Difference between revisions of "Honorary Doctorate Speech at Kyoto University"

Latest revision as of 03:05, 6 December 2017

so thank you very much I

don't understand

Japanese so it was very easy to

sit through the introduction

as

I mentioned a little while ago during the

ceremony this

day is a day in remembrance

of our friend Minko kambayashi

who just

a few days from now

five years ago died

suddenly of a cerebral hemorrhage and

shocked all

of us who knew him and loved him and

I

gave the title of this talk before

I prepared it and when I started

actually turned into a somewhat different talk

so hopefully you'll

like the talk i'm going to give rather

than the one I'm not going to give

and our

friend kambayashi San was a

man who improved

the present and looked into

the future and tried to help it grow and

besides

his great interest in children he was Dean

of informatics here and a computer

scientist for for many years and

one of the things that we wonder about in computing

is does computer science

itself have a future and

what I mean by that is not whether

computing has a future because it is obviously

inserted itself into every part of

the lives of ever

more people on the on the planet

but whether the

forward-looking

study and understanding and

invent of computing

that's another story because

in fact compared

to much of the work that was going on in the 60s

and 70s when people were actively trying to

find out what a computer science could

mean much less of that kind

of inquiry is going on today and

by this I mean little science

word in computer

science come to mean something like

the other real sciences or

will it be like the science and library science or

social science which means

a gathering of knowledge but not in

the principled way that physics and chemistry

and biology have

been able to revolutionize our understanding of

phenomena by using very different

techniques than simply the gathering of knowledge and

similarly

we have a term called software engineering

which was also an aspiration

in the late 60s and

the aspiration was to be

a discipline like that which

caused the empire state building

to be built in less than a year including

the demolishing of

the original site less than

a year demolish build the empire state building and

start occupying it done by

a little less than 3,000 people I

don't think there is any large-scale

venture we could do in computing today that

could organize 3,000 people to do something

as involved

in complex an artifact as the

Empire State Building so both these cases the

Science in Computer

Sciences and aspiration and the engineering and software

engineering is an aspiration and the question is

whether these aspirations can be turned into something

or whether what we have right

now will be deemed

good enough and I don't think it I don't think it

is good enough so here's a one way to

think about one or two of

the problems a

400-page book is about twenty thousand

lines of some language and

the foot of books is about

fifteen books and about

three hundred thousand lines in some

language and the Empire State Building is about

twenty two thousand books high or about

440 million lines in some language

and seems like a lot

but consider that just

Microsoft Vista and the microsoft

office together are

about 260 million lines

of code that's about thirteen thousand books

nobody has ever read those books I

don't think anybody at Microsoft

could tell you exactly what that code does

could tell you how to fix it could tell

you how to improve it and

so we take two extremes

looking at these figures one extreme

says well that's the size that's the actual

complexity of the functionality delivered

here that

in functionality by the way required me

to reboot my computer before

giving this talk because it couldn't adjust the software

on it couldn't adjust to the projector in

such a way to allow the application to display

the full screen correctly so

in those two hundred sixty million lines of code at least that

feature is not in there yet

so that will probably require an

a million lines of code or

the other extreme is that

this 260 million

lines of code is grossly

bloated

that very

little that contributes to any kind of functionality whatsoever

and it's really kind of a mess

n I'm not really just picking on

Microsoft because a very large worldwide

company I call it lfp for

large financial product company

their suite of products

comes to about 350 million

lines of code or about 17,000

books and their user interface has a

grand total of one hundred and forty five thousand

different screens they use

that term screens because i think they are still

living back in the days of the IBM 3270 where

that term was actually invented so

I doubt seriously whether the functionality

actually delivered is in any way

commensurate with the amount of code that

people have to deal with and of course that code is not

organized nearly as nicely as a book that

actually looks more like like this looks like kind of like a garbage

dump that's kind of what it is and

it's the kind of thing that would be condemned

if anybody could ever look at it but

if you look at a garbage dump through

a window about this big a little

part of it only looks slightly messy

if you could look at the entire garbage

dump of code I think we would be horrified

that's not really a dumb

because people have to live

with this people usually don't live on a dump so

perhaps this is a city after

a great disaster where people

are still forced to live

and part of this I think comes

from the small horizons that

our brains naturally

have our brains

are set up to worry about what's going to happen

the next few minutes and next few hours and we

can remember fairly vividly the things that happened a

few minutes ago and a few years ago

but everything else is kind of blurry and

because nothing

much happened 100,000 years ago

the evolution

didn't make our brains

to be in tune with the idea of progress as

our brains are made for coping not for

progression and so if there is a change it's a little

change one way or the other or a complete disaster but

the idea of making things remarkably better

in some way is not in our common

sense reasoning the problem is we live

in a time where this little window of

a few years behind 1970

and a few years ahead of 1970 it

actually part of a progression looks kind of like this

partly caused by Moore's Law

and partly caused by the Industrial

Revolution and a number of other factors

and the problem

is is that when you look at a small

exponential curve that can look linear and our

brains love linear things and so

we actually have to focus our will on

change

to try and understand whether

this change is actually really a linear change

like our brains wanted to or whether it's one of these horrible

so

again because our our brains are so

limited whenever we have

to deal with very large things

we're just overwhelmed by them and we start

turning large things into religions

so instead of thinking about the

content of 10,000 books we start believing

it so 200 million lines of

code is too large for us and most

people have never read a thousand

books so 20

million lines of something is a lot and

if somebody were to present

us with a hundred books worth of something that's a lot

10 books

we can get through 10 books is maybe what

we're supposed to read in a in a year at college

so that's one thousandth the 200

million lines there and one book is

omething we could probably handle twenty

thousand lines of something we could handle if you could imagine a large

software system

that could be expressed in twenty thousand

lines of something that

would actually be something like a Moore's law for software

right now we have an inverse Moore's law

for software because as the

positive Moore's law for software for

hardware is chugging along we have the interesting

property that but much of the applications

we use on this machine anyway actually

take longer now to load than they did 20 years ago

so what it means is is that

no matter

how much the

silicon factories in the world are

improving the speed and

capacity the software people are actually taking

that speed in capacity away

now of course this might be necessary I've

met many people who think this is a necessary thing

that actually what we have today is the best

of all possible worlds so we'll we'll

see whether that's really true now the other

problem is that 20,000 things

can be too much so here are 20,000

wires this is a real picture and on a staged

picture I'll not tell you where it this

picture was taken but the

person who started this off did it very innocently

one place had to be connected to another place and

person got a wire and connected it up and another

place had to be connected to another place and so

that person or another person got a wire and connected it

up and pretty much pretty soon they had this

and this is kind of the way most software grows

simple

little things can be handled but if

those things don't scale you wind up with this and

the

couple of years I've been asked embarrassing questions by

my wife who is a writer and a

an artist and here

is a actual screen

shot of her cinema sighs display

at her home and

I just segregated the software

questions about so as one kind of software on

the Left can sort of see what it is and there's

another kind of software on the right and her question

is well in these apps here

on the Left like Microsoft Word and

photoshop and so forth I can

she says I can see and do WYSIWYG authoring so

the stuff is right there and I

can make a change in it and it shows me the change and it's

fairly easy to work with but

right all the stuff that has to do with the web well

it isn't WYSIWYG and so

I have to go into special editing modes I have to

use the wiki I have to edit through a keyhole I have

to do many different kinds of

things well it's embarrassing enough for

her to ask me about that

what's even more embarrassing is the stuff

she likes was invented in the 70s the

stuff she didn't like was invented in the 1990s

about 20 years later so

in fact a large percentage of

software a lot of people used today because it's connected

to the web is actually inferior

to stuff that was done before and for

no good reason whatsoever

there was nothing about the

web that prevents

it from being WYSIWYG and easy to use

it was the people's attitudes

who did the web browsers that prevented this not

anything technical and so

we call lists reinventing the flat tire

in most sciences

your cautioned against reinventing the wheel

because you're supposed to have read

about what people have done before and

but it happens but

it's really bad when the reinvention of

stuff is worse than the original stuff because

people have paid attention to what's what's

been going on and so

basically we could say the present

doesn't compute we're in a

situation where the hardware capacity

allows a lot of stuff that

doesn't scale very well to barely survived

but in fact it survives

at the expense of being able to improve

it make changes understand

forth so this is what we want to look at in this

little talk today

when way of looking at this is when a new

idea appears like the printing press there

are two separate ways of looking at this new

idea one is as news news

is incremental to what you already know and so

when the people saw the Gutenberg

Bible their reaction

was oh this is I understand this

this is doing much more

cheaply a much more quickly what the monks did

by hand so this is an improvement to

what I already know and in fact the Catholic Church

did not think to suppress the printing press because

it seemed to be working

for them who's making more

Bibles but along with news if

the idea has some

potency there's also new and

what was knew about the printing press was going

to be in about a hundred years or so was

the invention of science and the invention of new ways

of governance so Kings were on their

way out and the power of the Catholic Church

was on its way out and nobody realized it because they

didn't think about what was knew about the printing press and

usually when something new comes along there's

a change in outlook it's not just

something incremental there's a different way of looking at things and

in this case the outlook shifted from a

world that believed things to

a world in which people started learning

how to argue about things how

to be suspicious of things rather than believing

things and the

problem with ideas is

that our brains are set up for News News

is simple news

is incremental it can be told very quickly we've

been doing it for a hundred thousand years but new

might require you learn something

for five years before you can truly appreciate

the new idea sonu

is not very popular most things

that are new are converted into news or done away with

entirely they're modified

in some form that tries to

kill off what's new and just use it as something

that's a mere improvement rather than a change

everybody likes change if

you talk to them everybody likes

change except for the change part so

people are talking about change all the time but the

the actual ability

face of a new I is very difficult for

most people in our favor

Canadian Marshall McLuhan said

this about goldfish I don't know who discovered

water but it wasn't a fish but he meant we are the fish

and the water the fish is swimming in is

our beliefs most

of our beliefs we can't see we can't even remember that

we have them because we treat our beliefs as reality

rather than

arbitrary things that

we just happen to learn because we grew

up one place and not another I McLuhan

was very good at pointing out the

difference between news and new

so one of the things he pointed out about news is

that the present for most people is only seen

in terms of the past it's

interpreted by what people already

know and it's thought of as reality

well it seems reasonable I mean that's

all we bring isn't it into each present

moment is the sum

total of what we have been how can we deal with

the President and nettie in any other way

one of his great lines was

he said until I believe it I can't even see it

another thing

McLuhan said is that you

shouldn't try

to worry about whether something is true or false

or good

or bad when you first encounter it because

if you do you're bringing your

past to make that judgment instead

of worrying about things are good or bad true

or false you should simply try and

find out what is going on and he meant try to look

at the present as it is

so this is a new idea for humans

because most humans are brought up in a culture and

they think of new as

something that is damaging

to their reality

rather than thinking of their reality

as a construct an ax and a perspective

and McLuhan pointed

out that artists and scientists can see some

of the present for itself and they do this in different ways

and because they can see the present a little bit they

