Eric Schmidt on technology, innovation & the global economy


Uploaded by Google on 24.09.2009

Transcript:
[clapping]
>> SCHMIDT: Audrey represents the very best of Pittsburgh.
The energy, the enthusiasm, the sense of what you can be when everything is so focused on
excellence. And I'm here because I like Pittsburgh.
I have a history with Pittsburgh because much of what I do today in computer science was
invented here many years ago. And I've always thought that as an industrial
model, and as a place of innovation, it's very hard to find other really good choices.
It's some combination of how the government here works, the universities work, and the
notion of the pragmatism born of Westinghouse and Mellon and all of the great people from
years ago are all imbued deeply in the culture. And what I wanted to do today was to talk
a little bit about sort of the preamble to the G-20 and a little bit about technology
and what's going to happen. And then have your questions and comments.
You know as...to put it in perspective if everybody shows up tomorrow and the meetings
are Thursday and Friday and then everybody leaves.
And there's...a lot of concerns about growth globally.
There's...a lot of concerns about many, many things.
We are all familiar with them. You've all seen all the stories.
And these are hard jobs. I don't envy the leaders, their tasks.
They have to sort all of this out in an even more difficult political climate, shaky markets,
lots of uncertainty and so forth. But I want to bring some light to this.
I want to bring some--a more realistic assessment as to where we are and also give you a sense
of what the future looks like. There is a quote from Alvin Toffler who...some
of you may remember who he is... he described the great growing engine of change, technology,
as the impact on economic growth. 02:05 And I think we often think that somehow
moving money around is economic growth. Or buying and selling established companies
that are growing for ever higher prices...well that's growth, isn't it?
Because we added the value that they did. Growth occurs because of private sector investment
and innovation, building new products and services that people care about.
That's how growth happens. It doesn't happen because of more government
spending or more inflation of the currency or whatever.
Real growth...sustainable growth comes from the investment in businesses and services
that we all take advantage of. And so part of that is a fundamental focus
on technology. And the kind of cutting edge technology that
makes you uncomfortable because it is so disruptive. Because it changes things.
It causes current models to be changed, right? They fall by the wayside and a new model,
a more interesting, more direct model emerges. And it seems to me that this is all happening
so quickly that we haven't really quite figured out how it's changing our society, and I want
to talk about that as well. As...so the venue of Pittsburgh, seems to
me, particularly appropriate. When Pittsburgh was originally chosen everybody
said, "Oh, why would you choose Pittsburgh?" Pittsburgh is a great story, right?
The whole notion of going from sort of a steel city to a producer of high-tech alloys, you
know titanium. My dentist, who I visited yesterday, explained
to me that dentists invented titanium because it's the one thing that your bones grow around
and everybody loves it, and they have been doing this for 50 years.
Titanium is this incredibly important thing. It didn't exist before.
The healthcare system that you all have here with UPMC, now the number 1 employer, the
unemployment rate being much lower than the national averages especially given the sort
of tough economic travails that you have been through over the last 10 years.
04:03 So it seems to me that the nice thing about this is you got a situation where the
rising powers...the sort of G-20 inclusive are now all here to see a success model for
a democracy, which blends both the old and the new in a particularly clever way.
So there are lots of laws that people talk about when they talk about technology, and
I would like to introduce a new one which I am going to call, "Gutenberg's Law" which
I'll make up roughly like this. History has demonstrated that there is a direct
correlation between the amount of information available to the average citizen and economic
growth and the progress of that citizen's country.
And almost everything we deal with in the negative or positive context has been because
of time compression because of all of this information.
And the more information that's available the more smart people can act on it.
And I'm one of these people who believes that there are smart people everywhere.
Here, there, in other countries, in other races, in other religions and so forth they're
all there too. And so in the global competitive model it's
not so much that the fact that you have an industrial policy...it's that you have an
openness about information, and that you use the information tools.
And the good news is that the statistics and so forth all favor what we're doing so well...so
very much. If you look at the printing press, for example,
the impact that it had on completely changing the understanding of...what was it at the
time a very small number of people who were even literate.
The telegraph...moving the original sort of Twitter model of the world.
Fast moving news...you know spreading wildfire, stocks going up and down, crises and so forth
and so on. And telephones and the gift of e-mail having
us always be in touch, even if we don't really want to be.
But all of these are basically communications, and if you look what's happened it is that
the cycles of society have accelerated. The ups and downs.
All because of the transparency and the connectedness of all of us.
It used to be that your regional model was local, now it's global.
We now respond to things in a global context that we would have never even known about
10 or 20 years ago because of the rise of all of this.
There's a lot of evidence that this is true. There's a study that the ICT sector, that's
essentially computers and telephony, represents in Europe 5% of European GDP, but it drove
40% of overall growth...sorry, 25% of overall growth and 40% of increase in productivity.
So not only is this happening, but the jobs that are being created are accelerants.
They're high paying. They have a lot of leverage.
If you have a choice of a job to fund if you're a government official you would do one of
these. In another World Bank study of 10 percentage
points of increase in high speed Internet access, economic growth was 1.3 percentage
points. And for countries where economic growth is
negative or 1 or 2% that's a huge number. So this stuff works.
And it works because of globalization: globalization of information, globalization of markets.
Now this is all happening in the backdrop of what the physicists are doing, and the
physicists continue to make our networks and our computers that much faster.
Japan leads in the speed of broadband. The average Japanese city dweller has 160
megabyte connection, which is inconceivable to us here.
The new DOCSIS 3 standard here in the U.S. is capable of generating 50 megabytes per
second, so that's pretty good. That manages to bring us to the 13th in the
world in terms of broadband penetration, not so good for America, the leader in technology.
07:45 But this is a wonderful revolution. Because everybody is benefiting, even Africa,
who are putting in new cables and so forth. And the data that is being generated as a
result of everybody being connected is overwhelming. An interesting number here is that...this
is an IDC number...487 exabytes of digital information...and I will tell you what that
is in a sec...was created in 2008.That's double from 2007.
Today we're generating about one zetabyte. A zetabyte is 10 to the 21st power of bytes,
which is a very, very large number. To put this in context, we estimate that the
data generated by all of humanity from the inception of humanity to 2003 was about 5
exabytes. So we have generated the entire world's humanity
of information between our birth as humans in 2003, in the last 2 days.
