Newspaper Association of America Luncheon
Remarks by Bill Gates
Tuesday, April 29, 1997
Chicago, Illinois

BILL GATES: Good afternoon. It's a great pleasure to be here. I'm anxious to do two things. First, to talk to you about the opportunity on how to use technology in a better way, not only out on the Internet but in the operations of your newspapers in a very broad way. The second is I want to undo a little bit of what you heard in the introduction. I want to point out what Microsoft is doing and why I think the newspapers being out on the Web is a great thing and something that you will use integral to your business in years ahead, perhaps with different competitive dynamics, but something that draws on all of your strengths in a very exciting way.

Now, the personal computer connected up to the Internet is a revolutionary device, and I think it deserves to be compared to other big advances in communications. People who think of it in terms of manipulating numbers are missing the whole point of the personal computer. Over the last 20 years the cost of computing has dropped by a factor of a million, and that's taking it from a device that only large organizations can use to something that individuals can work with. In the next 20 years we can say for sure that the cost of computing will drop again by a factor of a million. So everyone will own, for very little expense, computers hundreds of times more powerful than the most powerful computers we have today.

Well, what are we going to do with all that power? We're going to create computers that understand information -- computers that can speak, that can listen, that can recognize handwriting, computers that even with inexpensive digital cameras can see and know what's going on. And so, when you think of the PC of today, that is not the PC of tomorrow any more than the PC of five years ago is what we have right now.

The structure of the PC industry is very, very important. It's a hyper-competitive industry because there is so much specialization -- Intel making chips, but many people competing with it; Microsoft making software, a lot of competitors in that space; and then systems manufacturers and application manufacturers pursuing their goals. Driving that forward and getting those machines connected up at higher and higher speeds will change the world -- first the world of business and then next the way that people operate in their home and, perhaps most importantly, the way that people are educated.

Now, standards have been central to this. We know this is successful because they got a lot of applications and that drove more momentum. The biggest investment we've ever made is building the thing called Windows NT. It's a platform for integrated computing, and it's what's being used in most of the Web sites that are now being put out there. Again, it's very volume-driven. This is a scale-economic business where the marginal costs of selling an extra copy of software is very small, so we can afford to put billions into the R&D of these products and yet go out to a large mass market.

If you look at the newspaper industry as a user of technology, it was a pioneer. You were among the first to go out there and move to digital typesetting, and there's good news there in that you helped bootstrap a lot of the technologies. The bad news is that those systems are very fragmented today. Some of the systems are very old and they aren't very well integrated together; and as the world has gotten more and more digital, the richness of layout and typography and group coordination have gotten so good that your brethren in other industries are actually using digital technology in better ways.

A key point here is that, as the Internet comes along, the importance of integrating these systems, having common systems from city to city, having departments sharing information, easily taking a customer record and seeing what they're doing in print and what they're doing on the Internet in a sophisticated way -- that will require using technology quite differently. So this idea of a platform is very important. It's not just for business needs, but also the editorial needs; and by moving into the mainstream of PC technology, there's a very straightforward way to use inexpensive building blocks, systems that are dramatically less expensive than even maintaining the old systems that are out there.

Moving Windows into the publishing area has been a big focus for us. Color management, font standards, getting the proprietary systems moved over so there's some evolutionary path, and you saw the announcement where ATEX is moving onto this platform.

There are also new developers. A very focused software company is doing great, innovative things and I've just picked a few of those to mention: Unisys in Europe, Pantheon Builder. ISSI has a great editorial system. NewsMaker is the one that's out today, but their new one that's coming out this summer is much better because it really draws on PC technology, standard packages that have been extended. On the application side, people like GEAC have done a full set of applications that run together on this platform. And so we're very interested in working with leaders in the industry to show the way as to how technology can be applied.

Let me now shift and talk about the Internet and what I see newspapers having a chance to do there. The Internet has very low cost of entry. Anybody who gets a PC, and gets a little bit of software, can set up a Web site, and that means that everyone is a publisher. I was out at a very poor elementary school today, and these fifth-graders were telling me how they build Web sites, and they're actually funding the technology being used at their school by going to local businesses and doing Web sites for them.

