Remarks by Bill Gates
Sydney Conference Unit
Sydney, Australia, Sept. 11, 2000

PAUL HOUGHTON: My name is Paul Houghton. I’m the managing director for Microsoft Australia. On behalf of the Microsoft team, thank you for joining us this afternoon and we do appreciate you putting up with our security checkpoint out front. I hope you can understand that we have to take some necessary precautions.

It’s a very exciting time for us at Microsoft, a very exciting time for the industry. At Microsoft we are on the way to developing a generation of advanced software that really brings computing and communications together in a revolutionary way. We call this strategy Microsoft .NET.

At Microsoft we have a very strong history with developers and IT professionals. Through the Microsoft .NET strategy we’ll be bringing a set of tools and a software platform that will allow the developers and the IT pros, you folks who are represented here today, to transform the web and every other aspect of computing.

Today it’s my pleasure to be able to bring to you one of the founders of Microsoft Corporation, a true visionary, someone who is leading the Microsoft .NET strategy for us at Microsoft. So please join me in welcoming the Chairman and Chief Software Architect, Mr Bill Gates.

BILL GATES: Thank you. I’m excited to be back in Australia and not only will I have a chance to share with you some of the exciting things that are going on, but I’ll also get a bit of vacation while I’m here.

It’s a very interesting milestone for Microsoft. Just last week we celebrated our 25^th anniversary and 25 years seems like a very long time. I was 19 years old when Paul Allen and I started the company and we had this dream that was pretty crazy and of a computer on every desk and in every home.

In fact, that photo you see behind the text there, that’s the motley crew that comprised the first 12 Microsoft employees. Basically everybody wrote code. There was a pure group with engineers, we had no testers at the time, no marketers at the time and Microsoft Basic was our key product.

But the thing that really got us excited was the value of software, the miracle of the microprocessor and all the great hardware to come, would be brought to life by the software we would create.

The one debate I had with my cofounder was whether or not we needed to actually make the PCs ourselves. I said that we could restructure the industry so that the PCs would come from many manufacturers and yet our software would create compatibilities which would make them all alike That’s how we’d get millions of machines out there, because we knew it would take millions of machines for software companies to be able to write all the great applications to make the PC the tool we thought it could become.

So computing changed. It changed from being a tool of large companies, something where your bills were printed out by the big computer, the government had the big computer, to the cool tool that allows individuals to communicate in very rich ways.

We’ve had a lot of milestones along the way. The move to 16 bit computing in 1981, when we brought along MSDOS. The move to graphical computing that we bet our company on in the late eighties that finally came to success in the early 1990s. The bet upon high-end operating systems platform, Windows NT that is now getting out to the majority of business machines as Windows 2000.

Lots of key junctures there, but the key principle is that it’s about the PC being an incredible tool, about how software can bring it to life and that there should be a company whose total focus is building that platform. All of that has remained the same throughout our history.

We have now changed our slogan to talk about computing being available anywhere. This is because although the PC will remain the creativity device, the full screen device, people want to get at their information on all these new devices. Even though these other devices aren’t selling much today, particularly when compared with the PCs, we see that changing. This is simply because software will make it simple to use those devices and share between all the different things that are out there.

So how does this vision work, where does it lead to? Well, people’s whole lifestyles will be quite different. The way that you’ll organise a trip will be different. The way you’ll think about buying something will be different. All the media you work with will be digital. For example, if you wanted to organise your photos, that will be done digitally.

The same applies if you want to organise your music in a way that suits you. You could have it stored away, put it up on the shelf and you then have control over the experience, once this digital music gets to critical mass.

The music industry has missed the opportunity to be a pioneer here. They’ve created a little bit of a problem for themselves in that a lot of the digital music today is not paid for and so they’re trying to play catch-up.

Having seen that, the movie industry, on the other hand, is working very closely with us to make sure that from the beginning you can easily licence movies over the Internet. This establishes a habit of paying for intellectual property.