This is an explosion. And it's an explosion with very, very far
reaching implications for all of us. And you all know this.
Everybody here has a mobile phone with you? And you probably won't get it to...more than
a few inches from you, right? And you never let anyone use it, unless you
knew where it was, right? When did this happen to us?
When did we become so enslaved to our mobile phones?
In the last few years, huh? There were 77 million smart phones sold last
year. That number globally will eventually be in
the hundreds of millions of...these are sophisticated phones.
There's more than 3 billion mobile phones of any kind.
10 years from now when my grandson is 14, it will be possible that the systems will
be at least 100 times more powerful. It will be possible to have all music ever
recorded on the equivalent of an iPod. In 15 years, it will be possible to have 85
years of video on your iPod. Which means you'll die before you watch it
all [audience laughing] so you have the ultimate sort of frustration device, right?
So...and this is...another very...speaking of young people, a very alarming statistic.
The study that I read indicated that the average teenager is now going to send 10,000 text
messages per year. Now that's an awful lot of time texting, trust
me. And I guess now I sound like my parents who
were worried that I was watching too much television.
10:12 The scale of Facebook and social networks is phenomenal to be connected, and it gives
you this appearance of being connected to everyone.
Historically, you know you knew 100 or 200 people in your tribe, and now you feel like
you're a global citizen. So imagine what you could do with that.
You can now sit there with your mobile phones, and you could vote real-time for the performance
that you're seeing. At Woodstock of today, you'd have instantaneous
feedback as to whether you should get that guy off the stage or not.
And the community crowd sourcing would be used to determine an awful lot of things of
what would happen next, so what happens as a result is we have these resources that can
do amazing things. You can find interesting errors, and my favorite
current example is that in 1973 the Navy in San Diego decided to build a building.
And they built this building and is just happens to be the shape of a Nazi swastika.
Now this was...nobody noticed this for 25 years.
Because you know you can't really see the building from above it.
And Google Earth comes off and everybody says, "Oh my God!"
Right? We have built a building in the shape of this
horrendous and terrible symbol. So they then spend a whole bunch of money
to change the shape of the building from the satellites.
And my reaction, by the way, was like, "Didn't they notice this when they were doing the
plans in 1973?" [audience laughing]
But that's a story for another day. People were using Google Earth, and they were
looking, and they discovered a part of a jungle in Southern Africa that had never actually
been visited. And they sent in an investigation, and the
whole team went in, and they found all sorts of animals and plants that no one had ever
heard of. 11:59 This is unbelievable in the last year.
It gives you a sense of how big the world is and how powerful it is.
And in the same case, what's happening is all of these people are spending all of their
time contributing stuff to each other. And so the information that you get is now
more likely to be from your friend, or person to person, user to user.
We finally have now seen the breakdown of the traditional hierarchy of single individual
to broadcast. The majority of information that you get every
day, the way you hear and the heuristics and the judgment that you make are from what your
friends generate or what your peers recommend. And that's another permanent and important
change in how our society works. So imagine a situation now where you collect
all that information. I mean, I would like to see the equivalent
of Wikipedia for the doctors. Who are the doctors?
Instead of sort of saying, "Oh well I've seen 5 patients that have the same thing" they
have a resource which codifies all of medical knowledge in one place collaboratively which
they can then say, "These are the sum of all the cases, and this is what standard of care
is." You can do all sorts of crowd sourcing for
real-time information. Weather, pollution, crowds, this flash mob,
so forth and so on. And you can do truly personal search.
Remember what I read yesterday. Remember what I know.
Don't tell me things I already know. Tell me something I don't know.
And those technologies are available today. You can walk down...when you walk down the
street with your mobile phone...because the mobile phone, remember, isn't really a phone.
It's a GPS, a video camera, a still camera, a browsing device, and you can talk on it
too. Walk down the street.
It can tell you everything that happened on the street if you love the history of where
you are. It can suggest where you go.
It can predict where you will be. So this means a lot of changes for the way
government works, and I want to explore that a little bit.
If you look at the Obama campaign, and I was a little bit involved in that, they organized
both a powerful website and a powerful field marketing program.
And that model, right, of field organization as well as a powerful website, is probably
the new model going forward. 25% of the people who pulled a lever for Obama
on election day had already connected to the campaign.
Pretty interesting. So now you've got the same ability to connect
to things real-time. 14:27 So, for example, car auctions.
You really can find out the price of a car. eBay is now, for example, carrying car manufacturers.
You can find the true market price, as opposed to that negotiated price, with a car salesman
that you don't really like anyway. You can do all sorts of price comparisons.
There's an application for Android, which is a phone that we offer.
It's available for many people now, where you can take a picture of a barcode while
you're in the store and tell you if it's cheaper online.
Pretty interesting. A real consumer benefit.
Not such a good thing for the shopkeeper. And can you imagine a situation where the
government, and the people who write reports for the government, we could actually figure
out if anyone reads these reports. You know how you go, and you get...I produced
this report, and I produced this survey. And my thought is--like--nobody reads this
stuff. Well, now we can find out what they actually
read. Did anyone read all that research you did?
And if the answer is no, maybe you should get a different line of work.
All of a sudden this notion of real-time tracking becomes useful for a lot of things that we
take for granted. You can actually figure out who influenced
the outcome now with real data, as opposed to all of the heuristics that we always talk
about. And we're all convinced of things without
any facts whatsoever. And imagine a situation where the government
itself changed. And what it did is instead of, for example,
announcing the stimulus package, which in general I was in favor of over a 2-year period,
the government said, "We're going to do this for 3 months, we're going to measure the outcome
and we're going to iterate. And then we're going to iterate again. And then we're going
to iterate again as we see what works." Because we can do real-time data now of the
economy, of what people are doing, of what people are thinking and of what people are
reading. We really can know questions that we have
to estimate from our friends. And we can do it scientifically.
So when you instrument the world...another example, health care.
Contentious debates over and over again. We should know the answers to these questions.
If we got all the computers linked together in such a way that we knew the aggregate information,
not the personal details, but the aggregate information of what people were doing, we
could accurately estimate every one of these scenarios.
