Bill Gates Keynote
A Conversation with Bill Gates
San Jose, CA
January 27, 1998

Silicon Valley Venture Capitalist Bill Gurley was the host and questioner of an event at San Jose State at which Bill Gates answered a number of questions from people in Silicon Valley.

MR. GURLEY: Well, I'm certainly honored to be here. I'd like to thank both San Jose State, as well as the Tech Museum for inviting me to be the interviewer. Now, I know you didn't come to see the interviewer, you came to see the interviewee. I think it's a great opportunity for us to get to sit down and share some questions with Bill Gates, not only a world figure, but a great CEO. And he's also someone whose decisions, as in those of other executives at Microsoft, affect the people in the Valley, because there's so many companies here related to technology. I also think that despite the fact that Microsoft has 25,000 employees, Bill is still a bit of an entrepreneur. And I think that the company acts in very innovative ways. And I think we can learn from that, as well.

Now, we've got a fairly simple agenda. We're going to start off with some hard questions. We're going to start with the DOJ and get it out of the way. And then we're going to go in and talk a little bit about Microsoft and the Silicon Valley. And then we're going to talk about the future. And then we're going to follow up, and hopefully get some personal insights.

So, with that, let me kick it off with one of the hard questions. I was wondering if you might just give us an update on the situation with DOJ?

MR. GATES: Well, DOJ filed an action last year to try and block us from putting Internet technologies to Windows. And that was a surprise to us. We had included an Internet browser in Windows from the very first time we shipped Windows 95. And then we kept updating it. We went Internet Explorer 1, Internet Explorer 2, Internet Explorer 3, and recently Internet Explorer 4.0. And their view was that if the software had ever been available as a separate product, by itself, that then we can never put it into the operating system. And that principle is a big problem for us, because every feature we've ever put into the operating system, started out as something that we provided or someone else provided separately, whether it's memory management, or hard disk support, the TCP/IP stack, printer drivers, graphics interface, and now Internet types of capabilities. So we've got a real problem with the principles here.

We went through a bit of an intense period related to the preliminary injunction, where at first they asked us to remove the Internet Explorer, but because of the way it's tied into the system in the case of IE 3, it broke the system. And we had told them that's exactly what would happen. Then the government came back and said, well, just hide it. Even though you can get it to come up, all the features are there, all the files are there, just help us by hiding it, because that breaks less if you do it that way. And so we agreed to do that last week. That's all the preliminary thing, that really doesn't speak to the heart of the issue.

Now, we're onto the heart of the case, which is: Can we continue to innovate in the operating system? And unfortunately, it's so critical to us, in everything we do, and what's driven the PC industry, this is one we've really got to stand up for. And for the next, I don't know, six months or so we've got to explain what's the precedence here, and they're very much in our favor, explain what it is we're trying to do and get this behind us.

MR. GURLEY: Now, you've stated publicly that the central point behind your push is that you're standing up for customers, and you're wanting to innovate.

MR. GATES: Yes, that's right. We think Windows 95 was a very positive product to bring into the marketplace, that bringing graphical interface into MS-DOS, even though they had been available separately before, that that was a necessary step. And look what it did to the market. Look how it grew the market. Look how it made computers better. And we have a lot more to do there, putting speech into the system. But their approach would block us from ever putting speech into the system.This lawsuit is not about choice. People have always had a choice of whatever browser they want to run on top of Windows. The people download those products from the Internet. Here we've got products that are free, that are easy to get to. The vast majority of people do download upgrades across the Internet. And so choice is not the issue, and browser share is not the issue.

When Internet Explorer 1.0 was shipped with every copy of Windows we got zero share. With Internet Explorer 2.0 we got zero share. It doesn't matter that it's in there.Once we shipped Internet Explorer 3, started winning a lot of the reviews, and even more intensely when we came out with version 4.0, that won virtually all the reviews, then people started to pay attention. But, for us the big thing is simplifying the system, making it so that when you browse local information, it's the same as browsing Internet information. Making it so the way you navigate the help files, it's not some special thing, which is the way it's been in Windows. But, it's the browser, letting you go through those help files. Our forms technology is going to be built around the Internet standards.

That's pro-competitive.That's what customers want to see operating systems provide. They don't want to have to go out and buy everything separately, go get graphics interface separately, go get TCP/IP separately, go get the browser separately. And for people who are innovative, they can build on top of the system, that opportunity is very much there. We're very predictable in terms of what we do in the operating system.

MR. GURLEY: Going a little bit deeper on that, you talked a lot about your right to integrate, and that's been a bit word that's used with this case. Is there any limit to what you'd integrate? I mean, could we one day see Word or Excel in the operating system?

MR. GATES: We actually have a small word processor in there, Wordpad, it lets you do actually a fair bit of word processing. And some people can get by with that. We think that's a nice thing, so that people can compose electronic mail. Now, our basic criteria for putting things in the operating system has been consistent, for as long as we've been doing this. And that is that if something is desired by application developers, if they want to build on top of it, instead of making them all duplicate it and ship it with their product, we should get that in the operating system. Now, the graphical interface got out there early, and then we had to evangelize, and say please take advantage of this. And that was a tough period for us. A lot of people thought we wouldn't be able to get there.

With the Internet we're doing the same thing. And it hasn't been as tough. People are very anxious in their applications to call our HTML control, to call our Internet linking capability, all of which, therefore, need to be down in the operating system, so they don't have to recreate it or distribute it with their applications. So there's nothing surprising here. If there's anything that clearly should be in the operating system, it's got to be these Internet technologies. It is pro-competitive, not only for the computer industry, but for all industries where the Internet is a big change agent, allowing people to see what's available, more easily, and making business much more efficient than it has been.

