids have been up to.

STEVEN GUGGENHEIMER: Okay, go ahead.

WOMAN: Okay, let me send it.

STEVEN GUGGENHEIMER: Sure. So she’s going to send over a file. Okay, I’ll accept the file.

WOMAN: Okay, well we can’t wait for you to get home. We’ll talk to you soon.

STEVEN GUGGENHEIMER: Okay, great thanks.

Now, the thing is when I downloaded that file I didn’t download it to this Mira screen and I didn’t download it to the TV, I downloaded it again to the PC in the other room. But when I want to play this I can actually choose which screen I send it to, so I’m going to send it to the living room TV and you’ll notice it pauses the TV and it will go ahead and it will play the video.

(Children from video.)

STEVEN GUGGENHEIMER: Great. The babysitter has got too much time on her hands with the kids, but at any rate I think you get a really good notion of in the future it’s not just a PC, it’s not just a TV, it’s sort of how things begin to interact.

And some people say, well, if you can get your media on the TV, what’s the reason for the PC? Well, it’s nice to be able to display content but I still need to organize that content, manage that content, edit that content and changing a photo from P6397 to Jack and Hannah’s birthday is not something I want to attempt with a remote control. So the PC still plays a critical role in managing all this content but making it accessible is an important part of the experience.

Let’s take one more example. Let’s go to music. Here I have a PC running the new Windows Media technology, the Corona version, and on it what I’ve done is I’ve burned my own CD and using the new format I can actually burn a CD with about 365 songs or 22 hours worth of music. Now, that’s nice but if the only place I could play that CD was the PC it’s not that interesting an experience.

However, by working with the manufacturers of consumer electronics devices, and again working on the digital ecosystem I can take the same CD, plug it into my stereo here and they’ll bring it up. It will take a second; with 365 songs there’s a lot of macro data to read. But once it comes up I can go ahead and go into the tracks here and I’ll go to the top menu and you’ll notice there are the first ten songs, 20 songs, 30 songs, 40 songs. It takes a while. Most CDs have about 19 songs so getting to 365 -- I think I’ll just enter that. I’ll go to the 365th song and play it. It will take a second. (Music.)

Now, I could take that CD out, take it and put it in my car stereo. Here we have a Pioneer car stereo. We have a little personal portable player.

So again it’s not the PC as the standalone island; it’s how the PC in this case through music relates back to a CD/DVD player, a car stereo, a personal portable player. So it’s very important to start thinking about end-to-end experiences.

I’ll take this one more step. Obviously from music we go to video. You saw a little bit of an example earlier of sort of the Windows Media Pro format for audio and video. I’ll give you one more sample again where I think about things we want to do, the ability to stream high- definition quality video, six channel audio, have it accessible and then play it back on the screens in the home. We’re beginning to enable that through these new formats.

I’m going to bring up a short clip of the movie Dinosaur. It’s been encoded in the Windows Media Pro format, and I’ll just let it sit back and listen and watch a second.

(Video segment.)

So high-end video, clearly again a key point.

In summary, as we make the PC and the content more accessible throughout the home, as media becomes a more integral part, we really have an opportunity to begin to make PCs and the peripherals feel more like a natural part of the home, have a way to differentiate through the industrial design, really make people begin to identify with it as much more of an integral part of our daily lives.

Second, we really need to begin to think as an industry about the end-to-end experiences. The day of the PC, the TV, the phone, as we saw earlier, the stereo as standalone, independent devices that never work together is coming to an end, so we need to make sure that experience of how these devices work together is positive for the consumer, that they work well together and that we create an infrastructure to help people be more productive at home to be able to communicate in richer ways and really to enjoy entertainment, to have a more full entertainment experience.

With that, I want to invite Bill back up and thank you for your time.

(Applause.)

BILL GATES: Well, it’s great to see the Mira devices and how those are evolving. Those, of course, are coming out this year as the first step towards that overall integration.

We’re announcing today several partners in our Mira program. I’ve got them on the slide you’re looking at, the existing partners that were already announced and then today we’re adding Fujitsu, NEC and others as new partners that will have Mira devices.

That Mira idea is very interesting because it gets to the PC wherever you go, and, of course, that Mira capability we see that eventually being built into both the TV set and other types of screens you’re going to have around the home, so anywhere you go you’re able to connect up and have a home page that’s customized for that particular screen, so it’s not just being able to project but also customize for the size of that screen or the kind of activities that make sense in that location.

Well, now let me address a very important initiative at Microsoft called Trustworthy Computing. Earlier this year I sent out a memo to everybody in the company about how we were thinking about this as a priority and really saying that a high percentage of our development resources would be focused on this broad initiative.

