This week, CNET's Tom Krazit and Molly Wood interviewed Google CEO Eric Schmidt. This is a transcript of their conversation.
CNET (Molly Wood): Hi, I'm Molly Wood. Welcome to another edition of CNET Conversations. Today, I'm joined by CNET senior writer Tom Krazit. And we are here at the Google Campus, where we're very excited to be speaking with CEO Eric Schmidt. Thank you so much for having us!
Eric Schmidt: And thanks for having me on.
CNET (Molly Wood): Now, I was feeling, preparing for this interview, like we should ask you about search right out of the gate. But, honestly, it sort of feels like the search wars have been won. Do you feel like, with your competitors focusing on how to display, in order search results, is there an area where you're playing defense? Or do you feel like you kind of have this one well in hand?
Eric Schmidt: Well, we're always trying to make search better and we've got a lot to do. So we're certainly not done. And if you think about it, we still don't get the perfect answer to the right query every time. We're trying to go from, sort of, what you typed to what you meant. So we're using a lot more about you; where you are, if you have a phone, your GPS, your search history, those sorts of things, to really move from, sort of, text to meaning. And that's not a solved problem and it's something that's a great challenge.
We're also trying to make our search index much bigger. We don't have all the Web, although we believe that our search index is much larger than everybody else's. So we're constantly, constantly trying to get more data, more ranking, more information.
CNET (Molly Wood): And do you think going forward, it's realistic to sort of keep putting that 70 percent of resources towards search?
Eric Schmidt: We're certainly continuing with the 70-20-10 model, that makes a lot of sense to us. If you think about it, think of all of the information, the exposure that's occurred in the last few years, we have to get that into our index as well.
We have to get all of the real-time information and all the social-networking information, all of the breaking news information, that now drives so many people's use of the Internet, that's something we're adding.
So, over and over again, there's more information to be added, ranked, and otherwise, presented. So, we're not done and we're not going to be done for a long time.
CNET (Tom Krazit): Is there a law of large numbers there, though, where you have some sort of, you know, as you continue to add more and more data to the index, where putting the 70 percent of resources into it, it becomes harder and harder to do?
Eric Schmidt: Well, as Google gets larger, we have more resources to apply to it, so that's not a problem. What's very interesting is you could imagine, hey, we have enough of the Web now. We don't need to add more. You know, we have the important part. And the problem ... and that's not our view. And the reason it's not our view is when we look at what people are looking for, they're often looking for the tail information. You would think that, based on popular culture, that everyone cares about the stuff that's popular. But our data shows that people are looking deeper and deeper into the Web for even more specialized information. And that's why our job is so hard, but also what we're focusing so much of our resources on.
CNET (Tom Krazit): OK. So, to shift gears a little bit. This has obviously been a good year for Android. We were just talking a little bit about the pending Droid, everyone's waiting for theirs to arrive. And I sort of wanted to ask you, you know, now that you've been, this has been about two years right now, since they first announced the project. Where do you see things in three years?
Eric Schmidt: Well, it's very hard to know what happens, even in a year in our industry. But, Android was designed as an open platform that would create a whole bunch of things at once. There will be a lot of hardware choices, there will be a lot of operator choices, there will be a lot of networks that would use it, and there will be a lot of applications that would run on top of Android. It looks like we're on our way to getting all of that done in the next 12 months.
If you look at the number of operators, the number of networks, the number of hardware devices, you have more choices with Android than you do with any other, any other solution. And although we don't have the most number of applications, we're certainly working on that. And we're working hard to get even more people to build to the platform.
What's interesting about Android is because it's a powerful operating system, is it really is possible to write applications that are PC quality for your phone. And that's, I think, what's so revolutionary about it.
CNET (Molly Wood): So, given that fact, actually, since Android is taking off as well as it is and it does seem to be crossing multiple devices, what's happening with the Chrome operating system? And do you still feel like that's something that you need?
Eric Schmidt: We do, in fact, the Chrome OS is a different product in a different target market. One way to think about it is that Android is really targeted to people who are phone-centric. We all understand what phone-centric means. It's a mobile device. It has a Wi-Fi connection, you carry it. If it isn't a phone, it's something like a phone, a tablet or something like that.
The Chrome OS is really targeted at the PC Netbook-centric user and that's somebody who is at a desk, they've got a keyboard, or something like that. They've got a reasonable screen and they've got a good processor. The uses are different and we don't think that the two completely overlap.