MR. GURLEY: There's a recent article in Business Week that argues why you are always pushing innovation and throwing more stuff in, in certain cases, is that in markets where you have large market share you're actually raising prices. And they point to the OS. And I think from this concurrent licensing stuff that you probably read about. Is there any validity to that argument?

MR. GATES:Well, certainly, the price of productivity software has come down very substantially over the years. When 1-2-3 was introduced, it was $495, for a single application that wasn't one-hundredth as powerful as what spreadsheets have today. Today, you pay less than that $495 to get Word, Excel, PowerPoint, Access [together], all in a far more powerful form. Now, two years ago we said to our users that we were going to get rid of concurrent licensing, because you end up with very inconsistent pricing. Somebody who's using a concurrent licensing is paying a lot less than somebody who is paying for each user. And it was an artifact left over from years before.

So we started to phase out concurrency. We took it out of our standard products, and we created a special license that requires paying more money. And we made it clear that that was being phased out. Now, we're actually getting to the end of that two years. And for the people who have been using concurrency, which is a reasonably small percentage of the customers, it is a price increase for them, to make it consistent with the other people. But, the average price we get for productivity software is much lower today than it was a year ago, or two years ago, or three years ago. So that is an adjustment. And we need to work carefully with the people who are involved in that, particularly in areas like educational markets, where people are very, very price sensitive.

In terms of the operating system, the price of the operating system has always been less than 5 percent of he price of the machine. Now, if I had a monopoly, that is I wasn't subject to competition, believe me, the price for the operating system would be about 25 or 30 percent of the price of the machine. And we would actually make a boatload of money. But, because we're not optimizing for the short run, because we know any type of price increase above that 5 percent for the average PC would just make it that much easier for somebody to come in and take our place, we've kept those prices quite well.

They've gone up modestly as we went from DOS with Win 3.1 to Windows 95. They did go up somewhat. And for people who buy Windows NT, it's somewhat greater. Still, for a hardware manufacturer, they pay less than $100 for an operating system that we've put over $5 billion of development into. And it's the only comparable operating system to something like high end UNIX, MVS, where the prices are ten times as high as what we charge. So we feel very good that we are the software company that have brought prices down more than any other software company. And we're going to stick with that.

MR. GURLEY: What are your thoughts about the almost collaboration of your competitors with the government. Do you think that they are actually having an impact? Have you voiced that, I guess, with the government?

MR. GATES: Well, there's no doubt that we wouldn't have a DOJ dispute here if some of our competitors hadn't decided that battling it out in the marketplace was -- that their product wasn't going to do well enough on its own that way, that they were going to try and use the government to cripple us. And I use that word very carefully, because the idea of trying to tell us to ship products with features deleted, those are crippled products.

Now, that the DOJ has undertaken the thing, it doesn't really matter how it got started. We need to explain the principle involved, why it's good for customers to have integration and get through that, while not distracting ourselves from building the software and listening to customers.But, it is disappointing to have people back in Washington, D.C., acting like the PC industry is a bad thing, acting like the software industry isn't the most healthy, the most competitive thing going on.

America has a lot to be proud of with this industry. This area (Silicon Valley) has benefited immensely. The jobs have been created here, the wealth has been created here. You know, this is a story that everyone should feel good about. The industry should feel good about. And if we didn't have as many competitors, there wouldn't be competitors to complain about all this stuff.

MR. GURLEY: We'll only take one more question on the DOJ, and then we'll relieve you of that. There's been a lot of talk in the press about how Microsoft came out aggressive and then kind of relaxed a little bit. Is that overstated. Was there a learning experience on Microsoft's part in terms of interacting with the government?

MR. GATES: Well, the preliminary injunction is behind us now. We've come up with a set of options for the OEMs that, during the duration of the dispute, they have a lot of ways they can ship the system. All of them, they can either ship it full-blown or they can ship it in various forms of reduced capability.

What happened was that the government said, look, Internet Explorer and Windows, they're just two separate products, and all you have to do is take the files out and there's no problem. Now, the person who led Windows 95 development, David Cole, testified under oath that taking those files out in the case of IE 3 would break the system. The government said, no, he was wrong.

Now, the order was entered stating very explicitly to delete the retail IE files, and we did that, and we said, geez, this is crazy. We've told the government exactly what this meant. The preliminary injunction, when it's issued, you have to comply immediately. It's not like some grace period to go have a tea party--you do what it says.

MR. GURLEY: Correct.

MR. GATES: And so, we did what it says. We actually went back to the most recent version of Windows that runs with those files deleted, which is the retail version. We sell hundreds of thousands of copies of that every month. So, we also gave them the option of that version, which is the only one that runs with files deleted.

Now, then they filed a contempt motion against us two days later. They didn't call us up and discuss it. They just filed their motion. We went in, the government witness didn't have any other way that we could have complied, and then the government was saying, why don't you hide it. Well, they'd been very clear that that wasn't okay before. They said, just removing the icon was not. They wanted IE deleted, but then they changed their mind. So we said, okay, we will give you the hiding option.

MR. GURLEY: Right.

MR. GATES: And the thing that confused it all is, there's a command in Windows that's called the remove programs command. If you say to remove IE 3, it doesn't get rid of any of the browser files, the browser is still there. And that certainly confused the issue and made it look like, well, is Microsoft just being defiant or arrogant here. But because the browser is all still there, it wouldn't have met that judge's order.

