Remarks by Bill Gates
Technology Alliance Summit
Seattle, Washington
May 17, 2002

BILL GATES:  Well good afternoon.  I have the privilege of spending a little time talking about some of these opportunities. I mean, after all, will these breakthroughs really change our lives and improve productivity in the dramatic way they have in the past?  And the picture I want to paint is that this next decade will actually be the one of more advances, more improvements than many decades that have come before. 

Now, these advances will require investments in research, small companies taking novel approaches and large companies literally putting billions and billions of dollars into developing the advances.  Whether it’s in medical science or software technology the amount of information, the amount of great people that you need to bring together has really increased over the years, underscoring the importance of the pillars that the Technology Alliance believes in.  So let me say that I agree with everything that you’ve heard this afternoon about the focus and the choices that are required.

Now, there are a number of miracle technologies that are allowing software, computers, communications to advance rapidly.  The miracle of the microprocessor chip -- doubling in power every 18 months, that so-called Moore’s Law -- certainly will continue for these next 10 years.  It may continue beyond that, but there will have to be some incredible breakthroughs in dealing with the small scale involved.

The speed of communications, the amount of data one can send on a single optic fiber, again it’s very rapid improvement; in fact, so rapid that the capacity outran demand, making it a very tough business, but those improvements are coming full speed.

Likewise, the storage, the disks we have in all these machines; how can we dream about having movies and photos and all the different software we want?  Well, the fact is that it’s becoming very inexpensive to have that kind of capacity.

And so these hardware miracles allow us to build systems and build software that do neat things for individuals.

Now, when you think about this you can’t just be at the level of the technology.  You also have to think about, well, what’s important to people, and so how can you bring together new software, the hardware advances and meet end user needs to make things simpler for people that they’ll care about?  Certainly many products that seem neat technically did not meet this test, but fortunately if you really pay attention, you sit down with users and listen to them, there are many things that they would like to see software do far better.

Right now the main thing people are doing with computers is word processing, spreadsheets and electronic mail.  It’s been very nice to get people connected up that way, but even in those areas things like forecasting a budget, it’s still very difficult to go out and draw on the information that you’re interested in.  Keeping the software up to date, understanding what the error messages mean, being concerned about the security of those systems; there’s so much more even in improving the things that we're doing today and now I want to touch on some of these new things that will happen in the software domain.

Part of the key here is the ease of using the tools.  A PC today in many cases sits on your desk, you have your chair there and you have to go there and turn it on and wait for that to be accessible.

The portable machines have dropped down in size but they’re still very much based on using the keyboard.  And so at many companies when you walk into a meeting there’s always a question of what is the accepted protocol:  Is it okay to sit there with your screen up typing during that meeting or is that considered somewhat impolite because when you’re looking at your screen people feel like you’re not focused on the meeting.

We actually say in advance of any meeting whether or not people are just going to be face to face or we have a hybrid where the people who are sitting at the back can do all the computing they want but the principle participants need to be having those discussions.

Well, why is there that tradeoff?  Well, it’s because the keyboard is a limiting factor and so the idea -- the Holy Grail that goes back really to Vannevar Bush talking about the Memex, it goes back to Xerox PARC with Alan Kay talking about the Dynabook -- is to have a computer that instead of having the keyboard is more like just, say, a tablet of paper. 

And so here is this portable computer, and this is a machine that will be coming out in less than six months, is actually capable of being used as a tablet.  You simply turn the screen around and hold it this way and now you’ve got a surface here where you can sit and read documents and if you see information that’s interesting you can point at it.  You’ve got a little pen here that goes with this and you can do handwriting.  You can leave your notes, so if you go into the meeting and nobody’s going to tell you that you can’t write on this; it’s like a piece of paper.  And your notes, of course, are being saved.  You can search those notes.  You can convert them into typed text if you want to.  And so it’s a real change in terms of how you think of this device.

We’re doing pilot studies on it now, and it’s been amazing that not only do people use it in meetings and note-taking but there are things like photos.  You know, we all talk about photos going digital and are going to be on the screen, but as long as it was your desktop machine that you had to go to the chair and sit down, it wasn’t conducive to passing around your photos and sharing them with other people. 

Well, in this form factor, as we’ve done the pilots, people are doing a lot more photos and a lot more music because they can simply share it with somebody and pass it around and so that’s very advantageous.

The way that information moves inside companies today we can make that so much more effective where you’re not putting some things on pieces of paper, putting some things on the computer and some things in your back-end software and you’re constantly trying to make sure that those two things are working together.

I talked about reading; this idea of reading off the screen has been another dream of computing for a long time.  We had to improve the screens to get the incredible resolution but we also had to get something you could hold in your hands.  Our research has shown that as long as it’s in a fixed position that if it’s a very long document it’s just too fatiguing to sit there and read like that and so it takes this tablet form factor or these smaller devices in order for you to get to reading.