Actually, many of the product groups took several months to not develop any new code, but simply go back and really look at the processes for security and reliability, making sure the right kind of thinking, design and education was taking place there.

And I think this initiative goes beyond Microsoft in the sense that the idea of doing e- business in this rich way with Web services, the idea of saying that the way you share your family photos, the way you store those important memories are going to be using these digital approaches, people will be very reluctant to do that if the reliability and trust of these systems isn’t extremely high. And it means going beyond anything that’s existed before. When you have these systems out there on the Internet people who make security attacks are always probing and they’re going to find any weaknesses there. That wasn’t true when mainframes were just held inside the glass house and didn’t have these arbitrary connections to the outside world.

And so we’re having to pioneer the idea of how this security works, both making sure there’s not the number of flaws that there have been, making sure that whatever issues there are, that the fixes for those things are propagated very rapidly, making sure that whenever there is a problem that recovery from that problem is very straightforward; a lot of elements that go into this, but some really brilliant ideas that are going to make it achievable.

Things like even software verification, this has been the Holy Grail of computer science for many decades but now in some very key areas, for example, driver verification we’re building tools that can do actual proof about the software and how it works in order to guarantee the reliability.

If you look at the PC industry one of our great strengths is our diversity. If you just looked at the number of companies participating here at WinHEC, looked at the wide range of things being shown at this show it reminds you of what a global and broad industry this is, with small companies participating all the way up to the larger companies like Microsoft or Intel.

And it’s that work coming together that creates so many choices for the user, great innovation and attractive pricing, all the things we love about the PC.

But one thing that that ecosystem approach makes very difficult is -- how do you know that all the pieces are coming together? When you have a third party driver and an application on a system, what if on the boundaries there things aren’t working for that user?

And we all know anecdotally that there are many times that the pieces don’t come together, that the systems over a period of weeks or months doesn’t have the guaranteed stability that we’d like to have.

Well, it’s not so much those individual pieces as those boundaries. So with Windows XP we’ve put in the feature, the reporting feature that whenever an application wouldn’t work or whenever the system would go down that there was an option of sending that information back to Microsoft. And that was very eye-opening, a very important step to get on a statistical basis of what’s going on.

What you see is that the number of these reports tend to cluster around a really pretty finite set of events, several hundred, and that by focusing on those we’re able to take the overall experience and continually improve it and the combination of this monitoring, along with the Windows Update, means that we’ve now got a complete loop to make sure that we see what problems people are running into and to fix those things for other people before they run into those things.

But this has got to be something that we get everybody participating in in order to really have the best of both worlds, the kind of end-to-end trustworthy computing that I talked about and still preserve the kind of flexibility and openness that our industry has been known for.

So with that I’d like to ask Eric LeVine, who manages this effort for us, to come up and show us the latest on how we’re using the information. Welcome, Eric.

ERIC LEVINE: Bill, come on back and let’s take a look at a couple of quick things.

So what I’m going to show you here are two things. We’re going to take a look at the experience an end user has to go ahead and report an error to us and then we’re going to sort of step back and take a look at what that data looks like to you as an ISV or an IHV actually going to go ahead and look at it.

Now, of course, why crashes? I mean, obviously this is one of the most painful and frustrating experiences for a user. So what we wanted to do was just take advantage of connectivity and make it very easy for a user to anonymously share feedback with us when something has failed.

So the first thing I want you to notice in this user interface that comes up in here is by default is Don’t Send. We really want to respect people’s right to privacy and whether they want to choose to share this information with us. If they want to go ahead and drill down, they can go ahead and do that. This dialogue goes more into again just different people feel differently about these kinds of things. Some folks want to share; some aren’t so sure, so we let them drill down and down and go deeper and deeper to see the actual dump and decide for yourself do I want to share this or not.

I’m a trusting user so I’m going to go ahead and just with one click say let’s send this error report up. In this case it actually only moved about 20k of data so it’s very quick and boom the error report is up there.

Now, one of the things you’re going to see us doing more and more over time is being really more aggressive about closing the loop with users and actually gaining more and more accountability. So by default, while that was an anonymous upload, actually in the case of a serious error, a blue screen, we actually today will go ahead and let the user continue and choose to particularly make themselves non-anonymous to get some accountability and find out as time goes on what’s going on with the issue. This is a little privacy disclaimer, again it’s just very important, we don’t want anyone to get confused and send up data they didn’t intend to send up.

I’m going to log in with Passport and just give a quick little descriptor for my own memory of what it was that I think the problem had to do with and go ahead and click Finish, and at that point I’m set up to be notified if Microsoft figures out what’s going on, and I can go ahead and drill in there.

