埃裡克-施密特在麻省理工學院媒體實驗室 (Eric Schmidt at MIT Media Lab)

字幕列表影片播放

>> HOCKENBERRY: Please welcome to the 25th Anniversary of the MIT Media Lab, the CEO
and Chairman of Google, Eric Schmidt. >> SCHMIDT: Thank you. Thank you, thank you
very much. Thank you for having me. >> HOCKENBERRY: When I tell you smile, you
have a big smile, indeed. It's great to have you here. Thanks Eric for joining.
>> SCHMIDT: I [INDISTINCT], this place spawned a lot of what I live in everyday. Ten years
ago, a lot of our employees, when I visited a long time ago, it's like I said, "That'll
never happen." It happens everyday now. >> HOCKENBERRY: So if somebody comes to you
with Media Lab on their resume, that's a good thing.
>> SCHMIDT: We'll hire them. We'll hire them. >> HOCKENBERRY: That's a good--you'll hire
them, I see. >> SCHMIDT: Yeah.
>> HOCKENBERRY: You know, are you sticking around for the party?
>> SCHMIDT: Yes. >> HOCKENBERRY: Yeah.
>> SCHMIDT: We have a whole team here. >> HOCKENBERRY: I see. We'll you're going...
>> SCHMIDT: My guess is the demos that you guys are going to see later are the future
for the next 10 years. >> HOCKENBERRY: That's great. But you'll probably
have some attention at the party later, so I'd keep your entourage like right around
you. Eric, I want to talk about the institution of Google and institutions of change and entrepreneurship,
and I think we loosely use the terms entrepreneur, innovator and discover in this economy. And
I'm wondering if there are institutions that foster entrepreneurs and innovators, and people
who discover sort of basic deep problems that may not have immediate results or those best
worked out in separate institutions? >> SCHMIDT: I'm struck by a couple of things.
One, we have Marvin Minsky here; he's one of my personal heroes. There are people who
in the 60's foresaw--sort of creating what we have today when they created Kendall Square
and Kendall Square Research, and all of that, the progenitors of modern computing. And I'm
really struck when I spend time in our government with how much incumbency drives no change
and how this group and the culture that's represented basically in Cambridge, in Silicon
Valley, and a few other places in America, really are different. And we, because we spend
so much time with each other, we assume we're the norm, we're not. We're not the norm. That's
why television doesn't quite make sense to us. So what they're saying, "It doesn't quite
compute." Whereas in this case, the students are the model of the university, the funding
model, their R&D centers, which are linked pretty tightly together now, really do believe
in discontinuous change. So I would argue that the solutions to the problems that we
have in humanity, in government, in society as a whole will be not made by the incumbents
but rather by people like people in this room. >> HOCKENBERRY: Well, let's talk a little
bit about the difference between an entrepreneur who takes maybe existing intellectual property
and takes advantage of it in the market place and creates a sustaining model for it as a
business; an innovator is something that seems to be a hybrid between a discoverer and an
entrepreneur. And then of course, I think, you know, people who work in basic research
labs like Watson, like Kendall, like Xerox PARC, like the Media Lab, are basically working
on problems where there is not necessarily an expectation. What kinds of investments
foster those different individuals and do you separate that function at Google?
>> SCHMIDT: We try not to because it's so--one of this does happen--is this all gotten interlinked
and these distinctions are not as important as they were, it just involves doing amazing
things. And one thing you learn as an executive is when you walk through the hall, if you
ask people who are the--think people doing the most interesting things, everyone agrees.
So it doesn't really matter how we score it, we actually just know. And I'm really struck
now by--my entire career started with DARPA, NSF, and the kind of funding that people,
you know, generation before me figured was important. The history of American funding
of major Universities, including MIT, started in the post war period by understanding that
having a robust investment of basic science which precedes all of us, you know, the--what
allows us to get these extraordinary returns in terms of semiconductors; it's really physics,
right? All of those investments that occurred in the 60's and 70's, and 80's, we're now
the beneficiaries of. So, unlike our children, let's thank our parents, right. Let's actually
recognize that people actually worked very hard to create an opportunity that we all
now a benefit from. >> HOCKENBERRY: But let's explore the frame
there. They were motivated by a Cold War almost fear mentality, that we had to compete with
the Russians. Today... >> SCHMIDT: Now we can be afraid of our competitors
in the economic sense. Whatever works... >> HOCKENBERRY: Right, it doesn't...
>> SCHMIDT: ...that causes more money to come to basic research is a good outcome.