One of the things I think is very important is to be able to get people into a totally digital world. This is so users are not having to move back and forth between paper and the screen, or between the phone and the screen, that those two things totally get drawn in.

But in order to do that, we have to do something very dramatic in terms of the reading experience. If you go back more than 50 years ago to the original vision of what we now call the PC, the Memex, we read with it. It was a device that was comfortable to sit and read.

The PC screens today don’t achieve that. Sure, if somebody sends you a two or three screen piece of e- mail, you read it off the screen. For example the technology magazines such as PC Week or InfoWorld that I used to read on paper, I now read on the screen. However, when I want to read a fiction book or even if somebody sends me a ten page memo, that’s printed out.

We have to get the readability to be as good as paper. There are many elements that go into that; the design of the software, the screen, the form factor and yet we feel we’re very close to being able to achieve that. It will be a feature of every PC and there will be dedicated reading devices as well.

We also feel like we’ve got to get the PC into the collaboration loop when people are sitting down and meeting with each other. Sure, some people take a computer to a meeting today but most don’t. If you’re sitting there typing the keyboard is noisy, people are wondering, what you are doing, are you reading your mail, are you paying attention to what they’re saying? It’s not yet part of that standard way of getting together.

Even when you’re at a distance you don’t think about sharing an experience through the PC screen where you’re editing and talking together. Instead, messaging is catching on through products such as netmeeting. However, we haven’t moved that into the mainstream so we have to do that to make it a communications device.

Over the next five years between reading and real time communication, we can more than double the time people spend in front of the PC. This is an era that the use of the PC for creativity and communications will more than double even as we’re 25 years into this pursuit of the PC vision.

In the world of business, there is a high level of interest in what the Internet can do. People are asking, how can these digital approaches make them more efficient, make them more competitive? It’s at a real fever pitch. Companies’ stock prices are evaluated according to whether they have a great Internet strategy and there’s also some startups coming along that are challenging the existing businesses in areas like banking.

There is also the idea of matching buyers with sellers, the fundamental mechanism of capitalism. This can certainly be improved radically by the digital approaches. The Internet of tomorrow will solve that problem.

Today how would you distribute and advertise a product that around the world you might sell 1,000 units of it a year? How would you find buyers?

The answer is, you couldn’t. A business that has that kind of product simply doesn’t exist. But with the Internet, with the ability of those potential buyers to type in the right descriptive words and find you and compare your offerings to other people, with the new platform, you will be able to have that kind of company.

So the specialisation and richness of products and services will enable a dramatic change to the economy. As will delivery without distance as a barrier.

In the US, over the last five or six years, the employment figures, the growth figures, the lack of inflation have really confounded the economists. Literally the economy is doing things that according to the theoretical framework they’ve always had, it shouldn’t be able to do. They’ve had to admit there is a new external factor. The use of the PC technology that really changed during these last six or seven years, has changed things radically.

They were saying well, what is this computer thing really doing? Well now they’ve seen it and seen it in a much stronger way than they believed possible.

But it’s really scratching the surface. Collaboration, efficient buying and selling, elimination of paperwork, making jobs more effective, easily finding information inside your company, effective sales analysis, trying out a new idea and sharing it with people, knowing changes in the marketplace getting fast feedback from your customers -- this is only the beginning of what we can do here.

Customer service will be redefined. You can go out, check what’s going on and if you see something you disagree with on your screen, you can be talking to somebody who has your complete history and will help you out. So in this new world, it’s the best of the screen and the best of the phone.

For people who work in business, a lot of their time is spent inefficiently today - finding information, and sitting in meetings where they only are interested in part of what’s there. If you leave a meeting, you’re talking to somebody who wasn’t there, how do you share that information with them? It’s simple. The information should be a digital video on the Internet. You should be able to take that snippet, search the PowerPoint slides and mail that off to somebody to show them exactly what was taking place.

