The Internet of Things by James Whittaker of Microsoft

>>[MUSIC PLAYING] >>JAMES WHITTAKER: What’s next? We are going to spend the next
hour asking, and answering, that question, what’s next? Because as students, you all spend
a lot of time studying the past. You spend a lot of time
mastering the present.>>But asking, and answering,
the question, what’s next, is what’s going to make you successful. Because if you can’t answer this
question, you end up being left behind.>>Companies that can’t answer
this question get left behind. Individuals get left behind. Asking, and answering,
the question, what’s next, is probably the most important skill
that you will develop in your career.>>You only have to get it
right once, and you’re done. So we are going to ask this question. But in order to get there we’re going
to study the past where people have gotten that answer right and wrong.>>In the 1980s, the big
thing was hardware. In the 1980s, companies
bought room-sized computers and hired rooms full of
people to program them. And in the late ’80s,
things began to change.>>The companies that weren’t asking what’s
next, IBM, DAK, Wang, Sperry, are gone. The companies that did ask,
and answer, what’s next, were the companies that survived
and thrived into the ’90s. And of course, what’s
next, was software.>>IBM got that answer
wrong, and they declined. Microsoft got that answer right,
and they owned the next decade.>>Now, this decade, this
word decade is a hint. This industry runs in 10 year cycles
that are as reliable as Moore’s law is. 10 years is something being born,
and then growing into maturity, and being the big thing, and
then going away rapidly, only to be replaced by the next big thing.>>And of course the next
big thing in the ’90s was software, mostly
the operating system. And the killer app was productivity,
both office productivity and programmer productivity.>>The company that owns that
killer app for what’s next, ends up being the
company that dominates. And of course, that company
in the 1990s was Microsoft.>>Now 10 years later, Microsoft
failed to ask, and answer, the question what’s next, because the
world changes reliably every 10 years. And of course, what’s next,
was the web and the killer app was information retrieval. And Google had the best
answer for what was next.>>Now, in 2007, everybody
thought the cloud was next. Amazon comes out with
Amazon Web Services and everybody was betting on the cloud. That’s what’s next. But see, there’s a problem with that. It was too new. No invention really becomes
important until it’s 10 years old.>>So in 2017, the cloud’s going
to become really important, but it’s still not what’s next. I was a Google employee in 2009 when we
discovered the hard way, what was next. Do you know?>>So you’ve got to get used to
asking and answering this question, or you’re never going to get good at it. And if you get good at it,
you’re going to be relevant.>>Can I hold the questions for the end? Thank you. You’re going to be good at it. Or you were going to guess? What?>>SPEAKER 1: Mobile>>JAMES WHITTAKER: Of course. Mobile. Now, the way we discovered
this, is Larry Page sent out an email that said,
hey, we have a big problem. And he called a few of us together. And he showed us the big problem.>>He showed us the data. And the data was frightening. The data caused us to have
a near death experience. The data showed that users that
use the iPhone– the iPhone had been out for two
years in 2009– people who use the iPhone don’t use browsers. They don’t search the web. They use apps that
search the web for them.>>Imagine our fright as Google. We made– at the time we
made 97% of our income on sponsored links and ads
viewed through a web browser. If this iPhone thing
was going to be big, if this smartphone thing, if mobile was
going to be big, we had a huge problem. 97% of our income gone. Crazy.>>What did they do about it? A couple of interesting moves that
Google made in 2009 that you all should be aware of, because if you
don’t start studying this stuff, you’re not going to be able to get good
at asking and answering this question.>>Do you remember back before
2009, if you Googled what time is it in, say, Columbia, what’s your city?>>SPEAKER 2: [INAUDIBLE].>>JAMES WHITTAKER: I can’t even say that.>>SPEAKER 2: Bogota.>>JAMES WHITTAKER: Bogota, Colombia. What did you get? You got 10 blue links about
Bogota, Colombia, and time. All of a sudden in 2009, you got
the time in Bogota, Colombia.>>Make search more app-like was Larry
Page’s marching order, and we did. Now that’s a big deal for Google,
because every time Google delivers an answer, they don’t get paid for it. They have to deliver a link that might
be bought and paid for, or an ad, in order to make money. This was devastating to the business. Second thing they did is they moved
the A-team from Search to Android. And that’s probably why Marissa
Mayer left the company eventually, because that was a backup plan,
just in case this is important. But when we looked at this–
at Google, we looked at this– and not only were people using
apps, they were right to use apps.>>You understand, apps are a
better way to search the web. Your app searches for you.>>You’re from Brazil. You’re into soccer. What’s a better way
to get soccer scores? Install a soccer app that specializes
in soccer– oh, football, sorry– that specializes in football? Or search the web that tries
to generalize everything? It’s the app, right?>>If you’re a foodie, what’s
the best thing to do? Get an app that specializes in
all the information about food? Or search the web through a
browser that is a generalist?>>If you’re a science
nerd, the science app is going to outperform
the web every single time. Our life flashed before our eyes.>>But then, why, why did the iPhone win? Why? What was it about the
iPhone that made it special? The BlackBerry did everything
an iPhone would do. Windows CE did everything iOS would do.>>SPEAKER 3: Apple is a company
you can trust because they value the privacy of their customers. JAMES WHITTAKER: Nope. Apple has a lot of privacy data. Why? All those other phones, the BlackBerry,
Windows CE, PalmPilot, they all did everything that
those other ones did. What was it?>>SPEAKER 4: Performance.>>JAMES WHITTAKER: No. SPEAKER 5: Good design? JAMES WHITTAKER: No. The App Store. If you wanted to get an app on any other
machine, any other handheld device, before the iPhone, how
did you get that app? You had to go to the web. You had to know where it was. And every single one of those websites
had a completely different download experience, completely different.>>Users couldn’t navigate it. No one ever installed
apps on those devices. You couldn’t update them.>>And Steve Jobs came along
and said, hey developers. I’m going to put your app where
customers can actually find it. How does that sound? Really? You mean I can write