>> HOCKENBERRY: But do those motives pursue or produce the same kinds of outcomes? In
other words, do we--can we formulate the same kind of urgency about investing an innovation
today without a Cold War, as our parents did? >> SCHMIDT: Well, I would hope so. I mean
it, let's--if you watch enough television [INDISTINCT] spend enough time in Washington,
what's the future of America? Okay? Massive deficits, lost of manufacturing jobs overseas,
an increasing number of healthcare services jobs which are relatively low paid, declining
productivity and aside from the brilliant aspects of our leading universities, no American
leadership in anything. That's the sum of the message. So what's your answer? Your answer
is innovation. >> HOCKENBERRY: All right. But the motive
there is we're failing, let's stop failing. And...
>> SCHMIDT: Well, let's start with advanced manufacturing jobs. Here in Cambridge, you
have the largest cluster as far as I can tell of biotech and biologic related businesses.
That didn't occur by accident, it's not some random event that had occurred in Cambridge.
It occurred because people foresaw that a cluster of such investments built around the
innovation in Harvard, in MIT, in the various institutes that exist, would create whether
in fact the millions of jobs and leading position globally, that can be reproduced.
>> HOCKENBERRY: Not NSF though, not DARPA, not the institutions of government and policy
that were operating in the 50's and 60's. >> SCHMIDT: Well NSF and DARPA did other things.
They're--these guy's funding comes out of the--basically the National Institute for
Health. The important point is, I don't care how the money happens, what I recognize is
that you have young people who have an idea and they need the access. There's a problem
by the way in America that there's a value of deficit as it's called, that the leading
universities including MIT and Harvard, and another here in the Boston area, Boston University,
have ideas involving, for example, nanotechnology and they cannot get enough funding now. So
it's really a national emergency in terms of trying to get these businesses built.
>> HOCKENBERRY: Well, let's talk a little bit about what Marvin Minsky used to tell
me years ago and I thought it was so clever and advanced, and it was more evidence that--what
I've always believed is that Marvin is actually a visitor from another planet and it wasn't...
>> SCHMIDT: Would you like [INDISTINCT]? >> HOCKENBERRY: It wasn't reality. We'll talk
about that a little bit later. >> SCHMIDT: [INDISTINCT] Perhaps you can[INDISTINCT],
I'm sure. >> HOCKENBERRY: A little bit later. No, but
he describe that what was going on at the Media Lab was actually human evolution, it
wasn't a creation of tools. And the [INDISTINCT] actually on my program this morning referred
to this network nervous system that exists around us in the wireless domain, you've talked
about something called enhanced humanity. >> SCHMIDT: Yeah, [INDISTINCT].
>> HOCKENBERRY: Augmented humanity. What...? >> SCHMIDT: Let me make the argument as follows;
let's fast forward a few years. Everybody here has a mobile phone at the moment in your
cameras and so forth and so on, all of these phones will follow Moore's Law in these amazing
ways, the phones know where they are and they're highly personal. So what do I really want,
since I'm a tourist here in Cambridge? As I walked down the street, I want the phone
to, or whatever the device tablet to tell me, "Yes, you were here two years ago. Yes,
that store is a different store. Yes, you need some new pants, Eric. Actually, your
shirt is kind of dirty, maybe you should go over there, you could buy a really good bargain."
You walk into the store and it says, "By the way, just for you Eric, here's a 10% off coupon
if you buy right now." Now, this is all with my permission by the way, all something that
I chose to do and it remembers that. Now today, how does that work for me as an executive?
I have an assistant who remembers what hotels I like and where I'm going and sort of keeps
me and tells me, "You're late, as usual. Get over here." Why can't the device, why can't
the network sort of know that and help me if I want it to. Right? That's the possibility
that this new model describes. I'm quite convinced that what we call hyper locals, sort of local
mobile social, has a set of killer applications. I'm also quite convince that the Media Lab
has some very interesting ideas in the space that can help us with it.
>> HOCKENBERRY: And do you--to note the difference between an individual who has that capability
and an individual who doesn't have that capability at this point, just note the difference between
Eric and my shoes this morning. Eric [INDISTINCT] so, that's very easy.
>> SCHMIDT: And one of the com--one of the comments about computers is we always get
confused about what computers are good at. Computers are not very good at feeling, judgment,
emotion, although people are working on those, people--computers are extremely good at dealing
with billions of things of information and doing various forms of calculations that produce
interesting results. These computers can remember everything, and humans can remember almost
nothing especially as you get older, right? So the fact of the matter is that society
will change based on the fact that computers are very good at what they're good at and
computers--and people are still very good at what we're good at. And by the way, remember
that most of the machine learning algorithms learn from people, right, you train them.