That's behind us. It was a painful episode. I wish we could have articulated what we were doing better to the public at large. Believe me, when your own government sues you, it's not a pleasant experience, and I was not sitting there going, ha, ha, I'll do what I want. I was sitting there going, this is the worst thing that's ever happened to me. When can I get back to just focusing on developing software? And then to have the press come out and say, you know, what it said that felt bad.

MR. GURLEY: Well, if it was that horrible, let's move past it. Because I don't want to bring up those bad times again.

Could you talk a bit about the importance of partners to Microsoft, something that's actually quite relevant to this community, probably.

MR. GATES: Our business strategy from the beginning was quite different than all the computer companies that existed when we were started. We decided to focus just on doing the high-volume software, not to build hardware systems, not to do chips, just to do software. Now, for that model to work, the entire industry had to be restructured into a different approach. And, fortunately, that did happen. So you have companies that are good at various aspects, for example, Intel and chips, or a lot of consulting companies that have the building blocks to be put together. So, it was a strategy that required partners. I think the most successful partnership in the history of American business is the work we've done with Intel. When we started working with them, both companies were worth 1/100th of what they're worth today. And so, working hand-in-hand in a nice, complementary way, you know, with a little bit of friction from time to time because we're both pretty strong-willed companies, we built two of the most successful enterprises of the era.

Now, we have other partners, HP, Compaq, Digital, maybe we'll talk about that. The variety of partners is quite broad. Software developers are key partners. Here in the Valley we have a lot of great software developers. We have what we call solution providers, the people who train and install and integrate and consult. And those range in size from the very big EDS down to the smaller companies that are actually the most zealous about the new technology and working closely with us. And all of these partnership programs have been very, very key to our success, and whether it's being the best at working with developers, being the best at working with solution providers, whether it's helping some of these partners financially to get out there and make the aggressive investments, that is what's driven first Windows and more recently the success of Windows NT.

MR. GURLEY: What should a company do that wants to partner with Microsoft, just knock on the door in Redmond?

MR. GATES: Well, sometimes, you know, we have more people who want to partner with us than we have the capacity to deal with all of the things they'd like us to do. We have a specialized group at Microsoft for system manufacturers. We've got a group that works with them. For the software developers, we've got our developers relationships through DRG, our Developer Relations Group. And we've actually expanded that out into our fields with our applications developer customer unit that not only works with developers on technology, but that group is there to help them with the marketing of the product as well. For people who publish on the Web, we have our Site Builders projects. For people who are communications companies, we have a special group to work with them. And so, you can call your local Microsoft office, the people understand how to help you wind your way through domains at Microsoft.

We've worked with a lot of venture capitalists like your own firm to make sure that we're coming down on a regular basis and trying to make Microsoft approachable by these small companies, particularly the start-ups, because they are going to be the most aggressive about using new technology in exciting ways. And, you know, it's always the thing where we want to get feedback, we want to do better on those things. But I'm very proud of the way we are working with companies today.

MR. GURLEY: Right. Sometimes you take partnership a little bit further and you actually do acquisitions, and there have been quite a few large ones in the Valley recently. Could you talk briefly, just in general, about what Microsoft looks for in acquisitions?

MR. GATES: We have done a number of acquisitions in our history. Some have been smaller companies, where it's to get the technology and the great developers, and sometimes we integrate them into the development efforts at our headquarters. In other cases, it's a large enough group that they have a critical mass on their own. So, for example, Web TV, we decided leaving it here and having this be our center of excellence for all the things we're doing around the living room and the TV set that that made sense. So they're continuing to hire very aggressively down here, and we're pleased to have that diversity of hiring people in multiple locations.

Some of the smaller ones would have been like Dimension X, or there's quite a few that we bought, we ended up having them come up and join the relevant teams so that their software could get integrated in, in the right way. A lot of the time, the reason we do the acquisition is, when we see a market developing very rapidly, and we want to reduce the amount of time it takes us to get in there, get working with customers, get the feed back that's valuable. With Web TV, it was the box they had done and the user interface they'd built around that. With Hot Mail it was the customer base that they'd built up there. And I'm sure you'll continue to see a lot of that. Most of our developments we do in-house. We're not quite like the Computer Associates, which acquires big, big products. We mostly acquire bits and pieces. And we've done very well with those acquisitions, because we work with the people and make sure they're going to be happy with the jobs we give them.

MR. GURLEY: I think Web TV might have surprised a few people because they thought this whole set-top craze, or whatever you want to call it, might be competitive with PCs. So, what were kind of the three top things going through your mind as you said, boy, I've got to have this thing?

MR. GATES:The TV set will change. Instead of just being a passive viewing experience, it will be interactive, and there will be a digital box there that delivers higher quality video and audio, and gives you more channels to choose from, but also lets you interact so that you can get statistics on the game that you're watching, or go off and see what the weather report is, or exchange electronic mail. The miracle of Moore's law is going to make the box, that set-top box, that costs less than $300, very, very capable and we decided a long time ago to build the software technologies that would help the TV move in that direction.

Now, what the Web TV group had done is, they around the Internet, had built some good chips, and had built some good user interfaces that fit right into that area. And because we saw the cable industry starting to really think about designs, we thought if we brought together what Microsoft had done and what they'd done, we'd be able to work in partnership with the cable industry. So the products we sell through retail that connects to the phone network has been very successful. We weren't able to make enough of the new Web TV Plus for this Christmas, but we're catching up on that. And now we've just won a very major order from TCI and some of the other cable companies for a set-top box that uses Web TV technology along with our Windows-CE operating system.