CNET (Molly Wood): And so where do you see the Chrome operating system in three years?
Eric Schmidt: Well, of course, Chrome OS is not shipping yet. We hope to release the first open-source version of Chrome OS later this year, the year 2009. So it will ship sometime in 2010.
And in the open-source movement, the trick is to get it out at just the right amount of quality, so that people then extend it. They finish it with you, if you will. It becomes a better product because of the open-source movement.
So, sometime in 2010, we think we're going to have a very, very effective Web, Chrome-based operating system for Netbook and Netbook category devices. We think it's going to be very successful. It's too early to tell exactly what its usage power will be.
CNET (Tom Krazit): Well, what, I'm sorry Molly ... but one of the interesting things about Android though, is that it's being used in a variety of different devices, beyond mobile, right? You've seen some set-top boxes, you've seen other types of things. How do you avoid conflicts between what Chrome OS is, you know, winds up in and what Android does?
Eric Schmidt: I would prefer not to prejudge the success of products. My experience with products is that they tend to find their way. They tend to find where their niches are, what market they really do very well in. Their design structures, the two operating systems, are sufficiently different. They each will find the market for which it is most natural. So, sure, there are people building set-top box versions of Android. We love that. That's the purpose of open source. And I would assume that people do very interesting and novel things with Chrome OS. Again, a great thing about open source.
CNET (Molly Wood): So I guess, what I'm wondering is, I understand the differences between the operating systems, between the two versions of the software. Is there enough of a difference between the devices? I think there's a sense that, you know, the Netbook category is really going to become the smartbook category, where they are expanded smartphones. And then I wonder if there will be so much overlap.
Eric Schmidt: Again, I think you're smart people trying to think about the future. The future will unfold as it does. And we will have opportunities to decide and discuss what is appropriate and what is not. Let's not prejudge what these things are best used for. Let's build great technology, get the open source community going, have a great application strategy and our end users will ultimately judge. They are the judge of this outcome, not you, and me.
CNET (Molly Wood): The future will unfold and you'll be there.
Eric Schmidt: That's right.
CNET (Tom Krazit): So, speaking of future and things you have smart people working on, what's Google Wave? How would you define Google Wave? That's a question we get a lot from our readers.
Eric Schmidt: Google Wave is a similar, new effort within Google to try to redefine the way people communicate, primarily using e-mail and instant messaging. And what's neat about Wave is once you're in the Wave, your usage, based on our testing and our early use, is very different. It's the blend, if you will, of electronic mail, Post-its, updates, obviously, there's a lot of sharing, use videos and photos, and so forth. And it looks like it's going to be quite successful. Again, how would we map that against the existing e-mail systems? We don't know yet.
CNET (Tom Krazit): Do you think there'll be more growth with Wave, as, you know, people using the underlying technology to build systems, versus you know, the Google Wave UI and experience itself?
Eric Schmidt: Well the Wave team would say that both is going to happen, that they are both a platform for people to build applications that will use this information that's very communication-centric. And also, that they will have a lot of users. Again, at Google, we're happy to have both occur and it's certainly growing very quickly right now.
CNET (Molly Wood): Now, it's obvious that you're venturing into a lot of areas as a company, as we've just talked about. How has your, obviously, your lead in the search market gives you a lot of insights, a lot of data into consumer behavior, developer behavior. How do those insights help you in terms of just introducing new products? Do you feel like you know what people want?
Eric Schmidt: The link between scale size and search is not as obvious as you might think, because the search problem is so massive. So yes, we have Google Insights, which we, of course, publish and so forth, but ultimately, we make product decisions largely based on our own judgment and based on the feedback that we get with the local communities.
We've not yet been able to harness, if you will, all of that knowledge that's in there. And I also think it might be somewhat not OK for our customers, if we started to use that kind of information for other reasons. So we've been careful to keep the search information largely within improving our search algorithm.
There's no question that our search algorithms get better with more information; our ranking gets better, our knowledge gets better, and of course, our algorithms are going to constantly be in tune. We also use search history to, if you will, back solve or literally check the search algorithms to make sure that they're very good and they, we continue to make improvements on daily or weekly basis in search.
CNET (Molly Wood): And do you have a hard and fast line in terms of protecting that data? Is there, is it siloed so to speak, or is it kind of your best judgment as a company?