So, ideally what will happen is that in this sort of, I don't know how you want to call
it, but in this collaboration between this emergent amount of machine technology and
the human condition, the collaboration is helpful to both. Right? That human can train
computers and can--computers can help human live their lives better. I would argue in
fact that the goal needs to be set completely differently than ways normally the goal has
been set to use computers more. I would argue that the goals is to use computers less, right?
That in fact the computer is just around if you need it and otherwise you are free to
actually do what humans like to do which is to have a good time, be productive, care about
your family, or what have you. Right? And then in fact that the sort of the hours that
I spent reprogramming my PC in Window 7, you know, over the weekend is not a very good
use of my time. You know, why I chose to do that is a separate problem involving my judgment.
>> HOCKENBERRY: Yes. >> SCHMIDT: But the fact of the matter is...
>> HOCKENBERRY: Yes, I would--I would say... >> SCHMIDT: But, you know, and I would argue
that if you take a look, hit it with--at the iPad, part of the reason it's so successful
is precisely because it works so simply and it just works. And that the just works option,
which I think has been pioneered here and a couple of other places, is in fact a new
discovery for most computer scientists. And Google, of course, is also trying to do the
just works and its support in its businesses as well.
>> HOCKENBERRY: There are people who can help you with your judgment problem in this room
probably and you may hear from them later on today. Two questions, a personal one and
one I think that is more global, about the identity of Google today. I've always wanted
to ask you this Eric, what do you do when you absolutely don't want to be distracted?
>> SCHMIDT: Turn your computer off. >> HOCKENBERRY: But then--when you have to
actually do something as well. >> SCHMIDT: I'd read a book.
>> HOCKENBERRY: There we go. >> SCHMIDT: I know it's boring. I actually
also read a newspaper, a physical newspaper. I know it's going out of style.
>> HOCKENBERRY: Such a--such a chuckle head. Newspapers and books.
>> SCHMIDT: Newspapers and books. >> HOCKENBERRY: Love that. God, I love that.
>> SCHMIDT: You know, you actually read stuff. >> HOCKENBERRY: Yeah, the man needs a hug.
>> SCHMIDT: I also listen--I also listen to music.
>> HOCKENBERRY: Yeah. You're a public radio fan, too.
>> SCHMIDT: Yes. >> HOCKENBERRY: Yes, I'm glad of that.
>> SCHMIDT: Also to--I listen to public radio on--and on Podcast.
>> HOCKENBERRY: There we go. >> SCHMIDT: Of which there are--turns out
your show and virtually everything else, is now highly searchable.
>> HOCKENBERRY: Indeed, and the tipping points of whether people listen to things time shifted
or in the broadcast medium is rapidly approaching for each individual...
>> SCHMIDT: What's interesting is that debate is not a debate that's interesting to most
people. It's the debate that's fascinating to the media industry. The average person
now lives in a world of infinite choice of information, they carry it around with them
and they listen to when they want to. >> HOCKENBERRY: No, absolutely. I...
>> SCHMIDT: It seems pretty straight forward. >> HOCKENBERRY: So, the--no, and I agree but
today in your announcement of quarterly earnings, you reported a billion dollar revenue in mobile--in
the mobile space is that a tipping point? >> SCHMIDT: Well...
>> HOCKENBERRY: Advertising revenue. >> SCHMIDT: Anything that's a billion dollars
is good and anything that's growing fast... >> HOCKENBERRY: Oh.
>> SCHMIDT: And anything that's [INDISTINCT]... >> HOCKENBERRY: Let me just write that.
>> SCHMIDT: ...is even better. >> HOCKENBERRY: No.
>> SCHMIDT: A simple rule for those of you--those of you who are CEOs is that rising revenue
solves all known problems. So, I... >> HOCKENBERRY: Well, let's focus on that
thing for a moment. I think it's fair to say that in the 21st century you are one of the
few figures who we might compare with individuals from the classic Gilded Age of America a century
ago. >> SCHMIDT: Okay.
>> HOCKENBERRY: And that is both a responsibility and probably an annoyance.
>> SCHMIDT: Although, I would argue, Wall Street is top of the ladder--is there.
>> HOCKENBERRY: Probably, yes. In the case of the Gilded Age most of the individuals
who ran big institutions, big corporations were monopolies. I think, you, probably unique
to the individuals that we're speaking of in this room represent something that could
possibly be called monopoly but it's maybe something more [INDISTINCT] ubiquity. Google
has ubiquity. Monopolies are something--Microsoft is a monopoly, or was a monopoly at one point.
It'll argue that. >> SCHMIDT: Was found guilty of being a monopoly.
>> HOCKENBERRY: Yup. Exactly. It was found guilty of being a monopoly, actually.