There should be no limitations on people working together and collaborating. And yet even the most advanced companies today are just starting to do digital meetings. Just starting to have the learning that needs to go on in their company. We are not even half way towards achieving the full potential of the software world.

It was the same when Microsoft was founded. We assumed that the microprocessors would get more powerful. We assumed that disc drives would get inexpensive.

Again, as we plot this .NET course for years head. We’re assuming some fantastic hardware breakthroughs that our friends, our partners in those businesses will push ahead and break through the barriers.

We’re assuming the speed of Internet connections will go up, so-called broadband. Certainly in business that’s happening today and for consumers it will start to happen but not nearly as fast.

That’s an important element for the video scenarios. We’ve been able to take music and deliver it over dial-up connections but video, even with the best compression techniques, the quality you’ll get across a dial-up line will be very limited and so a lot of the scenarios assume broadband.

We assume wireless will be pervasive and wireless really comes in two forms. There’s the wide area wireless where you can go anywhere in the world and the commercial wireless network will connect you up.

In that environment the speeds will go up but they won’t be super high and you’ll pay per minute charges. It’s very nice to have that for last minute communications but you want to be careful about overusing it because the packet charges will be there.

The other form of wireless is what you’ll have in your home and in your place of business, where simply by installing a very inexpensive base station, you’ll have data rates that are good enough to do very, very high quality video throughout the place of business and throughout your house.

Therefore music files off your PC, can be played against any speaker in the house, the photos or the videos you’ve got can show up on a TV screen or a little screen on the refrigerator, that will be just standard in the home. This will be done through breakthrough wireless networking that many companies are working on today.

We’ll have smart cards for identification so that the password doesn’t become the weak link in knowing who is using these systems and their security.

The PCs will have the microphone built in so that not just text instant messaging but voice instant messaging, voice annotation, voice recognition, will be a standard part of how you work with all the different documents and how you interact with all the different websites.

I talked about this new tablet form factor where you’ll be able to write on screen and take it with you. Users will be able to use it for things that today you use paper for and so you gain a lot of efficiency.

The screen phone is obviously complementary, so to really bring that market to life we’ve got to make it easier to share the information between that and your other devices.

Running all of this will be these servers. Historically the big machines, the big servers, were mainframes. More recently they were expensive Unix systems. And yet the same thing that happened on the desktop is now happening with those servers. The PC architecture has achieved a level of performance that means that not only is the price performance better but the absolute performance is now exceeding the most expensive systems.

We’ve done one thing though that’s really radical, that those more expensive systems didn’t do. We’ve created a level of software reliability so that if you have multiple systems, if any one of them has a problem, it’s completely transparent to the people connected to that logical server that one of those systems has gone down.

In fact, it’s not a new idea. It’s new in the mainstream but there was always this set of specialised systems from people like Tandem who used what we call this software skill approach, where any number of systems can be combined together to give you arbitrary scale. However if one or a multiple of those systems went down it was not noticeable by the people who were connected up.

In this vision, websites are performing at very high scale and great reliability with all the price performance that the competitive part of the industry, the PC part, is able to provide. Just like it did on the desktop three years ago this transition is happening at the server level today.

A major milestone in that was, of course, the shipment of Windows 2000 and the ability to benchmark above the level of any of the Unix systems in terms of database, web speeds and all the other speed benchmarks that are out there.

So what I’m saying is that the Internet will be brought to a new level. It seems like a radical statement. Doesn’t the Internet do everything today, isn’t it the most wonderful thing? In fact, it’s already gone through several generations. There was a generation where the Internet was just used in the university. Even before I started Microsoft we had the Internet, we had FTP, we had virtual terminal capability, but it didn’t break out of that university environment until the price of communications dropped dramatically. We also had new formats like HTML, that Tim Berners Lee created when he was at Serne*, and has carried on the evolution of those formats through his incredible work at W3C, hosted at MIT.

So the second generation of the Internet was about browsing, it started really in 1994/1995. That’s an era where there were new startups like Netscape that said, hey, we’re really doing this browsing. People were saying Netscape’s moving at Internet speed, how can you ever catch up?