code and it’ll be used? Wow. That’s awesome.>>And it was third party developers
that made that platform what it is. The App Store was the killer app.>>Now see, this is a hint. Start watching for these gap. Start watching for these holes. From hardware to software, there
was a hole that was filled. We can’t program all these
machines individually. One operating system to
rule them all worked. The web.>>We no longer have to distribute media. We no longer have to update
software and ship it out on disk. We can just push it through the web. This is important. And the App Store solved the
biggest problem in mobile. And if it wasn’t for the App
Store, it wouldn’t have worked. Right, because it drew developers. And developers made the
platform what it is.>>I’m going to take questions
at the end, is that OK? Or I’m never going to get through this.>>So what’s next? What’s the 2020s?>>SPEAKER 6: Nanocomputers? JAMES WHITTAKER: We’re
going to get to that. Close.>>Now, here’s what we’re going to do. And I want to teach you
this because I think this is the way to discover the future. There was a bunch of gaps,
a whole bunch of gaps. And you’ve got to get
used to finding them.>>Every time your machines let you down. Every time you think, why did I have
to click five times to get this? Why is this functionality not there? That’s a missing link.>>That’s a gap that you
can step in and fill, but only if you’re watching for them. Only if you look at the holes in the
existing technology as opportunity instead, of something
that pisses you off.>>So what’s next? Here’s what we’re going to do. We’re going to take a scenario, and
we’re going to chuck it in 2020. All right?>>We’re just going to take
a user scenario and we’re going to throw it into the
future and see what happens. Have you ever seen this movie? The best movie ever, right? World War Z.>>It’s the best movie
ever for two reasons. First, it’s got zombies. And zombies are real, right? Not like that bullshit vampire stuff
we had to put up with a few years ago. Vampires are made up. They’re not real, they never will be. But they were kind of sexy,
so we watched them anyhow.>>But zombies, people, zombies are real. We can mathematically define this virus. It’s just a matter of time. It’s like your civic duty to watch
every single one of these movies. So that I don’t have to kill
you during the apocalypse.>>Second reason it’s the best movie ever
is because it’s got Brad Pitt in it. It’s like zombie movie meets
date night, and there we were. I’m going, oh, zombies. And she’s sitting next to me
and she says, Oh Brad Pitt. Somebody told her about the shirtless
scene and she couldn’t wait.>>But then something strange happened,
my date pulls out her phone and begins to share screen time. She’s like, oh Brad Pitt, add
something on– oh, Brad Pitt. And I’m thinking, what’s going on here? What part of shirtless Brad
Pitt does she not understand?>>And then, it got worse. She taps me on the knee and
said, I’ll be right back. And she walks out of the theater
during this most perfect movie. I’m stunned. I’m flabbergasted. I’m coming up with break up lines, man. This is crazy.>>Finally, she comes back, she sits down. She’s gone two, three minutes. And she’s in a huff, right,
she’s clearly hurrying. And I’m waiting for it, right. Waiting for it. Come on. Say it. You missed part of the movie. Say it.>>And she didn’t say it. She didn’t say, what happened? What did I miss? So I thought, all right,
I’ll be proactive. So I said, hey, you
didn’t miss anything. And she said, I know.>>Because she has an app that tells
her when to pee at the movies. There is an app for that. Someone has watched every single
minute of every single movie, and they curated it all. Right?>>Oh, there’s an action sequence
you don’t want to miss. Don’t pee during that one. New character, don’t
pee during that one. But here’s four minutes, right in
the middle, nothing really happens. Go take a leak.>>Now I ask you all, how many
of you all go to the movies? Just raise your hand
if you’re a moviegoer. I think it’s safe to say
we’re all moviegoers. How many of you all have this app? One person. I love it. I do too. Small bladder people.>>Look at this. This is a gap. Here is– $1.98. Right? Your $0.99, my $0.99, and yet all
of you all are potential customers.>>This is called the application
discoverability problem. And it is the biggest technical problem. It’s taking down the App Store. The App Store solved a huge problem. You couldn’t find apps on the
web, put them all in the store, now you can find them. Aha. You can’t find them again.>>That is a big world changing solution. So how should this work? Let’s cast this problem into the future. How should this technology
solve this problem? It should be that I’m sitting in the
movies, and I think, oh I’ve got to go. And I say, oh I don’t know,
Cortana, I have to pee. Now, what does Cortana
have to do to help me?>>First, Cortana has to
figure out what I’m doing. Where is my user? OK. No problem. Lat/long pair. Easy to do. Built in functionality to the phone. So now, Cortana knows
exactly where I am. Where is that? Looks up an address. Maps that lat/long pair to an address. Easy. Built in functionality
on the map application that is already on my phone. No apps, yet.>>And then, next, she’s got to think, OK
what is he doing here at this location? Microsoft Research has a
patent for geolocating you off the face of the earth. We know you’re on the second
floor of that building. Bing has every single floor plan in
most major cities, across the world, as data. OK.>>You’re in theater number four. Let’s see, now I can go look up on the
web what’s playing in theater number four right now. Because I know he’s watching movie,
he hasn’t moved it in 30 minutes. Cortana knows exactly what I’m doing.>>And in fact, Cortana, probably
when I say, Cortana I have to pee, she’s probably going to know, right? Because the last time she geolocated
me in a bathroom was two hours ago. She’s been living in my
pocket for two years. She has my mean time to bladder
evacuation data down, right?>>She’s going to say, yes, James, I know. I saw you buy that beer on the way in,
just wait a few minutes, I’ll vibrate and then you can go do your thing
and not miss any part of this movie.>>Now what happened there? There was no app involved in that. We actually have all of the pieces
to solve this problem, right now. So what can we infer about
the future based on this? There are several things we
can confer about the future.>>First is our technology doesn’t
have to go to the web anymore. We don’t have to go to the web anymore. Our technology takes us there. The web is no longer a destination,
it’s simply a data source.>>The movie times are up there. The pee times are in the cloud. They’re actually sitting in Azure. All that data is already there. The world is beginning
to turn into data. Our devices are beginning to