>> SCHMIDT: Just get the facts straight. >> HOCKENBERRY: Yeah. Something that you don't
like to point out in crowds of people, in public...
>> SCHMIDT: No, I could do a Google search [INDISTINCT]
>> HOCKENBERRY: Exactly right. Yeah, exactly right. Is ubiquity--because a lot of people
have described monopolies as being a non-innovative mode that monopolies tend to become risk averse
and have difficulty with innovation--is ubiquity something different? Does the social network
capability of being ubiquitous allow you an innovative capacity that has, perhaps, not
existed before? >> SCHMIDT: We don't phrase it that way but
I'm happy with the phrasing that you described. What we phrase it as just focus on the user
and all else will work. So we don't spend an awful lot of time--the classic thing that
happens is people who have MBA degree show up and they say like, "Where is the revenue
plan? Okay. Where is your business plan? Where is your ROI calculation," and so forth. And
those are hard to find. What you find is a very, very strong focus on getting more users
to use stuff. And if you can get enough users to use stuff, you can make money doing it.
And so philosophically, that aligns us with end-user interest, we hope. That it also aligns
us with scale. So you have an end-user focused business that scales rapidly; which, I think,
is ultimately the way you expressed the ubiquity point. The problem with monopoly is that monopoly
ultimately allows you to--puts you in opposition to the goals of your end-user because, ultimately,
end-users, in fact, do want choice. All right? They want--and, indeed, we do lots of things
to prevent us from being a monopoly. We allow people to take their data wherever they want
and if they become dissatisfied with us and it's an internal check and balance on our
own power. >> HOCKENBERRY: We want to open it to questions
from you, so prepare yours. [INDISTINCT] raise your hands, call out your questions. I will
repeat it so everyone can hear it. We'll do that in just a moment. Does Google have a
foreign policy? >> SCHMIDT: Well, you can see how well it
worked in China, you know, one day, somebody said, "Well, finally..." one of, in fact,
one of your politicians in Boston said, "Google versus China, finally an equal match." Well,
you saw what happened. We take a particular view on information which is, I think, a western
view and I think most western countries would agree with it. What I've learned is an awful
lot of countries that--there are an awful lot of countries that don't like our view,
China, obviously, because of the censorship rules, but there are others. Typical example
would be that a bad video gets uploaded; we take it down, and then YouTube gets blocked
for a year. Now that could be because we so annoyed the leaders of that country that they're
being--they're punishing us. An alternative hypothesis, you always want to have an alternative
hypothesis in a PhD program, is that, perhaps, those leaders are embarrassed by the multiplicity
of YouTube videos critical of their leadership, so we're never really quite sure. So we've
taken the position that, ultimately, if you want to play in the modern world, you have
to be pretty information transparent. We believe in transparency and open access, and that
gets us into trouble but it is what we believe. >> HOCKENBERRY: The world population in 1800
was about two billion which is estimated to be the amount of people or the number of people
who are actively online in the world today. What would that world look like when six to
eight billion are online? And what will Google look like in that case?
>> SCHMIDT: Well, one of the things that I am personally proudest of, and I hope you
are all as well, is the number of people who have come into the global conversation who,
literally, had no choice without the success of the Internet and telecommunication's revolution.
So I am beyond excited that there's on the order of four billion mobile phones, mostly
which are the feature phones, the less capable phones. I'm beyond excited that there are
800 billion--800 million to a billion smartphones. And I'm even more excited that another generation
of phones is coming--which is going to hit that 30, 50, $60 price point, which allows
you to get another billion of those phones out into the market. So what happens when
you have a powerful browser in the hands of people who've never seen anything except maybe
at television, in a shared model? We haven't heard from them. We don't really know what
they think. I personally believe they all care about Britney Spears and I think we're
going to discover that. But we don't--but we don't know. Right? So it probably means
greater global brands. It probably also means greater global access, and it also means that
we may hear from them and we'll hear what their lives are like in the same way that
CNN, in the '70s and'80s, exposed the horrific conditions of people that we could never see
before, and television bring those images directly to you. Now these people have phones
and video cameras and they can record both good aspects of their culture but also the
terrible situations that, for example, limit or--that are put in the third world. And I
think it leads to--the transparency leads to--the openness lead, ultimately, to making
the world a better place. >> HOCKENBERRY: So you would say, and this
is a very positive note that I wasn't necessarily expecting, that because of precisely that,
the possibilities for a huge corporation like Google to take steps in hearing those individuals
beginning to join this community and taking steps to improve their lives, the opportunity
to do good even as a ubiquitous near monopoly is as good as it's ever been.