There’s a simple answer. You move at double Internet speed and with things like Internet Explorer Version 3 and 4 and so on, we were able do that.

The era of Internet as a browsing environment is coming to a close now. That’s not to say that browsing won’t still exist but this era’s been characterised by all of the input coming off the keyboard, primarily being driven by PC screens only. The user has no control, if they want to see what information’s out there they’ve got to go to many different screens. For example if you want to do a financial forecast, you need to go to the government website to get inflation data, you need to go to some supply website to see their capacities. There are many different websites you need to go to.

If you go to them manually, you scribble that information down on a piece of paper and then go to your productivity applications and type that information in. If it changes you go back manually to the websites.

It isn’t done as a programmed type information exchanged. The web today is about presenting to the user, presenting to somebody who is reading the HTML.

This next generation is about changing the richness of the devices about putting you back in control so you get notified about the things you care about.

So in this next generation of the Internet that we talked about as .NET, XML is the key standard whereas HTML was for the previous. HTML will continue to evolve but only for the visual element. XML, on the other hand, has a level of flexibility that people have been talking about in academic circles for ages: object orientated, databases, heterogenous data. This, to some people, is the final arrival of the ability to have information be self-describing and exchange it very easily. So XML is the key standard here, just like we bet on graphics interface, we’re betting on XML.

We got onto the creation of a new standards committee, took some HTML work to create XML and now a lot of great things have come out of that. I won’t go through all the different acronyms but XSL, X-link, X-path, the syntax standards, the scheme standard which is now XDL, these things will be the plumbing that allow these systems to work with each other.

Let me be very concrete about that. When you sit down at a machine and say I want to buy a product with the following description, you will do a rich XML search and you’ll basically ask all servers on the Internet to look at their own self-description and see if they have something that matches what you’re asking for. So those servers that have that natural comeback, not to a visual interface but to the program you’ve written describe their offerings so your program can then rate those things and either complete a transaction or present some of the offerings to you. Therefore buying and selling is mediated in a much more powerful way through XML.

So we’ve got to have XML, we’ve got to have these servers being inexpensive with scalability and reliability that software scale provides. Software itself has got to change. People often say, why don’t we make things simpler?

It’s true, we need to make things simpler but we don’t want to take away the empowerment. Just saying to people that their whole job is just clicking on screens and not letting people do the kind of creativity software that really defines their work, that’s not the path forward.

We’ve got to make it easier to update software, easier to keep these systems going. If you ever have a problem why can’t somebody take over your system, see the screen do the right thing? That’s got to be built-in.

The Internet is the solution. All the things that have been complicated about the PC, the next generation of Windows, and the Internet can get rid of those.

And so the new version that comes of this is the anytime, anywhere, any device vision, with software being the magic thing that makes all the information show up when you want it and puts you back in control.

There’s a breakthrough interface here that is different than today’s interface. The idea of being able to type in requests, of being able to speak requests, those can’t just be layered on the interface that we have today.

So what are some of the benefits of this? You’ll be using these devices for real time communication. You’ll be collaborating with people in a way that goes way beyond today’s instant messaging. It will bring the best of e-mail and instant messaging into a more powerful framework.

The information you care about won’t be stuck on your PC. By copying the appropriate things out into the Internet and then with the right security making it show up on your other devices, things like your files and your favourites will always be available to you.

If you want to move to a new PC, it transfers requests without the horrendous effort that requires today.

What about your time and preserving your time? One of the hot issues today is this issue of privacy. If I give out my e-mail address am I just going to get flooded with a lot of junk, like I am with catalogues that come in the mail?

How would we prevent that from happening? The key idea is to take anybody who wants to use your time, which is a scarce resource, and have a piece of software which I call the agent that runs on your behalf manage it. The agent looks at information and based on your schedule says: What time of day is it, where are you, what kind of devices do you have with you, who is this person sending it, what’s it about and, based on that, using your preferences it decides okay, I’ll just throw this away, or I’ll notify them later, or it’s of great importance beep their pager or whatever device they happen to have.