The pocket-sized devices are also improving very rapidly. In fact, the boundary between what’s a phone and what’s a personal digital assistant, that boundary is kind of going away.  You see it with advanced Palm devices, advanced Blackberry devices and with the Pocket PC work that Microsoft is doing with its partners.  In fact, this one I’m holding here has this wonderful color screen, but it’s also, in terms of just being a phone where you can set up different calls and talk to people, it’s also far, far better than any of the existing devices.  You can log your calls.  It updates your address book.  If you’re talking with somebody, not only can you talk to them but also they can send information that shows up here on the screen, so a map or a schedule or something of that nature can all be made available.

And so we’ll look back on this time and we’ll say, "Oh yeah, those computers were so big, I didn’t have it with me.  It wasn’t something that I could just open it up and immediately the things I cared about were there."  In the future the idea of when you see an article that’s interesting to you that you’re reading off this screen, the idea of writing a little note on it and taking from your list of colleagues and friends and sharing it with them, you know, that will be just common sense.  Of course we all could share our thoughts on the latest articles about our industry or about things going on.  Was there ever a time you couldn’t do that?  Well, certainly in the paper-based environment that’s very inconvenient to do and so it’s not done.

I mentioned photos and music.  These are moving into digital form.  It’s been a challenge particularly because some of the content owners are worried about, will they get paid for their work when it’s in this environment.  There is a lot of effort going into that to make sure there are rights-management systems where you’ll have things that are free that are appropriate to be free, and yet things that the creators can get paid the royalties they’re interested in.

This is a scenario that I believe in a lot, both in terms of the software work that Microsoft is doing and the neat things that Corbis and people in that photographic area are doing just to make images accessible, so that whatever you’re interested in, either as a consumer who wants to create a poster or just look at things, or as a professional who wants to create a brochure or an advertisement, it’s just a few minutes of work to go out and see all that wonderful creative work that photographers do, be able to find the image that fits exactly the point you’re trying to make and very easily get it down and have that available, so it’s a scenario that really is going to explode over these next 10 years.

Communications today, even though it’s better, we have all these wireless devices, it’s easy to get a hold of people; in fact, it can be much, much better.  You have all these different phone numbers.  Well, you shouldn’t have to have different phone numbers.  You have times where people want to get hold of you and they can’t, times they get hold of you where you don’t want them to get a hold of you.  So right now you’re not in control. When somebody dials one of your phone numbers you don’t get essentially a piece of software operating on your behalf saying, "OK, who is it, what am I doing, how important is this thing? OK, is it worth that interruption or not?" 

And so a lot of people have got junk-mail problems, we’ve got phones ringing during meetings.  People are concerned, will they be in control of these systems,  and in fact with the right software design, the user should absolutely be in control, because after all the scarce resource is their time, and so being able to understand what their priorities are and what counts is very important.

It’s not just for communication; it’s also for notifying you, you know, say a flight is delayed or there’s a traffic jam.  Well, you don’t want during the course of your workday to hear about all the traffic jams taking place, but at the time you’re about to leave and go a certain route, certainly the particular information there should be brought to your attention and the system should simply know that from your schedule, from your preferences where you’re going, and be able to do that for you.  And so information when you want it, information that you don’t have to go and look for because it’s brought to you.

We all can say that meetings are one of the things that make work sometimes feel like, "Gosh, where did all that time go while I was sitting in meetings?"  Well, what part of the meeting was really very, very important to you?  Did you really have to be there throughout that whole meeting?  Did you have any options in terms of being in a different location or just browsing through what happened in that meeting afterwards, searching for the things that are of particular importance to you?

Our research group spends a lot of time studying why do people go to meetings, what do they think after it’s done, is there any way that could work better.

One of the things they’re inventing is called the ring camera.  You can see a little image of that here.  It’s the idea that you could actually record all the different actions during the meeting, and so somebody who’s in a remote location, they can see a panoramic view of everybody even if you’re around a table, or they can pick the speaker, and the system automatically knows who would be of interest in that interaction and that that is recorded and correlated with the slides and presentations that are done.

There are even some phenomenal things we do where we show the writing on the board, and, because we can see the meeting and how it progresses even when you're occluding the board, the way that the software works, it cheats and looks ahead and sees what you're writing and then it makes you translucent so you can see what the person is writing even when they’re blocking what’s there.

One of the neatest things about this is you can watch these meetings at whatever speed makes sense.  So unlike the people who are attending, we’ve shown that for a lot of meetings you can double the speed and miss absolutely nothing in terms of content that’s taking place.  And so just think of this alone as a pretty phenomenal improvement that could take place.