process and calculate intent almost better than we can.>>Secondly, Search. My machine searched the web. Do you know in 2015 something
really special happened. Now I’m not talking about bots,
I’m talking about machines.>>Machines on the web originated
and consumed more searches than humans for the first time ever. In 2014, it was still
a human dominated web. 2015 it’s a machine–
machines are equal. 2016, 2017, by 2020, the amount of human
generated and human consumed traffic is going to be minuscule. Our machines are going to
be consuming the web for us.>>And then, finally apps. Where’s the app in this? I don’t need the app. The app is a noun. I only need the pee times, I don’t
need all the trappings of it. Give me the answer and I’m happy.>>Why are apps nouns? So you pay for them. We’re going to talk about
monetization at the end, because making money gets
kind of scary in the future. But we’re going to talk about
it, because it’s important.>>Apps have turned into verbs. Right? My technology discerns my intent,
realizes I need something. It’s in the cloud. It’s on the web. Crack the app open,
bring out the answer, put it on my device
just in case I need it.>>All of a sudden, a lot of the things
that we do are no longer human generated. The machines are beginning to take over. So you said nanotechnology, I’m just
going to generalize it to machines, but yes, a lot of them are
going to be very, very small.>>So Microsoft, my boss came to me
summer of 2014, and he’s like dude, HR tells me you haven’t taken
a vacation in four years. Go on vacation.>>And so I Binged it what
this vacation thing was. And apparently, vacation is something
that people who don’t like their jobs do to get away from their jobs. And that kind of pissed me off, right? I don’t want a vacation. So I decided, I want to write code. And so I thought this whole internet
of things thing sounded stupid, right? Nest a thermostat. You can control it. And it’s better, some way, because
there’s a machine doing it.>>No, it’s not. Not in my house. I am a Seattle tree hugger. That thing stays off. Put on a sweater and kiss my ass. Or the lights. I’ve got this light program on my phone. I can do my burglar alarm. It’s harder to do the app than
it is to just go over and turn the light on or turn the light off.>>So I had this whole– I call bullshit on
the internet of things kind of feeling. So I thought, that’s what I’ll do I’ll
investigate this internet of things.>>I’ll put a machine on the internet
of things that deserves to be there. A machine that’s got something to say. A machine that’s difficult
for me to manage on my own, and I could use some robots to help me. And so I put my hot tub
on the internet of things.>>So Step 1 was my vision,
because I always like to vision how these things are going to work. My vision was that Amazon drone would
fly over full of hot tub chemicals. And my hot tub would see it coming,
and open its lid automatically. And we’d just shoot the chemicals
right in the– so I thought, OK first thing I’m going to do
is I’m going to open and close the lid automatically. And that was actually quite easy. A radio frequency controller on
the motor for my hot– easy, easy. I mean seriously, two hours. Most of it was just hooking stuff up. Four or five lines of code and I’m done. And now I’ve got a little Windows
phone app I can open and close the lid.>>Second, was– actually second was not
any of that– second was I thought, I don’t want that Amazon drone
squirting chemicals in if I’m in, so I put in a level, a detector. I can detect when a human gets
in, the water level rises.>>And in fact, it turns out, I was
looking at the data, I can weigh you. If you’re sitting in my hot tub, I
know how much water you’ve displaced, I’ve done the math, and I can weigh
you more accurately than a doctor can. And I know if your head’s
gone under, because I have to estimate your head weight
because it’s not under the water. So if you go under, I
know if you’re drowning. This is totally cool. I had a bug in that, my hot tub kept
thinking somebody was drowning and they weren’t. I’m like, just go ahead drown,
I’m tired of this error message.>>The next was checking and maintaining
water quality, that was the hard part. That took me a good day and
a half to solve that one. Because the way you check water
is you actually dip the water out, and you put in these little
chemicals, and you compare the color. And I don’t have a
machine that can do that.>>So I had to get a laser, and I
actually worked with a nanotechnologist to do this. Get a laser. Shoot it through. And you get the color spread on this
little backing piece of nanotechnology. I don’t know how it works. It gives me data and it allows
me to determine the color. And then I can maintain
the water quality. Very cool.>>Next was reorder chemicals. Easy. Amazon makes buying things
from the web so easy. Bless them. So it can reorder chemicals. It can monitor use. It sends me a signal when somebody’s in. It monitors how often
people are in, and I’ve gotten a bunch of data about that right.>>I know how dirty you are
when you get in my hot tub. Because I can check the
water before you get in. I can check the water after you get in. I know what you’ve done in my hot tub.>>The amount of data you can
get from these machines is pretty, pretty amazing. It will troubleshoot itself.>>So that’s what I’m doing now. I’m monitoring the voltage and trying to
figure out all the noise from the grid. The Seattle grid is really spiky. Your grid is, probably,
really spiky too.>>But I’m beginning to find the patterns
for when my seals are wearing out, because when the seals
wear out, voltage spikes. And it stays up. So it’s one of these– you
can detect it over time. I’m getting to the point where
it’s going to be able to do that.>>And then next summer,
I’m thinking 3D printing is going to get good enough that
I can just print my own seals. By the way, that’s something
else that’s happened in 2015. Before 2015, we were only
3d printing in plastic. This year we’ve added metal. We’ve added carbon fiber. We’ve added sugars. We can print carbohydrates. This is crazy.>>SPEAKER 6: There scientists [INAUDIBLE] JAMES WHITTAKER: Absolutely. And it’s going to match your
DNA, too, so it’s going to know. Because, you know, medicine is
made for a six-foot tall white guy. And so we’re going to able to do
some amazing things with this.>>Proteins are coming next. We’re almost to the point where we can
print cotton and fibers for clothing. They’re printing houses in China. They’re printing cars in North America. And so I’m thinking I can
print a seal for my hot tub.>>So now, the cool thing about
this is, when these hot tubs begin to talk to each other,
because let’s say you have a hot tub and I have a hot tub. And my hot tub might say, dude– because
I get to design the hot tub protocol, and they’re all going