16:47 Every one of the 535 modifications of the bill, we can do them instantaneously.
Or more properly, the press and the analysts can do--we get a more accurate idea of what's
really going to happen with these very, very important political and policy decisions.
Had we done the same thing we would've known this horrendous financial crisis.
We would have seen it coming. The computers knew that they were overextended.
We didn't ask them. And in some cases these computers knew one
thing, and these computers knew another, and the 2 computers weren't connected.
Had we been transparent and open we could have avoided the horrendous pain that the
globe has gone through, or at least some of it.
And let me tell you it has been painful for an awful lot of people.
It was a disservice to them, to put them through, when we could have done it differently.
So if you take the trends, the trends of instantaneous information, the explosion of information,
personal information in this real-time nature, it says a lot of implications for business
as well. Hal Varian is an economist at Google, our
chief economist, in fact. He calls this time a period of 17:51 calm
and editorial innovation. It's not any one thing that did it.
It's the system of innovation, and it's innovation here, here, here, here, here, here, all working
together and going as fast as they probably can.
18:07 The example that he uses, which I like, is that this is analogous to what happened
to standardizing of mechanical parts in the 19th century.
When you standardize mechanical parts all of a sudden you hit interchange.
You can repair things, and things like that. And it wasn't a single company that standardized
on mechanical parts, it just sort of happened because it made good sense to the smart people
who were alive at the time. Standardized electronic parts in the 20th
century--the same thing, We're doing the same thing with standardized
Internet components and that standardization has now made it possible.
So what happens is that this lowers the data 18:39 bearished entry enormously for innovation.
This is another point that is completely missed in the policy debate.
Now Audrey, who I said I view as sort of a national resource, articulates very clearly
something very fundamental. When you want to see innovation, you'll see
it in these creative communities. Typically near coffee shops.
Little companies that have banded together. Small groups of people who have a mission.
And they try something, and they try something, and they try something.
Pittsburgh is a good example of this, but there are many other places in the United
States that have this characteristic: parts of Seattle,
parts of San Francisco, parts of New York.
parts of Cambridge and Boston and so forth. You know the list.
But my point is that if you want to look for the face of innovation, don't look for a tower
with the sign "innovation" on it. Look instead for this messy, creative, unstructured,
interesting, full of creative people kind of a model because that's how innovation really
occurs. And that's where the innovative people will
ultimately go. So what happens is that this creates the opportunity
to do businesses...which I'm going to call micromulti nationals...these are 10 people
who see themselves serving a global audience. This was not possible until this generation
of Internet technology. Because you literally couldn't reach them,
you couldn't advertise, you couldn't get the points out, you couldn't make it happen.
And these are the businesses that are going to create the jobs that will get us to where
we need to go. And what's happening, of course, is that as
we lower the entry price, there're also advertising models...Google, of course has pioneered this...where
a lot of money is going to these small companies, which allows them to get to the next stage
and see how big they can really get. And another phenomenon has been the rise of
the cloud. The technology people call this cloud computing.
For those of you who don't know what it means it means that the computers are managed by
the professionals...and your computer you just turn it on and it uses the network and
it gets everything that it needs. And instead of spending all your time managing
your personal computer and wondering, "Did I backup that thing?" or something, it's all
taken for you. 20:59 So the opportunity that we all have
is to let people manage these data centers...Google was in this business, as are many others...and
you can do what you do best... which is to use computers to solve interesting information
and entertainment problems. So this rise of the cloud, and the opportunity,
is actually a large business in and of itself because companies are now moving one after
the other, after the other from the old model, the PC centric-mainframe centric model, to
this new cloud computing model where everything is managed by others, and they again spend
their time working on business processes and things like that.
The potential for cost savings, by the way, is a factor of 10 and so forth and so on.
People really care a lot about that. And what's interesting about this phenomena
as a result, is that...if you go back to the multinationals...is that size isn't such a
big barrier to entry before. An interesting statistic, that we have more
than 452,000 bloggers in the United States under...this is a Wall Street Journal count,
I suspect it's even low...and this is a...these are the number of people who say that blogging
is their primary source of revenue. That's pretty amazing.
What's interesting is that that's even more than the number of lawyers we have in our
country. So think about it.
And of course one is growing much faster than the other.
So this is a permanent shift in the economics of our world around us.
It means an awful lot of people are dependent upon these systems, and we're going to hear
their voices, and I think that is absolutely wonderful.
Now this of course is not unique just to the U.S., I've given U.S. examples, but you know
in India, for example, the mobile operator is using mobile phones and having farmers
to remotely monitor and switch on their irrigation pump sets, right, so they can water and so
forth. And you sit there and you say, "Oh, that's
no big deal." For them, it's life or death.
If they don't get the water, they don't get their farming, they don't get their farming,
they literally die because there is no...so it's very, very real.
You go on, and on, and on, and I've got example after example.
So what happens when this is done? So remember here is the model.
Mobile phones everywhere, and data everywhere, that's huge scalability.
What does the world look like? 23:12 Well I assure you something, which will
seem obvious. Humans will continue to do what humans do
best, which is to be human. All the things that we know, creative, intuitive,
paradoxical way that we are, and you know what I'm talking about.
What will computers be good at? They'll be good at remembering everything.
To recognize everyone, to know where everyone is, and to predict the short-term future and
to continue to beat us at chess, right? Think about it.
All those things involving rote memorization, calculation and so forth, we're not very good
at. And even the best professionals forget those,
but computers never forget. And what happens when this happens, when all
of this gets put together, is that our world becomes more ambiguous that the world becomes
more shades of gray. And the reason is that you can now finally
suck in all the perspective. It won't be a simple good guy-bad guy, us
versus them. And that the kind of leaders, and leaders
I think we have in this room that will do very, very well, are the ones that flourish
in ambiguity. We're not so sure we think we did this, we
thought this and so forth. Those are uniquely human characteristics and
the great leaders of the future will handle them in particularly good ways.
