Delivering the Future (1994)

From Viewpoints Intelligent Archive
Revision as of 22:13, 5 December 2017 by Ohshima (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
well thank you for inviting me here I like the title of this session I think
in the first year I was at Xerox PARC I
had an argument with some of the Xerox executives who are worrying about what
might happen and they wanted to defend themselves against the future it was so
wimpy I finally got angry at them and I
said look the best way to predict the future is to invent it don't worry about
what all those other people are going to do this is the century in which if you
have a vision and you don't violate too many of Newton's laws you can actually
build the thing that you have in mind now delivering the future is a more
subtle metaphor so it implies there's something to deliver like a baby and the
reason I like it is that there's all this exciting stuff that has to happen a
little earlier in the process in order to have a baby to deliver we scientists
call that research so I'm going to talk
to you a little bit about how the future
gets invented I think some of the some
of the stories and ideas I think may may help you now I'm not a professional
speaker I'm still a scientist as my
introducer said my title is a fellow we
decided at Apple that the mass noun for know there's a pride of lions we decided
that the mass now and for fellows should be jolly
so we have a jolly of fellows at at Apple we have three of them now twenty
years ago at Xerox PARC we invented six
things and I would not be here talking
to you if Xerox had accepted any one of
these it was in fact what happened after
we invented these six things that I got
interested in talking to people outside
of the computer field the six things were the first thing like a workstation
or a modern PC really the first thing like a Macintosh originally started
working in 1973 the first overlapping window user interface first modern OOP
system first laser printer and the
ethernet and then the the sixth thing I didn't have a really good picture
because I just show another Alto there which was what we call client-server
computing now in today's terms it's actually pierre-pierre computing because
clients see terms get corrupted I made
up the terms personal computer and I made up the term object-oriented and
both of those have gotten corrupted into an almost meaningless terms
client-server has gotten corrupted it's the new way of saying terminal mainframe
I can assure you that was not what we had in mind any more than we thought
that C++ would be called an object-oriented language or that the IBM
PC would be called a personal computer I was more meanings originally or that
windows 3.0 would be called a user interface as far as that goes but that's
the way it goes now what happened is
that around 1975 76 we had about 700 of
these machines both at Palo Alto and in
Rochester and we had I guess eight or
nine Ethernet linked together and then linked by long lines from Rochester to
Palo Alto and you could actually sit down and do something in Rochester that
is very hard to do today you could go to Rochester and sit down at an alto and
work as though that was your alto so
this was a what you what today is called a natural network virtual address space
the computations actually were able to float around the network they were not
bound to specific machines and what
happened was in August 1976 Xerox turned
down all of this technology even though there were hundreds of machines working
software the first desktop publishing system Microsoft Word which was
originally called Bravo was done at Parc it was the word processing system that
we used all of this stuff they turned it down of course that was a great shock
but what did we know we kept on plugging
away at at what we're doing in 1978 I met several of the executives who had
made that decision and was really surprised to discover that they were actually somewhat intelligent even even
for Xerox executives and then that was
when I started worrying because you see when they turned it down we just thought well they're stupid they're dumb what do
they know you know this is hopeless but when I
found out they were intelligent and they still turned it down I realized that something much more serious was going on
and one way of saying it as they were still doing Ptolemaic astronomy in
Connecticut and we had started doing Copernican astronomy out in California
when you get one of those where you get a complete miss between two ways of
looking at the world you have something really serious because there aren't even grounds for conversation and what's
worse the same words get used over again but in different ways in astronomy the
word orbit comes from one of the initial beliefs by the Greeks that the planets
are on transparent spheres you ever wonder why a planar orbit was called orb
in orbit because they thought they were on glass spheres then it got taken over
to mean planar orbits got reused by
Copernicus to me and a heliocentric system got really used by Newton and
Kepler and got reused by Einstein same word we use today the way people have
trouble with science science unfortunately decided to reuse the word
theory and people when they hear about
the theory of relativity they say oh that's just a theory or the theory of
evolution that's just a theory theory and science means something completely
different we separate out the two words
theory and conjecture it's a theory and science is something that has been set
up in such a way and beaten on for years so it is a ball stirred set of arguments
why don't we call it the truth the reason is is that science doesn't
believe that you can find truth it's a
cliche and a myth does science believes
that what you can find are better and better mappings of the world so that's
what I'm going to talk to you today about are because the way you get into
the future is by realizing that the way we look at the world in any given time
is always through our own mappings and we can come up with different mappings
and when we come up with different mappings we can look at the world in a different way now
the big problem with predicting the future is in large part because of the
funny ways that human like to think so here are three ways of getting from A to
B the first one is random walk you start off from place a there on the left and
at every corner you take a random turning and eventually you can prove
that you will get to B I always think of that as a perfect example of hire
management's main way of doing things
now lower in the organization you have a
more rational way of looking at it which is reduce the differences this is what
they teach in business school and in most elementary and high schools and
colleges is problem-solving problem-solving the way you solve a
problem here is that at each step of the way you see if you can get closer to the
gold state and then when you have reduced that difference to zero you've
achieved the goal now it turns out that both of these ways of doing things are
completely have been obsolete for many many years the way really good things
are done is at the bottom there and what
you do is instead of going from A to B randomly or by reducing differences what
you do is you go away from B for a while and find us find a superhighway because
you can go away from B for a while and you find a superhighway you can get very
very close to B be very very rapidly then get off find B now this is very
hard for people to do companies have the hardest time going away from their short
term goals in order to build structures
that will allow them to deal with many
many goals in the future rather than just the ones that they're working with now you could think of this third way of
looking at things as a architecture if you want you could think of it as a
paradigm that's going away from the goal
sometimes called research it's the kind of investment that you
make in order to find large-scale architectures that are going to help you
so if we want to predict the future we have to find some way that is not
extrapolate 'iv something that doesn't
rely on the future being a lot like the present where else we're just going to come up with improved versions of the
present so here is a dream called MMX is
done by Vannevar Bush who is president roosevelt's science advisor who wrote an
article in 1945 and in that article he said not too long from now everybody
will be able to have in their house a device the size of a desk in that desk
stored optically will be the contents of
about a small town library about five to
ten thousand volumes worth of materials and we'll be able to search that we'll
be able to look on look at this material
on various screens and we'll have pointing devices that we can put new
information in and keyboards and etc etc so this is Memex written in 1945 by a