can also see a little bit of the future that

is basically what we have to do in

our field and also for our children

now why is this difficult

it has to do with the makeup of human

genetics across the planet

we can learn new behaviors

but part of our behaviors come out of what

we are as human beings for instance if you look at

human beings across the planet and give

them present them with a new tool or a new idea

about ninety-five percent of them will

evaluate that new tool or idea as to

whether it contributes to

their current goals so most people are already working

things they have plans they have goals and

they stick to those plans and goals very strongly and

if the new tool or new idea doesn't contribute

to those plans or goals they will reject

it but about five percent of its look

at it in a different way five percent of

human beings might change their current plans and goals

in the presence of a new idea so it's one in

20 now if we look at human beings in

a different way eighty-five percent of

us do most things

because

other people deemed them and important doesn't

call outer directed the

goals that are chosen the rewards that are gotten

our gotten by living in a society makes sense because

we're social animals about fifteen percent

of us have more inner rewards

or less affected by what

other people think and if you combine these two

ways of looking at things you got a picture that looks a

bit like this where only one percent

of us is inter directed

and interested in ideas and tools and

the other extreme is that eighty percent

of us is outer directed and

very conservative

holding on to our goals and very instrumental in our thinking

these are the two

dominant reasons why new ideas have a hard

time penetrating and the newer an idea

is the more unusual it is the

more difficult it is because the eighty percent of us

actually doesn't think about the ideas so

much as looks for consensus and so

a typical time even

for trivial ideas to reach

consensus for the eighty percent is about

thirty years and more complicated

ideas may take 50 years or

even longer so for this group

change has to be more or less almost agreed

on by everybody before the change happens

and then four

percent of us think a lot of engineers

are in this group are interested in

ideas and tools and want

to work on problems the society deems important and

about fourteen percent of us is very

dangerous if you think

about what it means to be inter

directed and very stubborn in your own

pursuit of goals this group

makes a lot of politicians and

managers and so forth

so they this is a tough group to

with so this is a very simple way of looking

at human beings but it

gives a lot of explanation as to why though

is relatively simple to come up with brand new ideas

that are very powerful it is very very difficult

to disseminate those ideas

so we go back to 1970 and take

a look about what was news back then it was mainframes

the spinning tape drives

big computers owned by companies sort

of factory models of computing

and a strong belief that

you should always get hardware and your operating

programming language and your tools and your user

interface from a vendor they should

not try to make your own and many of these

ideas are still very prevalent today

for instance cloud computing and

dealing with vendors covers the space very

well and

then right at the same time about

two dozen people from a community of several hundred

at xerox parc were doing

things that were completely at odds with the news

so personal computing

distributed networks

making

your own tools doing everything

that was new and the

focus was not on factory computing but

on but on end users so

these are if you will this is the eighty percent versus

the one percent and the

outlook of

the

this group is

a little too complicated to try and sum

up in a talk so I just picked one

one part of the outlook that I think

is kind of a good way of thinking

about the way this research

community thought and that is it was a no

centers outlook that

is no discernible center no

hierarchy and

because this group was also responsible for

inventing the internet

simple way of thinking about this is that everything

is kind of like the internet so

the network's themselves were distributed

whether it was a ethernet or an Internet

inside the computers there was no operating

system there was what you might think of

internet all the way down it's kind of interesting to think about a

machine like this in

which if you calculate it

out it has about the capacity and the entire computing

power of the entire Internet in the

80s Moore's

law and so you'd expect if you looked inside

here you'd find something that had thousands

of virtual machines and dozens

of physical processors acting

very much like your own local internet

in fact being a cash for

the process he's running on the internet that's kind

of the way we thought of it back then but instead

what you find is something from the 60s called an

operating system that essentially stovepipes

most things in a very very bad

way makes it very difficult

to integrate things whereas it's very easy to integrate

on the internet

and of course if you have a distributed

physical

or virtual object system sending

messages around you don't need a programming language

because you're already sending

in a programming language and that context

is simply putting some sort of reasonable syntax on

the messages now just think if the

people who many years later did HTTP

realize what they were doing they

could have come up with some conventions for using HTTP

which would have made it the programming language of

the entire internet that would be incredibly

useful instead of the ad hoc set

of know 20,000

wires or the garbage dump that

we have today but they didn't realize

what it was that they were doing and

of course people talk a lot about mash mash

ups but in fact that was

the way things were done back in the 70s

because that's what you get when you have objects

that integrate through a user interface

so and the

inner the user interface does not care where

is that's actually generating these things it's a

way of presenting a whole bunch of things that

same I'm in

some relationship and these relationships

can be driven by by

making relationships between them and

so what we would call an application would

actually be just a set of useful objects working

together with no stovepiping like we have today

then finally the idea was let's not

have any mainframes let's just make zillions

of personal computers and

these personal computers themselves had

virtual processors so the original Alto at

xerox parc had 16 virtual processors

so the whole thing was done without

centers from top to bottom the

hardware and the software and of course some

of these ideas came out into the 80s

and we use them today but what's interesting is

what ideas did come out

which were mainly the ones that had no competitors like

the ethernet there just wasn't anything that competed with it and

to a much smaller extent the user interface

but in fact as

programming techniques and the architectural techniques

almost none of it came out what

we got is what we had in

the early 60s and pretty much what we have today

and underneath this is this

glorious thing called the internet which now has

several billion knows has never broken has

never had to be stopped in order to be fixed all of those

things that could be possible with the software

systems of today but because of the way they're made

simply aren't have never

happened and it'll be very hard to make happen in the future

so one of the

ways you get rid of 20,000 something's is

by finding an abstraction that

will deal with what what this

thing is doing what this thing is doing is simply getting

messages to go from one place to another and

the being able to string a wire from

one place to another means that at some point you have to

being able to send a message from any one node to another

node and by the way that's just what the

ethernet and the internet do but

they did it by virtualizing what

used to be switches into a message

passing situation and so

instead of having an increment that doesn't scale

you can almost always make something

that scales very very well by completely

virtualizing the computer you have into

some new kind of computer that's kind of the message

of this talk so

when computers came about

they had features let's

make more of them the

message of personal computing was the it's the end user

that's the important part of this because

a human is going to wind up using them and therefore

the hard part of the design problem has to do

with how you deal with the end user

one of my favorite

people on the planet is Jeanette wing

she is

great computer scientist theoretical

computer scientist the Dean of computer science

at CMU and she's currently the funder of most

he computer science funding in the United States and

you can see she has a very unique way

of teaching because she is a black belt in

karate and I understand the

students in her class really listen when

she talks and her

definition about computing

is it's the automation of abstractions which i think is a pretty

good sentence

but we have

to think about that for a second because

what if you were born in

10,000 BC with an IQ of 500

or what if you were born Leonardo

in the 15th century

leonardo was smarter than anybody in this room but

he couldn't invent a single motor for

any of his vehicles that he wanted so

even though he was smarter than most of the people who have ever

lived on the planet he wasn't smart enough he

to be born into a different century

here's a guy who wasn't nearly as smart as

Leonardo Henry Ford but

he was born at the right time when there is knowledge

that came out of science and

the improvements in engineering to

start inventing internal

combustion engines which had been invented a few years

before in Germany and all of a sudden that

plus the Industrial Revolution allowed

Henry Ford to make vehicles that Leonardo

could only dream of by the millions so

knowledge generally Trump's IQ

if you know a lot it helps

mart but if you're smart and you don't know a lot it

doesn't help much and finally

what's more important than knowledge is

outlook I realized when I was explaining

this talk this morning that I should have put Isaac Newton

in there because it was Newton more than any

other single person that made Henry Ford possible

by completely changing

outlook on what the physical universe was like as

far as European civilization

was concerned what outlook does

is to give you a stronger way

of looking at things by

changing your point of view and that point of view informs

every part of it it tells you what kind of knowledge to

get and it also makes you

appear to be much smarter so

a person with relatively normal IQ plus

the calculus is smarter than our committee Archimedes

was in ancient Greece

so I like to say knowledge is silver but

outlook is gold I dare

say that in most schools in the United

States most universities and most graduate schools

attempt to teach knowledge

rather than outlook and yet

we live in a world that has been changing out from

under US and its outlook that we need to deal

with that and by contrast to these two IQ

is just a big lump of lead

it's one of the worst things in

field that we have clever people in it because

like Leonardo none of us is clever

enough to deal with the scaling problems that that we're dealing

with so we need to be less clever and be

look at things from better points of view so

if we go back to Jeanette here what we really

want to say is computing is the automation of abstractions

in powerful outlooks we

must find those also now

if we go back to reinventing the flat tire question

what's in the browser

well JavaScript it's a programming

language the do em

and other graphics not very good

but what's interesting is the two of these together

is

more powerful

computationally than the alto was it park in the 70s

and the alto could do all of these things

user interface and the development

system and desktop publishing all of that stuff WYSIWYG

in about 10,000 lines of code

it's not a

lot of code this

code was virtualized and so if

we go back to the browser again we

could say well why don't we just do it Xerox

PARC let's let's forget about JavaScript and

just use JavaScript as an alto it's actually faster than

an alto but let's not let

JavaScript seep into the better

architecture we need to do

our WYSIWYG

environment inside a browser so

going to draw a line there and we're going to say okay this

is machine code here it's not

the way we want a program but we can make something out of it

so we get the guy who did this at

xerox parc originally dan ingles and on

top of that we put a real architecture

known to be able to build these kinds of environments and lo

and behold if you go to

lively

colonel type lively colonel into

google and do it in Safari especially you'll

get to see an entire WYSIWYG environment

done completely in the browser without any downloadable

executables this has always been possible

and I believe it is a

black mark against our field that even

though it has always been possible almost nobody who's working

with this modern software realizes

it's impossible or realizes how to do it this

don't think we quite have a real field now

I believe we're in danger of losing computer science completely

so one way of looking at it is

the simple way of looking at computers is they

can be programmed and people

are incrementing this by making systems using

programs that are kind of like the computer that are programming

language their programming with

but in fact the computer is more interesting

than that it can be programmed to be like any mechanism

you want including a completely different

kind of computer completely different kind of programming language and

that's what we should be doing I believe

most of the sludge most of the bloat has

to do with trying

to make very very large structures out of

just bricks rather than trying to invent arches

so

inventing abstraction here is something

that we have to be very careful with we don't

want to take the news the

abstractions that come to mind when you think of

aggregating blick bricks you get piles

and walls and what we need are

these non-obvious structures that

give us enormous ability to scale

without having

to use lots of material

and in fact abstractions that

have great power tend to be able to go on t-shirts

so Maxwell's

equations it's not it's not a complete test

but generally speaking if

you have a good idea you should be able to invent the mathematics

express that idea so Maxwell's

equations works pretty well equals mc-squared

works pretty well you

can put the American Constitution on a t-shirt and

you can put the

programming language and maybe still the greatest of all

time

something like a system psychology in that

Difference between revisions of "Honorary Doctorate Speech at Kyoto University"