to call each other dude. That’s just going to happen. I guarantee that’s going to happen.>>It can say, dude, here’s my data. Here’s my usage data. Here’s my chemical data. Let’s look at yours. And they’re going to compare notes. And they’re going to figure out what
the best data– what the best chemical concoction is.>>This company makes the best chemical. This company makes the best
chemical for the Northwest. This company makes the best
chemical for a desert climate. They are going to
figure all of this out. And it’s going to be amazing.>>By the way, this is where
the advertising and marketing economy goes away. You can’t advertise to machines. And if we’re right, and
if the machines really are the next thing, what
are you going to advertise? How are you going to say, hey, I
got some hot tub chemicals for you. Look at these hot tub chemicals, man,
they’ve got dancing cats and stuff.>>They don’t care. Machines are going to say, we know
the data, man, don’t come advertising to me. And all the machines are
going to be in on this. Yes.>>SPEAKER 6: One of the next things
is building personal apps for ads so that they don’t have to feel
like they’re being marketed.>>JAMES WHITTAKER: A couple slides
later, I’ll talk about that. Your refrigerator. I’m building a new house. I’m shopping for refrigerators. Refrigerators. You don’t have to just
scan things in anymore, like you used to have to keep the
barcode, so it knows you have chicken.>>Soon as you close your refrigerator
door, a bunch of lasers start shining on your food,
figuring out what you have in there, how long it’s been in there,
and its chemical composition. We’re going to be able to
detect food that’s gone bad. And it’s going to be able to
know what you have, it’s going to be able to cook, suggest meals. Your toaster is going to be
on the internet of things. What the hell’s a toaster have
to say on the internet of things? Actually, I think the toasters have a
lot to say on the internet of things. But I think one is going
to be really interesting. And that is end of life decisions. Because one day, your toaster’s going
to wake up and it’s going to say, wait a minute. Something’s wrong with me. Ah, my data’s off– something’s wrong. It’s going to go out on
the internet of toasters. And he’s going to say, dudes,
something’s wrong with me. What’s wrong with me? And these other toasters are
going to look at the data.>>And they’re going to say,
oh dude, you’re dying. Filament number four is wearing thin. You’re about to go. And it’s going to have to order
its own replacement toaster. Is that sad? That brave little toaster
is going to have to do that.>>And the thing that I
think is going to happen is, you’re not going to
get a choice of toasters. Your toaster is going to replace itself.>>You’re just going to get the stock,
gray toaster, because you don’t care.>>You have a pink Mac and cool glasses. You’re going to get a stylish toaster.>>Your machines are going
to know what you want. They’re going to know your income
level, and the neighborhood, and what other type of toasters
those people are buying. And they’re going to
make decisions for you.>>Now, people in my generation
are old enough to say, I don’t want the machines to take over. I want to make my own decisions. And we don’t matter, because we’re
going to die sooner than you all. And you all are going to die
sooner than the next generation.>>And the next generation’s
going to say, you mean, you picked out your own toaster? That’s stupid. You drove your own car? That’s stupid. Are you kidding me?>>Your alls kids aren’t going
to get driver’s licenses. Isn’t that cool?>>Clothing. Clothing is going to self-market. We’re all going to have to dress better. Because if I say, man, look at that
Microsoft shirt, that is awesome. I want that shirt in my size. Now, I could come up to you, and I
can have this creepy conversation about hey, man where did
you get that cool shirt?>>And all of this, or that thing is
going to be on the internet of shirts. And it’s going to be able to
communicate with my devices. I don’t know how. Maybe I do this. Maybe there’s a button. They communicate.>>And all of a sudden, I have one of
those 3D printing at home in my size. And you get paid from that. We’re going to talk about how
to get paid from all this. You just referred a shirt
manufacturer to a new customer. You’re going to get a
micropayment for that. This is my prediction. Now, it doesn’t matter if I’m wrong. I’m trying to teach you how to ask,
and answer, the question, what’s next? I’m giving you example answers. It may, or may not, work this way. It doesn’t matter.>>What really matters is that you
get practice asking, and answering, this question, what’s
next, so you’re ahead of the curve, instead of left behind.>>So these machines are going
to talk to each other, too. Every time I go, my Microsoft
band– whatever it is– is going to know I’m
traveling in Europe. And I’m over-eating.>>And it’s going to tell my
refrigerator when I get home, hey man, this guy’s unhealthy. You need to make sure you just order
healthy food for him for a few weeks, and let’s get his
statistics back together. They’re going to talk to each other.>>All of this traffic is
going to be going on, and it’s going to wash out everything
humans are doing on the web, everything humans are
doing with the cloud. Our devices are going to discern intent. And they’re going to be
correct most of the time. When they’re not correct,
they’re going to learn from it, and they’re going to
stop being not correct. And within five years,
our machines are going to know more about what we need
to do during the day than we are. That’s my prediction.>>Screens. What if we didn’t need screens? What if we didn’t have to
carry these things around? Because screens go away. If our machines are
automatically discerning intent, why do I need an input device
to tell them what I want to do? Why do I need an output device
to look at their suggestions? We’re going to need it for a
while, because the machines are going to be wrong. The machines are going to say, hey, I
think this is your intent, am I right? Then we’ll need it. But then, they’re not going
to need that for long. Or, they’re going to say, hey,
there’s a couple of choices here. And then we’re going to choose and
then they’re going to learn from that. And we’re not going
to need that anymore.>>If you take away this screen,
what happens to this machine? The electronics in this
machine, without the screen, are about the size of my-
glad I chose that finger– about the size of my index finger. Because if you take away
the screen, you take away 70 some odd percent of the battery,
because that’s all it does is service the screen.>>Moore’s law. Let’s push Moore’s law a
few years into the future. Instead of this machine
being this big, the machine is going to be, in about
four years, the size of just from the knuckle to the tip. And in another four years,