And what happens at the same time, and I don't want to be so completely Pollyanna here, is
that it gets harder because of sophisticated techniques to know exactly what the truth
is...now Google is in the business of trying to produce the very best search results, but
we don't claim to know what the truth is. 24:55 But legitimate truth groups, ones that
are actually trying to get a real message out, will have to battle with well-funded,
targeted, literally people trying to deceive. Because it will be in their economic incentive
to do...this is known as spin, by the way...it's not something that we're inventing any time
here; it's been around for 100 years. But in these new networks it will be possible
to spread disinformation as quickly as to spread information, and if you're a targeted
attacker, you might find that is particularly beneficial to whatever your cause is.
And what's interesting is that it will be possible to do clever forgeries and smart
people will figure how to trick people with these clever forgeries, and all of us will
have to learn how to deal with that, how to detect it, and hopefully how to not be influenced
by them. And what's interesting is that in a world
of essentially perfect copies, people will be quite happy with a copy of an original
because the digital copies will be so good. So that's a lot of implications for how the
... world work. And I'd love to tell you that in this new
world, because it's so brilliant, we have all these smart computers and all these smart
people, and we're all so incredibly educated and wonderful, that there will be fewer bubbles.
But in fact, there will be more bubbles. The bubbles will be higher and faster and
up and down...because it is true of all markets. Markets are not perfectly rationally stable.
But they will be up and down, and up and down, and up and down, and we'll all have to get
used to that. And we'll all have to figure out how we want
to get through life with it. I'm not saying it's good.
But I'm saying it's going to be true. So how...in this situation how could you,
how could we react to this? How could we...how do we feel about it?
I'll tell you that the only answer, I've been able to come up with...we at Google have been
able to come with...is about transparency. That in the situation that I'm talking about
open government, open activities, open standards, open networks are really the only way.
Any time you start to partition, or withhold things or so forth, you get yourself into
trouble...because information-hiding is ultimately a bad thing in these sorts of open networks.
There are lots and lots of examples of this, the growth of the Internet itself, the rise
of Linux, the rise of Wi-Fi. All the things that we take for granted today
really came out of this fundamental principal which has been around for a long time.
The Internet is a very, very good example of this.
As you know, the Internet was developed by DARPA originally for university components
and then eventually commercialized in a very clever way in 1991.
Another thing that people should do is they should stop consuming energy and start consuming
data. That if you think about it, all of these should
be faced with climate change can be addressed in many, many ways by a much smarter use of
information to control and use our energy. And climate change is clearly of...between
nuclear power so that...deadly nuclear power the two really great threats to the world
going forward. So here we have a situation if you go back,
we have real-time data, we have everything connected and so forth.
We can actually run real-time experiments. 28:13 We have the power to collect and analyze
data in real-time all over. And I think--I think this change is one that
we'll all have to get used to because I think it's a good one.
And we're going to use computers and use information to be more careful with how we use the most
precious resources we have all around us. Everybody is going to be a critic in this
new model. You know we went from sort of the Nielsen
box to everybody's a You Tube critic. People are using hybrid auto drivers that,
you know, you can have your little diagram of your gas efficiency.
So all of a sudden people are going to be watching and measuring these things at a level
we've never really seen before. And what happens is that you get these open
networks and plays, and you could do some amazing things.
So why are we not using the batteries that are in our cars to backstop the smart grid
in the middle of the afternoon on a hot summer day?
It was better than just sitting there idle. All that energy is sitting there, you can
charge them up at night. Well, the technology exists and the networks
are being built, and they're open enough that you can build...and people here in Pittsburgh
are, in fact, building the components of that power train.
The power train not inside your car, but the power train from your car back into the grid,
to help us with some of our climate and energy issues and save another coal-fired plant.
Another thing that we can do is take advantage of the fact that the network is more powerful
than your computer. Remember behind that network is literally
millions and millions of computers and servers that have a lot of interesting data that you
can figure out. 29:45 You can open up opportunities for businesses
and governments, entrepreneurs can build systems that can monitor and measure and detect new
things. And this is going to happen now with so many
people having mobile phones in very, very powerful new ways.
You're going to have to adapt and evolve...now I don't mean to say this in Pittsburgh, Pittsburgh's
story is a perfect example of this. In 2003, when the city was in very dire financial
straits, you actually reduced the size of your government sector.
You made really, really hard and painful decisions, probably, then setting up the growth that
has occurred since then. Was it worth doing it?
It was sure painful at the time. I remember when this was going on.
But what I like about this now is you're seeing the benefit.
You're seeing the uptick of all of that hard work that you did 5 or more years ago.
Let's make sure that the principle that's established in this dialogue is that extremists
don't win. And there are extremists in all categories.
I'm not picking any particular extremist group. Let's make sure that we use the collective
wisdom of us to understand what is good, and what is valid, and let's have both the decency
and the proper approach to try to get to the bottom of what the real truth is on any particular
issue. In our country and in many western countries,
we have a tradition of openness, which has really helped us and the government, in particular
the federal government, now has instituted many, many principles of trying to put real-time
information out there, so people can literally see what the government is doing.
The Obama presidency has, for example, already promised to fully identify all the aspects
of the federal budget. In addition to that, President Obama on Monday
outlined an innovation agenda that called for the full funding of NSF and DARPA and
the other things, the impact of those decisions over a 10 or 20 year period is phenomenal.
Because after all, where will the growth come from in America?
31:44 It's not going to come from high-volume, low-wage manufacturing jobs.
It's not going to come from the service industries selling to ourselves, unless there's an economic
structure that will generate the returns, that will generate the demand for...that can
satisfy the service industries. It's going to come from industries like advanced
manufacturing. High-tech manufacturing jobs here in America
using very sophisticated supply...basically manufacturing lines and not just in physical
goods also. There are technologies such as nanotechnology,
where such manufacturing plants, which look a lot like an office building, produce tremendously
interesting new products and also very high-waged jobs.
This is where the growth is going to be. So from my prospective, and I want to put
this into prospective for all of you and then take your questions, you have all of these
things put together, right? You have this change occurring very quickly.
I think, in many cases, more quickly than we've really assimilated.
The empowerment of individuals. The rise of real-time.
The collection of all of this data. The ability to answer the questions that I
am talking about in new and very, very fast, stunning ways.