pretty good scientist I had some good reasons to believe that this could
happen because he was aware of things that were going on at the time including
the development of computers and of various kinds of dry photography
microfilm microfiche and so forth we're all being developed then the critical
thing about this for the story today is that a number of the people who invented
the technologies that you're using today saw this Engelbart who was a towering
figure we'll meet him in a minute saw this in 1945 he was in the Navy and
he read it read this version in Life magazine had pictures on it I saw it
when I was 14 years old as a result of reading a science fiction book in 1954
so and other people saw it and when it
came time to actually do it there was already an image of what it would be
really nice to have it didn't look at all like ENIAC a huge
mainframe computer for solving ballistic problems it didn't look like a univac it
didn't look like a any kind of mainframe
computer in IBM 650 this looked like something that a person could use and
actually this became the rallying cry starting around 1960 ok so there are
several different ways of thinking about computing back then one was that the
mainframe was gone of a universal factory that was IBM's way of thinking
about it Engelbert thought of computing
as a kind of a vehicle now Engle Bart
was the guy who invented the mouse and that looks like a fairly modern picture
because let me ask the crowd here how many people have in their office right now a two-page display right not one of
these little guys but one of these guys you can put two two full pages of display okay couple people so there's
one right there that was the first one anybody ever used that picture was taken
in 1966 okay so in 1966 Engelbart had
thought through that computers were not about processing numbers they were
actually there to relate information now he was one of the people who invented
one of the first word processors but he didn't think computers were for word
processing he didn't think they were for outlying processing he thought that they were for
relating information and he realized something very very profound about
computing that is an ordinary physical
space that we live in if we have point a here and we have a point B down here it
isn't the same place
what's been generated in between them is that horrible thing called distance an
angle Bart realized something very interesting about distance and that is
if here's that this is a two dimensional space we have a two dimensional distance
but what's incredible about distance is if you give me one more dimension I can
get rid of the distance right just one
more dimension than the dimension of my information and I can get all of the intervention interesting information in
the same place in hyperspace and I can
do it for any number of them
that's what Engelbart did so he was the guy who invented what we now call
hypertext or hyper media he realized that what people really want is not an
imitation of paper they want something better they don't want to automate old
things what they need is to take all the interesting information about a wide
number of things and exits so that all of those things are in the same place
and Ivan Sutherland around the same time
did something wonderful he was the guy who invented computer graphics it was
invented around 1962 and again he just didn't do it to imitate paper to imitate
drafting he looked at and said what else can it do this is what we should always
ask most important question what else can it do and what it could do is it
could simulate things so on this thing this picture taken in 1962 Sutherland is
sitting in front of a bridge that he's drawn and sketchpad and that bridge is
dynamically sustaining a weight and showing him what the stresses and
strains on the bridge are so here are three completely different metaphors for
computing I think you'll agree that the
dominant metaphor today is still the Universal Factory
American business by and large uses
computers to automate accounting I'm sure that Wells Fargo does really do
they think of it as a vehicle for traversing hyperspace where all of the
stuff of interest to you should be available dynamically and rarely do they think of
it as a simulator for being able to deal with very complex problems and try and
understand what happens next now individuals do about 70% of the use
of spreadsheets on personal computers our what-ifs but IT organizations almost
never do IT organizations run post-mortems on already in bomb de right
by the time they get around to writing an application which is usually 18
months after it was requested the information is not of interest anymore
so the metaphor that we pick for
thinking about what computing is is very very critical now I started thinking
about I started in 1961 and I started thinking about this stuff and it I was
very intrigued with angle Bart's stuff and in 1967 I built this machine in the
center the top Center there it was called the Flex machine so it looks like
a personal computer of today it was a personal computer was a desktop computer
for very special deaths namely those made out of steel so weighed about 400
pounds and had a had a fan that was likened to
a 747 taking off C in those days the
integrated circuit packages integrated circuits were very very small they only
had about 16 or 32 gates on them and so this machine which was quite weak by
today's standards had more than 700 of those little black boxes in it and so it
generated a lot of heat and then I
happened to see some of seymour papert's early work with logo and it got me
thinking all of a sudden that computers were not Veals
computers were like media they were like media and I suddenly remembered
something quite striking I don't know
how many people here have ever been in a medieval library there's one still set
up in Florence and what they look like is there a room about four times this
size there are no shelves why because
most libraries had less than a hundred books one of the largest libraries in
Europe the Vatican Library in the year
1400 only had 392 books each Brook took
about 10 to 15 years to copy by a group of group of monks so when he went into a
medieval library what you saw were tables reading areas each book was
chained to the table table after table
after table about a hundred of them and so they it flashed into my mind that
looks just like a time-sharing bullpen right time sharing terminals using
something so expensive only an institution could own it
individuals accept that unless they were kings couldn't own this stuff and then I
flashed to Gutenberg's Bible which was
printed but it was printed in imitation of this stuff and I suddenly looked at
the Flex machine and said you know that thing looks just like a time-sharing terminal is this am I just imitating the
past what is this thing by the way Guttenberg
buy it the first Gutenberg Bibles were incredibly cheap costs only two years of
a clerks wages to buy one it was about
sixty thousand dollars today and if you think of that if you think of that as
about what a workstation costs expensive
workstation costs compared to the millions of dollars from one of these things it is rather cheap but what
flashes of my mind then was the turning
point in the development of the book which was this third stage and that was
done by a guy by the name of Aldous not Elvis PageMaker that could have been his
name because he was a printer in Venice that's where the company got their name
Aldus Manutius around 1495 decided books
should not be like this so the Gutenberg galaxy I mean the
Gutenberg Bible and manuscripts but he
decided that book should be the size they are today you've ever wondered why
books are the size they are today it could be any size it could be really
small it could be big this size I'll
tell you why because that is the size that saddlebags were in Venice Aldous
went out and measured saddlebags and took the average saddlebag size and
decided to print a library or she called the portable library in 1495 so that
people could take them with now this is a profound idea because Elvis was the
first person who really realized that books were now something that could be
lost it was not a crime against humanity
to lose a book because he could go buy another one okay so that happened around
1495 and suddenly I suddenly realized that a computer should look like this
should be something the size of a book should be something that you always have
with you and I made that cardboard model
in 1968 that's what a computer should
look like now just to give you an idea to try and understand why the the
situation we're in today here is a completely transitional one this is a
much Stabler form than this one you have
to realize that remember printing was invented around 1450 to 1450 for Aldous
was 1495 wasn't until 1516 more than 60