MR. GURLEY: We got a question through email. We had posted an email address so that we could take some questions from the people in the Valley. How is Microsoft going to reconcile its Hot Mail purchase with its Exchange Server strategy? I think that's a little bit off the kilter, but the thing I'd like to add to that is, do you think that corporations might start using out-sourced email as a service, and how will that affect your Exchange as an applications group strategy, if the answer were possible?

MR. GATES: Yes. Well, we have many email servers now. We've got the MSN email server, the Hot Mail email server, and the Web TV email server. So we need --

MR. GURLEY: I think the Hot Mail one runs on UNIX, isn't that right?

MR. GATES: Yes. Actually, Web TV still does as well.

MR. GURLEY: Oh.

MR. GATES: But, believe me, that's just a matter of time before that changes. We need to pull all three of those together, and we need to make sure that the features and protocols interact with the mail systems we sell. I don't think corporations are going to completely send their email outside, because they have special requirements in terms of security and setting up the directory, and having very efficient email flow within their global networks. But the idea they'll be able to exchange information like schedule information or files, or preferences out with the server on the Internet, the Hot Mail server, we think that's a great opportunity.

And so, we're making sure Exchange does the best job of interacting with that generic service that's out there on the Internet. There is some work that we have to do to make that work, and it will, I think, strengthen Exchange. Now, for small businesses, they'll have a choice of going just on the Internet, or using the Small Business Server, but we'll make sure that the Exchange email in that Small Business Server, if that's the route they go, works very well with the Hot Mail service.

You know, you wrote, I thought, a very good piece about this, which is, the boundary between what's software and what's a service is starting to blur, you know. What -- should we charge for the mail client, or should we charge a monthly fee for the fancier features against Hot Mail? The basic email will always be free, because that's how you get a lot of people to come in, but some of the fancier things, like coordinating a lot of schedules, we'll charge a small monthly fee and just give you the client software to do that.

And so, it will be very interesting within Microsoft to make sure we strike the right balance in making sure people who want to run their own networks can pay the software fees, but people who want to do it on the broad Internet can just view it as a service, and you can have all of those things still interoperating with each other.

MR. GURLEY: Sure. Let's shift gears a little bit and talk about protocols and tech stuff like that.

MR. GATES: OK.

MR. GURLEY: Java is obviously something that's got a lot of momentum here in the Valley and some companies have gone so far as to build Java in from the client through the object layer down into the storage structure and everything. And I guess my question to you is, you know, will those people's products run in a Microsoft world? Should they be thinking about changing that?

MR. GATES: Well, they will certainly run. There's a question of will they run competitively. When somebody has, say, a Windows machine, just to take an example, they want an application that takes advantage of that, that uses the user interface, that uses the clipboard, that fits in with the other applications. And so, the other application that comes with a duplicate operating system, its own graphics manager, its own scheduler, its own storage manager. And then you get a problem where the user has to say, well, wait a minute, how do I run these applications, how do I think of these? And also the size of that application is going to be a lot larger. And if you have multiple of those applications that have different versions of that second operating system, you're going to get a lot of conflicts there.

So, typically, if you want to target Windows machines, you call the native Windows services. And even people who started out saying, oh, I'm going to write a pure Java application and just use the Java runtime, most of the commercial companies have come around to saying, look, on Windows, we're going to do better than that, we're going to take advantage of the richness and the efficiency of calling through to Windows, while still getting the benefits of the Java language, and maybe on some other platforms not taking advantages of it as much. And there will be a nice equilibrium there.

You've got to remember, there's a lot of code out there in the world, and a new language is not a panacea, although it does nice things, you don't always want garbage collection. You don't always want the type of typing system that's in there. People have gotten used to using Visual Basic and C, which are far more popular than Java is. Yes, Java has come up and done very well, but it's still only about 10 percent of professional developers that are using Java at all, and so we think a strategy that supports all the languages, Visual Basic, Java, C, Access, COBOL, you name it, and lets them get at the rich services, that's a great strategy.

MR. GURLEY: Now, even if a start-up wanted to build solely on Microsoft's technologies, I guess because of the Internet and all the change, that's still been a little bit confusing. I mean, the ActiveX and HTML, and then now there's the DNA. If someone wanted to just be set for tomorrow, what tools should they be using from Microsoft's perspective?

MR. GATES: Well, it depends a little bit on the application they're building. We need to make sure our developer message is very clear. The overall framework of what we're doing in Windows we call Windows DNA, Distributed Internet Architecture, and that encompasses things like the Transaction Server we're building in, the Internet Information Server that's built in, the ways of getting at DHTML. For simple applications, Visual Basic does very well, even for some of the larger corporate applications, we've been able to scale that up quite a bit. People who do heavy-duty development are  using C today, and they're going to be using C five years from now. It just lets you get at a level, in terms of managing memory, managing type systems, that other languages don't let you get at. So we need to continue to support that.

And so what we need is a road map that says, for these types of applications, here's the tools that make sense. We have a thing called MSDN, which is a lot of information we send out to people, and we have our Systems Journal, and we've got our Web sites. I think we do a good job in getting data out to developers. If there's something in that road map they're finding confusing, we've got a lot of people dedicated to talking to them and helping them find their way through that. But the key thing is, if you want to sell to Windows' customers, whether it's Windows on the client or Windows on the server, natively calling some of those services, we think, is what will generate the most competitive application, no matter what language it's developed in.