And so this idea of agent means that you can say I want to track whether this flight is coming in on time and that’s very important to me, notify me. So I don’t have to go out and pull through that. Or I want to track some financial data, some internal system if it’s working well, if not I want to be notified. And so you, the user, get back in control and it’s across all your different devices how communication takes place.

In e-mail today, if you’re sitting working on something thoughtful and you get that beep that says that there’s new e-mail, there’s this tendency to go and look at it. I certainly do that and then I think oh, it’s some low level piece of e-mail. Well, that was inefficient, I interrupted my thought process for that. I should have set my system up so it said only a certain kind of e-mail while I’m doing this is worth the interruption. And by seeing my calendar the system will be able to do that for me with very little work.

As I mentioned there’s a barrier today between the world of creativity and productivity and the world of browsing. Those two things shouldn’t be separate. If I see a website and I have some comment where I think this is wrong or I think based on this we should do something, I should be able to create an annotation on that web page and share that with the people that I collaborate with.

It shouldn’t be the case you have to import it into something else. The world of the spreadsheet and the world of browsing shouldn’t be two different things.

That means that the system you wire really has to change. The browser essentially has to be part of that shell and everything has to be built on top of that. Things like viewings, annotations, the ability to type in questions wherever you are that has got to be central to the user interface.

When was the last time that we redefined the user interface? It was basically in the late eighties when we defined graphical interface. We’ve added new things onto it but we haven’t stepped back with a clean sheet of paper.

Now with the speech recognition coming in, with browsing being a central part, we can say that it’s worth the trouble to define this new interface and just like we did with the MSDOS interface, we will bring that along. This is something that won’t just be layered on it will be something new.

What kind of richness comes out of this? Richness of devices. A device that you can carry your music along with, a device that might be very game orientated and yet can access your information.

For developers, they have got to make sure their work isn’t proportional to the number of device types out there. If you have to write every website and say OK, now let’s write the lab on our website, now let’s write the lab to a website, now let’s write the HTML website, let’s write the XML website. That will never get to this scale. You’ve got to be able to create those website applications, supporting all the different devices where the marginal work per device is tiny.

And there is an answer to that. It’s literally to take XML and build the website for XML and then have software technology (that’s included in the new tools that we call visual studio.NET) to take the XML and map it into the various screen sizes with literally minutes of work. It essentially just forms design work when you want to map into a display of any one of these devices.

The beauty of this world is you get the rich part from up on the server, the people who are working with the information, can get the information at work, at home, online, offline, and if they have a smart device in XML. Users can then use the power on that device itself to do whatever additional work they want to do.

So how will this change things for developers? The whole arrival of the website application took place without a new generation of tools. So if you go and look at the source codes of websites today, they’re very complex. There are a number of .asp files and mapping files and script files in use. It is a fact that we’re using tools that were not designed for this scenario.

The first generation of tools that are designed to make rich websites are just coming out. We presented these at our developers’ conference in July. With this new Visual Studio you can see a dramatic improvement in what it takes to build one of these websites.

Not that everybody will start over. We don’t believe in that. They can evolve the code they have but the new work they do will be using this .NET framework.

With this framework you can deploy applications with a single click and they’ll go up and they’ll be hostable in a way that none of those applications will interfere with each other. In fact, when you say you host it, your application, based on the load, will either take one server or two servers or four servers Other applications that are running will dynamically take more or take less servers so you can have a common set of server pools that are used to make sure that everything gets the resources that it needs.

One part of this is that when these programs are talking across the Internet, there are some prearranged protocols underneath. You don’t want it to just be developers who program at that protocol level who can write these applications.

In fact, you’d like a business analyst, who understands the returns process or the pricing model, to be able to write the overall framework. If there’s a little bit of code that has to be inserted for say the shipping method or something, you’d like to just drop that in as a component.