So you've got to watch out for these fifth-graders. They've got printing presses in the forms of personal computers, and they're very creative in what they're doing. They're growing up taking this technology for granted and that's where you get incredible innovation, where people are pushing that forward.

If you want to see a microcosm of what the world would be like in 10 years through the changes the Internet will bring, you can go to certain university campuses. On these campuses today you can't sign up for a course without using the Internet. You stay in touch with your friends. You get class notes. You receive your grades. You submit homework. If your parents want to talk to you, they'd better learn electronic mail because that's how you communicate.

Once somebody gets involved in using the Web and takes it for granted, then for anything they're planning, any big thing they're buying, any information they want to get, they will view it as a resource -- not the exclusive resource, but a very important resource. I was in Detroit yesterday, and we were showing car buying on the Internet, and one of the nice things there is you can find out what the dealer paid to buy the car. So when you go in to do that negotiation, you're a little smarter than you used to be in that discussion.

You can actually buy the car through the Internet, get somebody to quote you that way. About 1 percent of cars are sold that way today and that will go up pretty dramatically. For about 15 percent of cars, the person visits the Internet before they actually go to do their transaction, and the information is getting richer and richer there.

Now, the primary role of Microsoft -- overwhelmingly -- is to provide these chief building blocks. Windows NT, Office, Front Page as an authoring tool. And things to help you manage your site -- software to rotate ads, software to see what's popular on your site, software to create communities where people can come in and chat about issues. That is becoming very turn-key. Even very small newspapers for a budget on the order of, say, $40,000 or $50,000 should be able to get the communications lines in and set up a wonderful Web site that covers a vast range of areas. There's no need to go to proprietary systems or get locked into anything or even to spend a lot of money. The building blocks are there. And once you have that, it's up to you in terms of editorial creativity and advertising relationships to see where that goes.

Of course, at first what people have done with the Web is primarily take print-based material and simply make it available, so the Web is just a distribution medium to get things out there with lower marginal costs and great timeliness. But as people are getting more experienced, their expectations are going up quite a bit and republishing is no longer enough. Some newspapers are doing wonderful things to go beyond that.

We've got to take advantage of this new medium, and I don’t think anybody knows yet the full potential. I think some of the key things are always to be up-to-date, have as much in-depth information as you want by following links, have an experience that includes audio and video as well as plain text, and most importantly, I would say, the ability to personalize the content. Microsoft feels this is fundamental, and so we're taking the idea of taking user preferences and building an engine that can generate pages for that person and providing that in our product technology. That's available to everyone to use for their applications.

Another key technology is letting people type in English sentences instead of having these key-word search things where you end up getting, you know, 50,000 hits and you're supposed to wade through that. We'll let you put in a sentence and we'll use that, and we'll give you the answers that really make sense for that sentence. Unless you check and say you're a librarian and you want every single site there, we'll find you the few that can really satisfy your interests. And as we've been studying these search requests, people's interests do repeat again and again there, and we can do a great job on that.

The linguistic technology we're applying here is the same linguistic technology that we've put into our productivity tools, into Microsoft Word to do grammar checking or to do spell checking. And so there's an engineering piece of this that is going to make it far more accessible and far more customized.

Now, the Internet is a gold rush. Everybody's investing in it, and, you know, people feel bad if they're not involved and your kid comes and says, "Dad, what's your URL," and you've got to say something. And so you go out there and do something without really understanding what the payoff for that is.

Because of this incredible over-entry where everybody's jumping in, it's pretty slim pickings right now in terms of incremental revenue. Even people who think of the Internet on the basis of not asking it to fund any of their other costs, because it's not yet cannibalizing the other activities -- most people are not at break-even on what they're doing. The only site we have that meets the test to be profitable is a travel site called Expedia, which is a transaction-oriented site. And there are a number of those like Amazon and some others that have gotten the profitability. Subscription-wise, the best number there is The Wall Street Journal, which just celebrated getting to 100,000 people.