Eric Schmidt: Well, it is, of course, technically siloed, because it's all in one place, but it's ultimately a judgment decision. And the sense that we use is we use the feeling of our user. How would our user feel if this information were used this way or that way? And all of our indications are that users are very happy to use their data to make search better, but they don't really want it to be used for much else.
CNET (Molly Wood): That makes sense.
CNET (Tom Krazit): So, I mean, one recent development that we've been talking about all year is sort of the combination of, you know, some of the increased scrutiny that you've faced from the federal government as well as the Google Books process. I mean we, you know, saw yesterday that the deadline has been extended for you guys to submit a revised settlement to the New York State, or excuse me, New York federal judge, following a meeting you had with the Department of Justice on Friday. And I was wondering if you could tell us, you know, what does the Department of Justice still object to about the revised settlement.
Eric Schmidt: I'd rather not go into the specifics of the DOJ's feedback. We understand that a lot of people care a lot about this settlement. It's pretty important to get this settlement right. My personal belief is that the settlement will be modified somewhat and that it will be approved. And so we're doing the work necessary to get the details right.
Whenever you have a settlement between roughly 20 or 25 parties, it just takes a fair amount of time for everybody to get comfortable with the details because the details really matter here.
CNET (Tom Krazit): Can you address some of those details that you're looking at revising from the original test.
Eric Schmidt: I would prefer not to. I think that this is a better discussion among the lawyers. It's highly, these are highly legal issues. From our perspective, I think it's worth saying that we were the ones who were sued. We were the ones who entered into a settlement. This is the way our system works and we are happy with the proposed settlement and we hope the judge will be happy with the modifications that we made.
CNET (Tom Krazit): Why is this Google Books matter so important to you as a company?
Eric Schmidt: Because the mission of the company is all the world's information and no one has attempted to go out and get all this sort of trapped information stuck in libraries until Google. And it set off a whole bunch of flares, if you will, people who are concerned about this issue or that issue. But for all those people who are sitting there consuming the, literally, millions of books of information that are now available in the Google Index, we think it's crucial that this activity be, both continue and be legally permissible and also not just in the United States, but also remember we have to deal with non-U.S. libraries as well.
CNET (Molly Wood): How, so, since there is this sort of increased level of government scrutiny in general, Google Books and generally speaking, how does that affect your product development strategy?
Eric Schmidt: We've decided to explicitly make sure that it does not affect our product development strategy because we believe that, as a company, we are acting in the best interest of our end users.
So the way we approach a problem is we say, is this in the best interest of our end users. And usually, the regulatory issues are resolved in a positive nature for the end user. Most of the complaints seemed to be inspired by either industrial structure issues or competitor issues, and those are appropriate complaints to take a look at. But as long as we're benefiting end users, we think we're going to be just fine.
CNET (Molly Wood): Now, that leads pretty nicely, actually, to I want to talk a little bit about evil. I often see you get asked about this notion that you can make money as a business without doing evil.
And it seems possible that, once you get to a certain size, you may always be evil to someone or you may have set up what is an impossible standard. And I guess, I was just wondering if you ever wish you hadn't put that in there?
Eric Schmidt: No, actually, we're very happy about the evil or the lack of evil comment. My own view on Google and where we are now is that we are facing the kind of criticisms that you're referencing because we are a company that is a disruptor. And we're also a company of significant scale. And I think we all agree to that. And we're in the information business and people care a lot about information for many legitimate reasons.
And, by the way, the rules differ from country to country, so we have to deal with different standards, different social standards, different cultural standards for what people want. So, from our perspective, the principle of do no evil was not a rule, in the sense that there was a defining line and you may do this and you may not do that. But it was more of a practice that if something appears evil, it's OK in our culture for an employee of any kind to say, anywhere, to say I think that's evil. And it forces the conversation.
So it's a way of our culture, if you will, correcting around, maybe we're making a mistake or maybe we're too aggressive here or we're too greedy here, or what have you. And as long as we're focused on end user benefit, we should be just fine.
CNET (Molly Wood): And it's so interesting because it feels like that's always been your position and yet, you make people kind of nervous. What do you think it is about Google, at this point, that makes people feel generally uneasy?
We had a lot of user questions just saying, "I'm scared, you have all of my stuff." You know, "are you Sky Net?" What do you think it is? Is it size, is it the information?