MR. GURLEY: I want to ask you a quick question about three companies, so we can keep it terse. First of all, Netscape, and I just want a simple answer. I'm just curious, when you look back from here, currently, today, at what's going on at the company, what would you say they've done right, and where do you think they've made some of their bigger mistakes?

MR. GATES: Well, I think Netscape is a very good company, and they've done a lot of good products. I think it's a little disingenuous of them to say they're surprised the browser is in the operating system. I mean, how can somebody come out and say, hey, by the way, we're going to wipe out Windows. These guys are completely obsolete because our browser is becoming an operating system, and then say, oh, they've hurt our feelings by putting a browser in the operating system. I mean, fair is fair. Come on here, do those things belong together or not. They say they do. We say they do.

MR. GURLEY: All right. We're going back to the --

MR. GATES: No. Anyway, you know, they're moving to do enterprise services. I think, you know, that is a big market. The interesting thing is that you've got IBM trying to do enterprise services. You've got Oracle trying to do enterprise services. You've got Sun now developing a large software division to do enterprise Java services there. And you've got Novell -- I mean, Netscape, sort of Novell but not really. And, so all of them are going after the same revenue stream, which is transaction management, building those server type applications. And, of course, don't forget us. And so, it's going to be a very crowded business that we're all aiming at.

MR. GURLEY: I've got a specific question about Intel. I was reading in your speech to the Consumer Electronics Show, and you seemed to speak kind of openly about the fact that Windows CE now runs on seven processors, and you said you hope there's more. Is Intel upset about that?

MR. GATES: Well, Intel is probably our most important partner. If you had to pick one company, we've done more with them. And it will be most important for us to keep working with them more than anyone else. What they're doing on Merced, we're very excited about. In fact, we have a big event coming up next month to talk about our progress on Merced, and give a message to the developers there. The work they've done on this product they just brought out, the higher speed Pentium, is fantastic. And so, across-the-board, we're very impressed with what Intel is doing.

Now, in this very low-end where CPUs have to have very low power drain, Intel chose not to participate in that. And so, we've talked to them again, and again, and again, you know, are you going to have a product in this space. Windows CE, the first processor it ran on is the Intel processor. But, customers for small things like this -- and I've just got one in my pocket, this Palm PC, they couldn't get a chip from Intel with that low power drain. And so, as Intel comes into these markets, which I think now they're considering doing, it will be great. We'll support that chip. But there certainly was an open dialogue there, and it's just a new part of the market that we decided to address.

But when you look at the PC and how the PC is evolving, when you look at making a PC that can do speech and vision and learning, we need Intel. We need the great work they're doing to make those processors faster, because we're writing software that requires the latest advances that only they, with their huge R&D, will be able to bring.

MR. GURLEY: Right. Following on that, I guess, I'd like you hear your comments on the recent merger between Compaq and DEC, and whether Alpha -- I'm sure they talked to you about this -- whether Alpha will continue to exist under Compaq's reign?

MR. GATES: Well, I grew up programming Digital computers, whether it was a PDP-1 that I wrote a baseball program for, or a PDP-8 that I wrote a Basic for, or a PDP-10 that I loved to work on the operating system and find problems on. So, I think Digital was a very special company. They were the company that was really bringing computing to the masses during that minicomputer era. And so, it's a bit of a milestone to have them absorbed by a PC upstart that's done so well. It just shows how much this business has changed.

Having said that, I think the combination is a very, very powerful one. The services capabilities that Digital built up over the years in terms of doing mail systems and networks on a global basis was very, very strong. And that's what Compaq was anxious to have. I have an event with Bob Palmer tomorrow. We're going to talk about some of the things that Microsoft and Digital have done together.

Now, the Alpha chip, I think, will be very interesting to see what Compaq does with that. It's kind of a fascinating situation where, because of the Digital-Intel settlement, Intel will be fabricating that chip and providing it to now Compaq at a fairly low price. And so Compaq will have a chance to fit that into their product line. At the same time, I know Compaq is very serious about Merced and wants to be a leader there, and we're very impressed with the work that Intel is doing with Merced.

So, how does Alpha fit in? I think it's a wild card, but I wouldn't write it off at this point.

MR. GURLEY: Okay. You talked about the co-evolution between you and Intel. You know, obviously, they built faster processors to run better software, your better software requires better processors, and you have a nice little feedback loop there. A lot of people, including Perry West, who sent in this email, are concerned that the bottleneck on the PC has shifted to the communications or the bandwidth, and that that could be a threat to the entire industry because the loop could break down. Is that realistic?

MR. GATES: Well, I think she's got a very good point. I know Andy Grove and I have talked about this quite a bit. And that's why you find Microsoft and Intel working with these communications companies to do whatever we can to help them get that bandwidth going. At one level, things look good. You know, fiber optic speeds are going up. These satellite two-way technologies will get into place. The routers that need to do this very high-speed stuff, those are being developed. But building out the infrastructure is expensive, and it's a regulated environment, so it's not clear the incentives are there to push this as aggressively as might be possible.

Now, I would say that in the business environment, I don't really have a concern. The bandwidth is going to be there. It's going to be very competitive. It's when you talk about getting out to residences and having high-speed two-way connections, there the only hope within the next five years are DSL, which is very early technology. We have several Microsoft employees hooked up, that's about half the people in the United States using DSL at this point, but there is a low-cost form that Compaq and Intel and Microsoft have gotten behind. And, although it's too early to predict, I'm optimistic about that.

PC cable modems are another great thing. They deliver high bandwidth. And so, if both of those can get developed and start competing with each other, then we'll see this bottleneck go away. But it could, if we don't get that going, it could slow down the home PC market.