But The Wall Street Journal is the best case. They're the most branded, unique, high-demographic site. People who use computers need up-to-date information, and it took some effort to get there. So that's going to take time to develop. If you look at something like news on the Internet today, there are so many choices out there. You can go to an AP site and get it directly. Don't go through any intermediaries, just go to AP. Go to dozens of newspapers. Depending on what news you're interested in, you have lots of choices.

Here in this medium, there's not the clear separation that there is in the physical world, so you're going to have local TV affiliates putting up news sites. You're going to have the Yellow Page people putting up directories and what are essentially classified ads. You're going to have software start-ups doing lots of things. And this gold rush-type atmosphere will stay around for the next four or five years.

At some point, things will shake out. People will figure out what are the revenue models, and they won't fund start-ups at quite the aggressive level that they have here. Now, for customers, this is great news because it means the amount of money going into building the tools and experimenting is accelerating the market in a way that a rational set of investments might not, and so we're going to go through what should have been 10 years of evolution in about three or four years.

The newspapers are in a very strong position here. People want depth. People want high-quality content. Advertising will be the primary revenue for most of these sites. People want someone who's got the discipline in publishing under time pressure. And the brand recognition is a very important thing that newspapers bring to all these aspects, I believe, particularly in areas like classified ads. The ability of the paper to offer a print-based offering that includes being up on the Web will be a dominant offering for a long, long time.

Although eventually there will be public kiosks around your community where you can go and browse the Web, it will be at least a decade before most classified advertisers would say, "No, I only want to sell to the people who are living this Web lifestyle." Now, maybe if you're near to some university where you're selling to the student market, it could happen well before that. But there's time to learn here, time to get these things right, and newspapers come in with a lot of very strong advantages in doing these things.

Microsoft's strategy here is to learn and to continue to play a primary role, which is building software technology. We are an over $12 billion-a-year software tool company, and that will be the primary source of our revenue, the primary source of our profits forever. That is what we are good at. Now, we're going out and pushing the boundaries here a little bit to learn how we can use technology -- primarily personalization technology, but also somewhat the linguistics.

We got in with NBC to do interactive news sites, and that's done very well. We're getting a lot of traffic there. It's a big investment. Part of the reason we worked with NBC was to be able to use some of their video. The bandwidth on the Internet doesn't make that very simple today, but if you take a 5- to 10-year timeframe, that will be far more common. And so there will be some benefits out of that.

We are doing an entertainment guide, a guide that tells people what they can do. If you're interested in it, you should go look at what we have out there. It's more of a competitor to a weekly entertainment guide than it is as to what goes on in the daily newspapers. It's called Sidewalk, and we're doing some innovative partnerships there with Sidewalk and we're learning. Over the next year, we'll be out in about eight different cities and see what kind of reaction we get there.

Expedia is the travel site, and as I say, that's come the furthest along. It's mostly based on technology. We've worked with American Express, who's selling that to the corporate market. So don't think of Microsoft as a primary competitor here. Think of us as somebody who can provide technology and in some of these areas like Sidewalk, you know, maybe there will be some overlap, but it's not core overlap. We're not doing local news, we're not doing classifieds, and we're seeing where this technology can go.

I listed a number of people we're working with. One example is a weekly paper in Seattle, The Seattle Weekly and Eastside Week. They do a very similar thing to what Sidewalk does, but they do it in print. And so we decided we could share information with each other and be a value to each other. So that's a very good partnership. It's getting us going there. Obviously NBC, and then I list [on a slide] a number of other people we're doing things with people on.

In fact, a lot of the partnerships we're doing now are outside the United States. There's a lot of flexibility to say, okay, in the first few years of this, we're willing to do it differently than we might do it in the long run and both learn together how the business is going to work.

One example of a newspaper that I think is doing a great job using low-cost technology, mainstream technology is Philly Online. One of the things you quickly learn on the Internet is those stories where you just take what's in the newspaper and put them up, people aren't very interested in that. Stories where you get the interaction and you say, "Okay, under this tax proposal, enter what your income is and see how it affects you," or "Type in your ZIP code and we'll show you whether they are giving out a lot of traffic tickets in your area." Things that are interactive and draw on what that person cares about are great.