I think it's wonderful. When I put it in a backdrop of innovation,
I see the story I'm telling as a key component of how America will grow out of this recession...which
we hope is ending and which many people think is already...the growth is going to come not
by moving more money around between the banks...which we're particularly good at by the way...the
growth is going to come from the investments, and the decisions, that are being made in
cities like here and by groups like this. There is no question that we face serious
threats in our economy. It's obvious.
But there are sensible strategies to address this, and I think one of the strategies is
optimism. 33:40 Literally, when people feel better they
invest, and they get better. So it's not as simple as just a math problem.
It is about feeling. It is about feeling confidence that you know
you can get through it. And at the low point, the stimulus package
and other things that were done by the governments around the world really did have a way of
jump starting, and it really did work. There's a quote that I like--it's from de
Tocqueville in 1831, "America is unique because of its abundance of land, its absence of a
king, and its democratic and equalitarian institutions and values."
From my perspective, this story...the story I'm telling, reflects the possibility of newness,
new places, new people, to have a chance to be heard, new institutions that must, sort
of, earn and re-earn their relevance. That's the challenge that everybody faces,
every government, every CEO, every citizen. And I think we have a lot of reasons to be
optimistic. Here we are in Pittsburgh, which has a great
story. I personally believe very much in the genius
of the American people. When I look at my own story, when I look at
my colleague's story, if I look at the story of my industry, what I see is creativity and
the power of technology and the genius of people working hard in small groups doing
clever things. What I would suggest is our message, as our
leaders show up basically tonight and tomorrow, is for them to look around and try to figure
out where's the growth going to come from, and it's going to come from exactly here And
exactly in this right way. And I'm excited to be part of that so thank
you very, very much.
[clapping]
I think...Audrey how did you...did you want to do questions?
>> AUDREY: I think we're good if we can just line up and there are questions, and we have
mics there.
>> Hello-test-okay I am wondering...
>> SCHMIDT: I can see you're just going to jump right in so please jump right in.
>> AUDREY: Yes, jump right in.
>> --so I've been wondering where Google's position on the long-term would be in the
string of innovation? Especially if I was an investor, I would see
the...how do you say it, the opportunity to promote your own products with the...yep to
promote your own products while you promote your search, and all the other services even
closer to at the customer, and now what I've been reading approximately, well, let's say
I think it was 2 months ago on the news, maybe you know about this...probably, I think, there
is said it was discussed that there might be the possibility of an antitrust case against
Google similar to the Microsoft...similar as it happened to Microsoft because of this
overall position close to the customer with the opportunity of promoting your own--let's
say, for instance, product search, which will give you a good opportunity to increase your
revenue. Or let's say promoting You Tube directly in
front of your search because also one of those opportunities even directly and closer to
the customer similar as to what happened with let's say Internet Explorer and the Microsoft
case. Maybe your comment on that?
That would be really great?
>> SCHMIDT: Yea, I think I understand your question.
Could Google do the kinds of illegal acts that Microsoft did...was found guilty of
[audience laughing]
for...this is the question you asked right?
>> No, I neither said that Google would be...that Microsoft did something illegal nor...I just
say that the opportunities there and...
>> SCHMIDT: They were found guilty, trust me.
>> Okay...
[audience laughing]
>>SCHMIDT: Use your favorite search engine you'll find the details.
[audience laughing]
You'll find them on Bing as well.
[audience laughing]
The rough answer is no. And the simple answer, the sort of longer
than no answer, is that we run Google on a set of principles, which have to do with end
user benefits, and we're careful not to advantage one property over another because it would
screw up consumer benefits, so we, in fact, show all the other properties.
We don't integrate particularly well. We're careful about that for the reasons that...the
reasons that you outlined. The antitrust question is actually slightly
different. Microsoft was found guilty of essentially
tying one thing to the other, and they were essentially blocking competitors.
Because we're not that kind of a platform, we don't have that kind of market power.
In Google's case, if we were to begin to do that, it would be trivial for you to switch
to a competitor, for example, Microsoft. It's much harder to switch from a platform
like Windows to somebody else, and Google, we're literally one click away.
>> Okay, thank you.
>> SCHMIDT: Okay, thank you. Next question.
>> Hi. I'd be interested in hearing your vision for Google ventures and also your thoughts
on the open Internet provisions that were introduced by the FCC chair a couple of days
ago.
>> SCHMIDT: Thank you. Google has created a venture arm called Google
Ventures. It's an experiment from our prospective.
We have lots of cash as you know, because we have a large cash portfolio, and we decided
that we wanted to invest in the technology area both to get a financial return, of course,
but really to help accelerate the kinds of technologies that eventually Google benefits
from. So it's too early to say how successful it's
going to be, but we're definitely investing...we're investing very similar to other venture companies.
So think venture company. 39:33 On Monday the chairman of the FCC, Julius
Genachowski, announced a set of principles around net neutrality and those principles
were very similar to what President Obama, then as candidate Obama, announced when he
was at Google about a year and a half earlier. The rough summary is that net neutrality is
about making sure that a particular kind of content is not treated differently from another
similar kind of content by the operator. And the scenario that people are worried about
is that the carrier, for example, has a vested interest in a particular video, and so they
discriminate in favor of that video and against another video from a third party who wants
to use the same transport. And in general, obviously Google supports
that restriction. We are careful to say that we want to do that
within the same category. So there's a very legitimate concern from
the operators that if they couldn't discriminate at some level, they could be flooded by bad
bits. We understand that.
We just want to make sure that if there is a particular category, say video as an example,
that they do not unduly favor their own over alternatives that are coming in.
If they were to do that, and I'm not saying that they have, but if they were to do that,
it could prevent startups from being able to come in and get there content on the net.
So we're strongly in favor of what the FCC is proposing and we're happy to work with
them to make sure that the regulations, as they are ultimately adopted, reflect what
I just said.
>> Thanks.
>> SCHMIDT: Yes sir.
>> So my question relates more to something that you said early in your speech, which
was kind of thought provoking. It seemed like it was a rather idealistic
solution proposed with the collection of all of the information--like it's a citizens responsibility
to sort of respond to things that they feel powerful about.
To collect all that data, to collect the statistics that say, "We want this to happen, and we
want this change." But I just wanted to clarify whose responsibility
is it to enforce or make that change as the government is in big corporations--like once
all these--once all this data is collected who--I like what you said earlier actually
where you said in your report, "Who is reading this information?"