MR. GURLEY: Okay. In your book, The Road Ahead, in Chapter 11, you talked about people in Gold Rush often get so caught up in the dream of quick riches that they over-invest in obvious areas and ignore subtler, long-term opportunities. I was wondering if you might highlight a few of both the over-zealous investments as well as maybe serious thoughts, you might not want to, these may be where Microsoft is investing in terms of the subtle, long-term opportunities.

MR. GATES: Well, I don't think there's anything subtle about what we're doing. When I said subtle opportunities, I meant, you know, take a company like Sienna that was doing optic fiber switching equipment. I think they, at least as far as I was concerned, sort of came out of nowhere, but they're in a great position, although that's competitive, too, they're providing the equipment that makes it all possible.

You know, when there was a big growth in population here in California, people thought the homebuilders would do well. Well, too many people thought the homebuilders would do well, so there were too many homebuilders, so they didn't do well. The savings and loans in that era did extremely well. So you want to be somebody who is providing the key component, but not that everybody is jumping in.

On the Internet, you have a lot of things where everybody is jumping in. I mean, how many travel sites are there on the Web?

MR. GURLEY: Well, you guys have one.

MR. GATES: We've got one. How many news sites are on the Web? We have one. You know, they are just out there, and the barriers to entry are extremely low. So, I would say the payoff for people who get into those and thinking about these solutions, for people who get into those very crowded areas, the payoffs are unproven, and they're pretty far out in the future. It's really going to take the Web to become a part of our everyday life, and there will be some windowing out, some of those people won't wait and be able to continue to invest in it, until the market becomes large enough. So there is -- there is a lot of competition there.

A little bit of what I'm saying is that we're -- because of our success, we can do very risky things. And some other people have that same risk profile, venture capitalists, probably do. But, if you're investing in Internet-related stocks, you're dealing with high-risk, high-reward type stocks. The volatility between now and the promised land, for even the Internet company that's doing the very best work … there will be a lot of ups and downs. And we've already seen a fair bit of that. You know, take Yahoo is viewed as doing things very, very well. Their stock has had a lot of volatility.

MR. GURLEY: Could you share with us your thoughts on what I would call your buyer aggregation sites, CarPoint, the travel site, and there's rumors, I don't know if they're official or not, of a real-estate site? Is this an area where you're going to go full bore?

MR. GATES: Well, to put that in perspective, the number of people working on those sites is not one-tenth of the number of people who work on Windows or BackOffice or Office. So relative to our mainstream businesses, that are very big businesses, these are still pretty small groups. Now, on something like Expedia (travel), we'll have 50 people. So by most people's standards, it's a very serious investment. Something like Sidewalk, we've got several hundred people involved with that. The site that is doing the best for us economically is Expedia, because it's got transaction revenue. It's not just waiting around for massive advertising revenue.

And advertising revenue on the Internet is still pretty small today, and it's being split across so many sites, that it's just not a lot of money. And people who are dependent only on ads, except for maybe some of the trade publications, anybody who is very horizontal in nature is not going to be making money today. If they say they are, they're doing their economics in some strange marginal cost type approach. But, you know, people are allowed to bet on the future, just like we are. And some of these people are going to do very well.

MR. GURLEY: Microsoft has despite -- you said you don't invest in verticals -- but you have spent a lot more time thinking about verticals. When you look at the Internet and the potential for disintermediation, and these buyer aggregation sites, which industries do you think are going to be effective first, and should really have, you know, their blinders and headlights on? Say, oh my, here comes the Internet.

MR. GATES: Well, the thing that the Internet does very powerfully is it takes mispricing and it exposes it. What do I mean by mispricing? If you have a travel agent, they can have a transaction where somebody calls up and books the ticket, and never changes it, and never causes a lot of trouble. On that transaction they've made a lot of money.

You could have another case where they call up and ask for the lowest fare, change their mind a few days later, and a few days later they cancel the trip. And so they've done poorly on that. Now, the Internet can let that person who wants to just do it the simple way go and do their booking, and even get some type of discount, against that travel transaction. Whereas, that other case, you may eventually have to have a service fee, if you're really taking a lot of people's time.

The same thing in banking. Banks today, they do a lot of services for you. Some of them are quite profitable to the bank, and some are basically loss leaders. And what you'll have is, through the Internet you'll have other banks, not Microsoft, we've said it 100 times, I guess we need to say it 100 more times, we're not doing any type of banking thing, where we're a bank and we get involved in banking, we're just giving -- providing software tools and services to banks. But, you'll have other banks come in and take those services that are very profitable and just go after those. So you'll have to really look hard at pricing things according to where the real costs are.

Now, the Internet is just starting to take off. You're seeing it for stock trading, you're seeing it for some banking, you're seeing it for some travel activity. But, if you take a market like books, it's a tiny part of the book market today. And the barriers to entry of people getting in and doing that are fairly low. And so there is a real question, will it be the early entrants, Amazon who was very innovative, or the more traditional players, like Barnes and Noble, who are now coming into the Internet a little bit later, about a year-and-a-half behind, but being very aggressive. And that's a struggle that will be fascinating to watch how it comes out.

MR. GURLEY: We got a lot of emails on this particular subject, a lot of people are curious about your thoughts with regard to foreign software development, and whether the Valley may be, with their high prices, and ambitions for stock, and everything, may actually see a competitive threat from off-shore development. I know Microsoft has done some stuff in India, BAAN has obviously done stuff around the globe. Is it a trend that's going to continue?