Latest revision as of 03:05, 6 December 2017

Navigation menu

Personal tools

Namespaces

Variants

Views

More

Search

Tool box

Tools

@@ Line 3: / Line 3: @@
 |id=tp9VbtLn2Jw
 |alignment=left
-|autoresize=false
+|autoresize=true
 }}
+<subtitle id="0:0:16">so thank you very much I</subtitle>
+<subtitle id="0:0:19"> don't  understand</subtitle>
+<subtitle id="0:0:22"> Japanese so it was very easy  to</subtitle>
+<subtitle id="0:0:25"> sit through the introduction</subtitle>
+<subtitle id="0:0:28"> as</subtitle>
+<subtitle id="0:0:31"> I  mentioned a little while ago during the</subtitle>
+<subtitle id="0:0:34">  ceremony this</subtitle>
+<subtitle id="0:0:37"> day is a day in  remembrance</subtitle>
+<subtitle id="0:0:40"> of our friend Minko  kambayashi</subtitle>
+<subtitle id="0:0:43"> who just</subtitle>
+<subtitle id="0:0:46"> a few days from now</subtitle>
+<subtitle id="0:0:49">  five years ago died</subtitle>
+<subtitle id="0:0:52"> suddenly of a  cerebral hemorrhage and</subtitle>
+<subtitle id="0:0:55"> shocked all</subtitle>
+<subtitle id="0:0:58"> of  us who knew him and loved him and</subtitle>
+<subtitle id="0:1:1"> I</subtitle>
+<subtitle id="0:1:4"> gave  the title of this talk before</subtitle>
+<subtitle id="0:1:7"> I prepared  it and when I started</subtitle>
+<subtitle id="0:1:10">actually turned into a somewhat  different talk</subtitle>
+<subtitle id="0:1:13"> so hopefully you'll</subtitle>
+<subtitle id="0:1:16"> like  the talk i'm going to give rather</subtitle>
+<subtitle id="0:1:19"> than  the one I'm not going to give</subtitle>
+<subtitle id="0:1:22"> and our</subtitle>
+<subtitle id="0:1:25">  friend kambayashi San was a</subtitle>
+<subtitle id="0:1:28"> man who  improved</subtitle>
+<subtitle id="0:1:31"> the present and looked into</subtitle>
+<subtitle id="0:1:34"> the  future and tried to help it grow and</subtitle>
+<subtitle id="0:1:37">besides</subtitle>
+<subtitle id="0:1:40"> his great interest in children  he was Dean</subtitle>
+<subtitle id="0:1:43"> of informatics here and a  computer</subtitle>
+<subtitle id="0:1:46"> scientist for for many years  and</subtitle>
+<subtitle id="0:1:49"> one of the things that we wonder  about in computing</subtitle>
+<subtitle id="0:1:52"> is does computer  science</subtitle>
+<subtitle id="0:1:55"> itself have a future and</subtitle>
+<subtitle id="0:1:58"> what I  mean by that is not whether</subtitle>
+<subtitle id="0:2:1"> computing  has a future because it is obviously</subtitle>
+<subtitle id="0:2:4">  inserted itself into every part of</subtitle>
+<subtitle id="0:2:7"> the  lives of ever</subtitle>
+<subtitle id="0:2:10"> more people on the on the  planet</subtitle>
+<subtitle id="0:2:13"> but whether the</subtitle>
+<subtitle id="0:2:16"> forward-looking</subtitle>
+<subtitle id="0:2:19">  study and understanding and</subtitle>
+<subtitle id="0:2:22"> invent  of computing</subtitle>
+<subtitle id="0:2:25"> that's another story  because</subtitle>
+<subtitle id="0:2:28"> in fact compared</subtitle>
+<subtitle id="0:2:31"> to much of the  work that was going on in the 60s</subtitle>
+<subtitle id="0:2:34"> and  70s when people were actively trying to</subtitle>
+<subtitle id="0:2:37">  find out what a computer science could</subtitle>
+<subtitle id="0:2:40">  mean much less of that kind</subtitle>
+<subtitle id="0:2:43"> of inquiry  is going on today and</subtitle>
+<subtitle id="0:2:46"> by this I mean  little science</subtitle>
+<subtitle id="0:2:49"> word in computer</subtitle>
+<subtitle id="0:2:52"> science  come to mean something like</subtitle>
+<subtitle id="0:2:55"> the other  real sciences or</subtitle>
+<subtitle id="0:2:58"> will it be like the  science and library science or</subtitle>
+<subtitle id="0:3:1"> social  science which means</subtitle>
+<subtitle id="0:3:4"> a gathering of  knowledge but not in</subtitle>
+<subtitle id="0:3:7"> the principled way  that physics and chemistry</subtitle>
+<subtitle id="0:3:10"> and biology  have</subtitle>
+<subtitle id="0:3:13"> been able to revolutionize our  understanding of</subtitle>
+<subtitle id="0:3:16"> phenomena by using very  different</subtitle>
+<subtitle id="0:3:19"> techniques than simply the  gathering of knowledge and</subtitle>
+<subtitle id="0:3:22"> similarly</subtitle>
+<subtitle id="0:3:25"> we  have a term called software engineering</subtitle>
+<subtitle id="0:3:28">  which was also an aspiration</subtitle>
+<subtitle id="0:3:31"> in the late  60s and</subtitle>
+<subtitle id="0:3:34"> the aspiration was to be</subtitle>
+<subtitle id="0:3:37"> a  discipline like that which</subtitle>
+<subtitle id="0:3:40"> caused the  empire state building</subtitle>
+<subtitle id="0:3:43"> to be built in  less than a year including</subtitle>
+<subtitle id="0:3:46"> the  demolishing of</subtitle>
+<subtitle id="0:3:49"> the original site less  than</subtitle>
+<subtitle id="0:3:52"> a year demolish build the empire  state building and</subtitle>
+<subtitle id="0:3:55"> start occupying it  done by</subtitle>
+<subtitle id="0:3:58"> a little less than 3,000 people  I</subtitle>
+<subtitle id="0:4:1"> don't think there is any large-scale</subtitle>
+<subtitle id="0:4:4">venture we could do in computing today  that</subtitle>
+<subtitle id="0:4:7"> could organize 3,000 people to do  something</subtitle>
+<subtitle id="0:4:10"> as involved</subtitle>
+<subtitle id="0:4:13"> in complex an  artifact as the</subtitle>
+<subtitle id="0:4:16"> Empire State Building so  both these cases the</subtitle>
+<subtitle id="0:4:19"> Science in Computer</subtitle>
+<subtitle id="0:4:22">Sciences and aspiration and the  engineering and software</subtitle>
+<subtitle id="0:4:25"> engineering is  an aspiration and the question is</subtitle>
+<subtitle id="0:4:28">  whether these aspirations can be turned  into something</subtitle>
+<subtitle id="0:4:31"> or whether what we have  right</subtitle>
+<subtitle id="0:4:34"> now will be deemed</subtitle>
+<subtitle id="0:4:37"> good enough and  I don't think it I don't think it</subtitle>
+<subtitle id="0:4:40"> is  good enough so here's a one way to</subtitle>
+<subtitle id="0:4:43"> think  about one or two of</subtitle>
+<subtitle id="0:4:46"> the problems a</subtitle>
+<subtitle id="0:4:49">  400-page book is about twenty thousand</subtitle>
+<subtitle id="0:4:52">  lines of some language and</subtitle>
+<subtitle id="0:4:55"> the foot of  books is about</subtitle>
+<subtitle id="0:4:58"> fifteen books and about</subtitle>
+<subtitle id="0:5:1">  three hundred thousand lines in some</subtitle>
+<subtitle id="0:5:4">language and the Empire State Building  is about</subtitle>
+<subtitle id="0:5:7"> twenty two thousand books high  or about</subtitle>
+<subtitle id="0:5:10"> 440 million lines in some  language</subtitle>
+<subtitle id="0:5:13"> and seems like a lot</subtitle>
+<subtitle id="0:5:16"> but  consider that just</subtitle>
+<subtitle id="0:5:19"> Microsoft Vista and  the microsoft</subtitle>
+<subtitle id="0:5:22"> office together are</subtitle>
+<subtitle id="0:5:25"> about  260 million lines</subtitle>
+<subtitle id="0:5:28"> of code that's about  thirteen thousand books</subtitle>
+<subtitle id="0:5:31"> nobody has ever  read those books I</subtitle>
+<subtitle id="0:5:34"> don't think anybody  at Microsoft</subtitle>
+<subtitle id="0:5:37"> could tell you exactly what  that code does</subtitle>
+<subtitle id="0:5:40"> could tell you how to fix  it could tell</subtitle>
+<subtitle id="0:5:43"> you how to improve it and</subtitle>
+<subtitle id="0:5:46">so we take two extremes</subtitle>
+<subtitle id="0:5:49"> looking at these  figures one extreme</subtitle>
+<subtitle id="0:5:52"> says well that's the  size that's the actual</subtitle>
+<subtitle id="0:5:55"> complexity of the  functionality delivered</subtitle>
+<subtitle id="0:5:58"> here that</subtitle>
+<subtitle id="0:6:1"> in  functionality by the way required me</subtitle>
+<subtitle id="0:6:4"> to  reboot my computer before</subtitle>
+<subtitle id="0:6:7"> giving this  talk because it couldn't adjust the  software</subtitle>
+<subtitle id="0:6:10"> on it couldn't adjust to the  projector in</subtitle>
+<subtitle id="0:6:13"> such a way to allow the  application to display</subtitle>
+<subtitle id="0:6:16"> the full screen  correctly so</subtitle>
+<subtitle id="0:6:19"> in those two hundred sixty  million lines of code at least that</subtitle>
+<subtitle id="0:6:22">  feature is not in there yet</subtitle>
+<subtitle id="0:6:25"> so that will  probably require an</subtitle>
+<subtitle id="0:6:28">  a million lines of code or</subtitle>
+<subtitle id="0:6:31"> the other  extreme is that</subtitle>
+<subtitle id="0:6:34"> this 260 million</subtitle>
+<subtitle id="0:6:37"> lines  of code is grossly</subtitle>
+<subtitle id="0:6:40"> bloated</subtitle>
+<subtitle id="0:6:43"> that very</subtitle>
+<subtitle id="0:6:46">little that contributes to any kind of  functionality whatsoever</subtitle>
+<subtitle id="0:6:49"> and it's really  kind of a mess</subtitle>
+<subtitle id="0:6:52"> n I'm not really just  picking on</subtitle>
+<subtitle id="0:6:55"> Microsoft because a very  large worldwide</subtitle>
+<subtitle id="0:6:58"> company I call it lfp  for</subtitle>
+<subtitle id="0:7:1"> large financial product company</subtitle>
+<subtitle id="0:7:4">  their suite of products</subtitle>
+<subtitle id="0:7:7"> comes to about  350 million</subtitle>
+<subtitle id="0:7:10"> lines of code or about  17,000</subtitle>
+<subtitle id="0:7:13"> books and their user interface  has a</subtitle>
+<subtitle id="0:7:16"> grand total of one hundred and  forty five thousand</subtitle>
+<subtitle id="0:7:19"> different screens  they use</subtitle>
+<subtitle id="0:7:22"> that term screens because i  think they are still</subtitle>
+<subtitle id="0:7:25"> living back in the  days of the IBM 3270 where</subtitle>
+<subtitle id="0:7:28"> that term was  actually invented so</subtitle>
+<subtitle id="0:7:31"> I doubt seriously  whether the functionality</subtitle>
+<subtitle id="0:7:34"> actually  delivered is in any way</subtitle>
+<subtitle id="0:7:37"> commensurate  with the amount of code that</subtitle>
+<subtitle id="0:7:40"> people have  to deal with and of course that code is  not</subtitle>
+<subtitle id="0:7:43"> organized nearly as nicely as a book  that</subtitle>
+<subtitle id="0:7:46"> actually looks more like like this  looks like kind of like a garbage</subtitle>
+<subtitle id="0:7:49"> dump  that's kind of what it is and</subtitle>
+<subtitle id="0:7:52"> it's the  kind of thing that would be condemned</subtitle>
+<subtitle id="0:7:55"> if  anybody could ever look at it but</subtitle>
+<subtitle id="0:7:58"> if you  look at a garbage dump through</subtitle>
+<subtitle id="0:8:1"> a window  about this big a little</subtitle>
+<subtitle id="0:8:4"> part of it only  looks slightly messy</subtitle>
+<subtitle id="0:8:7"> if you could look  at the entire garbage</subtitle>
+<subtitle id="0:8:10"> dump of code I  think we would be horrified</subtitle>
+<subtitle id="0:8:13"> that's not  really a dumb</subtitle>
+<subtitle id="0:8:16"> because people have to  live</subtitle>
+<subtitle id="0:8:19"> with this people usually don't live  on a dump so</subtitle>