So we have all this information that's been compiled, who's reading it?
Who's saying, "Oh, the people are expressing this as their goal?"
Who is going to make the change?
>> SCHMIDT: 42:05 Well, I would start answering by saying that I think we should have a moment
of pity or compassion for our elected leaders who we are busy sort of yelling at all the
time. Because imagine if you're one of them, and
imagine trying to figure out what the legitimate people's voice is today, in between the specialized
groups and the bloggers and the interest groups and the lobbying and the surveying, it will
give you a headache. And furthermore, it occurs on a 24-hour news
cycle so it occurs every hour, and you have to sleep some number of hours per day.
It is almost an impossible job. You're never satisfied.
You're never on top of the issue. And we need to develop a sense of where the
body politic is...and then the politicians would make that decision.
The principal of transparency says, "That we collect the data and illegitimate and legitimate
groups will use that data." But the data should be open--especially data
that government decisions are made of are made of everyone.
And I will tell you that all of the really important decisions are hard decisions, and
they're shades of gray. They really are.
And I think that if we look at issues like war, for example, which everybody has an opinion
about, we would be better informed with more information, more analysis, more public discussion,
more data analysis of what's going on, on this political issue, or this political behavior,
of this group, or they got threatened that kind of stuff, so I think you could replicate
that over and over and over again. We're better serviced by more transparency,
more people looking at it, more people measuring it.
>> : Thank you.
>> SCHMIDT: Yes sir.
>> Hi. When do you think cloud computing will help
non-IT industries focus on their core competencies without having to worry about security and
other issues like that?
>> SCHMIDT: 44:02 It's happening now. What happens today is many people are using,
for example, e-mail systems that are outsourced so they don't have to run their data servers
anymore. There are companies that do a VMware, as an
example of a company that does virtual machine outsourcing, where you can actually run your
computers on the Internet and have the image look like it's inside your computer.
Companies are generally trying to get rid of their data centers and have somebody else
own their data centers. This, by the way, is not a new phenomenon.
It was true for IBM personal computers 15 years ago when people would literally have
somebody else run their--excuse me IBM mainframes and called mainframe virtualization.
So I--my view that this is an inexorable shift from the old model to the new model.
It's highly beneficial to corporations. And they're doing it, by the way, not out
of some ... they're doing it because it's cheaper.
You get better reliability. The security stuff is largely now solved.
>> Thank you.
>> : 45:04 Hi. My question is on cloud computing too. So
I am actually writing a paper on how cloud computing is better for smaller businesses
and startups as opposed to bigger businesses and I was wondering if you could comment on
the same?
>> SCHMIDT: Well, if you were to start a company today, and maybe you will when you finish
writing your paper, what computers would you have to buy?
Well, you'd have to buy some laptops, I prefer the Mac, but you could also buy, you know,
PC's or what have you, and you would need an Internet connection.
And everything that you did you can do on a server.
Let's go through the list. You have to pay people...you have a payroll.
You have to get customers where you can use add Google AdWords--sorry for the plug...to
advertise your product...whatever widget that your building.
And that's much more efficient than hiring salespeople and building all the computers
to do all that. You do much of your manufacturing--manufacturing
can be outsourced. All your e-mails, all your services, all your
communications, all your marketing, and all your productions.
So if you're a small company, it makes no sense to have any computers except for the
computers that are on people's desktops. That's a huge change.
Huge, huge change. And I think that illustrates the point you
should make in your paper.
[all laughing]
>> Yes, but the thing is what I mainly see in all the marketing stuff that I saw both
on Google and the other cloud company providers is that all the authors don't focus in on
the niche thing that they actually offer, which is--ya know I saw that one cloud is
better for processing where as the other better in the storage which is why I came up with
a starting point, that particular startups need particular things.
And why is that the companies don't focus on those?
>> SCHMIDT: Well, I'm not going to criticize other companies marketing.
All I'll tell you is that the technical solution concludes that you don't need any computers
except your laptop. And so your job--your new job is to go figure
out which solutions when you sum them up together really work.
I don't think there is going to be one vendor of cloud computing services I think there'll
be many. Just as there are many IT vendors today.
>> Oh, and do you mind if I quote you on this statement?
[all laughing]
>>SCHMIDT: Yes everything, I do is transparent.
[audience laughing]
We are clearly, clearly on the record here.
[all laughing]
Thank you very much.
>> So I heard recently from a Rwanda engineer that Google has plans to open up a data center
in Rwanda, and he told me that the capacity--the demand for this data center will be about
45 megawatts and that is equivalent to a total installed electrical capacity in Rwanda today.
So it's effectively doubling capacity. You spoke quite a bit about the availability
of information, the availability of data and the fact that everywhere in the world, there
are--smart people exist. So it would seem to me that Google would have
an incentive to increase access to this information and to the Internet in Rwanda and in other
places in the developing world. So I was wondering if Google does have any
plans to facilitate the development of rural electrification and rural computing in Rwanda,
specifically and/or in other parts of the world?
>> SCHMIDT: You're getting my speech internally at Google so thank you, we agree. The rough
argument goes something like this. For the last 40 years there's been one initiative
about Africa after another. And Africa on a proportional basis is worse
off today than it was 30 or 40 years ago especially when you compare it to say India, China, the
rise of Asia. There are many, many reasons why this is true.
So what could we do now that's different? Let's wire the country.
Let's wire the continent. So you dig into it, and you discover that
until maybe last year, there was only one fiber optic cable that served the whole continent.
It went from Lisbon all the way down to Cape Town on the left side--I guess it's the west
side. After a lot of work, a lot of telecommunication
carriers and amazing, amazing acts of courage have laid a couple fiber optic cable now also
one on the east side. And they've now connected a land line cable
that will actually go all the way from Darussalem basically into Rwanda and Uganda.
It's interesting that if you're operating Kenya, Rwanda and Uganda, all your Internet
was connected by satellite...so incredibly expensive and slow and so forth.
And that becomes a pretty big deal if you're trying to build an outsourcing regime.