On the NBC site now, we take your ZIP code and we show you how your congressman voted on the issue that's being covered in that story. We'll put it right there automatically just because we know where you're located, and then we'll put a link in -- you can click on that and send e-mail telling your congressperson or senator whether you're with them or against them on that particular area. So the idea of being innovative and going further is very, very key, and Philly Online is somebody we've worked with to provide the technology there.

So in summary, I really see an opportunity for you to use technology. An online newspaper is not enough. We've got to take it further. You've got some fantastic strengths. I think in terms of local information, newspapers are going to lead the way and the Internet will be part of your portfolio of businesses. It won't replace the other things you do, but it will be in there and be important.

There will be a lot of competitors, particularly in these next four or five years, but you shouldn't get overly paranoid thinking that somebody's a broad competitor and it's not possible to work with them in some ways when that's not what they're doing. If somebody starts hiring local reporters, okay, then you should get worried. Then they probably are trying to duplicate the whole thing. I dare somebody to do that. It just wouldn't make any sense.

But a final point and perhaps the most subtle one is that to be ready for this new world, you've got to use technology in your current operations. If your employees don't have electronic mail and aren't sending mail around to each other, if they're not going out to the Internet to research things and find things out, both on the business and editorial side, you're missing a great opportunity.

We're at the beginning of the information age. I'm sure many of the people here are going to be the pioneers who show us what the potential is and we're looking forward to working together on this. Thank you.

[Moderator is Arthur Sulzberger, Jr., publisher of the New York Times.]

MR. SULZBERGER: Thank you, Bill. Based on the fifth-graders you met, we clearly need to rethink our newspaper business. Bill has graciously agreed to answer some questions. We're going to go until this audience runs out of questions, he runs out of answers, or we hit 2:25, whichever comes first. … Since I've got the mike, I'm going to ask the first question.

Bill, close your eyes for a second and imagine if you can that you are not the chairman of a multi-billion-dollar software company, but rather the publisher of a newspaper somewhere in the United States, maybe even in your hometown of Seattle, that we're your friends. What would your electronic strategy be?

BILL GATES: Well, I'd certainly have a presence out on the Web today. I'd certainly want to be able to become a place where the community turns to chat about things or to find things out. I'd certainly want to think about how to present traffic and weather, those things that are often changing and people are going to come back on a regular basis to get.

The frequency of how people visit makes a big difference in terms of whether something can be advertised or funded or not. If you only come to see something once a week, it's almost impossible to make the business case. If it's something that people look at multiple times a day, then it's great. Weather, in a sense, is the most profitable thing on the Web because you get it for free and people look at that quite a bit.

The thing I would do is try and share costs between my print side and my Internet site. I wouldn't have an Internet division and I wouldn't have an us-versus-them type thing. I'd make sure everybody in the paper used PCs, used software, and probably in the process take a lot of cost out of it and put a fair bit of efficiency into the thing. So then I can dial in how much more investment I'd put on the Internet site as the business grows.

I'd go ahead and put classified ads up there right away because anytime you enter the classified in for the paper, you can just boom, have it automatically go up onto the electronic site. And I would do some fun things. I would create visibility for my Web site by labeling things as experiments and then being willing to try those out.

MR. SULZBERGER: I'm going to have to trust that Lincoln Milstein (sp), the vice president of new media for -- (inaudible) -- is out there. Lincoln?

Q. Thank you. Mr. Gates, now that you're one of us in the publishing business, so many of the decisions that we make every day is based on our own personal case and our own intuition. I was wondering if you would indulge us and tell us a little bit about your personal tastes as a news consumer. Would you take us through a typical day or a week in the life of Bill Gates? Would you tell us whether you read newspapers regularly and, if so, which ones? Will you tell us whether you watch the news on TV or if you read magazines? But more importantly, will you tell us as a consumer how you would change those products to meet the challenge of the future?