+<subtitle id="0:8:22"> perhaps this is a city  after</subtitle>
+<subtitle id="0:8:25"> a great disaster where people</subtitle>
+<subtitle id="0:8:28"> are  still forced to live</subtitle>
+<subtitle id="0:8:31">and part of this I think comes</subtitle>
+<subtitle id="0:8:34"> from the  small horizons that</subtitle>
+<subtitle id="0:8:37"> our brains naturally</subtitle>
+<subtitle id="0:8:40">  have our brains</subtitle>
+<subtitle id="0:8:43"> are set up to worry  about what's going to happen</subtitle>
+<subtitle id="0:8:46"> the next  few minutes and next few hours and we</subtitle>
+<subtitle id="0:8:49">can remember fairly vividly the things  that happened a</subtitle>
+<subtitle id="0:8:52"> few minutes ago and a  few years ago</subtitle>
+<subtitle id="0:8:55"> but everything else is  kind of blurry and</subtitle>
+<subtitle id="0:8:58"> because nothing</subtitle>
+<subtitle id="0:9:1"> much  happened 100,000 years ago</subtitle>
+<subtitle id="0:9:4"> the evolution</subtitle>
+<subtitle id="0:9:7">  didn't make our brains</subtitle>
+<subtitle id="0:9:10"> to be in tune  with the idea of progress as</subtitle>
+<subtitle id="0:9:13"> our brains  are made for coping not for</subtitle>
+<subtitle id="0:9:16"> progression  and so if there is a change it's a  little</subtitle>
+<subtitle id="0:9:19"> change one way or the other or a  complete disaster but</subtitle>
+<subtitle id="0:9:22"> the idea of making  things remarkably better</subtitle>
+<subtitle id="0:9:25"> in some way is  not in our common</subtitle>
+<subtitle id="0:9:28"> sense reasoning the  problem is we live</subtitle>
+<subtitle id="0:9:31"> in a time where this  little window of</subtitle>
+<subtitle id="0:9:34"> a few years behind 1970</subtitle>
+<subtitle id="0:9:37">  and a few years ahead of 1970 it</subtitle>
+<subtitle id="0:9:40">actually part of a progression looks  kind of like this</subtitle>
+<subtitle id="0:9:43"> partly caused by  Moore's Law</subtitle>
+<subtitle id="0:9:46"> and partly caused by the  Industrial</subtitle>
+<subtitle id="0:9:49"> Revolution and a number of  other factors</subtitle>
+<subtitle id="0:9:52"> and the problem</subtitle>
+<subtitle id="0:9:55"> is is that  when you look at a small</subtitle>
+<subtitle id="0:9:58">exponential curve that can look linear  and our</subtitle>
+<subtitle id="0:10:1"> brains love linear things and so</subtitle>
+<subtitle id="0:10:4">  we actually have to focus our will on</subtitle>
+<subtitle id="0:10:7">  change</subtitle>
+<subtitle id="0:10:10">  to try and understand whether</subtitle>
+<subtitle id="0:10:13"> this  change is actually really a linear  change</subtitle>
+<subtitle id="0:10:16"> like our brains wanted to or  whether it's one of these horrible</subtitle>
+<subtitle id="0:10:19"></subtitle>
+<subtitle id="0:10:22"> so</subtitle>
+<subtitle id="0:10:25">  again because our our brains are so</subtitle>
+<subtitle id="0:10:28">  limited whenever we have</subtitle>
+<subtitle id="0:10:31"> to deal with  very large things</subtitle>
+<subtitle id="0:10:34"> we're just overwhelmed  by them and we start</subtitle>
+<subtitle id="0:10:37"> turning large  things into religions</subtitle>
+<subtitle id="0:10:40"> so instead of  thinking about the</subtitle>
+<subtitle id="0:10:43"> content of 10,000  books we start believing</subtitle>
+<subtitle id="0:10:46"> it so 200  million lines of</subtitle>
+<subtitle id="0:10:49"> code is too large for  us and most</subtitle>
+<subtitle id="0:10:52"> people have never read a  thousand</subtitle>
+<subtitle id="0:10:55"> books so 20</subtitle>
+<subtitle id="0:10:58"> million lines of  something is a lot and</subtitle>
+<subtitle id="0:11:1"> if somebody were  to present</subtitle>
+<subtitle id="0:11:4"> us with a hundred books worth  of something that's a lot</subtitle>
+<subtitle id="0:11:7"> 10 books</subtitle>
+<subtitle id="0:11:10"> we  can get through 10 books is maybe what</subtitle>
+<subtitle id="0:11:13">we're supposed to read in a in a year at  college</subtitle>
+<subtitle id="0:11:16"> so that's one thousandth the 200</subtitle>
+<subtitle id="0:11:19">  million lines there and one book is</subtitle>
+<subtitle id="0:11:22">omething we could probably handle  twenty</subtitle>
+<subtitle id="0:11:25"> thousand lines of something we  could handle if you could imagine a  large</subtitle>
+<subtitle id="0:11:28"> software system</subtitle>
+<subtitle id="0:11:31"> that could be  expressed in twenty thousand</subtitle>
+<subtitle id="0:11:34"> lines of  something that</subtitle>
+<subtitle id="0:11:37"> would actually be  something like a Moore's law for  software</subtitle>
+<subtitle id="0:11:40"> right now we have an inverse  Moore's law</subtitle>
+<subtitle id="0:11:43"> for software because as the</subtitle>
+<subtitle id="0:11:46">  positive Moore's law for software for</subtitle>
+<subtitle id="0:11:49">  hardware is chugging along we have the  interesting</subtitle>
+<subtitle id="0:11:52"> property that but much of  the applications</subtitle>
+<subtitle id="0:11:55"> we use on this machine  anyway actually</subtitle>
+<subtitle id="0:11:58"> take longer now to load  than they did 20 years ago</subtitle>
+<subtitle id="0:12:1"> so what it  means is is that</subtitle>
+<subtitle id="0:12:4"> no matter</subtitle>
+<subtitle id="0:12:7"> how much the</subtitle>
+<subtitle id="0:12:10">  silicon factories in the world are</subtitle>
+<subtitle id="0:12:13">  improving the speed and</subtitle>
+<subtitle id="0:12:16"> capacity the  software people are actually taking</subtitle>
+<subtitle id="0:12:19"> that  speed in capacity away</subtitle>
+<subtitle id="0:12:22">now of course this might be necessary  I've</subtitle>
+<subtitle id="0:12:25"> met many people who think this is a  necessary thing</subtitle>
+<subtitle id="0:12:28"> that actually what we  have today is the best</subtitle>
+<subtitle id="0:12:31"> of all possible  worlds so we'll we'll</subtitle>
+<subtitle id="0:12:34"> see whether that's  really true now the other</subtitle>
+<subtitle id="0:12:37"> problem is  that 20,000 things</subtitle>
+<subtitle id="0:12:40"> can be too much so  here are 20,000</subtitle>
+<subtitle id="0:12:43"> wires this is a real  picture and on a staged</subtitle>
+<subtitle id="0:12:46"> picture I'll not  tell you where it this</subtitle>
+<subtitle id="0:12:49"> picture was taken  but the</subtitle>
+<subtitle id="0:12:52"> person who started this off did  it very innocently</subtitle>
+<subtitle id="0:12:55"> one place had to be  connected to another place and</subtitle>
+<subtitle id="0:12:58">person got a wire and connected it up  and another</subtitle>
+<subtitle id="0:13:1"> place had to be connected to  another place and so</subtitle>
+<subtitle id="0:13:4"> that person or  another person got a wire and connected  it</subtitle>
+<subtitle id="0:13:7"> up and pretty much pretty soon they  had this</subtitle>
+<subtitle id="0:13:10"> and this is kind of the way  most software grows</subtitle>
+<subtitle id="0:13:13"> simple</subtitle>
+<subtitle id="0:13:16"> little things  can be handled but if</subtitle>
+<subtitle id="0:13:19"> those things don't  scale you wind up with this and</subtitle>
+<subtitle id="0:13:22"> the</subtitle>
+<subtitle id="0:13:25">couple of years I've been asked  embarrassing questions by</subtitle>
+<subtitle id="0:13:28"> my wife who is  a writer and a</subtitle>
+<subtitle id="0:13:31"> an artist and here</subtitle>
+<subtitle id="0:13:34"> is a  actual screen</subtitle>
+<subtitle id="0:13:37"> shot of her cinema sighs  display</subtitle>
+<subtitle id="0:13:40"> at her home and</subtitle>
+<subtitle id="0:13:43"> I just  segregated the software</subtitle>
+<subtitle id="0:13:46">questions about so as one kind of  software on</subtitle>
+<subtitle id="0:13:49"> the Left can sort of see  what it is and there's</subtitle>
+<subtitle id="0:13:52"> another kind of  software on the right and her question</subtitle>
+<subtitle id="0:13:55">  is well in these apps here</subtitle>
+<subtitle id="0:13:58"> on the Left  like Microsoft Word and</subtitle>
+<subtitle id="0:14:1"> photoshop and so  forth I can</subtitle>
+<subtitle id="0:14:4"> she says I can see and do  WYSIWYG authoring so</subtitle>
+<subtitle id="0:14:7"> the stuff is right  there and I</subtitle>
+<subtitle id="0:14:10"> can make a change in it and  it shows me the change and it's</subtitle>
+<subtitle id="0:14:13"> fairly  easy to work with but</subtitle>
+<subtitle id="0:14:16">right all the stuff that has to do with  the web well</subtitle>
+<subtitle id="0:14:19"> it isn't WYSIWYG and so</subtitle>
+<subtitle id="0:14:22"> I  have to go into special editing modes I  have to</subtitle>
+<subtitle id="0:14:25"> use the wiki I have to edit  through a keyhole I have</subtitle>
+<subtitle id="0:14:28"> to do many  different kinds of</subtitle>
+<subtitle id="0:14:31"> things well it's  embarrassing enough for</subtitle>
+<subtitle id="0:14:34"> her to ask me  about that</subtitle>
+<subtitle id="0:14:37">  what's even more embarrassing is the  stuff</subtitle>
+<subtitle id="0:14:40"> she likes was invented in the 70s  the</subtitle>
+<subtitle id="0:14:43"> stuff she didn't like was invented  in the 1990s</subtitle>
+<subtitle id="0:14:46"> about 20 years later so</subtitle>
+<subtitle id="0:14:49"> in  fact a large percentage of</subtitle>
+<subtitle id="0:14:52">software a lot of people used today  because it's connected</subtitle>
+<subtitle id="0:14:55"> to the web is  actually inferior</subtitle>
+<subtitle id="0:14:58"> to stuff that was done  before and for</subtitle>
+<subtitle id="0:15:1"> no good reason whatsoever</subtitle>
+<subtitle id="0:15:4">there was nothing about the</subtitle>
+<subtitle id="0:15:7"> web that  prevents</subtitle>
+<subtitle id="0:15:10"> it from being WYSIWYG and easy  to use</subtitle>
+<subtitle id="0:15:13"> it was the people's attitudes</subtitle>
+<subtitle id="0:15:16"> who  did the web browsers that prevented this  not</subtitle>
+<subtitle id="0:15:19"> anything technical and so</subtitle>
+<subtitle id="0:15:22"> we call  lists reinventing the flat tire</subtitle>
+<subtitle id="0:15:25"> in most  sciences</subtitle>
+<subtitle id="0:15:28"> your cautioned against  reinventing the wheel</subtitle>
+<subtitle id="0:15:31"> because you're  supposed to have read</subtitle>
+<subtitle id="0:15:34"> about what people  have done before and</subtitle>
+<subtitle id="0:15:37"> but it happens but</subtitle>
+<subtitle id="0:15:40">  it's really bad when the reinvention of</subtitle>
+<subtitle id="0:15:43">  stuff is worse than the original stuff  because</subtitle>
+<subtitle id="0:15:46"> people have paid attention to  what's what's</subtitle>
+<subtitle id="0:15:49"> been going on and so</subtitle>
+<subtitle id="0:15:52">basically we could say the present</subtitle>
+<subtitle id="0:15:55">  doesn't compute we're in a</subtitle>
+<subtitle id="0:15:58"> situation  where the hardware capacity</subtitle>
+<subtitle id="0:16:1"> allows a lot  of stuff that</subtitle>
+<subtitle id="0:16:4"> doesn't scale very well to  barely survived</subtitle>
+<subtitle id="0:16:7"> but in fact it survives</subtitle>
+<subtitle id="0:16:10">  at the expense of being able to improve</subtitle>