In this case actually today the system is near real time. You’re going to actually see us making it so that really right when I enter that site we have a solution and we tell you. In this case it’s actually for real, it’s the real thing out in the cloud, analyze the dump, so I’m going to go ahead and drill in and it’s gone ahead and taken a look at this error and it is exactly what it thinks is at fault, who the vendor is, and in this case I can go ahead and click a link or ultimately just get the fix automatically from Windows Update.

So again that’s the end user experience. For us this is all about getting the solution to the end user to what traditionally has been a very frustrating problem.

Now what I want to go ahead and do is actually pop over to a fairly new portal that we’ve created for ISVs and IHVs to go ahead and get at this data. And so if we could bring up PowerPoint on the center screen let’s take a look at that curve. And again you actually have to log into this. There’s this little logo program site, although you don’t have to have a logo to get at this data, but you have to basically prove what you own and what’s yours because we don’t want people to be able to see data that doesn’t pertain to their components. The user’s anonymity and privacy is really tantamount throughout this.

So there’s a new link now on the logo site, the Windows error reporting, and I can go ahead and drill in here, and what this is showing me is literally for my particular set of components the failures that people have reported. They are broken down into two areas. The upper table is basically maps to blue screens or kernel views, bugs and drivers, and the below we have the same thing for an application crash, not one that took the system down but just an individual application.

And here’s the thing I want you to notice. As you look at this, we’ve basically been able to classify these unique issues and I can go ahead and just click a button and it’s going to go ahead and send to me a sampling of the dump file that account for that issue right to a prearranged FTP site, but if you look at these numbers this shows how many times each of these issues have been hit. And so at the top here you see a 1,061 dropping off to 151.

And if you look at that curve, that’s exactly what’s going on there, because really for you as the developer you want to spend your time on the issues that people are hitting most so you can get the biggest return and get the most stability for your user right away, so clearly you’re going to spend a lot more effort on this one versus this one. And in the past, you know, we sort of just had this random system where crashes might make their way back through us, although for support organizations this has been very frustrating, but now we’re just using connectivity to make it very easy for a user to just get the right data to us and let us take action, so really to ask all of you to start looking at this data, a one bug at a time kind of problem, but we want to start on the back end of the curve and work our way down and make a lot of headway very quickly.

So anyway, Bill, thank you very much.

BILL GATES: It looks good. Thanks, Eric. (Applause.)

So this is the error reporting and that portal site just getting off to a start, but I expect that next year at this conference we’ll be able to report the results, that is how many fixes were made by third party vendors going up into that portal and how we propagated those out through Windows Update. And I think we can really surprise people that taking the ecosystem, improving it very substantially and letting people have the best of both worlds, the flexibility they have today and yet this closed-loop analysis that Internet connectivity and the software you just saw makes available.

Well, the industry has come together around a lot of initiatives over the years. Many of those things are features, key features we just take for granted on the PC today. USB is probably the most important of those. That’s been a phenomenal success. It took many years, tough to bootstrap, but it’s great to see the scenarios that just wouldn’t be possible without it.

Power management, legacy reduction, getting these digital interfaces to the LCDs for higher quality; now Wi-Fi, 802.11, we’re just at the beginning of that one, so a lot of pieces coming together there but already many portables shipping with that built in. Music, imaging, the embedded operating system work that applies for specialized devices; all those things are fully underway and very important.

Today what are we looking at as the key priorities? Well, of course, the evolution of USB, getting it so things like video work there with capacity, the evolution of UP&P for not just PC to PC but other devices in home, like that screen discovery scenario that I mentioned where you just bring it in, it’s recognized, that’s done through those protocols, are key there.

Bluetooth: That’s I think going to be a very rapid one, a lot of key milestones here. New form factors where tablet is on the top of the list and then real time communication.

Overall it’s about making the PC an even more empowering tool, more hours of the day, less frustration, a lot more you can get out of it, which is what’s exciting about this industry.

So this decade is the digital decade. Across all these segments there is room for innovation: The home, the workplace, the server; every one of those things I’d say we haven’t even done half the work that we need to do to get the ideal device.

The focus on the innovation I think is very healthy right now. People are really thinking about things that are good, long-term contributions, a lot of cooperation with the industry understanding that this is what it takes to get growth going, some tough architectural challenges in the areas of the trustworthy computing but again a key element to bolt this together.

And so for all of us in this industry it takes a long-term view to do these tough things but the payoff both in a business sense and the impact we have is really quite phenomenal.

So I thank you for your efforts and look forward to the cooperation that will come out of these initiatives.

Thank you.

(Applause.)

 

© 2009 Microsoft Corporation. All rights reserved. Contact Us |Terms of Use |Trademarks |Privacy Statement