>> SCHMIDT: Well, the goal of the company is to do good and so it's only a good thing
when people have access to broadband. So an example would be the role of 4G in the western
world, in Europe and in the United States. Trust me, the people at the Medial Lab will
figure out what to do with 10 megabits but low latency that's continuously streamed to
your tablet and phone. Trust me, we can use that. I'm not worried about that. Same argument
about people who've never had access before, what will they do? Well, I mean, we think
they're all going to spend their time checking for crop yields and so forth. A much more
likely scenario is they're going to be heavily obsessed with entertainment because they're
human beings and people are the same everywhere. So that again is--in my view, a positive effect
of this. And our business approach is to get them wired and then not worry too much about
what happens after that because they're clever and these platforms work. And eventually,
there will be business opportunities five, 10, 20 years from now, which will be massive.
Right? >> HOCKENBERRY: But we have [INDISTINCT] listening.
>> SCHMIDT: Well, we better be. >> HOCKENBERRY: And that's the key.
>> SCHMIDT: But, more importantly, we have enough capital as a corporation to invest
into that market and we have the patience to understand that people are the same everywhere.
People, actually, want connectivity everywhere and that connectivity will benefit them in
education, in access. It will force their governments to become more transparent even
in these horrific dictatorships that exist. And it's an overwhelmingly positive message.
But what happens is, people say, "Oh, well, as a result, some evil person will have voice."
Yeah, well, the principle of free speeches that evil voices are overwhelmed by the good
voices. And I live in a world--and I spend an awful lot thinking [INDISTINCT] everyday.
I'll tell you that the overall human condition is a very positive one. That the average person
I deal with really does want to make the world a better place and we can all participate
in delivering that. It's a great message. >> HOCKENBERRY: Erich Schmidt, thank you so
much. Hands up. Stand please. >> [INDISTINCT]
>> HOCKENBERRY: What's Google's education policy? Will we see Google high school, we
already have Google university? >> SCHMIDT: We are working--we're working
on our strategy there. As you know, the--if you go back to the Nation at Risk book from
1982, 1983 I believe, the problems of education in America are profound. They're complicated.
I would observe that much of the education establishments seems to be organized around
the education establishment and not measurable outcomes with [INDISTINCT] to the children.
So speaking as a businessman, I would start by measuring the outcomes and then iterating,
in trying to make things better. The contribution that Google can provide is a number. We worked
hard to get the schools wired. And I, on an earlier job, did that as well, to getting
the schools wired, getting the teachers empowered with access to the Internet, getting mobile
phones into that, are all very, very important. A further thing that we can do is try to promote
the use of video and other information sources to help advance education especially for people
who don't have access to very good teachers. There are, for example, people who are remarkably
generous and they're funding YouTube videos of the best teachers so that you can actually
see that and see that for free anywhere in the world. And what I want to think about
is you've got some brilliant student who's isolated in some rural part of our country
who has a broadband connection and not very many people to talk to, I want that person
to be able to get that same level of education that I could get. And I think that's a good
goal. So those are some of the things that we're working on. I'm looking for even stronger
technologically intensive education ideas and if you have them, we'd like to hear them.
Most of the ideas that are--had been discussed about education, I think, are pretty much
dealing with the incumbencies and the existing things rather than thinking of completely
new ways of affecting education. >> HOCKENBERRY: Question from up here. Speak
up now. Hey. >> I wanted to thank you for all of your comments
but specifically, the ones about incumbency and how to bring innovation and that how current
forces kind of fight against it. One of the things has made the Media Lab really a great
place since its start is small research groups working together, fostering innovation but
still communicating as a whole. Does that model break down when you get to something
the size of Google and how do you see how that kind of grassroots innovation can really
be fostered all the way up to the top where it can be visible and make a difference?
>> SCHMIDT: I think the Media Lab is, certainly, one of the first institutions that I knew.
It came back and, basically, argued that you had to break down the stovepipes that existed
in academia. And the Media Lab, to me, 25 years ago, you know, Nicholas and sort of
the usual suspects, sort of foresaw the need for breaking that down. And my guess is, in
other industries, there are other people who also saw that. But I think it's actually [INDISTINCT]
now that you're not going to get something interesting unless you build these teams.
The problem with organizations is there has to be a hierarchy, there has to be a boss,
there has to be a building, and so forth and so on. We tried to break that down inside
of our culture but we have the same problems. They're not as obvious and we talked about
it a lot. But I think that ultimately the reason incumbencies breakdown is because of
strong entrepreneurship, strong leaders, and people who are willing to hear 'no' a lot.