This idea of high level design known as case where programmers only had to write little bits and the analyst could do the work, has around for a long time. But case always suffered from the fact that the high level design and the low level design diverged and the low level design was the real one. So you ended just focusing on that and the high level design didn’t have the importance it should.

Now we have a chance to bring those together, to have high-level design and programs code co-exist. To enable this we have something we call BizTalk. BizTalk is very important to this new world. We’re telling people that all the processes in companies will be digital and so the things about how something’s approved or how you price for a customer, all of that has got to be done, and be enabled to be changed on a very rapid basis. This shouldn’t necessarily involve writing complex codes.

So there’s a need here for a tool that’s matched with this new demand, something that we call BizTalk. I’d like to ask Dave McNaughton to come on out and give us an example of how somebody might use BizTalk in this new .NET world where they’re doing business in a digital fashion.

PRESENTATION

GATES:So I think a key point here is that as businesses go digital, the ability to see what’s going on is invaluable. It is not only necessary to build those kinds of processes, but track what the status is, track how many people are waiting for their shipments, track if there is any problem in terms of people ordering and it not being available.

All of those things should be accessible to the non-developer, so that you can use an agent and set a threshold to say okay, if it gets over a certain amount I want to be notified exactly what’s going on there, or I actually want some code to run on my behalf.

This year is where we put out a number of foundation products for this vision. This is the year when we say the richness and maturity of the server actually is now not only at the level Unix is at, but actually going beyond it with things like the software scale capability.

It’s required us not only to define a family of products but also all the elements that go around it, the consulting service, the 24 hour support, the process in terms of any updates that take place and all of that work.

And I’d say it’s been these last six or seven years we’ve been building to have exactly that capability.

What are some of the server add-ons? Well, SQL Server, as I’ve said we’re very proud of the performance we’ve gotten out of this. Of the transaction benchmarks, the top five now are all on Windows and most of those are done with SQL Server. A few are done with another database, IBM DBII, so those two on Windows are providing a level of scale that’s beyond what any company actually needs today.

In the future they’ll need that on the Internet but again, it’s just a matter of connecting another service.

Commerce Server lets people build e-sites quickly. Application Center is the thing that deploys the software across many servers and lets you monitor it. BizTalk is where we’re bringing the visual ability to define how all these software pieces work together. Host Integration Server has always been our product for the mainframe. Exchange through collaboration where we take a big step forward, letting you develop collaborative apps, and even deploy Exchange on a department by department basis and finally the Internet connectivity product which has got our proxy and firewall capability where we’ve got some dramatic enhancements. It’s the way these pieces go together that create these high-end server solutions.

I mentioned earlier that XML is at the centre of all of this .NET activity and no product remains untouched by that fact. The tools will be much richer.

The languages will be much richer. Contrary to one of our competitors who think that there will only be one computer language and everybody should spend billions to redo their work in that language, there will be more new languages in the next five years than ever before. XML will have a profound impact on even the existing languages. XML will ensure that those languages get extended in some very rich ways.

One of our collaborators on this is Bertram Myer at Monash University. He and his colleagues have taken a .NET framework and they’ve been able to do some very interesting things, putting the Eiffel language on top of it to show that we’ve got total language independence. From Eiffel to Cobol, to Java to Visual Basic to C, those things are running in this framework, supporting XML and so your interoperability doesn’t depend on what programming language you wrote in.

It simply depends on wrapping the data in the XML and then you get the full benefit of this platform.

We thought we’d give you an example of using SQL Server with some of its .NET features. So I’d like to ask Terry Clancy to come on out and show us what we’ve got new here and how it ties into .NET.

PRESENTATION

GATES:There’s one simple message here, it’s that this world of technology and how it’s changing business, learning and even the way that consumers will be entertained, it’s not standing still, it’s actually accelerating.