years after the printing press was invented that anybody thought to put page numbers on a book think about that
how could they have meant notice Elvises Brooke here has no page numbers how
could they possibly not have done the reason was is that when you only have 392 books in the world's largest library
and it if you're really rich and you may have at home five or ten books why have
page numbers you know where everything is books are a celebration of what you already know so when we think about
computers today one of the things we have to really understand is that we are
in a transitional phase and that the use of computers by humans is going to be
completely different than it is today just as the use today is completely
different from what it was 20 years ago that's the importance of making sharp
boundaries between these different ways of thinking about things now this all
would have been fantasy and science fiction except for people like Gordon
Moore Gordon Moore was a physicist in 1965 he published the white part of this
diagram which said you know from 1959
whoops sorry from 1959 up to about 1965 things have
been improving at about the rate of factor of two every year in integrated
circuits he said I wonder if that can continue so he started looking at the
physics of integrated circuits that gave him this line
a few years later he said this line is too ambitious I'll predict that it will
grow by a factor of two every two years which gave him this line and just for
your enjoyment here the purple line is
what actually happened right so it's grown slightly better it's grown about a
factor of two every eighteen months since 1965 and Gordon Moore's prediction
was a 30-year prediction now in order to
make this prediction he had to back away he had to do remember my diagram he had
to go away from B because the silicon
that people were using was not a predictable silicon it was not easy to
make predictions about it so he picked a simpler way of doing things which is thought to be too slow and made his
predictions based on that that slow silicon happens to be a silicon we used
today MOS silicon this single chart if
well understood explains why IBM and Dec are in such trouble today explains why
we have millions and millions of personal computers today it explains why
people were not able to predict the future in a reasonable way because this
is an exponential curve this is talking about a factor of 10 million change in
20 years so almost everything that we're
doing today is based on this now the
thing is you have to realize about the human nervous system is that factors of
10 and factors of a hundred change our qualitative to us they may look on paper
quantitative but a factor of 10 change in speed is qualitative to us what
happened to IBM because they were aware of this is they couldn't believe that
this meant anything other than they would be able to make cheaper mainframes
and make more profit on them they could not imagine the actual implications of
being able to put an entire computer on a chip and that there would have to be
millions of them so they lost they missed so these three
ways of looking at things I gave names to I called the mainframe
thing institutional computing I called the desktop machine personal computing
and I called this handheld Aldus type computer an intimate computer and in the
late 60s I sat down and tried to write down some of the conclusions that a
person would reach for in each of these belief structures now another word
that's been devalued almost beyond repair is paradigm right paradigm is a
belief structure so you could just as easily say this is the institutional
paradigm the personal computing paradigm or the intimate paradigm doesn't matter
scientists like paradigms because it is
so hard to think about things that aren't already around so we play all
kinds of tricks and mind games and one of the ones that we do is to drive artificial boundaries between things to
see if we can find differentiation so this is one of those things saying these
are three different ways of thinking about the world and what are the implications whoops sorry about that
now what types one of the ways of thinking about it is what types of data
do you use institutional uses letters and numbers I discovered just a few
years ago that my Dennis still had those green and white print out sheets needless to say I changed Dennis so I
was wondering what other old technology he might be using there's law by the way
does anybody know who toward more is here who is Gordon Moore who did Moore's
law yeah so Moore got a lot of criticism for
that curve and so he decided to do this
thing well the best way to predict the future is to invent it so he said let's found a company to work on MOS silicon
happened to be Intel he's one of the richest guys in California right now
he's worth more than a billion dollars from that because of course see when the
astronauts got out in space and they look back at the earth you know what they saw they saw something that looked
just like a globe that had been done in the 19th century that's what science is
all about no surprises you get out there and if
you've done your work properly you should see what you expect to see so
Gordon Moore was right and he was able to get profit from it and if you think
about the implications of Moore's law you're going to start using bitmap
screens it's one of the first things we invented at Xerox PARC
because now we could have bits bits weren't just for computation anymore and
then later on all other media was going to have to come along and one of the
implications of this is that when you get into this third stage you're not
going to print printing will go away now people have been saying that for years
and we have more paper than yes yes I know but think of it this way when your
information is related in such a way that all important things are together
you're not going to print it out how
many people here do forecasting by spreadsheets
okay who does anybody print out the spreadsheet template no you know why
because what ones you're going to print out you're going to print out a hundred of them every one of those things you've
done you change a number and the whole spreadsheet changes so the multi is to
think about in multimedia are not combining voice and images the multi is
to think about our multi-dimensional and multi process so as we go more and more
towards multiply multi dimensional representations and simulations printing
will kill both of those things and what
we have now is synthetic paper that comes along just in the nick of time to
save us we can now have all of our dynamic information structures with us
at all times so printing is finally going to go away by my friend Nicolas
Monte thought of it this way thought of the worlds of publishing entertainment
and computing imploding on each other because now every single one of them
could have everything that they do represent it digitally when you have everything represented in
the same form you basically have the same business they use the same
transmission everybody who has access to the Internet is a publisher you can
publish just as well as mcgraw-hill can in fact your if you have access to the
Internet you are actually a miniature FCC I don't know whether you know it but
people send television programs back and forth on the on the internet and radio
there's a constant internet radio between scientists and stuff and there
are no channels were requested from the FCC because the internet does not care
it's just sending bits back and forth and the bits can be anything so this is
one again that the publishers in the entertainment industry only discovered a
couple of years ago and yet it's been true once more
Wrote is liya this is why this stuff is scary to people because it seems like
all this stuff is suddenly happening the reason it's suddenly happening is
people will not believe a disaster until it's actually in front of their eyes
the whole reason we have eyes is so we
don't have to jump off the cliff to see if it's dangerous right
we have long distance sensors and we need the equivalent of long distance
sensors when we're dealing with the future user interface probably the
biggest change when you're over here you
have a few thousand users perhaps using an airlines reservation system or a
banking system you can train a few thousand people there's a factor of a
thousand between these two panels so millions can't train millions of people
over here there's another factor of a thousand because this intimate computer
is a lot like a phone in fact it will have a phone in it so these are billions
of people millions of people it put an enormous pressure on trying to
understand what it would be like to
actually get computers to do things and we realized that the big transition
would have to go from being a command environment like a control panel of a
nuclear reactor that's sort of what a 3270 terminal is like to a learning