And if you look at the, again, the Silicon Valley sort of archetype, it's a person who
is young, rejects existing orthodoxies, and is willing to put heart, body, and soul into
making something happen. And we celebrate the winners but remember there's lots of losers
too and that's part of the system. The typical venture investment is one in ten is successful,
the typical Google scale investment is one in a 100 to one in a 1,000, which is just
how the economics work. So it's important to remember that we have a lot of people who
are trying very hard. A few succeed but we want to celebrate everybody who tries to change
something. It's that changing--that yearning for change that, I think, drives societies
so fundamentally better. >> HOCKENBERRY: How many in this room have
heard the word 'no' a lot? Clap if you have. >> SCHMIDT: All the entrepreneurs.
>> HOCKENBERRY: Here you go. >> SCHMIDT: Just try to raise money in the
last year, year and a half. You understand what I'm talking about?
>> HOCKENBERRY: Sure. Sure. Right, indeed. Question. Right here in the front.
>> [INDISTINCT] >> HOCKENBERRY: How is expensive journalism
going to be paid for if, as Eric has suggested, people are going to be hugely interested in
Britney Spears, perhaps, in the future and other entertainment models?
>> SCHMIDT: We all benefited from a unique combination of events; strong classified revenues
for newspapers, local pseudo-monopolies where the local news was only really under the hands
of one or two providers, the emergence of print advertising and so forth, and the fact
that people just like to read and read physical newspapers, all of those trends and magazines
as well, all of those trends as you know have been under tremendous attack. And it's a real
lost for, I think, a vibrant society. So what happened is those models would essentially
throw off a public good which is the ability to fund investigative journalism. Even a couple
of proposals have had to replace that. A number of, again, very generous people have funded
nonprofits which attempt to fund that public good. A number of the newspapers are doing
well enough that they'll clearly survive any--they've become national essentially. You know, clearly
have enough to continue to fund that public good. But there's a clear loss of voices.
When you talk to the folks in the media industry, there's a lot excitement about subscription
models because they're used to subscription models. So from a Google perspective, we want
to empower these subscription models and we want to let the--between the publisher and
the consumer, they can figure out, "Is this information that's going to--people will subscribe
for it or whether it's going to be advertising supported." So the restructuring is dramatic.
It's incredibly painful. There are models that are successful. Huffington Post is an
example of a Google partner which I think people here know which appears to be doing
quite well in an advertising model. So there are mechanisms which do work which ultimately
I think will generate enough profit that they'll be able to fund that public good.
>> HOCKENBERRY: Is that hyperdizing (ph) though the subscription model and the advertising
model in the sense that people talk about journalistic institution as communities, and
PR for instance being a community? It's a kind of a subscription model. You'd sign up
to be a part of the community as this revenue as a result to that Huffington Post has a
community aspect to it in addition to being this advertising--are those merging ultimately?
>> SCHMIDT: They are. But I would rather--again, we always talk about the incumbents. I'd rather
talk about the end-user, right? Because the end-user will ultimately determine the outcome
of this and let's talk about what the end-user is going to do. Today, I read the physical
newspaper. Five to ten years from now I'll have this incredibly intelligent tablet which
will have every aspect of the newspaper that I care about expect maybe that grainy feeling
and the stuff on my hands. But what's important about that content is it will know who I am.
It will remember what I read. It will suggest things that I should know. We can actually,
using algorithms now, suggest serendipity. We can find related stories and stories that
you should know about and furthermore we can do very sophisticated advertising on such
a device. We can do advertising that's targeted at you today. People here--those of you who
read a newspaper this morning--I did, right? Does anybody remember the actual physical
ads in the newspaper? Typically not. It's hard to sort of remember. And they are there.
You just went by them. Whereas we can generate ads that are so good, so targeted, that you
will remember and we hope that they will help pay for the quality of the content that you're
getting. >> HOCKENBERRY: What's the responsibility
of an institution like Google, though, to lower the cost of entry? A newspaper has a
low cost of entry. People can be civically engage and get the information about their
world at a low cost. If everything's on a tablet, certain people will be able to pay
that cost of entry, others will not. Is that Google's responsibility to lower that cost
of entry? >> SCHMIDT: Well, we would argue that since
almost everything on Google is free for end-users, that's a pretty good responsibility right
there. >> HOCKENBERRY: That should be.
>> SCHMIDT: I mean, the fact that basically we don't charge for access to this information,
and we're not planning on, and we--and I suspect we never will, is probably a pretty good starting
point. >> HOCKENBERRY: Question right here.