A lot of the Asian countries became successful because they had a lot of cheap labor, and
they were able to outsource jobs to them, but if you can't connect to them, you can't
do it because the costs were too high. So we prioritized getting everybody connected,
getting those connections in place and accelerating the information and the production of local
information as our key initiatives. With respect to Rwanda we actually signed
an agreement with the President Kagame to actually offer Google's App services essentially,
which are cloud based. The issue fundamentally in all these countries
is the one you laid out. There is not enough electric power to run
the data centers. The telecommunications networks are in the
cities only. They don't reach the rural thing.
And those are systemic issues, and it will take many years for this to get addressed
and we're going to help as best we can.
>> Okay, but there aren't currently any products or programs that you're developing or are
you in the process of developing a project that would facilitate rural electrification?
>> SCHMIDT: We've looked at it. We don't do the electrification ourselves,
so we have to have the government working with the local providers.
One of the good stories is that their local operators, those are MTN and those kinds of
guys, have to have their local power anyway for the telephone system.
So the--one of the great stories, and again we really appreciate how powerful this is,
that the spread of inexpensive mobile phones in Africa has changed their world in a way
that is inconceivable anywhere else. Yes sir.
Thank you.
>> TORRES: Dr. Schmidt thank you for spending your time
with us. I'm Miguel Torres and I'm a PhD student at
the University of Pittsburgh. Right now healthcare is very expensive and
this of course has driven perform and you know their ongoing efforts.
However, one of the challenges is there is a disarray of, you know, different...the different
components within the providers of healthcare. How do you see Google being part of the solution
to this? Do you have any take on that?
>> SCHMIDT: You probably know more about this than I given your research.
Let me make the following argument. Imagine if each one of us had the equivalent
of a USB stick that had our medical history since birth.
And when we walked into the doctor's office we could plug that in and the doctor could
see, Oh, I see 10 years ago you had such and such an operation or such and such a disease
or such and such a drug, and then powerful servers could go over that data and then make
recommendations to the doctor based on the knowledge of that specific core letter or
pattern matching around that patient. That's a pretty interesting idea.
And it's enabled by the general availability of open patient health records.
There's legislation and in fact, there's a requirement for a standard to come out in
December of this year, again through an office of the White House and a series of legislative
activities in HHS, which are trying to standardize on what that format looks like.
52:45 And of course we all understand that you wouldn't carry the USB stick out, in fact,
that information would be held in some cloud-based server somewhere so you wouldn't have to carry
it with you all the time. So what could you do with that kind of information?
Well, not only could you do much more accurate healthcare diagnosis by computer because you
would have it all in some reasonable computer health record and make recommendations to
the doctor, but you could also do some very sophisticated drug trials because you could
really follow the impact of this combination, or that combination, in things that we couldn't
do. So from my perspective, the use of information
to do two things is possible, and Google is doing the best we can to help there.
One is that more information just empowers the consumer.
And one of the voices that's been lost in the healthcare debate is the consumer voice
because consumers are confused and so forth, and they don't remember things very well,
but remember computers do. And then the second one is making that information
cycle much faster. Thank you.
>> My question is about your ideas about clean jobs and high-tech manufacturing.
What specifically gives the U.S. a competitive advantage over India, China and developing
countries in this sector given that in India and China and these other developing countries
you can find qualified people who are willing to work for less money, and therefore, it
makes the products cheaper? What specifically give the U.S. an advantage
in these fields? And do you think that these fields will ever
completely replace the high volume of manufacturing jobs that have been lost over the last 30-40
years?
>> SCHMIDT: That's a very good question. The simple answer is that the only competitive
advantage we have in America is the quality of our higher education system and the graduates
that it produces. It's already the case that China is on its
way to being the world's leader in green manufacturing technologies because they've decided--because
they're more or less command of control organized...to invest there.
So the only game we get to play--because the people in China and India are just as smart
as we are if not smarter--the only game we get to play is to use the tremendous resource
that's represented here, for example, by Pitt, by Carnegie Mellon and other universities,
is true throughout the United States. That tradition of innovation, the one that
I've highlighted so much, really comes from there.
The other question you asked is over the 30, 40 years, can they replace the jobs?
If you look at the history of--a typical example would be the loom in 1860, 1870 if I get my
dates right. The development of the mechanical loom ultimately
was one of the key components of the rise of the Industrial Revolution that ultimately
built things like Pittsburgh and so forth and so on.
When they study what happened with the loom it displaced a very large number of people
who were previously doing it by hand. The bad news for those people was that they
did not get new high paying jobs. Their children did.
So the real problem with all of the changes that I'm talking about is a generational one.
I'm very confident that an innovation agenda around advanced manufacturing and the things
that America does well will replace the jobs that we've lost, for example, high-volume,
low-tech manufacturing, but the jobs will be numeric jobs.
They will be the children of the people who lost the jobs.
The people who lost the jobs will probably not come back and that's a real tragedy for
them.
>> Okay, thank you.
>> SCHMIDT: Yes sir.
>> Thank you for coming, I really enjoyed your talk.
I typed my question into my phone so if I'm staring at it that's why.
>> SCHMIDT: I think this is the new correct behavior. [audience laughing]
Thank you for modeling the correct behavior for all of the older people in the audience.
>> My pleasure. Your vision of a future, sort of enlightened
by high-tech, is persuasive and attractive. I think up to this point the tech sector has
enjoyed a lot of good faith from people that has helped it to grow, and it's produced promising
results, but do you think that we're at risk of a social backlash founting from the ongoing
rumors about cloud computing being dangerous? About companies like Google collecting too
much data on us and turning into Big Brother? If you think this kind of skepticism is a
real threat to continue growth of tech, how could it be overcome?
>> SCHMIDT: I do believe it's a threat. I believe that most of us our ignorant of
history, and we are too stovepipe to really look at everything correctly.
It seems to me that the phenomena that you're describing has existed in America for hundreds
of years. That the rise of every technology has brought
people who were naysayers and in some cases they were generally terrified of it.
In some cases, they were part of an institution that was threatened by the rise of the technology,
and they were doing their best to delay it. And the reason, by the way, it's so messy
is that the encumbrance fight tooth and nail against the change because it's in their economic
interest to do so. It's not that their bad people.
They're essentially incentivized to fight progress.
And what happens is eventually they give up because they retire, or their companies go
bankrupt, or nobody cares anymore. And there's example after example of that.