BILL GATES: Okay. The first thing I want to say is that my personal tastes are not typical, and in terms of how we design our products, we're not aiming for this market of one. When we do software products, we're aiming for a very broad marketplace. I do read a lot of print material. I read The Economist front to back. I read the Journal, not every article but a lot of it every day. I read the local Sunday newspaper, The New York Times. I read most of the business magazines. I read science, Scientific American. At work I read a lot of trade journals. I only get four trade journals at home, and I get about six at the office.

I'm a little split now in terms of how much I read off the Web versus how much I read on paper. Take the Journal, I can get it electronically, but I just have the habit of reading it on my exercise bike, and I need to get a LCD up there and see that. There's a key point, which is that screen technology, as that improves, it will change this trade-off between what you read in paper and what you don't read in paper. And 10 years from now, screen technology will be dramatically better than it is today. There's also a theory of "distribute electronically and print locally," which somebody like HP, who's in the printer business, of course, will try and push that as much as they're able to.

But if you take computer trade journals, I now have switched to read, say, PC Week much more online than I read it in print. And it's a little disconcerting, though, because when I read it online it marks which articles I've read and everything, and when I get to the print it's enough in a different format, I find myself paging to an article and saying, "Oh, no, that's one I already read." So I think maybe for trade journals, I'll flip over and at some point be primarily electronic in doing that.

But if you take The Economist -- that's a pleasure thing. That's nothing to do with business. That's just goofing off when I read that. So I still do that in print, except they can't get it delivered to me until Monday or Tuesday, and so I do go up on the site, their site, on Friday and search to see what they have about businesses I care about or technologies. So that one slice of it I'll go and read there.

I'm an avid reader of Slate Magazine, which is an online thing that Michael Kinsley runs at Microsoft, and I use Expedia to do travel and Amazon to buy books, and I've found a lot of nice gifts for people out on the Internet.

MR. SULZBERGER: Yeah, we all goof off reading The Economist. Somewhere in the back, I've been assured Bob Ingle, the vice president of new media of Knight Ridder, exists and wants to ask a question.

Q. Mr. Gates, I was interested in your remark that perhaps we shouldn't be so paranoid about working with a company on the one hand of software products and platforms and so on, but who may be competing with us in other ways. And your friend, Andy Grove, of course, says only the paranoid survive. So I think that many of us would cotton to being paranoid. You also said we shouldn't worry until we see companies going out and hiring local news reporters. And as far as I can tell in Sidewalk, that's exactly what you're doing.

BILL GATES: No.

Q. You're not?

BILL GATES: No.

Q. You're staffing each of these sites.

BILL GATES: Not with reporters, no.

Q. Well, it's a thin line between people who collect listings information and reporters. They are --

BILL GATES: Boy! Do reporters know that they're just list gatherers? I mean --

Q. Are critics reporters?

BILL GATES: Well, a critic is not -- a critic is something where you can get an overlap between an entertainment guide, which is what we're doing, and what's in the newspaper. Certainly you have the movie listings and what their times are and things like that. So that portion of the newspaper which overlaps the weekly and overlaps other things, Sidewalk intersects with that, don't get me wrong. But that is not -- and this is about people who are interested in going on the Internet and having us recommend things to them based on a very database-driven approach, and a very profile-driven approach there.

Q. Well, in any case, however you look at that part of it, selling advertising and directory listings is very, very closely akin to classified advertising. So I guess my question would be, since there are alternatives to using the Microsoft platforms, many times as good and in some cases superior, why should newspaper publishers contribute to your principal revenue stream when they see you coming in as a competitor in the sale of advertising?

BILL GATES: Well, there are many people who sell advertising. I assume some of you watch TV. If your question is, should you make irrational decisions, my answer would be no. You should buy the best product to get things done. I come from a business where everybody is a competitor with everybody else, and everybody cooperates with everybody else. And people are very smart about knowing when they should do a partnership with somebody and use their technology and when they're in competition. That's just how it works every day.

And so I think in picking technology, it would be unwise to pick inferior technology just because of that. If you think something is better, great. I think that's fine. You shouldn't tilt in the other direction. But the opportunities for taking advantage of our technology, I think we have a very strong case there, and that's our primary business. I mean, if you look at the employees of Microsoft, that is what we're out there doing every day. And the only rational decision where you wouldn't buy the best product is if you thought you could, by denying your purchase dollars, put a company out of business. And that just doesn't come up very often.