MR. GATES: Well, Microsoft has done very little. We may have 3 percent of our development outside the United States. And that will go up, but I don't see it in some high percentage. There are companies that are doing a lot of their development in India. India has been the big place for this, because they have an educational system, that the British got started, that includes some technical universities, IIT, that do a very good job. And, of course, the people who go to those schools have good English skills. And so you've got sort of a critical mass there, particularly in Bangalore.

That's an interesting phenomenon, but there's going to continue to be a shortage of great developers. And for a company like Microsoft, we want to do the product as best we can. Even if there was some way to do it that was half the cost, if it was six months later, or a little lower quality, or a little less in touch with our customer needs -- it doesn't make sense for us. We sell in such high volume, that we can afford to get the very best developers, get them on our campus, and really have them working to hone to product to be the best it possibly can be.

So the kind of software that you're most likely to see out in India in big numbers are applications development products that corporations want to get done. GE has moved a lot of development out there. Software development that's not quite as demanding. Now, I think there is no doubt that there will be a shortage of people to do these jobs, you know, for many decades to come. No matter what happens in India. India is already seeing problems, where the salaries have gone up a lot, and people are raiding each other, because even there it's a finite pool of people, and a lot of them, once they get going, they decide to come to the United States.

MR. GURLEY: To get the money, obviously. Let's move on to some of the personal insight questions, here, to close up. There's been a lot made about Microsoft's turnaround to the Internet, the 180-degree turn. Inside the company, was it that big a deal? I mean, did you go through a huge learning experience as a CEO, in making that happen?

MR. GATES: Yes, I think it's fair. I mean, it wasn't like one day nobody was discussing the Internet and the next day everybody was. It's been dramatized a little bit. But, we were very focused on getting Windows 95 done. And we were looking at the Internet and kind of saying, wow, it's really starting to heat up. There were a few people, like Steve Sinofksy and J. Allard, who were sort of agitating. You know, Steve had gone back to Cornell, seen what was going on, and these were people who said, hey, this is really spreading beyond just the university environment.

But once we got Windows 95 done, we took all those people who had been saying that, went out on a retreat, and we heard it loud and clear. And we said, wow, we better get going, because this -- the dream we always had of all these machines being connected together, it's happening. And if we don't do the right thing, it's going to happen without us. In fact, it could replace all of our products, because this is such a key capability. And so we gave everybody access to the Internet. We said, get out there, play around with it. Try it out. Form your own opinion. And we had several months where it was kind of chaotic, where everybody was saying we should do different things, everybody had their own thoughts.

We did a few retreats, to pull the ideas together. Then I wrote the memo, which is the Internet Tidal Wave, that condensed all of it together, and it was sort of a rallying cry. We were able to go public by the end of '95. And set some stakes in the ground, say to people by certain dates we'll have a browser, by certain dates we'll have the Web server. And then we could be measured against that. And it was exciting to see the energy that was developed inside the company, because when you have something like that, something that's an opportunity and a death threat, it kind of brings out the best in people.

MR. GURLEY: You wrote two articles on your New York Times column about a good manager and a good employee. I was wondering if you could maybe just give a few high points of each of those, in terms of how Microsoft thinks about hiring?

MR. GATES:We're very big on managers who are very much in touch with doing hands-on work, who appreciate the work that people underneath them are doing, and retain the skill sets to jump in and do some of it themselves. So they can understand what is the load like, what's hard, how's that going on, and pitch in when there is something that's particularly tough. We're big into managers that believe in a lot of communication. It's awful when you get a group that's kind of drifted, and the morale has gone down, and you wonder, why didn't you find out early, you know, you should always know that as soon as possible. And so managers really have to be in touch with all of their people.

We like people who have got an enthusiasm, for the product, technology, who really believe that it can do amazing things. We're very big on hiring smart people, so you'd better be comfortable working with other smart people, and kind of having the debate, and questioning that goes along with that. We're very big on using our own tools, sharing information that way, so that we can make it feel like a smaller organization, even for the parts of the company that are not, up at our main campus.

And we've been very lucky, because good managers tend to hire other good managers. And so we've had a tradition that at least so far has built on itself.

MR. GURLEY: You also discuss this infatuation with bad news, that I found intriguing. I wish all of our portfolio companies had the same infatuation. Could you talk a little bit about that?

MR. GATES: Actually that's, of all my columns, that's my favorite column. And if somebody was going to read one, that's the one I'd say to read. There is a tendency in companies to let good news travel fast. Oh, we just won this account. Oh, things went so well. But, the thing about good news is, it's generally not actionable. You just say, oh, we should keep doing what we're doing.

Bad news, on the other hand, is actionable. This customer is not very happy. This competitor is doing something very well. This project is behind. These people aren't all that enthused about their work.The sooner you get the bad news, the better off you're going to be, in order to kind of absorb it, to change your product plan, to go back and talk to the people, really dig into it, and so when somebody sends me mail saying, you know, we won XYZ account for Exchange. I send mail back and I say, does that mean we lost every other account, because you only sent me mail on one account? Tell me about the ones we lost, and why? And so that's really gotten into our culture.

If a project is behind, let's talk about it early on. And that way you don't have to have it build up to be such a dramatic thing, by the time you reveal it people are kind of ashamed of it, or they're worried that somebody might get blamed about it. You can see what's happening there and all deal with it, as a team. And Microsoft has retained this very, very well. You know, it means when you go out into the different product groups, they tell you what they're worried about. When you go out to the different geographies, they tell you what they're worried about. And the electronic mail is just full of things that, you know, we need to change and do better in some way.