environment that's what forced us to invent what is now called the the
Macintosh style interface now what happens in the future here is something
even more cataclysmic because this third panel is energized by pervasive
networking think of the driving forces driving force for mainframe computers
was actually business IBM didn't want to build them they wanted to stay in the
punch card business was it was American
business that forced IBM to become a big company driving force for the second
frame was integrated circuits as we've been talking about in the driving force
for this third frame is pervasive networking which is happening to us now this transition is
going to be complete pretty much by the year 2000 billions of people are going
to have to be there and so one way of thinking about this is here you have to
train the user in this one the user
interface teaches the user that's its main job and in this one the user
interface learns from the user in other words it actively is trying to
understand the user and what it's what it's trying to do so one way of thinking
about this is the two ways that we've had of amplifying our reach have been
tools Macintosh style interface is a
tool based interface it's designed to work with hand and eye and the other way
we're used to amplifying our reach is through agents agents are creatures that
can take on our goal structure and carry out goals in our behalf and what's even
stronger about agencies they can transmit your goals to other agents
imagine how powerful that is some person has a few ideas in the middle-east 2000
years ago gets 12 agents to be interested in it and half the half the
earth is now interested in those same
ideas that's the power of agents over
tools so one of the big transitions that's going to happen in the next five
years will be the transition from tool based interfaces which we've just
transitioned into to this softer more
dynamic way of doing things that that networks are going to force us into
integration of information it's usually done painfully with files using
relational databases it's it's incredible to me that after more than 20
years almost 25 years now the less than
half of the databases in America's fortune 500 companies are relational
I'm sure everybody is aware of that it's
unbelievable most of them are still flat files and cut and paste in the Macintosh
and then to active components and if we have time we can talk a little bit more
about that as you go to more and more complex systems the parts of things have
to take more and more of the initiative for the linking what kind of software
airline reservation system in institutional generic tool like a
spreadsheet or a word processor or desktop publishing system because you
have millions of people buying them the users have to complete the application in order to use it and in the future
most of the applications in business
will be done by the end users now it happens that today in the fortune 500
companies more than half of the applications that are used in Fortune
500 companies today are done by the end users not by the IT organization
spreadsheets are the big culprit IT
organizations hate them because they are and personal computers because they have
to give up control but in fact there is a way of leveraging the productivity of
users and we'll talk about that a little bit later each of these 3 belief systems
has a different way of wanting to be programmed data structures and
procedures are an extremely weak way of
doing things this is COBOL and see one of the problems is if you send a data structure
a thousand miles between point a and point B you pretty much have to matter
transmit a programmer along with it in order for people to use it the reason is
is that the data structures themselves are dumb dead things that don't know how
they should be used so objects particularly at Xerox PARC were invented
so that when you transmit something from A to B you're also transmitting the
knowledge of how to have how to use it so the amount of code you have to write
at the other end is very minimal and networks are going to force a completely
different kind of jamming paradigm that we call agent oriented just to give you an idea of
this is about maybe 50 years old to go
back to Von Neumanns original design this is about 25 plus years old I did my
first language in 67 this one is more
than ten years old they've never heard of this stuff but in fact people have
been working away in research labs because looking ahead we could see that
pervasive network was coming and we realized that classical objects could
not be transmitted willy-nilly around the world and be able to be components
of things without a lot of work that calls for a new way of thinking about
programming and this agent oriented way is the way we think it will work
analogy to architectures the Greeks were really smart but they never invented the
arch so they were constrained to build
rather small but beautiful temples if
you want to build something bigger you have to have a different architectural principle the Gothic arch allowed
cathedrals to be built that were 10 times the size of the Greek temples
interestingly enough there's only slightly more material in Chartres
Cathedral than there was in the Parthenon all right because Gothic cathedrals are
mostly air and windows and then geodesic
domes using a completely different theory can you can make a geodesic dome
that will go over all of the Gothic cathedrals made on earth using a
different architecture the way we improve ourselves is by changing
architectures we have the same IQ we had ten thousand years ago imagine having an
IQ of 250 in the year 10,000 BC what
could you do not a lot all right Leonardo probably had an IQ of
250 he did a lot of drawings but he
could hardly build anything because he wasn't in a place where he could
leverage his IQ you leverage IQ with knowledge and architectures these are
the thing most important things the human race comes up with our different ways of thinking about things that allow
us to appear much smarter than we did years ago so here's a my favorite
metaphor for the future Nicholas Negroponte II said once we'll know it's
the future when my left cufflink communicates with my right cuff link via
satellite this sort of sums it all up
now let me ask you do you think Nicholas was being facetious no in fact by the
year 2000 you will be able to have a cuff link size thing that can go up to a
satellite and come down you ever called
at dinner if you have a cellular phone and your friend has one and you've
talked to each other across the dinner table with your cellular phones do you
have any idea of where the information is going in order to come back okay so
this is a way of thinking about I like
this one because it's so so you tray it's very Nicolas like it's so tray that
it gets people thinking about wow this is not the computing I thought we were
doing so particularly the future is easy
it hasn't changed in 20 years that's the number one thing you have to understand
it takes so long to get this stuff out of labs it takes even longer to convince
venture capitalists and other people to support it every technology that's going
to affect you in the next 10 years is already in somebody's lap so one of
things you have to ask yourself do you pay somebody even half time to go around
and visit the 40 or so labs in the world that count probably not so how do you
know you're gonna wait until Businessweek discovers it they don't understand half of these
technologies most scientists like myself spend part of our time just visiting
other people's research labs because we can't wait for journals to come out
journals are always two years behind and
so most of the stuff virtually everything that is going to affect you
already exists in some demonstratable form and you should know about it
because those are the actual tools that we are going to make it possible for you
to build the future that you want okay now here comes so that was the uplifting
part now the problem with this picture
is that every time a technological advance comes along almost everybody
tries to automate all of their old processes we've already talked about that
that is not the way to do it because often the technology is not even as good
as the old technology when it starts automating ever tried to read a couple
of paragraphs off of CRT I think CRTs
are good at automating what books do they're terrible so what you have to do
is ask Ivan Sutherlands question which is what else can it do you have to
understand how to leverage the new properties of this technology that will
lead to new processes in your business and unfortunately for everybody it's
also going to have to require some learning on your part this is the thing
that humans hate especially Americans schools teach us to treat classes as
vaccinations against dread diseases right remember saying thank God I'm done