+<subtitle id="0:16:13">  it make changes understand</subtitle>
+<subtitle id="0:16:16">forth so this is what we want to look at  in this</subtitle>
+<subtitle id="0:16:19"> little talk today</subtitle>
+<subtitle id="0:16:22"> when way of  looking at this is when a new</subtitle>
+<subtitle id="0:16:25"> idea  appears like the printing press there</subtitle>
+<subtitle id="0:16:28">  are two separate ways of looking at this  new</subtitle>
+<subtitle id="0:16:31"> idea one is as news news</subtitle>
+<subtitle id="0:16:34"> is  incremental to what you already know and  so</subtitle>
+<subtitle id="0:16:37"> when the people saw the Gutenberg</subtitle>
+<subtitle id="0:16:40">  Bible their reaction</subtitle>
+<subtitle id="0:16:43"> was oh this is I  understand this</subtitle>
+<subtitle id="0:16:46"> this is doing much more</subtitle>
+<subtitle id="0:16:49">cheaply a much  more quickly what the monks did</subtitle>
+<subtitle id="0:16:52"> by hand  so this is an improvement to</subtitle>
+<subtitle id="0:16:55"> what I  already know and in fact the Catholic  Church</subtitle>
+<subtitle id="0:16:58"> did not think to suppress the  printing press because</subtitle>
+<subtitle id="0:17:1"> it seemed to be  working</subtitle>
+<subtitle id="0:17:4"> for them who's making more</subtitle>
+<subtitle id="0:17:7">  Bibles but along with news if</subtitle>
+<subtitle id="0:17:10"> the idea  has some</subtitle>
+<subtitle id="0:17:13"> potency there's also new and</subtitle>
+<subtitle id="0:17:16">what was knew about the printing press  was going</subtitle>
+<subtitle id="0:17:19"> to be in about a hundred years  or so was</subtitle>
+<subtitle id="0:17:22"> the invention of science and  the invention of new ways</subtitle>
+<subtitle id="0:17:25"> of governance  so Kings were on their</subtitle>
+<subtitle id="0:17:28"> way out and the  power of the Catholic Church</subtitle>
+<subtitle id="0:17:31"> was on its  way out and nobody realized it because  they</subtitle>
+<subtitle id="0:17:34"> didn't think about what was knew  about the printing press and</subtitle>
+<subtitle id="0:17:37"> usually  when something new comes along there's</subtitle>
+<subtitle id="0:17:40"> a  change in outlook it's not just</subtitle>
+<subtitle id="0:17:43">something incremental there's a  different way of looking at things and</subtitle>
+<subtitle id="0:17:46">  in this case the outlook shifted from a</subtitle>
+<subtitle id="0:17:49">  world that believed things to</subtitle>
+<subtitle id="0:17:52"> a world in  which people started learning</subtitle>
+<subtitle id="0:17:55"> how to  argue about things how</subtitle>
+<subtitle id="0:17:58"> to be suspicious  of things rather than believing</subtitle>
+<subtitle id="0:18:1"> things  and the</subtitle>
+<subtitle id="0:18:4"> problem with ideas is</subtitle>
+<subtitle id="0:18:7"> that our  brains are set up for News News</subtitle>
+<subtitle id="0:18:10"> is  simple news</subtitle>
+<subtitle id="0:18:13"> is incremental it can be  told very quickly we've</subtitle>
+<subtitle id="0:18:16"> been doing it  for a hundred thousand years but new</subtitle>
+<subtitle id="0:18:19">  might require you learn something</subtitle>
+<subtitle id="0:18:22"> for  five years before you can truly  appreciate</subtitle>
+<subtitle id="0:18:25"> the new idea sonu</subtitle>
+<subtitle id="0:18:28"> is not very  popular most things</subtitle>
+<subtitle id="0:18:31"> that are new are  converted into news or done away with</subtitle>
+<subtitle id="0:18:34">  entirely they're modified</subtitle>
+<subtitle id="0:18:37"> in some form  that tries to</subtitle>
+<subtitle id="0:18:40"> kill off what's new and  just use it as something</subtitle>
+<subtitle id="0:18:43"> that's a mere  improvement rather than a change</subtitle>
+<subtitle id="0:18:46">  everybody likes change if</subtitle>
+<subtitle id="0:18:49"> you talk to  them everybody likes</subtitle>
+<subtitle id="0:18:52"> change except for  the change part so</subtitle>
+<subtitle id="0:18:55"> people are talking  about change all the time but the</subtitle>
+<subtitle id="0:18:58"> the  actual ability</subtitle>
+<subtitle id="0:19:1">face of a new I  is very difficult for</subtitle>
+<subtitle id="0:19:4"> most people in our  favor</subtitle>
+<subtitle id="0:19:7"> Canadian Marshall McLuhan said</subtitle>
+<subtitle id="0:19:10">  this about goldfish I don't know who  discovered</subtitle>
+<subtitle id="0:19:13"> water but it wasn't a fish  but he meant we are the fish</subtitle>
+<subtitle id="0:19:16"> and the  water the fish is swimming in is</subtitle>
+<subtitle id="0:19:19"> our  beliefs most</subtitle>
+<subtitle id="0:19:22"> of our beliefs we can't see  we can't even remember that</subtitle>
+<subtitle id="0:19:25"> we have them  because we treat our beliefs as reality</subtitle>
+<subtitle id="0:19:28">  rather than</subtitle>
+<subtitle id="0:19:31"> arbitrary things that</subtitle>
+<subtitle id="0:19:34"> we  just happen to learn because we grew</subtitle>
+<subtitle id="0:19:37"> up  one place and not another I McLuhan</subtitle>
+<subtitle id="0:19:40"> was  very good at pointing out the</subtitle>
+<subtitle id="0:19:43"> difference  between news and new</subtitle>
+<subtitle id="0:19:46"> so one of the  things he pointed out about news is</subtitle>
+<subtitle id="0:19:49"> that  the present for most people is only seen</subtitle>
+<subtitle id="0:19:52">  in terms of the past it's</subtitle>
+<subtitle id="0:19:55"> interpreted by  what people already</subtitle>
+<subtitle id="0:19:58"> know and it's  thought of as reality</subtitle>
+<subtitle id="0:20:1"> well it seems  reasonable I mean that's</subtitle>
+<subtitle id="0:20:4"> all we bring  isn't it into each present</subtitle>
+<subtitle id="0:20:7"> moment is the  sum</subtitle>
+<subtitle id="0:20:10"> total of what we have been how can  we deal with</subtitle>
+<subtitle id="0:20:13"> the President and nettie in  any other way</subtitle>
+<subtitle id="0:20:16"> one of his great lines was</subtitle>
+<subtitle id="0:20:19">he said until I believe it I can't even  see it</subtitle>
+<subtitle id="0:20:22"> another thing</subtitle>
+<subtitle id="0:20:25"> McLuhan said is  that you</subtitle>
+<subtitle id="0:20:28"> shouldn't try</subtitle>
+<subtitle id="0:20:31"> to worry about  whether something is true or false</subtitle>
+<subtitle id="0:20:34"> or  good</subtitle>
+<subtitle id="0:20:37"> or bad when you first encounter it  because</subtitle>
+<subtitle id="0:20:40"> if you do you're bringing your</subtitle>
+<subtitle id="0:20:43">  past to make that judgment instead</subtitle>
+<subtitle id="0:20:46"> of  worrying about things are good or bad  true</subtitle>
+<subtitle id="0:20:49"> or false you should simply try and</subtitle>
+<subtitle id="0:20:52">find out what is going on and he meant  try to look</subtitle>
+<subtitle id="0:20:55"> at the present as it is</subtitle>
+<subtitle id="0:20:58"> so  this is a new idea for humans</subtitle>
+<subtitle id="0:21:1"> because  most humans are brought up in a culture  and</subtitle>
+<subtitle id="0:21:4"> they think of new as</subtitle>
+<subtitle id="0:21:7"> something that  is damaging</subtitle>
+<subtitle id="0:21:10"> to their reality</subtitle>
+<subtitle id="0:21:13"> rather than  thinking of their reality</subtitle>
+<subtitle id="0:21:16"> as a  construct an ax and a perspective</subtitle>
+<subtitle id="0:21:19"> and  McLuhan pointed</subtitle>
+<subtitle id="0:21:22"> out that artists and  scientists can see some</subtitle>
+<subtitle id="0:21:25"> of the present  for itself and they do this in different  ways</subtitle>
+<subtitle id="0:21:28"> and because they can see the  present a little bit they</subtitle>
+<subtitle id="0:21:31"> can also see a  little bit of the future that</subtitle>
+<subtitle id="0:21:34"> is  basically what we have to do in</subtitle>
+<subtitle id="0:21:37"> our  field and also for our children</subtitle>
+<subtitle id="0:21:40"> now why  is this difficult</subtitle>
+<subtitle id="0:21:43"> it has to do with the  makeup of human</subtitle>
+<subtitle id="0:21:46"> genetics across the  planet</subtitle>
+<subtitle id="0:21:49"> we can learn new behaviors</subtitle>
+<subtitle id="0:21:52"> but  part of our behaviors come out of what</subtitle>
+<subtitle id="0:21:55">  we are as human beings for instance if  you look at</subtitle>
+<subtitle id="0:21:58"> human beings across the  planet and give</subtitle>
+<subtitle id="0:22:1"> them present them with a  new tool or a new idea</subtitle>
+<subtitle id="0:22:4"> about ninety-five  percent of them will</subtitle>
+<subtitle id="0:22:7"> evaluate that new  tool or idea as to</subtitle>
+<subtitle id="0:22:10"> whether it  contributes to</subtitle>
+<subtitle id="0:22:13"> their current goals so  most people are already working</subtitle>
+<subtitle id="0:22:16">things they have plans they have goals  and</subtitle>
+<subtitle id="0:22:19"> they stick to those plans and goals  very strongly and</subtitle>
+<subtitle id="0:22:22"> if the new tool or new  idea doesn't contribute</subtitle>
+<subtitle id="0:22:25"> to those plans  or goals they will reject</subtitle>
+<subtitle id="0:22:28"> it but about  five percent of its look</subtitle>
+<subtitle id="0:22:31"> at it in a  different way five percent of</subtitle>
+<subtitle id="0:22:34"> human  beings might change their current plans  and goals</subtitle>
+<subtitle id="0:22:37"> in the presence of a new idea  so it's one in</subtitle>
+<subtitle id="0:22:40"> 20 now if we look at  human beings in</subtitle>
+<subtitle id="0:22:43"> a different way  eighty-five percent of</subtitle>
+<subtitle id="0:22:46"> us do most things</subtitle>
+<subtitle id="0:22:49">because</subtitle>
+<subtitle id="0:22:52"> other people deemed them and  important doesn't</subtitle>
+<subtitle id="0:22:55"> call outer directed  the</subtitle>
+<subtitle id="0:22:58"> goals that are chosen the rewards  that are gotten</subtitle>
+<subtitle id="0:23:1"> our gotten by living in  a society makes sense because</subtitle>
+<subtitle id="0:23:4"> we're  social animals about fifteen percent</subtitle>
+<subtitle id="0:23:7"> of  us have more inner rewards</subtitle>
+<subtitle id="0:23:10"> or less  affected by what</subtitle>
+<subtitle id="0:23:13"> other people think and  if you combine these two</subtitle>
+<subtitle id="0:23:16"> ways of looking  at things you got a picture that looks a</subtitle>
+<subtitle id="0:23:19">  bit like this where only one percent</subtitle>
+<subtitle id="0:23:22"> of  us is inter directed</subtitle>
+<subtitle id="0:23:25"> and interested in  ideas and tools and</subtitle>
+<subtitle id="0:23:28"> the other extreme  is that eighty percent</subtitle>
+<subtitle id="0:23:31"> of us is outer  directed and</subtitle>
+<subtitle id="0:23:34"> very conservative</subtitle>
+<subtitle id="0:23:37">holding on to our goals and very  instrumental in our thinking</subtitle>
+<subtitle id="0:23:40"> these are  the two</subtitle>
+<subtitle id="0:23:43"> dominant reasons why new ideas  have a hard</subtitle>
+<subtitle id="0:23:46"> time penetrating and the  newer an idea</subtitle>
+<subtitle id="0:23:49"> is the more unusual it is  the</subtitle>