It’s one of the things that have happened in this industry. Stock prices go up and down, new companies come and go, but those gyrations shouldn’t confuse people from realising that there is progress that’s taking place. The benefits, fundamentally, to buyers and sellers of this are stronger than any advances we’ve ever seen in the history of business.

For IT professionals there’s a lot of opportunity here. The things that have been taught; the cost of management, keeping servers running, being able to deal with the demands that come up, making sure the software stays up to date without having to visit the desktops, being able to support people by being able to take over their system rather the infrastructure makes a huge difference here.

For knowledge workers the vision that’s been talked about for a long time, at one point called information at your fingertips, is right here in front of us. The office environment itself will let you browse through this rich XML data, it will let you browse through your notes, and use that with the other modules in a very powerful way.

A key point is that without first needing to become a programmer, this will put you in control of exactly what notifications come in. Therefore without polling, you’ll know the new things that are going on and yet it won’t be things that are taking your time when you’d prefer not to be interrupted. Consequently the idea of control is central to people really wanting to be involved in this new environment.

The industry will change as this vision rolls out. Software will be thought about less as a packaged product and more of as a service. People who want to software will pay a very small monthly fee and not only get the rights to use it but the support that is also available makes sure they’re getting the benefits out of it.

For developers there are some new paradigms here, not just XML but the tools that have built around it.

For people who take the standard industry building blocks that Microsoft and others provide and craft those into solutions, a lot of the work they’ve had to do in the past that required writing very complex code, and almost recreating what the building blocks were trying to do, now have the ability to customise. These will be much richer because you’ll simply take the XML and add new properties. It’s designed for the customisation and that’s the kind of value-added that customers expect out of the services part of our industry. The demand for those services right now exceeds supply so the efficiency of the new tools will not only deal with that problem but also scenarios that wouldn’t have been possible before.

People who want hosted software can now choose what information their customers are able to access. There is still not the kind of customisation there used to be when it was controlled out of the data centre, however, we are able to separate out the question of who runs the software, versus who has the administrative authority to change various customisation parameters.

For the companies that have built the devices, the people who built the PCs, not only will the volumes be going up but, you’ll have new generations with the microphone built in, with a camera, with the new level of graphics, with new levels of performance.

There will be a whole spectrum of new devices, all of them characterised by having a flat screen and connecting up to the wireless world. Imagine starting with the device that fits in your pocket, up to a device called an e-book, to a device called a tablet, to a device that would take over the entire surface of your screen. You would interact with these devices by touching things directly and editing things with the pen on the surface. You can browse information just like you do on a real desktop where you leave a lot of papers around to have that kind of easy access.

This is the environment where we need the new user interface. This exploits the kind of graphics and multimedia that are in these systems. As a consequence PCs will continue to change and have companion devices that have similar technology built in.

For the companies that provide the connectivity, we need them to also rise to the challenge. Fortunately, optic fibre is at the pure level of capability. The kind of speeds we can get means that we shouldn’t run into any kind of bottleneck. If we take that fibre and deploy it and figure out how to deploy these technologies through the last mile to get into consumer spaces, (including using very specialised wireless technologies), we should be able to get to broadband at prices that aren’t much different than we have with narrowband today. This may not happen overnight but through years of investment and the kind of volume demand that comes out of these markets, this will take place.

When the microprocessor first came along people thought then the computer industry could collapse because there’s more supply than demand. Yet as the software arrived and the software industry emerged, demand met that and so the elasticity in terms of what people were interested in was much higher than expected.

The same, I believe, is true here of telecommunications capabilities. That’s why we’ve partnered up with a lot of those people, to do pilots and really encourage them to make those big investments.

So for people who have been in this field, it’s, great to know that things are not going to get stale.

I feel confident that in the next 25 years, the kind of software we will do create will be a lot more interesting than what’s come before. We call this .NET. There’s a lot of hard work to make it a reality. Microsoft can only do a part of that so we look forward to working with you to make sure all the pieces come together.

Thank you.