with that class I never have to take it again this is like getting a smallpox
shot thank God I'm done with that vaccination now I'm safe I never have to
learn anything yet the problem is it doesn't work like that anymore in order to leverage the new we have to
learn what the new is in order to do that we have to understand how the mind works
my favorite saying in McLuhan's is I don't know who discovered water but it
wasn't a fish this is our number one problem what he meant is that we exist
in a goldfish bowl filled with water that we don't know exists and that water
is called our beliefs to us we are in reality those people
halfway across the world that think differently than we do have to be wrong and therefore we should kill them this
is the way human beings work because we're not conscious that we don't live
in reality we live in beliefs in fact we
can't ever get to reality can't ever get
to reality there is no reality that we can get to yeah beliefs and the beliefs
can be good or bad Maps so this this
image I think is perfect if you always think of yourself as being in a goldfish bowl you have to ask yourself what color
water am i swimming in right now so one
of the things that we do culturally is that we teach people are special
goldfish bowls so we have like a different goldfish bowl and people in
India do or people in Lebanon
okay now take a look at this face upside down face and see if you can see ting
interesting about it who's got what do you think the mouth is
what mouth is upside down see anything
else about the picture okay let's let's
pretend we're in a typical American school or let's say a Wells Fargo training center
okay the dominant belief in school is
that knowledge is a fluid and the way we
teach it is we transfer drop by drop
from the full student full teacher vessel to the empty student vessel okay
that fluid is in terms of a language like English so let me give you a
sentence about this picture what we've done here is we've taken a picture of a young girl extracted her mouth and her
eyes turn both of them upside down put
them back into the picture and then turned the entire picture upside down
okay so now you should be quite prepared
for what it looks like right side up
now I'm going to turn this back because I've discovered nobody will listen to me
when it's on the other way now you have
to realize I didn't give you any more information the second time right what I
described in that sentence was exactly what you saw and yet you had a reaction
on the second picture you didn't have one on the first who knows what that
means who's got a theory don't be shy
why is it worth it why is it worth a thousand words because but why yeah but
why why couldn't you why couldn't you picture what it was
going to look like from my words okay so
lots of different theory the real what
what is actually going on here and it's been confirmed in a particularly striking way as the last four or five
years is that when you're listening to English sentences a part of your brain
over here and a part of your brain over here is active okay and when you're
looking at pictures a part of your brain back here and a part of your brain down here particularly for faces is active so
you're thinking about words completely differently and with a different set of
mechanisms than you're thinking about pictures think about that this is why
McDonald's will never run a print ad saying if you eat a McDonald's hamburger you'll become a better-looking person
right because it's absurd in English but what do they do they run ads with
gorgeous people holding hamburgers and a
different part of your brain a part of your brain that remembers very well but
isn't so good at thinking makes the inference the advertisers want them what
you want you to McLuhan was once asked why don't you read psychology books and
he said well psychologists study white rats he says I look to see what advertisers are doing they're the ones
who are studying humans
this is why the Presidential Debates which were three hours long a hundred
years ago are now two minutes on each point on television right because there
is nothing to think about from what television shows you and none
of the politicians want you to think about any of the issues that's why they
come on television and you don't want to think about the issues which is why you watch them on television itself and I
there's an ni there's no one of us there
are actually lots of us in here thinking
different ways and most of the time we're not aware of it at all
so Brenner was American psychologist who studied this stuff he took a six year
old kid six year old kids when you pour from a squat glass into a tall thin one
we'll say there's more water in the tall thin glass because they're there in
what's called an iconic reasoning stage and this is pretty constant from kid to
kid around age five or six then Brenner
did something interesting he's got a kid who would say there's more water in the tall glass and he covered up the tall
glass with cardboard and asked the question again the kids say oh wait a
minute there can't be more water where were to go where would it come from and
Brenner took the cardboard away and the kid said no look there's more water and Brenda put the cardboard back and the
kid changed his mind again so if you have any six-year-olds you'd like to
okay what was Brenner exploring who's got a theory yeah
visual versus language visual versus symbolic so he said you know there are
at least three ways we have of knowing about the world and probably a lot more
how many people here have ever remembered a route better when they were
driving the car versus when they were a passenger why huh yeah we have a
kinesthetic mentality this is the one if
you're a musician you know much a lot about this one this is the one that you have to get it this is the one that
learns how to ride the bike so we have
that's a completely separate one that's why you can't just watch somebody playing a sport and go out and do it
it's in a different part of the brain it's in a different set of symbols we
have a visual mentality one of its great
tricks is how many people here now admit it I'll admit it how many people click
through channels how many people here have ever clicked
into the middle of a movie they haven't seen for 20 years how long did it take
you to recognize that movie a couple of
seconds think about that now compared to
school twenty years ago you were with a
significant other most likely you did not have your mind on the movie
particularly didn't know you're going to be tested 20 years later the test was
random into the center of the thing and yet within about 50 frames your visual
mentality had retrieved that you'd seen it before and about half of the time in
tests that have been done people remember what's going to happen next think about this in the context of your
entire life and think of why there are icons on a user interface if you put
pictures of animals up on the wall you can find the elephant four times faster
than pic then words for the animals and here's a beautiful one if you put boxes
around the words you can find the word elephant twice as fast then if there
isn't a box there why
yeah but basically what happens is the Box icon eise's the word the words are
actually perceived in a different place in your brain then then images that
aren't words see why am I am I telling
you this stuff because the world is not as it seems right we go from day to day
from moment to moment thinking that what's out there is reality it isn't all
it is is interpretations by that three pounds of oatmeal we've got between our
ears and that three pounds of oatmeal
can be taught to perceive things any way the culture chooses let's look at
creativity for a second here we are plotting along in the pink context you
can think of this is a goldfish bowl we're in a goldfish bowl and when
somebody wants us to do something the best thing we can think of is extrapolation or if going from A to B
problem solving so it's very hard to be creative when you stay inside the
goldfish bowl moments of creativity happen when something happens oh yeah
here's a good example is if you take frogs and paralyze their food with a
little chloroform flies the frog will
not eat them the Frog cannot see flies
that aren't moving that's incredible but it will try and eat little oblong
shapes of cardboard that are moving this
cyan't we scientists love this one because we we think of the you know the ideas they're all around us if you think
of the Flies as I did we just can't see them they're just there all the time we
can't see my nervous system won't let us do it we're keep on snapping at pieces
of cardboard but then something great
happens something makes us look at the idea in a different way and bingo we
have an idea this is why I say terrific ideas usually hide behind good ones good