+<subtitle id="0:23:52"> more difficult it is because the  eighty percent of us</subtitle>
+<subtitle id="0:23:55"> actually doesn't  think about the ideas so</subtitle>
+<subtitle id="0:23:58"> much as looks  for consensus and so</subtitle>
+<subtitle id="0:24:1"> a typical time even</subtitle>
+<subtitle id="0:24:4">  for trivial ideas to reach</subtitle>
+<subtitle id="0:24:7"> consensus for  the eighty percent is about</subtitle>
+<subtitle id="0:24:10"> thirty years  and more complicated</subtitle>
+<subtitle id="0:24:13"> ideas may take 50  years or</subtitle>
+<subtitle id="0:24:16"> even longer so for this group</subtitle>
+<subtitle id="0:24:19">  change has to be more or less almost  agreed</subtitle>
+<subtitle id="0:24:22"> on by everybody before the change  happens</subtitle>
+<subtitle id="0:24:25"> and then four</subtitle>
+<subtitle id="0:24:28"> percent of us  think a lot of engineers</subtitle>
+<subtitle id="0:24:31"> are in this  group are interested in</subtitle>
+<subtitle id="0:24:34"> ideas and tools  and want</subtitle>
+<subtitle id="0:24:37"> to work on problems the society  deems important and</subtitle>
+<subtitle id="0:24:40"> about fourteen  percent of us is very</subtitle>
+<subtitle id="0:24:43"> dangerous if you  think</subtitle>
+<subtitle id="0:24:46"> about what it means to be inter</subtitle>
+<subtitle id="0:24:49">  directed and very stubborn in your own</subtitle>
+<subtitle id="0:24:52">  pursuit of goals this group</subtitle>
+<subtitle id="0:24:55"> makes a lot  of politicians and</subtitle>
+<subtitle id="0:24:58"> managers and so forth</subtitle>
+<subtitle id="0:25:1">  so they this is a tough group to</subtitle>
+<subtitle id="0:25:4">with so this is a very simple way of  looking</subtitle>
+<subtitle id="0:25:7"> at human beings but it</subtitle>
+<subtitle id="0:25:10"> gives a  lot of explanation as to why though</subtitle>
+<subtitle id="0:25:13">is relatively simple to come up with  brand new ideas</subtitle>
+<subtitle id="0:25:16"> that are very powerful  it is very very difficult</subtitle>
+<subtitle id="0:25:19"> to disseminate  those ideas</subtitle>
+<subtitle id="0:25:22"> so we go back to 1970 and  take</subtitle>
+<subtitle id="0:25:25"> a look about what was news back  then it was mainframes</subtitle>
+<subtitle id="0:25:28"> the spinning tape  drives</subtitle>
+<subtitle id="0:25:31"> big computers owned by companies  sort</subtitle>
+<subtitle id="0:25:34"> of factory models of computing</subtitle>
+<subtitle id="0:25:37"> and  a strong belief that</subtitle>
+<subtitle id="0:25:40"> you should always  get hardware and your operating</subtitle>
+<subtitle id="0:25:43">programming language and your tools and  your user</subtitle>
+<subtitle id="0:25:46"> interface from a vendor they  should</subtitle>
+<subtitle id="0:25:49"> not try to make your own and many  of these</subtitle>
+<subtitle id="0:25:52"> ideas are still very prevalent  today</subtitle>
+<subtitle id="0:25:55"> for instance cloud computing and</subtitle>
+<subtitle id="0:25:58">  dealing with vendors covers the space  very</subtitle>
+<subtitle id="0:26:1"> well and</subtitle>
+<subtitle id="0:26:4"> then right at the same  time about</subtitle>
+<subtitle id="0:26:7"> two dozen people from a  community of several hundred</subtitle>
+<subtitle id="0:26:10"> at xerox  parc were doing</subtitle>
+<subtitle id="0:26:13"> things that were  completely at odds with the news</subtitle>
+<subtitle id="0:26:16"> so  personal computing</subtitle>
+<subtitle id="0:26:19"> distributed networks</subtitle>
+<subtitle id="0:26:22">making</subtitle>
+<subtitle id="0:26:25"> your own tools doing everything</subtitle>
+<subtitle id="0:26:28">  that was new and the</subtitle>
+<subtitle id="0:26:31"> focus was not on  factory computing but</subtitle>
+<subtitle id="0:26:34"> on but on end  users so</subtitle>
+<subtitle id="0:26:37"> these are if you will this is  the eighty percent versus</subtitle>
+<subtitle id="0:26:40"> the one  percent and the</subtitle>
+<subtitle id="0:26:43"> outlook of</subtitle>
+<subtitle id="0:26:46"> the</subtitle>
+<subtitle id="0:26:49"> this  group is</subtitle>
+<subtitle id="0:26:52"> a little too complicated to try  and sum</subtitle>
+<subtitle id="0:26:55"> up in a talk so I just picked  one</subtitle>
+<subtitle id="0:26:58"> one part of the outlook that I think</subtitle>
+<subtitle id="0:27:1">  is kind of a good way of thinking</subtitle>
+<subtitle id="0:27:4"> about  the way this research</subtitle>
+<subtitle id="0:27:7"> community thought  and that is it was a no</subtitle>
+<subtitle id="0:27:10"> centers outlook  that</subtitle>
+<subtitle id="0:27:13"> is no discernible center no</subtitle>
+<subtitle id="0:27:16">  hierarchy and</subtitle>
+<subtitle id="0:27:19"> because this group was  also responsible for</subtitle>
+<subtitle id="0:27:22"> inventing the  internet</subtitle>
+<subtitle id="0:27:25"> simple way of thinking about  this is that everything</subtitle>
+<subtitle id="0:27:28"> is kind of like  the internet so</subtitle>
+<subtitle id="0:27:31"> the network's themselves  were distributed</subtitle>
+<subtitle id="0:27:34"> whether it was a  ethernet or an Internet</subtitle>
+<subtitle id="0:27:37"></subtitle>
+<subtitle id="0:27:40"> inside the  computers there was no operating</subtitle>
+<subtitle id="0:27:43"> system  there was what you might think of</subtitle>
+<subtitle id="0:27:46">internet all the way down it's kind of  interesting to think about a</subtitle>
+<subtitle id="0:27:49"> machine  like this in</subtitle>
+<subtitle id="0:27:52"> which if you calculate it</subtitle>
+<subtitle id="0:27:55">out it has about the capacity and the  entire  computing</subtitle>
+<subtitle id="0:27:58"> power of the entire Internet  in the</subtitle>
+<subtitle id="0:28:1"> 80s Moore's</subtitle>
+<subtitle id="0:28:4"> law and so you'd  expect if you looked inside</subtitle>
+<subtitle id="0:28:7"> here you'd  find something that had thousands</subtitle>
+<subtitle id="0:28:10"> of  virtual machines and dozens</subtitle>
+<subtitle id="0:28:13"> of physical  processors acting</subtitle>
+<subtitle id="0:28:16"> very much like your  own local internet</subtitle>
+<subtitle id="0:28:19"> in fact being a cash  for</subtitle>
+<subtitle id="0:28:22"> the process he's running on the  internet that's kind</subtitle>
+<subtitle id="0:28:25"> of the way we  thought of it back then but instead</subtitle>
+<subtitle id="0:28:28"> what  you find is something from the 60s  called an</subtitle>
+<subtitle id="0:28:31"> operating system that  essentially stovepipes</subtitle>
+<subtitle id="0:28:34"> most things in a  very very bad</subtitle>
+<subtitle id="0:28:37"> way makes it very  difficult</subtitle>
+<subtitle id="0:28:40"> to integrate things whereas  it's very easy to integrate</subtitle>
+<subtitle id="0:28:43"> on the  internet</subtitle>
+<subtitle id="0:28:46"></subtitle>
+<subtitle id="0:28:49"> and of course if you have a  distributed</subtitle>
+<subtitle id="0:28:52"> physical</subtitle>
+<subtitle id="0:28:55"> or virtual object  system sending</subtitle>
+<subtitle id="0:28:58"> messages around you don't  need a programming language</subtitle>
+<subtitle id="0:29:1"> because  you're already sending</subtitle>
+<subtitle id="0:29:4">in a programming language and that  context</subtitle>
+<subtitle id="0:29:7"> is simply putting some sort of  reasonable syntax on</subtitle>
+<subtitle id="0:29:10"> the messages now  just think if the</subtitle>
+<subtitle id="0:29:13"> people who many years  later did HTTP</subtitle>
+<subtitle id="0:29:16"> realize what they were  doing they</subtitle>
+<subtitle id="0:29:19"> could have come up with some  conventions for using HTTP</subtitle>
+<subtitle id="0:29:22"> which would  have made it the programming language of</subtitle>
+<subtitle id="0:29:25">  the entire internet that would be  incredibly</subtitle>
+<subtitle id="0:29:28"> useful instead of the ad hoc  set</subtitle>
+<subtitle id="0:29:31"> of know 20,000</subtitle>
+<subtitle id="0:29:34"> wires or the garbage  dump that</subtitle>
+<subtitle id="0:29:37"> we have today but they didn't  realize</subtitle>
+<subtitle id="0:29:40"> what it was that they were doing  and</subtitle>
+<subtitle id="0:29:43"> of course people talk a lot about  mash mash</subtitle>
+<subtitle id="0:29:46"> ups but in fact that was</subtitle>
+<subtitle id="0:29:49"> the  way things were done back in the 70s</subtitle>
+<subtitle id="0:29:52">because that's what you get when you  have objects</subtitle>
+<subtitle id="0:29:55"> that integrate through a  user interface</subtitle>
+<subtitle id="0:29:58"> so and the</subtitle>
+<subtitle id="0:30:1"> inner the user  interface does not care where</subtitle>
+<subtitle id="0:30:4">is that's actually generating these  things it's a</subtitle>
+<subtitle id="0:30:7"> way of presenting a whole  bunch of things that</subtitle>
+<subtitle id="0:30:10">same  I'm in</subtitle>
+<subtitle id="0:30:13"> some relationship and these  relationships</subtitle>
+<subtitle id="0:30:16"> can be driven by by</subtitle>
+<subtitle id="0:30:19"> making  relationships between them and</subtitle>
+<subtitle id="0:30:22"> so what  we would call an application would</subtitle>
+<subtitle id="0:30:25">  actually be just a set of useful objects  working</subtitle>
+<subtitle id="0:30:28"> together with no stovepiping  like we have today</subtitle>
+<subtitle id="0:30:31"> then finally the idea  was let's not</subtitle>
+<subtitle id="0:30:34"> have any mainframes let's  just make zillions</subtitle>
+<subtitle id="0:30:37"> of personal computers  and</subtitle>
+<subtitle id="0:30:40"> these personal computers themselves  had</subtitle>
+<subtitle id="0:30:43"> virtual processors so the original  Alto at</subtitle>
+<subtitle id="0:30:46"> xerox parc had 16 virtual  processors</subtitle>
+<subtitle id="0:30:49"> so the whole thing was done  without</subtitle>
+<subtitle id="0:30:52"> centers from top to bottom the</subtitle>
+<subtitle id="0:30:55">  hardware and the software and of course  some</subtitle>
+<subtitle id="0:30:58"> of these ideas came out into the  80s</subtitle>
+<subtitle id="0:31:1"> and we use them today but what's  interesting is</subtitle>
+<subtitle id="0:31:4"> what ideas did come out</subtitle>
+<subtitle id="0:31:7">which were mainly the ones that had no  competitors like</subtitle>
+<subtitle id="0:31:10"> the ethernet there just  wasn't anything that competed with it  and</subtitle>
+<subtitle id="0:31:13"> to a much smaller extent the user  interface</subtitle>
+<subtitle id="0:31:16"> but in fact as</subtitle>
+<subtitle id="0:31:19">programming techniques and the  architectural techniques</subtitle>
+<subtitle id="0:31:22"> almost none of  it came out what</subtitle>
+<subtitle id="0:31:25"> we got is what we had  in</subtitle>
+<subtitle id="0:31:28"> the early 60s and pretty much what we  have today</subtitle>
+<subtitle id="0:31:31"> and underneath this is this</subtitle>
+<subtitle id="0:31:34">  glorious thing called the internet which  now has</subtitle>
+<subtitle id="0:31:37"> several billion knows has never  broken has</subtitle>
+<subtitle id="0:31:40"> never had to be stopped in  order to be fixed all of those</subtitle>