Another thing that was a dream of a few years ago, when the Internet startups were really exploding and everybody was saying these things would happen overnight, the dream was that business processes, things like purchasing and payments, would move away from the very manual paper-based systems we have today to being able to really track things, be able to be notified of things in a purely digital form.

Well, as we’ve gone in and sat with companies and watched the way they use faxes, and the purchase orders come in with errors and they try to get their back-end systems up to date, it’s been very painful to see that these things are, in fact, quite inefficient in how they work.

And it turns out the research advances to build the infrastructure so that these systems can be secure, so that you can find other people who you might be interested in doing business with, those are very tough problems.  And during the Internet hype, many of the companies who talked about this happening quickly were ignoring the fact that big advances had to take place, and so those advances had been taking place, the companies that have big research budgets have been focused on some new things, all of which come under the heading of Web services, and the tools around Web services, that are going to make this company A and company B working together something that’s very, very efficient compared to what we have right now.

So as we think of all these improvements, the productivity opportunity is far greater in this next decade than everything that PCs and the Internet have done so far.  And everything I’ve talked about here is very concrete.  It’s not something where we need some miracle for it to take place. The R&D path is clearly laid out.  The customer interest and scenarios are clearly laid out and so those benefits will be coming along.

One of the cautionary notes that people always raise is, well, if we're going to use these systems even more than we do today, what about all these concerns, privacy, somebody coming in and pretending to be me and seeing my information or a system not being available?  After all, this infrastructure will be almost like electricity is, where without it you simply can’t get things done.

And so a lot of the extra power that I mentioned at the beginning is going into making these systems fault tolerant, making it so that if one of the systems goes down that you don’t even notice that, that automatically the work is taken over by the other systems, making sure that the information is redundant, so if a disk fails or a power supply fails, that there is absolutely no effect at all.  And so we're taking the advances and pouring them into this idea of what we call trustworthy systems.

Now, security is a challenge because the weakest link is always your problem.  For example, today people use passwords to log in and people use passwords that are very easy to guess.  They write them down.  So it’s simply not sustainable to do things that way.

Over time people will have a so-called smart card.  You can see this is one here.  And you’ll use it just like you do for an automatic teller machine.  You’ll have to put this into your system.

We’re running a pilot of this next week.  We have 100 CEOs coming in to Microsoft and we’re giving them Tablet PCs, letting them do little notes and we’re giving them all a smart card, and so we’re taking an audience that is fairly non-technical hands on, and we’re seeing if inserting this smart card is going to work for them.  And I think if it works for them it will work for everyone.

There’s a lot of energy going into this area including some very deep algorithms to make these systems work well.

Microsoft, of course, is proud to be in this space.  This is a chart that shows our employment here.  Over the history of Microsoft we’ve had about half our people in this state and over 80 percent of our R&D is done here.  In fact, we spend over $4 billion a year in basically one location, which is the most R&D spend by any company in one location anywhere in the world.

When you think about this, and you try and call up an image of it, there is no big equipment, there is no chip fabrication or steel molten furnaces or anything like that.  This is people sitting in offices, thinking up new ideas, collaborating with other people about those new ideas.

One of the key collaborations we have is with the University of Washington, and so we’ve been pleased to see that the computer science department there has gone on to add new people and add new capabilities, and so it’s not only improved dramatically in absolute, it’s also become certainly one of the strongest departments in the world.

Also when I look at this chart, I’m thankful to think about the spring weather that we have here.  Most of the recruiting that we do is in the spring, so we get people up here and they see the nice weather, and by the time winter comes they’re very committed to the work that they’re doing here.

So there are some policy challenges that are faced.  I think investing in a very strong, specialized, world-class department, particularly at the University of Washington and other four-year institutions, you know, for me that rises up as a very, very critical issue.  It’s very heartening to see how many people have come to this luncheon, and the kind of top talent that’s really coming together to advise the state and raise the issues all the way out to the level of the voters to make sure that these things get the discussion and the consideration they should have.

Many of these things are the toughest issues to really get people to think about politically because they’re investments you make today where the payoff is a decade from now.  You know, the kind of great things that will be coming out of the UW over the next five years, of which there will be many, and great things coming out of Microsoft Research, the foundation has been laid for those things based on what happened the last five years.  But once you get beyond that horizon the decisions we're making today will make a very big difference.

The reason I think this is all worthwhile is that the sky is the limit in terms of medical technology, the deep understanding now that we’re getting of biological systems, partly using software to track that information and understand it, taking the PC, which has been the greatest empowerment tool yet invented, and taking it to a whole new level in terms of ease and use and power, you know, I have the confidence that can happen and I think with the efforts of the people here a large part of that is going to happen right here in Washington State.

Thank you.