ones are the ones that you get by logic in the same context you're improving
something this is what engineers like to do take something it already exists and
make it better okay you want to get a
completely new idea you have to have a bingo and this bingo operates in
different areas so Arthur Kessler said you know if you're telling a joke this
is a joke isn't it the joke takes you down one path and suddenly reveals
you're in another context so if it's a joke you laugh out loud and sometimes
you do in science as well science it's
in humorous ha-ha and science it's a ha and an art it's aa because each one of
these things is designed and is set up to take you from the context you're in
to another context you ever thought of having art classes at Wells Fargo ever
thought of having art classes for your kids in school after prop 13 ever wonder
why people are dull and boring
yeah aha that's the whole point of this talk so
if we look at Kessler's way of looking at things general education is incredibly important all that stuff that
is non pragmatic is actually what we need to be creative the reason is that
if we know lots of things and we're sort of meandering our way in the pink
context where I can we usually are just our subconscious mind makes little
ripples into these other things that we know it's like a little ripple into the
blue context for a second and then flashing back out because the the pink context said oh no that's not what we're
talking about so these little distractions that are going on people
who are good creative thinkers know how to make those distractions appear they
get them out of their subconscious and can make them explode into these ideas
the more you know the more context you have to change from the way you're
thinking into some new way here's
another problem I love frogs they're just the best example we probably all of us had front
have frog blood in us there's another thing you can do with a frog you can put
it in a pot of water and turn on the heat really low and that frogs nervous
system is not set up to detect absolute temperature so it will never climb out
of the pot you'll just sit there and
gradually boil now the Frog the previous frog with the Flies I think of as
science this one I think of his business [Laughter]
let's be really careful and deliberate
and we'll really think this out and meanwhile okay here's for about a
hundred thousand years ago here are some hand axes so here's here's one of the
earliest tools that humans used you notice it has a flaw in the design when
you pound on it this goes into your palm you pound on it it pounds back into your
palm and later on they made an improvement is they flattened out this
part so when you pound on it it wouldn't pound back into your palm and the thing
I like especially about this is that the interval of time between the picture on
the left and the picture on the right is about two hundred thousand years you can
imagine every once in a while somebody in a village would say you know if we just flattened this thing off and they
take this guy and burn him at the stake going against tradition the way it's
always been done against reality because
the problem is that every time somebody does something like this they're actually violating somebody else's
notion of reality that is why people hang on so much they're willing to do almost anything rather than be pulled
out of what they're comfortable with so basically what we are today are cave
people with briefcases
we're the same genetically as we were a hundred thousand years ago but instead
of being able to learn our culture by age 7 or 8 and then committing to it for
the rest of our life we actually have major changes every three or four years
this is what is screwing us up and when
a crisis comes along how do we do it we use our guts one of my favorite things
in the paper a couple of days ago in the New York Times maybe you saw in the
business section it said the Fed uses
its instinct to set the the interest
rate they said we use our instinct to
set the ants they're talking about an economy with millions and millions of
different parts and they're trying to use their instinct yeah this is like
voodoo another one I saw today and the USA
Today was good there was a spokesman from the tobacco Institute saying they
did a survey that showed that many more people in America knew that cigarette
smoking was dangerous to their health than new who gave the Sermon on the
Mount it's only one in three knew it was Jesus that gave the Sermon on the Mount
but 80 percent of the people in America and almost everybody who smoked knew it
was dangerous so why don't they quit my
cardiologist I talked to a couple of months ago said 300,000 Americans will
die this year from cardiovascular
disease that was easily preventable by just changing that we're told by their
doctors that all I had to do is change their diet and exercise patterns why not
they quit the reason is our nervous system is not set up to be interested in
long-term events it's set up to be wary of saber-tooth Tigers we love short
quick remember Jessica McClure little girl in the well in Texas 58 hours she
was in there and many Americans watch that almost continuously see what happened during those 58 hours a hundred
and five thousand children in the world died from preventable causes every year
in 94 in 93 there's a holocaust worth of
children dying from preventable causes 10 million dying from starvation and
dehydration and so forth we can't even deal with it we can't imagine it and so
we fastened on to one little girl in a well who wasn't even hurt right this is
nuts this is not sane
our mind is tiny we look at an epidemic
when it's starting off through a little
keyhole you see nothing that's what
happened when aids was announced every scientist in this country knew that if
you have a disease that is contagious
fatal and you don't have a cure you have one of these guys but nobody could see
it and so nobody did anything you know many people are going to die in Africa
in the next 12 years 35 million young
Vinney in the Pacific Rim 42 million according to the World Health
Organization they're politicians still don't get it because nobody is dying
like that now we can't use our common
sense to deal with the complexity of the world that are in today we have to use
something beyond common sense we have to use symbolic models of the world because
we have to do something before we just walk off the cliff so the whole whole
problem that I'm talking about here has
to do with dealing with things that are more complex than our common sense was
set up for problem solving which is what these teach in school is a terrible way
of getting ahead mosquitoes solve it with DDT wind up with more
mosquitoes because you're not just working with mosquitoes you're working
with an ecology the DDT kills the fish mosquitoes breed faster than the fish
the mosquitoes breed resistant strains faster than the fish you wind up with
less fish and they were eating most of the larvae of the mosquitoes and all of
a sudden you have more mosquitoes and the DDT doesn't work anymore this is what happens time after time because the
truth is is that we're involved in very complex systems and you should be aware
of this I'm sure being in the banking business it's nonlinear it's not
something that you can sit and turn a couple of dots and know what's going to happen like the Fed tries to do with the
quality doesn't work three recent winners of the Baldrige award General
Motors IBM and Dec why does equality
work because the only way you can get Six Sigma good at something is to have
done it for a long time and in this high-tech world that means you're
obsolete long before you ever get there Bill Gates doesn't want Microsoft to
ever get Six Sigma good at what they do that's not the thing what you have to be
is be fitted to the times as they are not getting really you can only get
really great at something if it's not going to change on you so total quality management is a farce
in high tech arenas so coming back you
have to look at McLuhan again I don't know who discovered water but it wasn't
a fish we have to try and make that work for us this time I'm going to actually
going to skip over the video and just go to this so the trick behind this stuff
is you can only learn to see when you realize you're blind you have to admit
it this is what scientists do in order
to deal with science you have to ignore what your hands tell you about touching
things what your eyes tell you about the world and you have to try and dig deeper
than what your senses and common sense are telling you it is really hard to do
so one of things you can do is you can make a better goldfish ball this is what Wells Fargo should do difference between
four hundred years ago coming out of the Middle Ages and the Renaissance and now
is that we're now self-conscious about these goldfish bowls and we can