>> [INDISTINCT] >> HOCKENBERRY: What mechanisms do you use
to listen to users? Do you employ particular strategies to make you're getting the right
signal? >> SCHMIDT: There's ultimately an inventor
and it goes like this. The inventor walks into my office and says, "I have a brilliant
product and I built and it's going to do great." And I say, "Wonderful. I completely agree
with you." And then we measure them to the nanosecond. And if the slope is a hockey stick
up like that, they're a brilliant inventor. And if the slope is like this, they get cancelled.
Right? And then by the way, we asked them to do it again because they're--it's great
to have an inventor who's failed once, right? Because they really are a lot more humble
to start with, and they listen a lot better. You need the input from the end-user but you
need the inventor with a passion. If you listen to the end-user, they'll just give you a jumble
of feedback. You need that person who has that creative spark. And by person, sometimes
it's a team. Sometimes it's a group. Sometimes it's a set of things. In our case, we do a
tremendous amount of, basically, blind testing along the way, so we really insist on the
state of the art in terms of user testing, in all sorts of ways in which we do that.
And then the most important thing is we typically do what we call 'dog food', you know, eat
your own dog food, so we use--we test this stuff internally and then we roll it out to
friends. And my guess is, if we took the union of the--of you all in this room and ten of
your friends, if you all and ten of your friends like something, that's a pretty good predictor
for pretty much everybody, right? If we could invent such a great product.
>> HOCKENBERRY: But if--okay, [INDISTINCT] right here.
>> [INDISTINCT] >> HOCKENBERRY: Will Google ever create its
own content? It, up till now, rides on mainstream media in presenting news.
>> SCHMIDT: Yeah. The term 'ride on the back of' is a sort of loaded question.
>> [INDISTINCT] >> SCHMIDT: Yeah. And so, when I'm asked that
question, it's usually in the context of, "And therefore you should pay me a lot of
money for my content." So--no, no, I'm saying--but if that's--but just so, I want to be clear
on that. What we do with the mainstream media is we send a tremendous amount of traffic
to their websites and so that each of them has an opportunity to actually not accept
that traffic. They can actually choose not to do that. But they all choose to do it,
so we have this conversation all the time. Furthermore, what we do is we provide advertising
model and to some degree, subscription models, which allow them to then sell that content.
And that's, I think, how we do it. With respect to doing our own content, we've typically
found that there's a line there that we should not cross, that it's better to let the content
folks do the content and have us basically provide the platform, the monetization and
so forth. Partly because the company is run by computer scientists, and as arrogant as
we are, we're not necessarily so good at these other businesses. And the second thing is
it's better to have the competition among the content players for the end-user voice.
As I said we've typically not gone across that although we enable end-users to create
a lot of content and let the market sort it out.
>> HOCKENBERRY: Questions on the mystifyingly silent middle section here? Anything? Any
questions from--no? Would you call denying access to content that has been criticized
in countries where you've worked a form of content creation in a sense that you're negatively
selecting what gets seen? >> SCHMIDT: Well, again, I'm not sure what
you're referring to because if you're referring to China, we actually took a very pro-content
position and then… >> HOCKENBERRY: Actually, I'm not--[INDISTINCT]
much more generally, if you were in the position of taking stuff down or having to make that
choice… >> SCHMIDT: We only do…
>> HOCKENBERRY: …is that not content creation and so [INDISTINCT]?
>> SCHMIDT: No, it's not. It's under duress and under threat of a gun. You know, one way
to understand it is that every country--I've now learned this the hard way--every country
has the same debate we have about free speech and what is permitted content and so forth,
and at least in the sort of democratically or the pseudo-democratically thing, they have
a process which involves rule of law. And so we have learned the hard way is that we
have to respect their rule of law even if we disagree with it, but we do our very best
to publicize when this is occurring. So if you disagree about a policy in a foreign country
because the country is--we are subject to their laws, they can actually arrest and imprison
our employees, we are forced to operate under their laws even if we disagree with them.
>> HOCKENBERRY: Okay. It's been great, the time that you've spent with us, it's really
thrilling. I think we're going to sort of wind up with one more question right here.
>> Is there [INDISTINCT] >> HOCKENBERRY: Is there a class of information
that you expect never to be able to find on Google?
>> SCHMIDT: Well, hopefully, the information that you choose not to give Google which should
be the information--right? That would be the simplest answer.
>> HOCKENBERRY: Well, what about the number of people who died today from malaria?
>> SCHMIDT: Well, presumably, that's a publicly knowable question.