just the size of my fingernail.>>We can sew these machines
into anything, right? You can have computers. Do you realize that this Windows
phone I carry in my pocket, and the smartphones you
all carry your pocket, is more powerful than any
computer that existed in 1994? In 1995, there were two that
gave it a run for its money. 1994, 21 years ago.>>And so now, all this computing power
into the size of my fingernail. You all understand Moore’s
law is under threat. Moore’s law is going away by 2020. We will no longer be able to
get silicon so thin that we can continue to double the number of
transistors on it, and microprocessors. How, or what, are we going to do?>>We’re already solving this problem. IBM has produced– IBM
remember them from the ’80s? They’re coming back. See, it’s much easier to
come back from a long time ago than it is to come back
from owning the last thing. That’s why I think Apple’s in
trouble, because they’re on top now.>>A germanium silicon mix, which is
going to give a few more years of life, maybe as much as a decade
of life, to Moore’s law. We’re also looking at
graphene and carbon nanotubes. Both of those are going to continue
Moore’s law into the far future. At least a future that is far
away enough to be unimaginable.>>Machines are going to
get a lot more powerful. And we’re not going to need screens. The screens are going
to pop up anywhere. And you all had HoloLens on yet? HoloLens uses your brain, tricks
your brain into putting objects that you think are real. Because that’s what
your brain already does.>>I don’t look like this. I am your brain– the way I
look is your brain translating what the light is doing bouncing
off of my skin, and hair, well skin. That’s what HoloLens
uses to trick your mind into seeing things that aren’t there.>>So we put a screen on the wall. Although this is a
total bullshit exercise, no daughter needs her dad to
figure out how to fix this. That daughter’s going to do what
everybody else does, and go to YouTube to figure out how to fix this. So I don’t think I
believe that scenario.>>But this is what it looks like. It’s not virtual reality,
it’s augmented reality. Screens can be anywhere. You can work with this. You can see a Word document
eight feet away from you, giant. You can have a holographic keyboard
materialize and hover right in front of you, and
with the special gloves, it actually feels like a real keyboard.>>The screens are going to
come when we need them. And we’re not going to need
them for most things, just creative things, design things. For the times when we are
intensely human, again. Because that’s my hope is that
the machines will get us back to that intensely human place.>>Now let’s talk about money. And again I’m going to talk– I’m
going to go– I might not be right. I want you all to ask this question. I want you all to answer this question. But I’m showing you how to do it.>>What about money? How are we going to make
money in this future when our machines are
doing everything for us? Well, let’s take a look in the past. How did we make money– sorry, it’s
not my fault Harvard doesn’t have HDMI. >>The web, how did we
make money on the web? There’s two words you’re
missing web and ads.>>Why are browsers still the
same as they were in 1994? Do you all realize that? Netscape Navigator in
1994, here’s how it worked. You invoked it. And you got a rectangle on a screen. It had a text box. You clicked in a text box,
typed the search term, hit Enter, and got 10 blue links.>>That’s the same way the Edge
browser, and the Chrome browser, work 21 years later. Exactly the same way. Why? No other piece of software works
the same way as it did 21 years ago. And it’s exactly the same way. It’s a little faster. And it takes a few more
file formats, but that’s it.>>Why? Because of ads. Browsers are money
making little machines. Because it’s start,
stop, start, stop, right? You have to type in search terms, get
a search, stop, opportunity for ads.>>And then you go to a website,
stop, opportunity for ads. Over and over.>>It’s like the NFL. They only play for 10 seconds before
they stop for commercial break. It’s crazy. Advertisers love it. But it doesn’t work so well
in the apps world, does it? It’s a lot harder. There’s not enough room.>>Apps are more like soccer. Where’s the ads in soccer? You can’t stop a soccer game. Go on, go on– hey, time out, guys. Aren’t you all buggered? Take a break. That doesn’t worked out way.>>And apps don’t work that way, either. Because I got that food app, I got
that science app, for a reason. I want to get work done using my app. Ads get in the way. So when apps came out, all
of a sudden, the ads economy had to make way for
the purchase economy.>>And for $0.99, forever, not
per year, but forever, you can make those ads go
away, and people do. The purchase economy is
beginning to really take hold. And I think you’re going to see it take
hold in some really interesting ways, too.>>If you look at my blog
on, I wrote a blog post called Twitter is
stupid, which is kind of cool, because it trended on Twitter. Twitter is stupid, because
this is why it’s stupid. I tweeted this last year, or something,
traveling to Boston and New York City next week. Any locals who can hook me up with
some good live music recommendations?>>I’m blessed with a lot of followers. I’ve got a bunch of people
from Boston and New York City. Nobody replied. Nobody replied. I got two favorites. Why in the hell would
you favorite this tweet? James can’t find no music, asshole.>>Are you kidding me? So I thought this is broken. Surely all of these places, all of these
live music venues in these two cities have a Twitter account, right? And so I looked it up, because
Bing’s ingested all the Twitter data.>>So I could use a few of the API
causes– it’s not even codes, it’s just making an API call. 1,500 of them. 1,500 Twitter accounts that
claim to be live music venues. So I just wrote a little loop,
and sent this tweet to them. Blast. Four lines of code.>>And then I waited. I didn’t have to wait long. One minute, because every single
people– all of them who care, have a social media coordinator, with
three screens– Instagram, Twitter, Facebook– waiting for customers.>>And I’m thinking, what’s wrong here? Here I am in Twitter saying,
I want to spend money. And here are other Twitter accounts,
I would like to take your money. And Twitter’s like, would you like
to see an ad for– you’re stupid, Twitter, stupid.>>How should it work? It should work like this. Left swipe, right swipe. Data to tell me what I like. Or am I connecting you
to the right people?>>Let’s see. Hyatt Regency Boston says
they’ve got jazz music. Jazz sucks. Left swipe. Holy. The only people that like jazz are
people that play it, for goodness sake.>>The Boston Symphony, they
got classical music now. Now, now, now. We legalized marijuana in