structure them when you do something like an programming language you're
making a goldfish bowl for people it's a paradigm it's a collection of beliefs
it's a way of doing things these things are incredibly powerful because of the
way our minds work we can't change the way our minds work in any direct way but
we can change the context in which they do work
we see what we know if you take most
people who did it drawing like this would be thrown out of art class in fact
this person was this is what they did after just a couple of months later a
self-portrait of themselves and this is another portrait done by the person who
did this picture on the top after about three months in fact you can become an
incredibly good artist in almost any area in just a few months our ability to
learn things is much much greater than we think in fact most people are more ignorant
most people think of themselves as stupid but they're actually smarter than
they think they are the problem is that most people know less than they think
they do all right so we're constrained much more by what we know about the
world than what our latent abilities are we can make these multiple mentalities
which are always fighting each other work for us for instance this is this is
sorry this is almost illegible but this is a coherent essay a page long done by
a five-year-old and she'd only been writing for about three months three or
four months of course second and third graders often don't write very coherent
essays so how do how did the teacher get her to write this coherent essays well
it happens that children especially can't sequence symbolically the symbolic
part of their mind isn't set up for sequencing but they can do really good
sequencing iconically so what the teacher had the child do is to draw a
movie of the story she had in mind all children even we've tested children down
to the age of even three and a half and four can draw a coherent storyboards try
this with your children you'll also notice something really
strange and wonderful when they're drawing the storyboard they can draw better than when they draw a regular
drawing there's something about the focusing of it we still don't understand
then having done a coherent story board
over there the teacher then says write one or two sentences about each of the
panels and when the child does that they get a coherent essay that has a
beginning middle and end now is that cheating is it I don't think so I mean
if that's cheating calculus is cheating right the whole idea is we have to cheat
we're not smart enough not to get by
without finding these scaffolds we can
work our buns off in the wrong context
well three mentalities how about a way of grabbing things for user-interface
how about a way of using our ability to remember things in terms of images and
how about a way of dealing with symbolic renderings of those things and if you
put those together you get this machine in this interface
that's how it came about we took a psychological model about the
way people actually seem to think spend a couple of years trying to understand
it in the terms of communication between human and machine and came up with a
much better way of doing a user interface so the whole point here is
that by changing the context in which
you're in which you're thinking about things you can change the results not in
quantitative ways but in qualitative ways so the strongest weapon that you
have for exploring the future is the one
between your ears providing it's loaded thank you
okay now we have time for some some questions
and I love to answer questions so please
don't be shy yes
well I think I think everybody heard that didn't they basically is saying
from experiences that I've had being on
advisory boards and so forth and talking to large companies what mistakes do
people make well mistake is the wrong word because
the it's very it's quite rare for companies to actually make mistakes what
they do is to do things perfectly logically in the wrong place right and
those aren't really mistakes they're ill-chosen paths so the one of the major
things that companies don't do is to even look at this country as a model
this country doesn't have zillions of rules compared to say Germany it has a
constitution the constitution is kind of the DNA and it's designed to provide
maximum diversity and still allow us to
argue with each other right science what happened when science was invented is
people basically started saying this is
coming out of the 15th and 16th centuries was saying when Galileo's
telescope showed the four moons of Jupiter orbiting around a center in the
heavens that wasn't the earth it toppled the church's cosmology because this was
dogma right this was actually the Word of God that there was only one Center of
motion in the universe and that was the earth so that was devastating basically
the the conclusion that people came from that is we can't get involved in dogma
if we're going to get ahead the problem is if you don't get involved in dogma
how can you do anything all right isn't everybody's opinion the same right so
this is misplaced democracy everybody's opinion is say no everybody's opinion
isn't the same so how do you resolve that the way he resolved it is setting up a system for argument science is a
system for argument as Richard Feynman likes to say science means you don't
have to trust the experts right because there's
no scientific claim that you can't test you don't have to believe anybody but if
you want to play in science you have to be tough on the argument so what most
companies don't do is to set up a situation where they can argue well most
companies are terrible at arguing because they take arguing personally
right and that is no case and so what you get is party lines and dogma and all
that the paraphernalia of the past what you need to do is to have a way of
separating out personalities from the
fact that there are very difficult issues that need to be feared long and
hard and they have to be taken down to the essence or else you're just going to
get another wave another five year plan like the Russians used to do yielding
nothing because they refuse to look at the basis of Marxism and ask whether
there was anything there or not right so I think the thing is is to try and
convert from a dogmatic hierarchical system because if you think about
American businesses are distressingly undemocratic fact I think of them as
being like hunting and gathering societies as compared to agriculture my
American businesses try to put their competitors out of business
whereas farmers never do why because a
farmer needs all the other farmers to come help him build his barn so farmers
do limited competition on Saturdays they go out and help each other on Tuesdays
and they spend the rest of the time trying to get their farm going a lot so a lot of agricultural metaphors have
never come into business in any strong way one of the ones I used at lunch is
an agriculture you need to let your fields even in the Bible it says let
your fields lie fallow one year and seven that means you're giving up 14
percent of your return in order to be there 20 years from now right that's a
research budget 14 percent that's what it means American businesses can't
understand they're always top soil stripping and what do you get when you
topsoil strip you get a Dust Bowl and that is what we have got
so the only way these changes can happen is if the reward system has to be
changed by the top-most management to
investing some part in the future in a serious way not in a token nice way and
to I believe that people will not change unless there is some reward for failure
tell you a funny story I was I gave a
talk at the Stanford 25th anniversary of computer science and there 75 Silicon
Valley companies in the room and I was talking about told the story at Xerox
PARC when I when I first did the overlapping window interface I use a demo up to various Xerox executives and
one guy I went shoe and showed how the windows came up and this is so much
easier and expands the effective area of the display and this and that blah blah
blah and at the end as I said what's really great about this stuff is it only has a 20% chance of success we're taking
risk just like you asked us to he looked at me and he said boy that's great but
just make sure it works right the American businessman is idea
of risk being that 20 percent 100 percent of the time but it isn't it's
like baseball Ty Cobb's lifetime batting average was 367 two out of three times
he made out right they have to invest
when you're doing things that involve some risk it's not possible to hit a baseball perfectly every time it's a
very hard thing to do it's not possible to be right every time but the thing is
you can assure yourself of being wrong every time if you don't do the investment because then you just wind up
with a dust ball another question yes