>> HOCKENBERRY: In a month, not today. >> SCHMIDT: Well, but in theory we'd like
to be able to… >> HOCKENBERRY: There are limitations that--because
of your restrictions, but there are also limitations because of…
>> SCHMIDT: No, I mean, maybe there--there are all sorts of laws about this. There is,
for example, a law that limits the aperture of the satellites that we can launch to take
pictures for Google Earth, and that's a limit. That's one that's imposed on us by U.S. law
which we're just subject to, so there's a category of those. But I think the most important
message is anything that's publicly accessible that's, again, legitimate and legal, we would
like you to be able to find out instantaneously. That's about information access. Information
that is private should remain private. And I hope that that's a clear answer.
>> HOCKENBERRY: Google, when the company was formed, was a huge number, right?
>> SCHMIDT: At Google. >> HOCKENBERRY: Yes.
>> SCHMIDT: Ten to the 100ths. >> HOCKENBERRY: Yes. Now...
>> SCHMIDT: Still a large number. >> HOCKENBERRY: ...now, it's a verb in addition
to being a number. >> SCHMIDT: It's still also a very large number.
>> HOCKENBERRY: Right. Right. What will it be in 25 years, do you think?
>> SCHMIDT: You mean--it'll still be a very large number. We did the math. We did the
math. >> HOCKENBERRY: That--that much is true.
>> SCHMIDT: We did the math on how big 10 to the 100th is and, just for those of you
who are not so technical, it's not--it's not the same thing as--it's like multiplying.
It's a very, very large number. It's much larger than the number of proteins and electrons
and so forth in the universe. >> HOCKENBERRY: So, if a billion is a--always
a good thing, at Google is always a really, really…
>> SCHMIDT: It's very [INDISTINCT] >> HOCKENBERRY: …really good thing.
>> SCHMIDT: Yeah, so what happens is it's a billion--it's basically exabytes and petabytes
and zettabytes and gigabytes and so forth, and we got a long way to go before we get
to Google. I think in--one of the things to remember is to do a little bit of compounding.
And humans are very, very bad with compounding, so let's do a little math. Here we are at
MIT and so forth and so on, so Moore's Law is doubling every 18 months. Most people believe
Moore's Law can continue for another 10 to 20 years. We're always projecting the end
of Moore's Law but the physicists and chemists and so forth are doing amazing things with
parallelism. So 10 to the 18th is roughly an increase of a thousand--let's say let's
do the math, roughly a hundred and five years which means roughly 10,000. Did I get that
right? No? So I think it's 10,000 in 10 years, so in 15 years how much time is that? Do the
math. So, all of a sudden, the underlying computation, everything we see today in 10
to 15 years will be that many more orders of magnitude of information and access and
speed. So if everything is 10,000 times faster, cheaper, more prevalent, what does that all
[INDISTINCT]? That's why I come back and I argue about, computers will remember pretty
much everything that you let them remember, will pretty much know what everything is up
to in terms of things that you want to publish with it. There'll be a very, very large number
of devices. These are activities that the Media Lab is working on a lot. And that it
will sort of take all of this for granted. And I think all of us are planning on being
alive for the next 10 to 15 years, so hopefully we're all going to be part of this and we're
all going to see it. And I would argue that society--society has to really think about
what it means to have all of these things. This is a question far beyond the pay grade
of Google or of anybody here in the room. What does it mean to have this kind of access,
this kind of information? I'm quite convince it's all positive but I suspect it has a lot
of surprises. >> HOCKENBERRY: You're roughly how old?
>> SCHMIDT: Fifty-five. >> HOCKENBERRY: Fifty-five. You will retire
when your job ceases to be blank. >> SCHMIDT: Well, we actually agreed that
I have to stay at Google, with Larry and Sergey, for a very long time.
>> HOCKENBERRY: But just personally, I'll stop doing this as a job when it stops being
blank. >> SCHMIDT: My current plan is to do it when
I keel over which hopefully will be a long time from now because the opportunity to empower
people with information is the opportunity of a lifetime. I would argue that within your
career and the things that you do, you do it--you don't have to do this anymore, you
do it because you enjoy it and because you have a huge impact. And for me personally,
the opportunity to serve--to serve the interest of broad access to information for the 6 billion
is a tremen--I mean, Megan, who's one of your stellar graduates here, and I were talking
about Google app engine--Google Earth engine basically, as an opportunity to take Google
Earth and instrument it and make it dynamic, right? And then take programs and feeds from
everyone. So you can see the impact of what's happening to the world that we--that is the
only world we have, and the world that we all love. It's a tremendous opportunity for
all of us, and I love to be part of it. >> HOCKENBERRY: And you're innovating from
within at Google to make it possible. >> SCHMIDT: Absolutely.
>> HOCKENBERRY: Eric Schmidt, thank you so much.
>> SCHMIDT: Thank you so much. Thanks, John. >> HOCKENBERRY: Well, great.
>> SCHMIDT: Thank you so much.