+<subtitle id="0:31:43"> things  that could be possible with the software</subtitle>
+<subtitle id="0:31:46">  systems of today but because of the way  they're made</subtitle>
+<subtitle id="0:31:49"> simply aren't have never</subtitle>
+<subtitle id="0:31:52">happened and it'll be very hard to make  happen in the future</subtitle>
+<subtitle id="0:31:55"> so one of the</subtitle>
+<subtitle id="0:31:58"> ways  you get rid of 20,000 something's is</subtitle>
+<subtitle id="0:32:1"> by  finding an abstraction that</subtitle>
+<subtitle id="0:32:4"> will deal  with what what this</subtitle>
+<subtitle id="0:32:7"> thing is doing what  this thing is doing is simply getting</subtitle>
+<subtitle id="0:32:10">messages to go from one place to another  and</subtitle>
+<subtitle id="0:32:13"> the being able to string a wire from</subtitle>
+<subtitle id="0:32:16">one place to another means that at some  point you have to</subtitle>
+<subtitle id="0:32:19">being able to send a message from any  one node to another</subtitle>
+<subtitle id="0:32:22"> node and by the way  that's just what the</subtitle>
+<subtitle id="0:32:25">  ethernet and the internet do but</subtitle>
+<subtitle id="0:32:28"> they  did it by virtualizing what</subtitle>
+<subtitle id="0:32:31"> used to be  switches into a message</subtitle>
+<subtitle id="0:32:34"> passing  situation and so</subtitle>
+<subtitle id="0:32:37"> instead of having an  increment that doesn't scale</subtitle>
+<subtitle id="0:32:40"> you can  almost always make something</subtitle>
+<subtitle id="0:32:43"> that scales  very very well by completely</subtitle>
+<subtitle id="0:32:46">  virtualizing the computer you have into</subtitle>
+<subtitle id="0:32:49">  some new kind of computer that's kind of  the message</subtitle>
+<subtitle id="0:32:52"> of this talk so</subtitle>
+<subtitle id="0:32:55"> when  computers came about</subtitle>
+<subtitle id="0:32:58"> they had features  let's</subtitle>
+<subtitle id="0:33:1"> make more of them the</subtitle>
+<subtitle id="0:33:4"> message of  personal computing was the it's the end  user</subtitle>
+<subtitle id="0:33:7"> that's the important part of this  because</subtitle>
+<subtitle id="0:33:10"> a human is going to wind up  using them and therefore</subtitle>
+<subtitle id="0:33:13"> the hard part  of the design problem has to do</subtitle>
+<subtitle id="0:33:16"> with how  you deal with the end user</subtitle>
+<subtitle id="0:33:19"> one of my  favorite</subtitle>
+<subtitle id="0:33:22"> people on the planet is  Jeanette wing</subtitle>
+<subtitle id="0:33:25"> she is</subtitle>
+<subtitle id="0:33:28"> great computer  scientist theoretical</subtitle>
+<subtitle id="0:33:31"> computer scientist  the Dean of computer science</subtitle>
+<subtitle id="0:33:34"> at CMU and  she's currently the funder of most</subtitle>
+<subtitle id="0:33:37">he computer science funding in the  United States and</subtitle>
+<subtitle id="0:33:40"> you can see she has a  very unique way</subtitle>
+<subtitle id="0:33:43"> of teaching because she  is a black belt in</subtitle>
+<subtitle id="0:33:46"> karate and I  understand the</subtitle>
+<subtitle id="0:33:49"> students in her class  really listen when</subtitle>
+<subtitle id="0:33:52"> she talks and her</subtitle>
+<subtitle id="0:33:55">  definition about computing</subtitle>
+<subtitle id="0:33:58"> is it's the  automation of abstractions which i think  is a pretty</subtitle>
+<subtitle id="0:34:1"> good sentence</subtitle>
+<subtitle id="0:34:4"> but we have</subtitle>
+<subtitle id="0:34:7"> to  think about that for a second because</subtitle>
+<subtitle id="0:34:10">  what if you were born in</subtitle>
+<subtitle id="0:34:13"> 10,000 BC with  an IQ of 500</subtitle>
+<subtitle id="0:34:16"> or what if you were born  Leonardo</subtitle>
+<subtitle id="0:34:19"> in the 15th century</subtitle>
+<subtitle id="0:34:22"> leonardo  was smarter than anybody in this room  but</subtitle>
+<subtitle id="0:34:25"> he couldn't invent a single motor  for</subtitle>
+<subtitle id="0:34:28"> any of his vehicles that he wanted  so</subtitle>
+<subtitle id="0:34:31"> even though he was smarter than most  of the people who have ever</subtitle>
+<subtitle id="0:34:34"> lived on the  planet he wasn't smart enough he</subtitle>
+<subtitle id="0:34:37">to be  born into a different century</subtitle>
+<subtitle id="0:34:40"> here's a  guy who wasn't nearly as smart as</subtitle>
+<subtitle id="0:34:43">  Leonardo Henry Ford but</subtitle>
+<subtitle id="0:34:46"> he was born at  the right time when there is knowledge</subtitle>
+<subtitle id="0:34:49">  that came out of science and</subtitle>
+<subtitle id="0:34:52"> the  improvements in engineering to</subtitle>
+<subtitle id="0:34:55"> start  inventing internal</subtitle>
+<subtitle id="0:34:58"> combustion engines  which had been invented a few years</subtitle>
+<subtitle id="0:35:1">  before in Germany and all of a sudden  that</subtitle>
+<subtitle id="0:35:4"> plus the Industrial Revolution  allowed</subtitle>
+<subtitle id="0:35:7"> Henry Ford to make vehicles that  Leonardo</subtitle>
+<subtitle id="0:35:10"> could only dream of by the  millions so</subtitle>
+<subtitle id="0:35:13"> knowledge generally Trump's  IQ</subtitle>
+<subtitle id="0:35:16"> if you know a lot it helps</subtitle>
+<subtitle id="0:35:19">mart but if you're smart and you don't  know a lot it</subtitle>
+<subtitle id="0:35:22"> doesn't help much and  finally</subtitle>
+<subtitle id="0:35:25"> what's more important than  knowledge is</subtitle>
+<subtitle id="0:35:28"> outlook I realized when I  was explaining</subtitle>
+<subtitle id="0:35:31"> this talk this morning  that I should have put Isaac Newton</subtitle>
+<subtitle id="0:35:34"> in  there because it was Newton more than  any</subtitle>
+<subtitle id="0:35:37"> other single person that made Henry  Ford possible</subtitle>
+<subtitle id="0:35:40"> by completely changing</subtitle>
+<subtitle id="0:35:43">outlook on what the physical universe  was like as</subtitle>
+<subtitle id="0:35:46"> far as European civilization</subtitle>
+<subtitle id="0:35:49">  was concerned what outlook does</subtitle>
+<subtitle id="0:35:52"> is to  give you a stronger way</subtitle>
+<subtitle id="0:35:55"> of looking at  things by</subtitle>
+<subtitle id="0:35:58"> changing your point of view  and that point of view informs</subtitle>
+<subtitle id="0:36:1"> every  part of it it tells you what kind of  knowledge to</subtitle>
+<subtitle id="0:36:4"> get and it also makes you</subtitle>
+<subtitle id="0:36:7">  appear to be much smarter so</subtitle>
+<subtitle id="0:36:10"> a person  with relatively normal IQ plus</subtitle>
+<subtitle id="0:36:13"> the  calculus is smarter than our committee  Archimedes</subtitle>
+<subtitle id="0:36:16"> was in ancient Greece</subtitle>
+<subtitle id="0:36:19"> so I  like to say knowledge is silver but</subtitle>
+<subtitle id="0:36:22">  outlook is gold I dare</subtitle>
+<subtitle id="0:36:25"> say that in most  schools in the United</subtitle>
+<subtitle id="0:36:28"> States most  universities and most graduate schools</subtitle>
+<subtitle id="0:36:31">  attempt to teach knowledge</subtitle>
+<subtitle id="0:36:34"> rather than  outlook and yet</subtitle>
+<subtitle id="0:36:37"> we live in a world that  has been changing out from</subtitle>
+<subtitle id="0:36:40"> under US and  its outlook that we need to deal</subtitle>
+<subtitle id="0:36:43"> with  that and by contrast to these two IQ</subtitle>
+<subtitle id="0:36:46"> is  just a big lump of lead</subtitle>
+<subtitle id="0:36:49">  it's one of the worst things in</subtitle>
+<subtitle id="0:36:52">field that we have clever people in it  because</subtitle>
+<subtitle id="0:36:55"> like Leonardo none of us is  clever</subtitle>
+<subtitle id="0:36:58"> enough to deal with the scaling  problems that that we're dealing</subtitle>
+<subtitle id="0:37:1"> with so  we need to be less clever and be</subtitle>
+<subtitle id="0:37:4">look at things from better points of  view so</subtitle>
+<subtitle id="0:37:7"> if we go back to Jeanette here  what we really</subtitle>
+<subtitle id="0:37:10"> want to say is computing  is the automation of abstractions</subtitle>
+<subtitle id="0:37:13"> in  powerful outlooks we</subtitle>
+<subtitle id="0:37:16"> must find those  also now</subtitle>
+<subtitle id="0:37:19"> if we go back to reinventing  the flat tire question</subtitle>
+<subtitle id="0:37:22"> what's in the  browser</subtitle>
+<subtitle id="0:37:25"> well JavaScript it's a  programming</subtitle>
+<subtitle id="0:37:28"> language the do em</subtitle>
+<subtitle id="0:37:31"> and other  graphics not very good</subtitle>
+<subtitle id="0:37:34"> but what's  interesting is the two of these together</subtitle>
+<subtitle id="0:37:37">  is</subtitle>
+<subtitle id="0:37:40"> more powerful</subtitle>
+<subtitle id="0:37:43"> computationally than  the alto was it park in the 70s</subtitle>
+<subtitle id="0:37:46"> and the  alto could do all of these things</subtitle>
+<subtitle id="0:37:49"> user  interface and the development</subtitle>
+<subtitle id="0:37:52"> system and  desktop publishing all of that stuff  WYSIWYG</subtitle>
+<subtitle id="0:37:55"> in about 10,000 lines of code</subtitle>
+<subtitle id="0:37:58">  it's not a</subtitle>
+<subtitle id="0:38:1"> lot of code this</subtitle>
+<subtitle id="0:38:4"> code was  virtualized and so if</subtitle>
+<subtitle id="0:38:7"> we go back to the  browser again we</subtitle>
+<subtitle id="0:38:10"> could say well why  don't we just do it Xerox</subtitle>
+<subtitle id="0:38:13"> PARC let's  let's forget about JavaScript and</subtitle>
+<subtitle id="0:38:16"> just  use JavaScript as an alto it's actually  faster than</subtitle>
+<subtitle id="0:38:19"> an alto but let's not let</subtitle>
+<subtitle id="0:38:22">  JavaScript seep into the better</subtitle>
+<subtitle id="0:38:25">  architecture we need to do</subtitle>
+<subtitle id="0:38:28"> our WYSIWYG</subtitle>
+<subtitle id="0:38:31">  environment inside a browser so</subtitle>
+<subtitle id="0:38:34">going to draw a line there and we're  going to say okay this</subtitle>
+<subtitle id="0:38:37"> is machine code  here it's not</subtitle>
+<subtitle id="0:38:40"> the way we want a program  but we can make something out of it</subtitle>
+<subtitle id="0:38:43"> so  we get the guy who did this at</subtitle>
+<subtitle id="0:38:46"> xerox  parc originally dan ingles and on</subtitle>
+<subtitle id="0:38:49"> top of  that we put a real architecture</subtitle>
+<subtitle id="0:38:52">known to be able to build these kinds of  environments and lo</subtitle>
+<subtitle id="0:38:55"> and behold if you go  to</subtitle>
+<subtitle id="0:38:58"> lively</subtitle>
+<subtitle id="0:39:1"> colonel  type lively colonel into</subtitle>
+<subtitle id="0:39:4"> google and do  it in Safari especially you'll</subtitle>
+<subtitle id="0:39:7"> get to  see an entire WYSIWYG environment</subtitle>
+<subtitle id="0:39:10"> done  completely in the browser without any  downloadable</subtitle>
+<subtitle id="0:39:13"> executables this has always  been possible</subtitle>
+<subtitle id="0:39:16"> and I believe it is a</subtitle>