MR. SULZBERGER: Not nearly enough. We're going to move to questions from the floor now.

Q. What is your view of what the print product will look like in 10 years and in 20 years?

BILL GATES: Well, I don't think my crystal ball is that much better than anybody else's. I doubt the print newspaper will look dramatically different 10 years from now than it does today. It will probably have more URLs. It will probably have more subscribers who are talking to you about what they want through electronic mail. You'll probably treasure the fact that you have the electronic mail addresses of a very high percentage of your subscribers, since most of them will be on electronic mail. I think you will have gotten to the point where your Internet revenues had better be funding more than just the Internet marginal costs. It had better be funding some part of the basic costs of the business as well, when you get to that point. But I don't think the print newspaper will be substantially different than it is today. Everything you create there you will be using in electronic form as well to make money.

MR. SULZBERGER: Sorry, was there a question that follows on that?

Q. Yes. I mean, why won't it be different? Given the opportunity of the Internet, why won't it look different?

BILL GATES: Well, you're asking me about the print, why the newspaper would look different. But it's like saying, "Will plays be different after movies?" or "Will TV be different?" I mean, the mediums don't change just because a new medium comes in. People like the newspaper today. They like what they do with it. So I'm not even sure why you're saying the paper would be different just because some information is coming in interactively.

MR. SULZBERGER: Well, you've made the newsprint manufacturers very happy.

Q. Mr. Gates, you recently addressed a convention of the broadcast industry and commented that their drive towards HDTV might make -- (inaudible) -- because Microsoft has a better idea. Could you recapitulate that argument?

BILL GATES: Okay. Well, I didn't speak at the broadcasters convention. There was a person from Microsoft who spoke at a subpart of it, which was the multimedia element. There's a bootstrap problem for high-definition TV, which I think you can probably appreciate very well, which is, when you're funded by advertising, a high-resolution Coke ad does not sell more Cokes than a low-resolution Coke ad. So when a TV station thinks, "Oh, great, we get to buy all this new equipment and do all these new things in order to transmit this better signal for no additional revenue," it's tough to get the bootstrap. Plus which you have a very high percentage of the people who get that local station getting it over the cable, and there's an old approach to doing this, a way of encoding those signals, that was coming along for a long time ignoring what was going on in the computer industry. And so now there's a question of, should that old approach be the one that those guys broadcast or should they use this other approach. It has to do with interlace versus non-interlace. And it's very much an engineering kind of debate.

PCs like screens [on which] all the lines are together rather than first odd, then even. And so there will be a lot of PCs that can receive that progressive scan capability and can't receive the interlace. And so we're talking to people about which one they should broadcast. Should they broadcast multiple, should they include the ones that the PCs happen to support. But this will be worked out by engineers, believe me.

Q. Incidentally, you'll be happy to learn that we are NT users at our newspapers, and one of our problems we're facing -- by the way, we do have a Web site, and we also run our classifieds, and we've done very well financially on the Web. My question is, we're looking for software that we can seamlessly move the print display ads onto the Web without a lot of human intervention. Are you working on a software package to do this?

BILL GATES: Absolutely. Between ourselves and third parties, we'll make that very easy for you. In fact, the whole way that your advertisers submit display ads to you, there's a way to make that far more electronic and take a lot of your costs out of the system and, at the same time, get it up onto your Web site in its full richness. And so we should chat with you about how that's done, because that will be standard procedure.

Q. I write a column for editors and others online. I'd like to ask you to clarify -- by the way, thank you for lunch, and I can report in my column today that the condemned media ate a fine meal. Your MSNBC site, which is one of my favorites for getting news, satisfies my desire for serendipity when I make a personalized version, because no matter what I do with your site, it forces me to have something called weird news, which I find wonderful because if I would structure my personal edition, I would eliminate all the things I think I don't want to see and therefore miss that story on something when I'm reading The Times and The Journal -- (inaudible). However, your MSNBC had a poll a month ago, a month and a half ago, that says that your viewers, 65 percent reported, that by the year 2000, 65 percent of the surfers to MSNBC expect to get all their news from the Internet, which flies in the face of some of the kind words you've had.