first of all I was really delighted to hear you talk about Doug Engelbart oh
yeah me too trite but I'm a fan one of
the things that he's recently been working on is what he calls
bootstrapping corporations into the 21st century yeah and you touched on this you
talked about how businesses can improve themselves and one of the things he was
trying to get across without a key to the long-term vitality of an
organization is to get better and better at improving itself and I was curious
what you thought about his concept of for example most Fargo being a bank
naturally keeps things very private from
Bank of America and other competitors one of the engl Barts positions is that
we should actually be collaborating with one another yeah to improve the process
for everyone so I'd like to hear your thoughts on them yeah well the trick there of course our antitrust laws
so we have a harder time doing caresses than the Japanese do but basically I
mean here is a really you can throw tomatoes if you want but I mean the the
crudest statement I can think of is that in science you want to help your buddy
your competitor because there was a way
of doing it and getting credit for it namely they they get really shat on if
they don't put your name as a source of this part of the idea so in science you
tell everybody everything and there's a way of not losing in the reward system
and science has been one of the fastest
growing things in business the idea is let's screw our buddies because
especially in the short term when we're measuring in the short term we can always see little ways of cheating in
order to make the short term better and if there's no long term theory then I
think there's a there's a super problem I mean the truth is the companies do
collaborate by for instance sending people to standards committees of
various kinds there's a whole layer of company Apple is involved in quite a
number of consortiums there are 50 and 60 companies big and they include people
that are competitors of each other but for instance everybody knows that unless
we get network standards worked out in a reasonable way and we have a theory of
object protocols that work out where it's actually going to hold things up for another decade and so there's this
whole thing about how can we get the state-of-the-art stuff here and not just
try and beat everybody else to market but have some ecological theory now
angle Bart by the way invented the term
bootstrapping and invented it back in the 60s and he was giving that talk back
when I first met him which is a 1966 so he's been constantly trying to do
this and he it's a similar talk affects some of the parts of my talk I've adapted from his great old talk that he
used to to give about these we improve ourselves by improving our tools and the
tools McLuhan like to say we shape the tools and then the tools come around and
reshape us so you have to be very careful about the tools that we make and
how they reshape us know that yes Ian's standards and I was working in
marketplace that you have standards and you also want empower the individual to do their own thing how do you see that
balance being able to give a person freedom but also want to have standards
so everybody can move quickly it sounds like it's different directions but we have to find out it's that's that's a
really good question and it goes back to the Constitution idea I I was one of
about 20 people who worked on the
ARPANET 20 30 people who worked on the ARPANET and it's the thing that became
the Internet and one of the things we problems we had back then was the ARPA
projects had different computers he networks basically and heterogeneous
computer that's just shafted that way
you're going to get different mixes and so you have to sort of face up to it and
that was in 1968 it was just when I was starting to design OOP systems and I
realized that objects by their very nature of having a separate inside and
an outside we're perfect to go from different machines a different machine they could even carry different code
along with them and nobody was the wiser they could carry an interpreter with
them like PostScript does in case there isn't something that knows how to compile them and so when you think about
it just the notion of an object or the
notion of a cell think of how many different cell types we have they all run from the same DNA right so the
compatibility is done by not allowing gratuitous information outside of the
cell right so if you send objects around so people can be infinitely they can do
anything they want put it inside of an object and as long as the object can be
recognized by various protocols and api's that object can run on a wide
variety of different machines and still be completely idiosyncratic to what the
creative person wants to do so you can think of that as you know it's a
biological most of the answers I have to questions like that came out of I used
to be a molecular biologist a long time ago and thing you have to realize is a
biology is so much more complicated than anything computer people do it's by you
know tens of orders of man for instance a bacteriological cell which we have
lots in our but our body is about a hundred gigabytes worth of information
processing machinery and the processing
pattern matches are done in in sub microsecond speed all right so you
happen it's just so typical one of those things is like fifty thousand four
megabyte despot desktop computers that's one bacterium one bacterium is one five
hundredth the size of a typical cell in our body right we have ten trillion
cells in our body now the thing you have
to realize is that what's happening in our body is hundreds and hundreds of
thousands even millions of cells are dying every day
do everybody know that the atoms in our
body don't last for more than seven years even in our bones they're replaced
so we humans biological organisms are patterns moving through time and space
material comes into us and it goes out of us the pattern stays behind right so
there's a constant dynamic organization now why are we still alive the answer is
that more things are going right with us and are going wrong compare that to your
COBOL program where one line of code now let's see it wasn't well as far go but
what Bank was at Chemical Bank one line
of C code right and all of a sudden
people were having every time they took something out it was twice as much as
taken out how can you do that come on we
shouldn't you think of that kind of stuff as I mean if you're gonna use transaction systems can't you imagine
transactions at the code level like it's
like do you really want to program in a procedural language when you should be
programming in terms of constraints that's the way you do that when you do
that you wind up with something that's like a biological it's tolerant of error because the
constraints are gonna make the overall system get to the goal anyway so it's
like all biological systems when they're following a light they don't get perfect
information back and they're not doing things perfectly but the feedback mechanism will get them to the goal
that's the way we should be programming another question yes who makes it yeah
actually I always throw that in I thought you know I've all my
professional career starting from 1968 I've worked with children the Mac was
designed for children not for adults in fact some of the drawbacks in the user
interface are because of that fact we
did that interface so we could teach children some very complicated things and the to me the because of the way our
minds work because as humans we're
inescapably drawn and thank goodness in a way to learn our culture think of the
an arc think if we weren't social they'll be frightening
so we're drawn to learn a big set of beliefs and if we want to change those
beliefs it's much easier to change them in children than in adults and so I you
know I'm sort of the opposite the Roman Catholic Church and you know I want to save the children first and they are the
ones that I can't find any way of blaming them when something is going wrong with them whereas adults I can
find infinite lists of things to blame adults for they should know better etc
etc but for children good grief so the
answer is you just keep on throwing this stuff I think the Chinese water torture
treatment is you just kept people to keep on trying to notice the things are
infinitely more complex than television can ever portray and once people start
thinking in terms other than in television where you're dealing with
individuals and start thinking in these larger abstract terms that's that's when
things start happening anybody can save one child the most
important thing is to realize that if you save one child and somebody else say is another child
and all the children will be saved yes
time to quit I would love to stop on that note okay matter of fact thank you
very much [Applause]