+<subtitle id="0:39:19">  black mark against our field that even</subtitle>
+<subtitle id="0:39:22">though it has always been possible  almost nobody who's working</subtitle>
+<subtitle id="0:39:25"> with this  modern software realizes</subtitle>
+<subtitle id="0:39:28"> it's impossible  or realizes how to do it this</subtitle>
+<subtitle id="0:39:31">don't think we quite have a real field  now</subtitle>
+<subtitle id="0:39:34"> I believe we're in danger of losing  computer science completely</subtitle>
+<subtitle id="0:39:37"> so one way  of looking at it is</subtitle>
+<subtitle id="0:39:40"> the simple way of  looking at computers is they</subtitle>
+<subtitle id="0:39:43"> can be  programmed and people</subtitle>
+<subtitle id="0:39:46"> are incrementing  this by making systems using</subtitle>
+<subtitle id="0:39:49"> programs  that are kind of like the computer that  are programming</subtitle>
+<subtitle id="0:39:52"> language their programming  with</subtitle>
+<subtitle id="0:39:55"> but in fact the computer is more  interesting</subtitle>
+<subtitle id="0:39:58"> than that it can be  programmed to be like any mechanism</subtitle>
+<subtitle id="0:40:1"> you  want including a completely different</subtitle>
+<subtitle id="0:40:4">kind of computer completely different  kind of programming language and</subtitle>
+<subtitle id="0:40:7"> that's  what we should be doing I believe</subtitle>
+<subtitle id="0:40:10"> most  of the sludge most of the bloat has</subtitle>
+<subtitle id="0:40:13"> to  do with trying</subtitle>
+<subtitle id="0:40:16"> to make very very large  structures out of</subtitle>
+<subtitle id="0:40:19"> just bricks rather  than trying to invent arches</subtitle>
+<subtitle id="0:40:22"> so</subtitle>
+<subtitle id="0:40:25">  inventing abstraction here is something</subtitle>
+<subtitle id="0:40:28">  that we have to be very careful with we  don't</subtitle>
+<subtitle id="0:40:31"> want to take the news the</subtitle>
+<subtitle id="0:40:34">abstractions that come to mind when you  think of</subtitle>
+<subtitle id="0:40:37"> aggregating blick bricks you  get piles</subtitle>
+<subtitle id="0:40:40"> and walls and what we need are</subtitle>
+<subtitle id="0:40:43">  these non-obvious structures that</subtitle>
+<subtitle id="0:40:46"> give  us enormous ability to scale</subtitle>
+<subtitle id="0:40:49"> without  having</subtitle>
+<subtitle id="0:40:52"> to use lots of material</subtitle>
+<subtitle id="0:40:55"> and in  fact abstractions that</subtitle>
+<subtitle id="0:40:58"> have great power  tend to be able to go on t-shirts</subtitle>
+<subtitle id="0:41:1"> so  Maxwell's</subtitle>
+<subtitle id="0:41:4"> equations it's not it's not a  complete test</subtitle>
+<subtitle id="0:41:7"> but generally speaking if</subtitle>
+<subtitle id="0:41:10">you have a good idea you should be able  to invent the mathematics</subtitle>
+<subtitle id="0:41:13">express that idea  so Maxwell's</subtitle>
+<subtitle id="0:41:16"> equations works pretty well  equals mc-squared</subtitle>
+<subtitle id="0:41:19"> works pretty well you</subtitle>
+<subtitle id="0:41:22">can put the American Constitution on a  t-shirt and</subtitle>
+<subtitle id="0:41:25"> you can put the</subtitle>
+<subtitle id="0:41:28">programming language and maybe still the  greatest of all</subtitle>
+<subtitle id="0:41:31"> time</subtitle>
+<subtitle id="0:41:34"></subtitle>
+<subtitle id="0:41:37"></subtitle>
+<subtitle id="0:41:40"></subtitle>
+<subtitle id="0:41:43"></subtitle>
+<subtitle id="0:41:46"></subtitle>
+<subtitle id="0:41:49"></subtitle>
+<subtitle id="0:41:52"></subtitle>
+<subtitle id="0:41:55"></subtitle>
+<subtitle id="0:41:58"></subtitle>
+<subtitle id="0:42:1"></subtitle>
+<subtitle id="0:42:4"></subtitle>
+<subtitle id="0:42:7"></subtitle>
+<subtitle id="0:42:10"></subtitle>
+<subtitle id="0:42:13"></subtitle>
+<subtitle id="0:42:16"></subtitle>
+<subtitle id="0:42:19"></subtitle>
+<subtitle id="0:42:22"></subtitle>
+<subtitle id="0:42:25"></subtitle>
+<subtitle id="0:42:28"></subtitle>
+<subtitle id="0:42:31"></subtitle>
+<subtitle id="0:42:34"></subtitle>
+<subtitle id="0:42:37"></subtitle>
+<subtitle id="0:42:40"></subtitle>
+<subtitle id="0:42:43"></subtitle>
+<subtitle id="0:42:46"></subtitle>
+<subtitle id="0:42:49"></subtitle>
+<subtitle id="0:42:52"></subtitle>
+<subtitle id="0:42:55"></subtitle>
+<subtitle id="0:42:58"></subtitle>
+<subtitle id="0:43:1"></subtitle>
+<subtitle id="0:43:4"></subtitle>
+<subtitle id="0:43:7"></subtitle>
+<subtitle id="0:43:10"></subtitle>
+<subtitle id="0:43:13"></subtitle>
+<subtitle id="0:43:16"></subtitle>
+<subtitle id="0:43:19"></subtitle>
+<subtitle id="0:43:22"></subtitle>
+<subtitle id="0:43:25"></subtitle>
+<subtitle id="0:43:28"></subtitle>
+<subtitle id="0:43:31"></subtitle>
+<subtitle id="0:43:34"></subtitle>
+<subtitle id="0:43:37"></subtitle>
+<subtitle id="0:43:40"></subtitle>
+<subtitle id="0:43:43"></subtitle>
+<subtitle id="0:43:46"></subtitle>
+<subtitle id="0:43:49"></subtitle>
+<subtitle id="0:43:52"></subtitle>
+<subtitle id="0:43:55"></subtitle>
+<subtitle id="0:43:58"></subtitle>
+<subtitle id="0:44:1"></subtitle>
+<subtitle id="0:44:4"></subtitle>
+<subtitle id="0:44:7"></subtitle>
+<subtitle id="0:44:10"></subtitle>
+<subtitle id="0:44:13"></subtitle>
+<subtitle id="0:44:16"></subtitle>
+<subtitle id="0:44:19"></subtitle>
+<subtitle id="0:44:22"></subtitle>
+<subtitle id="0:44:25"></subtitle>
+<subtitle id="0:44:28"></subtitle>
+<subtitle id="0:44:31"></subtitle>
+<subtitle id="0:44:34"></subtitle>
+<subtitle id="0:44:37"></subtitle>
+<subtitle id="0:44:40"></subtitle>
+<subtitle id="0:44:43"></subtitle>
+<subtitle id="0:44:46"></subtitle>
+<subtitle id="0:44:49"></subtitle>
+<subtitle id="0:44:52"></subtitle>
+<subtitle id="0:44:55"></subtitle>
+<subtitle id="0:44:58"></subtitle>
+<subtitle id="0:45:1"></subtitle>
+<subtitle id="0:45:4"></subtitle>
+<subtitle id="0:45:7"></subtitle>
+<subtitle id="0:45:10"></subtitle>
+<subtitle id="0:45:13"></subtitle>
+<subtitle id="0:45:16"></subtitle>
+<subtitle id="0:45:19"></subtitle>
+<subtitle id="0:45:22"></subtitle>
+<subtitle id="0:45:25"></subtitle>
+<subtitle id="0:45:28"></subtitle>
+<subtitle id="0:45:31"></subtitle>
+<subtitle id="0:45:34"></subtitle>
+<subtitle id="0:45:37"></subtitle>
+<subtitle id="0:45:40"></subtitle>
+<subtitle id="0:45:43"></subtitle>
+<subtitle id="0:45:46"></subtitle>
+<subtitle id="0:45:49"></subtitle>
+<subtitle id="0:45:52"></subtitle>
+<subtitle id="0:45:55"></subtitle>
+<subtitle id="0:45:58"></subtitle>
+<subtitle id="0:46:1"></subtitle>
+<subtitle id="0:46:4"></subtitle>
+<subtitle id="0:46:7"></subtitle>
+<subtitle id="0:46:10"></subtitle>
+<subtitle id="0:46:13"></subtitle>
+<subtitle id="0:46:16"></subtitle>
+<subtitle id="0:46:19"></subtitle>
+<subtitle id="0:46:22"></subtitle>
+<subtitle id="0:46:25"></subtitle>
+<subtitle id="0:46:28"></subtitle>
+<subtitle id="0:46:31"></subtitle>
+<subtitle id="0:46:34"></subtitle>
+<subtitle id="0:46:37"></subtitle>
+<subtitle id="0:46:40"></subtitle>
+<subtitle id="0:46:43"></subtitle>
+<subtitle id="0:46:46"></subtitle>
+<subtitle id="0:46:49"></subtitle>
+<subtitle id="0:46:52"></subtitle>
+<subtitle id="0:46:55"></subtitle>
+<subtitle id="0:46:58">something like a system psychology in  that</subtitle>
+<subtitle id="0:47:1"></subtitle>
+<subtitle id="0:47:4"></subtitle>
+<subtitle id="0:47:7"></subtitle>
+<subtitle id="0:47:10"></subtitle>
+<subtitle id="0:47:13"></subtitle>
+<subtitle id="0:47:16"></subtitle>
+<subtitle id="0:47:19"></subtitle>
+<subtitle id="0:47:22"></subtitle>
+<subtitle id="0:47:25"></subtitle>
+<subtitle id="0:47:28"></subtitle>
+<subtitle id="0:47:31"></subtitle>
+<subtitle id="0:47:34"></subtitle>
+<subtitle id="0:47:37"></subtitle>
+<subtitle id="0:47:40"></subtitle>
+<subtitle id="0:47:43"></subtitle>
+<subtitle id="0:47:46"></subtitle>
+<subtitle id="0:47:49"></subtitle>
+<subtitle id="0:47:52"></subtitle>
+<subtitle id="0:47:55"></subtitle>
+<subtitle id="0:47:58"></subtitle>
+<subtitle id="0:48:1"></subtitle>
+<subtitle id="0:48:4"></subtitle>
+<subtitle id="0:48:7"></subtitle>
+<subtitle id="0:48:10"></subtitle>
+<subtitle id="0:48:13"></subtitle>
+<subtitle id="0:48:16"></subtitle>
+<subtitle id="0:48:19"></subtitle>
+<subtitle id="0:48:22"></subtitle>
+<subtitle id="0:48:25"></subtitle>
+<subtitle id="0:48:28"></subtitle>
+<subtitle id="0:48:31"></subtitle>
+<subtitle id="0:48:34"></subtitle>
+<subtitle id="0:48:37"></subtitle>
+<subtitle id="0:48:40"></subtitle>
+<subtitle id="0:48:43"></subtitle>
+<subtitle id="0:48:46"></subtitle>
+<subtitle id="0:48:49"></subtitle>
+<subtitle id="0:48:52"></subtitle>
+<subtitle id="0:48:55"></subtitle>
+<subtitle id="0:48:58"></subtitle>
+<subtitle id="0:49:1"></subtitle>
+<subtitle id="0:49:4"></subtitle>
+<subtitle id="0:49:7"></subtitle>
+<subtitle id="0:49:10"></subtitle>
+<subtitle id="0:49:13"></subtitle>
+<subtitle id="0:49:16"></subtitle>
+<subtitle id="0:49:19"></subtitle>
+<subtitle id="0:49:22"></subtitle>
+<subtitle id="0:49:25"></subtitle>
+<subtitle id="0:49:28"></subtitle>
+<subtitle id="0:49:31"></subtitle>
+<subtitle id="0:49:34"></subtitle>
+<subtitle id="0:49:37"></subtitle>
+<subtitle id="0:49:40"></subtitle>
+<subtitle id="0:49:43"></subtitle>
+<subtitle id="0:49:46"></subtitle>
+<subtitle id="0:49:49"></subtitle>
+<subtitle id="0:49:52"></subtitle>
+<subtitle id="0:49:55"></subtitle>
+<subtitle id="0:49:58"></subtitle>
+<subtitle id="0:50:1"></subtitle>
+<subtitle id="0:50:4"></subtitle>
+<subtitle id="0:50:7"></subtitle>
+<subtitle id="0:50:10"></subtitle>
+<subtitle id="0:50:13"></subtitle>
+<subtitle id="0:50:16"></subtitle>
+<subtitle id="0:50:19"></subtitle>
+<subtitle id="0:50:22"></subtitle>
+<subtitle id="0:50:25"></subtitle>
+<subtitle id="0:50:28"></subtitle>
+<subtitle id="0:50:31"></subtitle>
+<subtitle id="0:50:34"></subtitle>
+<subtitle id="0:50:37"></subtitle>
+<subtitle id="0:50:40"></subtitle>
+<subtitle id="0:50:43"></subtitle>
+<subtitle id="0:50:46"></subtitle>
+<subtitle id="0:50:49"></subtitle>
+<subtitle id="0:50:52"></subtitle>
+<subtitle id="0:50:55"></subtitle>
+<subtitle id="0:50:58"></subtitle>
+<subtitle id="0:51:1"></subtitle>
+<subtitle id="0:51:4"></subtitle>
+<subtitle id="0:51:7"></subtitle>
+<subtitle id="0:51:10"></subtitle>
+<subtitle id="0:51:13"></subtitle>
+<subtitle id="0:51:16"></subtitle>
+<subtitle id="0:51:19"></subtitle>
+<subtitle id="0:51:22"></subtitle>
+<subtitle id="0:51:25"></subtitle>
+<subtitle id="0:51:28"></subtitle>
+<subtitle id="0:51:31"></subtitle>
+<subtitle id="0:51:34"></subtitle>
+<subtitle id="0:51:37"></subtitle>
+<subtitle id="0:51:40"></subtitle>
+<subtitle id="0:51:43"></subtitle>
+<subtitle id="0:51:46"></subtitle>
+<subtitle id="0:51:49"></subtitle>
+<subtitle id="0:51:52"></subtitle>
+<subtitle id="0:51:55"></subtitle>
+<subtitle id="0:51:58"></subtitle>
+<subtitle id="0:52:1"></subtitle>
+<subtitle id="0:52:4"></subtitle>
+<subtitle id="0:52:7"></subtitle>
+<subtitle id="0:52:10"></subtitle>
+<subtitle id="0:52:13"></subtitle>
+<subtitle id="0:52:16"></subtitle>
+<subtitle id="0:52:19"></subtitle>
+<subtitle id="0:52:22"></subtitle>
+<subtitle id="0:52:25"></subtitle>
+<subtitle id="0:52:28"></subtitle>
+<subtitle id="0:52:31"></subtitle>