So from our perspective we believe that the principles of consumer focus, consumer benefit
and transparency will ultimately win over the skeptics.
That this is not an industry that operates in secret.
This is an innovation--this is an industry that operates in declining costs, not increasing
costs. And so it's always better next year.
Which is a nice message from the standpoint of the political operatives and so forth and
so on. So as long as the industry doesn't do anything
stupid, I think we'll be just fine.
>> Thank you very much.
>> SCHMIDT: Go ahead.
>> So no one's asked about the recapture acquisition, so I will.
So you might say recapture belongs to a larger class of services including Amazon's Mechanical
Turk or Crowd Flower from Dolores Labs where there is a large task such as OCR clean up
in all New York Times you defined into very small pieces, and you could distribute it
to thousands of people around the globe and you arrogate back to accomplish the task.
What are your opinions on this general phenomenon some people call it crowd-sourced-labor whether
... or more generally?
>> SCHMIDT: It's clearly one of the industries of the future, and it's enabled by all the
platforms that I'm describing. What I like to think about is--think about
a world where you have a billion people who have smart phones, with ideally Android phones
but there will be others as well, with very powerful applications running on their smart
phones where computers behind them can both take that information in real-time and also
do interesting problems. It's obvious, for example, that not only can
we if you wish, and again I want you to be able to get permission on this, not only do
you tell your friends where you are, but we can also predict where you're going using
these kinds of algorithms. Now again you can turn that off, but for a
lot of people I would like to actually know what is the probability of my friend showing
up within the next 10 minutes. Or shall I go back and go do something else
for awhile. So there's lots and lots of applications where
you have points of information, people. You have servers and then you can either disaggregate
the problem which is the examples that you described, in other words, distribute it out,
or you can centrally answer it. And this is a problem of computer science,
and it's one that I think is a very, very open one for a long time.
>> AUDREY: Hey Eric. Time for one more question.
>> SCHMIDT: Okay maybe we could--we have four people let's do this quickly.
I don't want to...can you just each four of you tell me your questions, and I will try
to answer them all at once?
[audience laughing]
So Pennsylvania has a particularly urgent economic and environmental interest in the
cost of renewable energy generation dropping below that of coal and Google has obviously
been involved in this and, in the name of transparency, I was wondering if you could
give us any insights or updates into the innovative renewable energy innovations and investments
that Google has particularly made, like high altitude wind?
>> SCHMIDT: Perfect. Tell me the next question please.
>> The question is about motivation. When a startup is in its initial phases there
are two things that motivate it. One is if it fails its completely destroyed
it's forgotten about, and if it succeeds it becomes wildly successful often.
If the employees of a startup were to join a big company like Google both of those motivations
are dampened. If they fail, there's their paycheck and their
company can keep them there, and if they succeed the success is dampened because of a lot of
the credit is taken up by the company. I don't know if Google is less of a place
like that. I'm not entirely convinced that, that problem
has been solved at Google in that there is still good motivation to innovate, but I am
sure you are trying to address that problem. And that would be interesting.
>> SCHMIDT: I think I can answer that question. Yes ma'am.
>> Hi, my question has to do with something you said in the beginning of your speech.
You said that growth was driven by technology--private investment in technology.
However, in the beginning of the 20th century it was brought--firms with a broad technological
base that introduced major radical innovations. Also with the current focus on short-term
information, firms are much more narrow and specialized.
Aren't we--don't we face the danger of just seeing more incremental innovations instead
of big radical innovations?
>> SCHMIDT: I really hate to end this on a quazi-confrontational notes, but a lot of
people have criticized Google's efforts to collect information on [indistinct] as Orwellian.
How would you respond to that?
>> SCHMIDT: Okay perfect. A great set of questions let's see if I can
remember. Let's start with green and renewable energy.
The goal here is to get the cost of any form of renewable energy roughly equal to or better
than that of coal. And we started with electricity.
Wind is getting very close. The other ones are still relatively expensive.
We are working hard to invest in--we Google are working hard to invest in companies which
we think will show that way. Even if we don't achieve that objective, and
I hope we will, there are so many other things that we can do for climate change insulating
buildings, better efficiency, better energy management and so forth.
But we can make a pretty--becoming carbon neutral yourself.
Getting your building carbon neutral. Getting your university carbon neutral.
All of those things are important. Let's see there was a question--I'm going
to forget all four, so I apologize. There was a question about Google as sort
of a 1:04:14 wellyen[??] and that's easily written, and I understand the concern.
Google is run on a set of principles as we discussed earlier about information, and I
would remind you that if we were to do something evil with all the data that somehow we're
getting people would leave us very quickly because they would go somewhere else.
They would find another choice for their--or they would just stop using search engines
in their entirety. So the real check on our behavior, besides
our hopefully good morals and good leadership and good decision making, is that our customers
would flee. The press would hate us.
We would be in big, big trouble. You had a question about the industrial model
and the focus on short-term profits and so forth.
In evolutionary versus revolutionary investments in the area of revolutionary investments innovation
leads to both evolutionary as well as revolutionary innovation.
Much of what we see today is incremental if you look in the computer industry.
Many of the products that I use today are really a wonderful refinement of something
that was pioneered 10 years ago, 15 years ago.
And then, every once in awhile, there is something revolutionary.
I would argue, for example, that in its current use Twitter, as an example, of something that
just sort of happened, right? And there are plenty of examples of that,
and so it's easy to decry the short-term focus of the industry.
But I would say that long-term innovation is alive and well.
The incentives still work, the money can still be raised, and it's a good message.
I'm sorry and there was one more question which I forgot.
Oh, this was the culture question. Yes, on acquisitions.
It depends on how what you think people work for. I believe that they work to change the
world. Nobody gets up in my view and says, "Oh, I'm going to work, and I'm going to get
my so many dollars per day and you know I really just wish I didn't work here at all."
At least not in my world. People work for passion and so as we become
larger we try to keep it going. We try to remind everybody why we do what
we do, and it works and I think whether it's a small company or big company, I think you
can motivate around passion and innovation. It produces much better leadership.
Thank you so much, thank you very much. Thank you again Audrey.
>> AUDREY: Thank you so much.
[audience applauding]