Eric Schmidt: I think it's a combination of size and scale and we work hard to communicate our values as a company. We also make it easy for people who are dissatisfied with Google to leave Google. We have the strangely named Data Liberation Front Group, whose job is to get information out of Google and into your, our competitors' systems.
So we think that consumers not only tend to trust us because of our public statements plus if we violated them, we would be sort of destroyed in the media and our brand, but also, we wouldn't do it for other reasons, good moral reason.
And finally, we want to keep them as a customer and we make it easy for them to switch.
CNET (Tom Krazit): I want to just go back to the evil notion for a second and just ask real quick, has your cultural definition of evil changed as you've grown from a smaller company, focused almost exclusively on search, to a company that is doing so many different things right now?
Eric Schmidt: The significant change, I think, has been the globalization of the company, because there are things, which are not evil in America, which are evil in other cultures.
CNET (Tom Krazit): And vice versa, vice versa.
Eric Schmidt: So there have been a number of cases, which I'd rather not go into, where we've been, we've had robust conversations, if you will, where an employee or a group says look, this is just wrong, but it's OK with American sensibilities. And vice versa.
And we've taken the position that we are a global company, not just an American company and we have to represent and respect that.
Certainly, the China decision, which was very controversial at the time, but I think ultimately, the right one for us, is another example of a tortured internal discussion, which ultimately came to roughly, the right outcome.
CNET (Molly Wood): I think this is sort of a user question. We had one user say, and I'll just quote him directly. "I use Gmail and SketchUp, Google Earth, Google Voicemail and more, and then of course, there's Google Search, iGoogle, and much more than that. I would be dead in the water if Google were to go down or start charging fees or get hacked or start trading my data."
And really, it sort of leads to this question among users, is Google too big to fail?
Eric Schmidt: Well, the good news is we're not applying for a too big to fail bailout from the fed.
( Laughter )
CNET (Molly Wood): That's a relief.
Eric Schmidt: The most important thing is, we don't want Google to go down. We work very hard to keep it reliable. We don't, we always want to be available to this user. And we want to make those products even more useful.
There, in every one of the products that you named, we've got significant investment around new features, new functionality, which is consistent with developing the brand. And from our perspective, we're able to fund this incredible investment from our very successful advertising business.
And we also believe in advertising, that advertising provides real value, that ROI based or pay per performance based ads are useful, whereas generic ads are probably not very useful. And that's, it's the twin pillars, if you will, of focusing on the end user and then having relevant ads that are user-focused.
And I think that's really compelled Google's success.
CNET (Molly Wood): Now obviously, in the Google database, almost everyone probably has one or two Google search results that they sort of wish would go away. I'm sure that you have a very stern policy about this but users wondered, you know, if you're, is it sort of like dating a police officer who can do something about your traffic tickets?
Have you ever had to enforce a policy against, you know, altering search data?
Eric Schmidt: I'm sure that in the scenario with the police officer, fixing the police ticket would be illegal. We have a similar rule at Google, which is we don't do those sorts of things. We're also very careful to forget, if you will, search information after a specific time, generally about 18 months.
The question of your, if you will, information being retained by Google, is not really, at this point, a Google decision, it's really a political or public policy decision, enforced by different governments in different ways. Our view and our general approach is to let you have control over whether this information is retained and that's what we want to get to. The laws may or may not allow us to do that.
CNET (Tom Krazit): So as you move forward with this goal to organize and make the world's information accessible, I'm sort of wondering how you feel about the possibility of, can you digitally index every single piece of information in the world?
And if you do, where do you take that from there?
Eric Schmidt: You could have imagined indexing all the pieces of information in the world until the real-time information explosion. Mathematically, with the growth of the Web, it could have been possible for us to reach all of that information and get into sort of the event horizon of the scale of the Internet.
It's pretty clear to me now that every single person is a broadcaster, a publisher, a blogger, and a paparazzi and the scale of that problem, right, with their mobile phones, and with all of the real-time information and real-time location information and so forth, it will be incredibly difficult to get all of that information in any one place.
And I'm not sure it's even possible.
CNET (Molly Wood): So you're going to need that 70 percent after all.
Eric Schmidt: Yes.
CNET (Molly Wood): We want to thank you so much for joining us today on CNET Conversations.
Eric Schmidt: Well, thank you, thank you so much for having me.
CNET (Molly Wood): It's been nice being here at Google. And you can, of course, get in on the conversation yourself at CNET.com/conversations, as well as watch this interview and all the previous ones.
Thanks so much for joining us.