BILL GATES: Wow. That must be a strange early adopter audience there. I mean, when you think of news, you're going to turn on the TV set and watch Tiger Woods make that last shot. You're going to pick up that Sunday paper and see what's going on there. You're going to pick up that entertainment guide and flip through it to see what you happen to want to see there. Because the Internet is a gold rush, the opportunity for somebody to over-hype it and say, "Oh, overnight everybody goes out of business, no one will ever go to a store and buy anything ever again," or the opportunity to write the article that says, "Oh, it's worthless; it's hopeless. It means nothing. It will never be popular." You know, we're just going to go through those cycles.

Those articles, both of them, will be written thousands and thousands of times between now and 10 years from now, which is how long it will take to get the content better, to get people used to it, to get the devices easier, to get natural language into it, to get the prices down of all these things, and make it so, at that point, it will all be taken for granted. [Then] no one will ever write an article about it again, and everybody will understand that you have all these skills that carried over into that new form: Okay, of course, that was destined to happen. And we'll move on to the next thing. Probably discussing artificial intelligence.

MR. SULZBERGER: A point of clarification, we paid for this lunch. We may be slow, but we're not cheap. Is there another question out there?

Q. Hi, Matt Cohen, chief technology officer for New Century Network. As we build new products for the new medium, new news and entertainment and other products, we take forward lessons that we learned in other media, and you, as well as most of the companies that print them, are in the process of learning those things. And as we move forward, we also bring forward things that we learned that were important because of restrictions and limitations of the old medium, and that's the natural process. You mentioned that the first thing that happens with a new medium is you re-purpose what's going on in the old medium. So, as someone who is experimenting with new methods of delivering information, my question to you is, what things -- well, let me back up for a second. …

Somebody mentioned to me that their daughter had asked them why call it dialing a phone number. And that makes clear how some of the things that we take completely for granted, that are very, very ingrained in the way we think, will vanish and become completely invisible to the next generation. So my question to you is, what are the things about the way that news in particularly is done today, from where you sit, that will vanish and become invisible in the next medium?

BILL GATES: Well, every day msnbc.com is our experiment in seeing what people like. And the reader feedback thing where you can go in and rate a story, and you can see how the other people are rating stories, that's incredibly popular. Having audio clips is just incredibly popular. Now, you have to be a little bit careful because when people first use a new medium, there's kind of a fascination with it that isn't projectible out to the next audience that uses it. Even these early users are going to stop goofing around in some ways.

For example, I think the percentage of people who go out and use search engines and type in crazy queries, I think that will go down a lot because we'll be able to customize for people a much higher percentage of the information that they're interested in. So, in a lot of ways, I'd say it's too early to tell. The news ticker that we've done, where you can have the stock prices and the news stories, that's incredibly popular. We get a heck of a lot of our traffic that way. We have a site called investor.com that costs us about a 20th as much to create as our news site, and has traffic about the same. In fact, Investor is very close to being a break-even site, which says to you the news site is not. But, when you want to learn about these things early, it requires that kind of investment.

But I'd say it's really too early to tell. The average age of the news reader of MSNBC is much lower than the average age of a newspaper reader. And that, again, does that continue, or is it just that early audience phenomenon?

MR. SULZBERGER: Well, perhaps to no one's surprise, we ran out of time rather than out of questions and answers. So I want to thank you, sir, for joining us today. But before we say good-bye, I'd like to perform one small experiment, about Sidewalk. How many people in this room publish some form of a weekly entertainment guide, raise your hand?

BILL GATES: No, but here's the point. In the only city we're in, we're in partnership with the print-based entertainment guide. We found a way to share costs to help each other out. And we are not going to do a print entertainment guide. And so, we'd love to talk to people who do that about whether there is a way to help each other.

MR. SULZBERGER: Good. Love it. Thank you.