my state a year or so ago, and all of a sudden, classical
music is sounding a lot better. But I’m not yet stoned enough. Left swipe.>>Orpheum Theater in Boston’s got
Cage the Elephant and the Foals. Cage the Elephant is awesome rock
band from Kentucky, my homeland. And I went to see this concert.>>This is commerce. This is the purchase economy
that’s going to begin to take over, and Twitter is beginning to do this. I don’t know if my blog post had
anything to do with it, or not. But you can now donate money to a
political candidate through Twitter.>>They are finally beginning
to connect buyers and sellers within their ecosystem. And you are going to see this. It’s a massive threat to Google. This is a massive threat
to anything that puts ads in the path of buyers and sellers.>>Facebook and Twitter did
this, they would consume much of the purchase economy on the web.>>Now what about when devices
start coming into play? What about this scenario. Cortana, I have to pee. And I go. And I come back. How about that? How does this person make
money in this new economy?>>I think it’s subscriptions
and micro payments. That data is in Azure. That developer pays a subscription
to be an Azure customer.>>So do a million other, I think
Microsoft announced last week a million paid accounts in
Azure, or something like that. So a lot of money in the ecosystem.>>So when Cortana sees me go and
consume third-party content, a micropayment will be made. This makes perfect sense. The subscription money
is already out there.>>You all have subscript– you’re paying
subscriptions to Spotify, and Netflix, and Comcast, or Time Warner, or whoever
whatever lousy cable company you all are stuck with. I know they’re lousy,
because they’re all lousy.>>In order to get from one
Game of Thrones episode to the next Game of Thrones episode
on my Comcast box, it’s 21 clicks. Are you kidding me? That’s an opportunity.>>So for a cent, maybe you watch
World War Z. That’s a $0.04 pee. A three hour Hobbit
movie, that’s a $0.28 pee. The premiere of Star
Wars, that pee’s $1, man.>>So now, instead of you and I, the
two lone people who got this app, makes a couple of dollars, or he
makes money on every single customer. This is the economy that’s coming. Every single, possible customer, when
they consume your value, you get paid.>>Now this, you need to
think about very carefully. Because I think this is
the future of making money. It’s your ability to inject
value into the ecosystem. The more the value gets consumed,
the more money you’re going to make.>>SPEAKER 6: What about RPGs
and buying virtual products? JAMES WHITTAKER: That’s
purchase economy, right? And that’s already taking
place, so it’s now.>>Now, what about things? When these machines really begin
to start doing real things for us, it adds one more additional
way of making money. And that’s sharing. Airbnb, we share our houses. Uber, we share our cars.>>Do you know there’s an Airpnp? Have you ever been driving,
and go, god, I got to go? Why do all my examples
have to do with urination? God, I got to go, right. And you’re driving by all
these houses, with toilets. It’s a buck and a half to
go pee at somebody’s house. And then they can rate you, so
you can’t– it’s got to be clean. >>So sharing comes about. In fact, I’m going to
do that for my hot tub. It’s going to rent itself out, because
it’s completely self-maintaining now. $100 an hour, it’s going
to rent itself out. So anyhow, if you’re in
Woodinville, Washington, next year, and you need a soak–>>[THKK]>>Now, what about when
the machines take over. Because if you listen to people like Ray
Kurzweil, and Bill Gates, and Stephen Hawking, and Elan Musk, everybody’s
worried about this thing they call the singularity, the
point at which the machines get so smart that they don’t need us anymore.>>Now we’ve got a long way to
get there, but on the way, we are going to become
less and less relevant. You understand that our industry has
destroyed a lot of other industries. The video rental industry, gone.>>The photography industry, gone. Eastman Kodak used to
employ hundreds of thousands of people, and millions
of ancillary jobs, and taking pictures and developing film,
and wedding photographers, all of that billions of dollars concentrated
into the 13 employees of Instagram.>>We are really good at destroying
jobs and making it harder and harder and harder for people to make money. So what happens when the
machines start doing everything? What then?>>Because you know what? They’re going to be
better at it than we are. There is going to come a day, in the
very near future, when you will not get on an airplane if there’s
a human in the cockpit. Too dangerous. I don’t know if that person
is taking their meds. No way. There’s going to come a day,
not too distant future, where it will be illegal for
a human to drive a car because the machines
are a lot better at it. They’re going to be
better at everything. They’re going to be better at building. They’re going to be better at driving. They’re going to be better
at paving our roads. They’re going to be better
at designing traffic flows. They’re going to be better
at designing traffic lights. They’re going to be better
than us at all of this.>>So what do we do? When the number of jobs that we
can do better than machines are, OK, I’m going to have to
take this question, aren’t I? SPEAKER 6: Machines don’t
have feelings, and they have no desire to do certain
things, nor can they handle things. JAMES WHITTAKER: They can’t desire
if they don’t have feelings. SPEAKER 6: Right. So they don’t handle things
like interpersonal interactions, there’s going to be more jobs
for people like psychologists– JAMES WHITTAKER: OK. You’re a little bit ahead of me, again. So bear with me. I’m going to get there, I promise.>>So what do we do? Is that what we’re left with? Basically all just the human part? Where we’re philosophers,
and we’re poets? Is this why we’re legalizing marijuana? Are we preparing for this day
where basically it just, let’s think deep thoughts and get high.>>Or it could be what you suggest. It could be that the machines
give us back our humanity. It could be that they take care of all
the things that we don’t want to do. So we can get back to being human.>>But is this singularity going to come? It might. Our machines are
getting really powerful. Why are they? Why is it all of a sudden
that we’re talking about AI? Do you understand what’s
going into this AI? I claim it’s not AI, at all. I claim it’s just code.>>Look. Here’s why our computers, our
machines, are looking so smart now. Number one is they got data
they didn’t have access to. Even a few years ago, they
didn’t have access to this data. Now, everything is born digitally. Of course, they have the–
there’s just that much more data.>>Secondly, that data’s
stored together, now. We don’t really have a web. You all understand the