MR. GURLEY: That's great. Could you talk about other CEOs that you admire, maybe a little bit about why?

MR. GATES: Okay. Well, here in this area, I think there are some fantastic CEOs. I think Andy Grove is an incredible CEO. He's big on picking objectives, and driving the company towards that objective. He's big on clarity. I mean, he is an engineering manager, par excellence. And he's actually been willing to write about that in his books, and I think those are very worthwhile. He also teaches a little bit, at some of the business schools. And I think that's a real privilege for the people who get a chance to hear him talk about it.

In terms of an inspirational leader, Steve Jobs is really the best I've ever met. I mean, he can make people work, you know, more than they should. He's got to be careful, it's such a strong power, he can overuse it. You know, I always say to him, he's a first-class magician, and I can recognize him, because I'm kind of a second-class magician. It doesn't mean I can do what he does, but I can kind of tell, wow, that's powerful stuff. And he -- you know, he's got a belief in the excellence of products. He's able to communicate that. And in the creation of the Macintosh, you know, I'm sure he wants to equal that and go by it. But, as far as I'm concerned, you know, what he did there was just unbelievable. He drove that team to do something that was a fantastic contribution.

MR. GURLEY: Are there any non-tech CEOs that you --

MR. GATES:One of the best thinkers, and best business leaders and, you know, most fun people that I've ever gotten a chance to meet is Warren Buffett. He's kind of the opposite of a high-tech CEO. He sort of thinks about things. He doesn't like investing in businesses where the rules can change on you overnight. So he doesn't invest in any of the companies we've been talking about. He likes businesses like Coke, where he sort of understands, people liked Coke, you know, 50 years ago, they're going to like it 50 years from now. And you can sell it everywhere and, you know, you always are ready for more.

MR. GURLEY: Has anything you've learned from him affected the way you run Microsoft?

MR. GATES: Absolutely. I think Warren has had more effect on the way I think about my business and the way I think about running it than any business leader. He's got, you know, a way of thinking long-term, a way of analyzing the business fundamentals that -- you know, the way he does it, he makes it all seem so simple. Well, of course, you know, he's analyzing in all these factors, and he's thinking way ahead of everybody else. But there's an immense value gotten out of that, as well as enjoy him as a friend. He's just an incredible person.

MR. GURLEY: Just a couple more questions, and then we'll wrap up. We've got a lot of emails about your philanthropy process, and what your thoughts are. You said you're going to wait. People want to know why you're waiting. That kind of thing. I know it's a tough question, but we got a lot of emails.

MR. GATES: Well, that's a very fair question. You know, I'm 42 years old and, at this point, I've given a little over $500 million to two foundations, and one of those foundations is called the Gates Library Foundation, and it's got a simple goal, which is that every person, particularly every child in the country, should be able to go to their library and not only get great books, but also to have a state-of-the-art PC connected to the Internet, so that they can reach out to that world of knowledge that certainly privileged kids are going to have access to. Everybody should be able to do that.

And that foundation has picked some of the poorest states in the nation to start out with. I'm going down to Alabama in a couple of weeks, because the installation in all the libraries in Alabama has been done over these last four months. And so, we're going to move on state-by-state until this vision that everybody ought to have equal access to that capability to pursue their curiosity and be part of the great things going on there, until that's done. So that's one of the foundations, and that's a fun one.

(Applause.)

MR. GURLEY: All right. The last question comes from Cheryl Christian through an email. And maybe ties into the bit Warren talked about. She wants to know what your 10-year goal is for both Microsoft and your private life?

MR. GATES: Well, I don't know about goals --

MR. GURLEY: Is that in your PC?-

MR. GATES: -- I'm sure it's in here somewhere. I just have to download it. Ask my wife what she has to say. You know, in my personal life I don't set big goals. I'm sure 10 years from now, I'll probably have a couple more kids, because I'm enjoying having a daughter who's almost two years old now, that's the most fun thing I've ever done, and look forward to more.

In the business side, the vision of the company was set a long time ago, the computer on every desk and in every home. Ten years from now, that computer will have a completely different shape than it has today. It will be a tablet that you carry around. It will be a screen on the wall that you just call up information. It will be something small like this that's far more powerful, and the way you interact will be very different. And, you know, the goal of having Microsoft software helped make that come true and helped that be used to make businesses more competitive, to help people learn in better ways. Basically, a broad set of things that improve the quality of people's lives.

That goal, which has been driving us forward 20 years now, is unchanged. And I'm really pleased with the progress that's taken place. Machines under $1,000, great graphics, a big software industry, and we're moving at full speed. We're counting on our partners to also move at full speed. That's been a big part of it. But I'd say this is the most fun business to be in, and my job is probably the most interesting that there is. So the next 10 years are going to be a lot of fun for me.

MR. GURLEY: Great. Now, before we wrap up, I want to invite Marshall Burak up here. I think he has something he'd like to present to you from the university.

(Applause.)

MR. BURAK: Bill, we thought you could use something warm on those cold, rainy winter nights up north. (Bill was presented with a San Jose State University sweatshirt)

MR. GATES: Okay.

MR. GURLEY: Great. Thanks, Bill. We appreciate it.

(Applause.)

MR. BURAK: And we thank you for coming to our university and sharing your thoughts. Particularly from the college of business perspectives, we thank you for the support we've received from Microsoft, and literally giving us our Windows to the world.

MR. GATES: All right. Thanks again. Bye-bye.

(Applause.)