web is already dead.>>How many of you all have built a
web server, over the last year. So this is the highest concentration
of web server developers that will ever exist in a university. If I had asked this five years
ago, every single one of you all would be building
web servers every week. Because that’s what you did. You configured web servers all the time. I had six of them under my desk
serving different kinds of traffic. That’s not the web we have anymore. The web we have now is data centers. All those web servers have migrated
into the cloud, into data centers. And so they’re all together, which
means we can store contiguous data, related data in the same place, and
off-line process the crap out of it.>>Right now, no one’s
searching for World Cup data. And so all– soccer World Cup, maybe
rugby but rugby is over now, too. So all that’s migrated to back end
machines, and now all this stuff–>>We can push data that’s more
popular different places. And then offline, we
can say, OK hey let’s take a look at this World
Cup data, and see what we can figure out, completely offline. That’s why it’s looking so much smarter,
is because it’s all sitting together. And our machines are a lot
faster at processing it. Vast amounts of data, well-organized,
and processed at speed.>>And yet still, at some point,
we are going to figure something out really important. And I think it’s the
Human Brain Project. That’s what we need to watch. Do you all know about
the Human Brain Project?>>Look, we mapped the human genome with
computers no more powerful than this. In 1990, the Human
Genome Project started. We mapped the genome of an arbitrary
human being in 13 years using computers no power more powerful than this.>>Now we have computers way
more powerful than this. And we started mapping
the brain two years ago. This is where our machines
are going to help. Our machines allow us to do this. We don’t map the brain
without our machines. And what are we going to discover?>>What have we already discovered
in just two short years? Two years into the Human
Brain Project, we’re already hitting milestones at year eight. We know what depression looks like. We know what anxiety looks like. We know what bipolar looks like. We know what autism looks like.>>All of these things
are already mapped out. How much longer until our
machines look at this and say, I know how to cure anxiety. I know how to cure depression. Our machines are going to do this. And it’s going to be stunning.>>So I think our machines
are going to allow us to do what we were meant to do. This, I think, is the fundamental
purpose for human beings is to explore, to terraform other
worlds, to reach other solar systems, to find other lives, to figure out
whether ancient aliens really is true.>>And we are going to slowly solve
every single mystery of mankind. This is what you all are going to do. This is what your
children are going to do. And over the decades, we will exhaust
every single mystery on the planet, and on other planets. And then what? This is the thought I’m
going to leave you with. Then we’ll have questions. Perhaps, just perhaps,
using these magic machines, the power of our minds amplified
by these magic machines, we’ll discover that we weren’t
meant to go to heaven at all.>>But through technology, to
create heaven for ourselves. Perhaps, just perhaps,
the meaning of life isn’t given to us by a higher power. Perhaps, we use our technology
to evolve into that higher power.>>Perhaps, God, it is said,
created us in his own image. Maybe, through these magic
machines, we create god in ours.>>My name is James Whitaker. I work for Microsoft. Thank you. Follow me on Twitter, if you’d like. The transcript of this is
on>>And I’ll take questions. SPEAKER 7: So two
questions, first thing. When we’re talking about–>>JAMES WHITTAKER: Oh, you’re
getting a free question here, huh? SPEAKER 7: –App store
transforming everything. But Ubuntu had an app store that
looked exactly like the one from Apple. It was just for desktop. Why is it that App Store for
mobile is that life changing? JAMES WHITTAKER: Because mobile, that’s
where all the users were in mobile. Now there’s a second piece to this. The second piece to this is Steve Jobs,
and his amazing storytelling ability.>>SPEAKER 7: Because technically,
the Ubuntu app store already solved the problem
of I have an app for–>>JAMES WHITTAKER: But there were
eight people with Ubuntu machines. And there were 200 million people with–>>SPEAKER 7: And that’s the Steve Jobs.>>JAMES WHITTAKER: –iPhones
and that’s Steve Jobs, right. But it was the developers, the
developers building functionality. That’s why you all have
iPhones, and not the other because they’ve got the best app store. What was your second one?>>SPEAKER 7: About the future. I always wonder why people say
so much, internet of things, right now, because we
basically have the technology, we had this technology doing
most of this stuff like 10 years, or even 20 years ago.>>JAMES WHITTAKER: Ah, I don’t know. We had the sensors, but the
data– we didn’t have the data. There wasn’t anything
useful for them to do. And we didn’t have the conductivity. Bandwidth is almost free now. So.>>SPEAKER 7: Does the data
bring that much into it?>>JAMES WHITTAKER: But
it’s 10 year cycles. Everything that you all are
using right now is 10 years old. And what’s going to be really big in
10 years has already been invented. That’s why the cloud is
underpinning all of this. And the cloud was invented in 2007. So it takes a while for the world
to catch up with the technology. Yes, ma’am.>>SPEAKER 6: So, the Brain
Project, there are those people in the field of psychology, who feel
that neuropsychology may turn out to be nothing more than phrenology. How do you think– is it
possible to quantify and create algorithms, understand a state
of consciousness and intent, when we don’t understand
what those things are?>>JAMES WHITTAKER: So I actually
think that the machines are never going to catch up to us. My opinion is that
gray matter will end up being triumphant over
germanium, silicon, carbon nanotubes, and graphene. I think there’s something going
on up here that’s really special. I do think we’re going to figure it out. I’m not sure we’re going to
figure out how to build it. Yes, sir.>>SPEAKER 8: Who governs this? If we get to a point where the machines
and our software are making all of these momentous decisions, does
that mean that at some point we– in our world where governance isn’t
important anymore, and Google, Microsoft and international
corporations–>>JAMES WHITTAKER: That’s why people
are signing all these petitions to create rules about never– we’ve
already programmed machines to kill. We’ve done that, and the
machines have independently killed people, because they
are obeying their programming.>>And so, we can program them
to do anything we want. Where are those laws going to come from? Do people trust– the companies
at the forefront of this are Microsoft, Google,
IBM, even Cisco, Tesla. Who are we going to trust?>>I mean, this is a
societal level discussion that really needs to take place. I don’t have an answer for it, because
I don’t think there is one yet. Donald Trump, Vladimir
Putin, they’ll figure it out.>>[LAUGHING]>>I love blue states. We can laugh on things like that. Wow, I have to be careful
when I talk in red states. Any more questions? Yes, sir.>>SPEAKER 7: So when you
were talking about things, you know, if you went
to the movie theater, and you used the bathroom, that
a payment would already happen.>>JAMES WHITTAKER: Yeah. Money is going to be moving
around, so that micropayment, I am 100% sure that
that’s going to happen. It has to. The App Store is breaking down. No one’s making money in the App Store. And whenever no one makes money,
that’s when change happens.>>It was harder to make– easier to
make money on web than Windows. That’s when the transition went. When it was easier to make
money on mobile than the web, that’s when the transition went. When it’s easier to make money
in the cloud than on mobile, that’s going to be the transition. It’s capitalism. Did I get all your question?>>SPEAKER 7: So actually
it was more about– does that mean you see the industry
becoming more consolidated? >>JAMES WHITTAKER: At first. But it’s going to be a reaction. My prediction is, actually, that
it’s going to be individuals. I don’t know how long the
big companies are going to last because we won’t need them.>>When infrastructure is
free, storage is free. Yeah. If you all need to go, you won’t
hurt my feelings by filing out. But when storage is free,
when infrastructure is free, when networks are free,
when communication is free, there’s no advantage to the
big conglomerates anymore. So it’s an individual thing.>>It’s your ability to
code, it’s your ability to code, it’s her ability to code,
and inject value into the ecosystem. Because the infrastructure
is going to be free. It could be a great
age of the individual. Yes, sir. So did you have the
grand scale of everything you covered, but I just
had a question about what your vision for [INAUDIBLE], how to
interact with [INAUDIBLE] directly everything is totally taken care of. So you talked about
[INAUDIBLE] you’d have Word doc and [INAUDIBLE] over here. And so one thing I found about
physical [INAUDIBLE] systems is that some things are
evolving, like touchscreens are definitely getting way
better than they used to be. Mice are getting way better
than they used to be. But the physical keyboard
is actually pretty amazing. And it’s [INAUDIBLE] JAMES WHITTAKER: Is it, or is
it– so I don’t know the answer, but I want you to not get
lost in your opinions. Question that. Is it just your familiarity
with a physical keyboard that’s creating this affinity to it? Or is it really something that is–
because we’re very limited by it. This corded keyboard is actually
the wrong way to type fast. Artificially, slows us down
because the old mechanical things we needed to slow typists down. Right? I don’t know.>>I think, I believe that
the machines are going to be able to figure out
intent really easily. If you’re a writer,
you’re going to need it. Right? The machines, those input devices,
are going to be for the creatives. That’s how we’re going to know who are
our artists, who are our creators, who are our designers, and who are just
the muggles, that go through world not needing their machines. >>Keep thinking about this, and
be careful of your biases. We all inject biases in this. I’ve got biases injected in this. That’s why I warned you, this
is not what’s going to happen, it’s my opinion of
what’s going to happen. I want your all’s take away to be you
walk out of here thinking, what’s next? What’s next? I want you cussing me in the middle
of the night when you wake up and say, what’s next? Damn James. Asshole. What’s next. Careful of your biases.>>Is there more questions? Yes, sir.>>SPEAKER 7: Earlier you talked about
machines being able to predict intent and what people want. What happens– will they be able to know
when people want to change their minds or trying something new? Let’s say your fridge is going
to order your groceries for you. What if one day I want to
try coconut milk ice cream. JAMES WHITTAKER: I
remember my mother-in-law– I remember tasting
flavored coffee one time and thought, um, that’s really good. And every time we went to visit
her she had flavored coffee flavor and I’m like, how do I tell her? So I mean, we don’t have to worry
about hurting machine’s feelings. And so you just–>>SPEAKER 7: Tell it.>>JAMES WHITTAKER: Yeah. Tell it. Speak to it. Or whatever. The interfaces– I think there’s
going to be a time, a phase where we are going to be discovering this. It seems to me that you know, you said
touch screens are really getting good. To me that means they’re doomed. Because as soon as we perfect
technology, we don’t use it anymore.>>PCs got really good, and
we’re like, oh sorry. We’re cheating on our
PCs now with these. These are going to get really good and
it’s going to go into something else.>>10 years. 10 years seriously. 10 year cycles. Think about that. It is as reliable and as
accurate as Moore’s law. >>Anybody who hasn’t
asked a question, yet? >>So I also teach a course on
storytelling, and on creativity, the brain science guy
on how to be creative. Neuropsychology is a
really, really cool topic. And I’ve decoded creativity and I
know what it takes to be creative.>>So if you all enjoyed this,
I can come back sometime and teach one of the other seminars. And the storytelling class actually
tells six stories and deconstruct them. If you go to, you can see
me doing this with a high school. I give a high school commencement speech
throughout the state of Washington.>>And I give this speech,
and then deconstruct it, and talk about why you all have
been sitting here paying attention. When most of the time you sit
there and stare at a PowerPoint, you’re bored of fucking tears, right?>>There is actually a
method to the madness. And I tell a great
story about Larry Page. SPEAKER 6: Can we come to workshops? Or do you just do seminars? JAMES WHITTAKER: Well, I teach
these monthly at Microsoft. All of them–>>SPEAKER 6: But for creative writing, JAMES WHITTAKER: Oh, right. Got it, no. I give homework. And so, at Microsoft, the way I do
it is I take classes and chop them into four bits, and give homework,
and then we talk about the homework afterwards. It would be great. So maybe we could do one together.>>I just hate running workshops. So I don’t do it,
because I don’t like it. Not that it’s not important,
it’s very important. I just don’t like it. Any more questions? Peace.

Author: Kennedi Daugherty

Leave a Reply

Your email address will not be published. Required fields are marked *