Alan Kay Talk at 40th Anniversary of Dynabook (2008)

From Viewpoints Intelligent Archive
Revision as of 00:28, 25 November 2021 by Ohshima (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
evening everyone welcome
glad to have you with us tonight what a big crowd this
is fantastic I don't often do
this by the way I'm John Haller the CEO of the Computer History Museum for
of you I haven't met and I don't often do this but
I really have to give a giant thanks to
our staff to get ready for tonight
because it just so happens that the victory
party for the Silicon Valley branch of the Barack
Obama for president campaign
decided to rent the Computer History Museum
for its victory party last night and so we had we
had 2,000 people partying
until 1:30 or 2 o'clock in the morning and
as many of you were I'm sure and so
I just have to say that they did a wonderful job getting
everything ready for tonight and I just have to really salute them
so there's hardly anything better than a
reunion and tonight really has the feel of a reunion and
so for all of you who are being reunited around the 40th anniversary of
the Dynabook let me just say welcome it adds great
energy to the museum and we're really happy to
you here you know I've discovered in the short time
that I've been here that people get emotional when they come to the museum and
they do it for various reasons some people get
emotional when they see a trs-80 or some people get
emotional when they see the Utah teapot and
admit when I first came to the museum I got emotional about
something too and I was very surprised about it it wasn't an object
it was a text panel and
I read something on a text panel that I had
been quoting for many many years and there
it was attributed right there on this
panel and it said the best way to influence the
future is to invent it and
it was Alan Kay of course who said
that so when I saw that I thought now that is
really fantastic and I got a little emotional about it and
the reason obviously that is so
reason we're all here tonight is that the Dynabook embodies
the very best of that kind of thinking because it
was people coming together imagining
and better future and then inventing that future through
a computer that has changed the world and
the people who are here tonight did that and
it's exactly the kind of story that the museum
enjoys telling and I heard something a
minutes ago in the cocktail party that really hit hit
because one of our trustees had Feigenbaum
was relating a conversation he'd had with someone about why
he likes the museum and he said I like it because it's
as interested in collecting people as it is in boxes
so Alan and Chuck Thacker
two people we've collected over the years because
they are both fellows here at the museum and we're
to have them with us here tonight and I want to say pay a special
tribute to them as fellows because fellows are
important when they come back and it's always great to have fellows in the
museum now there are other ways of influencing the
by inventing things and of course
that is certainly true of One Laptop Per child
and dr. Mary Lou Jepsen who is also with
us tonight o LPC has generously underwritten
tonight's program and I want to take a moment to pay
tribute to OLPC n to dr. Jepsen because there
is a very strong genetic link between the
original vision of the Dynabook and the vision that
OLPC has for well really for
the world both share a vision of how computers
can be used to help children learn both are
dedicated to instilling in children a passion
exploring the world and sharing their discoveries with each
other and although we gather tonight to
to reunite and celebrate 40 years of the Dynabook
that idea is as fresh today as
the dream of OLPC is to bring education
and computing to all the world's children it's
a big idea big ideas or the hallmark of the computing
industry and of pioneers like Alan
Kay and Chuck Thacker in particular so I want to say thank you
to OLPC for helping this night be possible and for
all you're doing to bring access to children
in the 21st century
and now
it's my pleasure to introduce our moderator for tonight
Steve Hamm of Business Week Steve is a senior writer
for the magazine I think he has the best beat in the world
because he covers innovation globalization
and leadership that's about as big as it gets he writes a
wonderful blog on Business Week com if you haven't
seen it you should and if you were here early perhaps
lucky enough to get a signed copy of the book that
he has just written called the race for perfect
inside the quest to design the ultimate
portable computer I'll bet you can guess what
he ultimate portable computer was and he's here to moderate the
event tonight so please join me in welcoming Steve
one thing that occurs
to me is that just based on the
past half hour of my life here in the museum
there are probably more people in the
audience who know more about portable computing
in the two years that I spent researching it
so I think there just said this is like a wealth of
knowledge and experience here and it's
really very impressive to see all the people who've turned out
to celebrate this anniversary
in the past few days
something remarkable happened in it wasn't
it wasn't politics
and it was it went almost totally unnoticed
but in the computer world it was something that was kind
of a silent shift what
happened was in the third quarter of this year the
number of portable portable computers sold
overtook the number of desktop
computers sold there's something that's been
anticipated for a long time as so there's no big surprise but
it is a remarkable moment and of course
at the time that that that happens just
as quickly as that happens
there's another shift that's happening this that's you know even
even Wilder which is the shift to handheld
devices and then the fact that those numbers those number
the hundreds of millions but in the billions and these
days those are computers too so it's
kind of on this anniversary it's
great to notice that and it's great to have these
panelists here two of whom are
some of the great pioneers of computing
not just portable computing but computing itself personal computing
and another who who has already made
a great name for herself for Mary Lou but I think
represents in many ways the potential of the future
portable computing and personal computing
this is
what I found as I started working on my
this been a 40-year quest its
people are possessed with this with
these ideas with with these goals of making better
and better more personal computers and
the portable computer of course is the most personal
it's with you well it could be with you
all the time now that would iPhone other things like that but it could
be with you wherever you go and it
can be a true companion so
these are the the empowering computers
now when we were talking we
were setting up this event doing our plans and we've
quipping on the phone about about things Chuck
Thacker said that the Dynabook
is the most influential machine had never built
so I think right
so in fact Allen
in the past year had been trying to build
prototype he's been working on it I think it's well along the way and I think I'll tell
you more about it when he speaks but this
the Dynabook has been
something a something of a talisman there's something about icon
for decades and it's
it's something that inventors and
engineers and designers have have looked
at and said well you know this is kind of an ideal
we can aim for and they've
worked individually and in groups
many of them anonymously they're
you know they're there there are certain lots of famous names
that we know but there are people you know I just met
a few minutes ago the
guy who is the principal designer of the note-taker
who we the first portable computer and at
Xerox PARC had never met him before and
he gave me his car keys now always a wedding photographer so
he's graduated to career
but but just great to meet him a
wonderful guy so so
we've gone from those early days
as early the early concepts the
early personal computers at Parc and then did the commercial
computers you know from the locker bowls the laptops
the handhelds and now did
the smartphones we've seen this incredible evolution
you like to call it a revolution but 40 years seems
more revolutionary than revolutionary but they've been
certainly these great leaps ahead during the time and
where we are now you can it's just amazing to think back how far
we've come and the
quest is of course very far from over
there just so many things that you can imagine that can be done with with
portable computers in the future so I'm
going to move right along and introduce Alan
very well-known guy but just a just
a few high points but here's a guy who you
know grew up in various places in
the East it's but his team years in New York City I
think some formative times in the Queensboro
Public Library he told me where did the
tremendous amount of reading and Truman you know one of
important things happened to him and his was
he did a lot of educating himself and books
were a big part of that and you
know you can see how the Dynabook vision the vision of of
a computer book that the
children could use to amplify to give
themselves to empower themselves to
to to strengthen
their ability to learn and ultimately
to assert themselves in the world you can see how
that vision came it's not just metaphor it's
more than that you know he
got his start
in in portable computing when he was at
Utah University of Utah
studying computer science worked on an
early personal computer called the Flex machine
was inspired to to
think about portable computers
when he went to Lexington mass
and visited some computers in
the classroom projects run by Seymour Papert of MIT
and on the on
the flight home from from seeing that demonstration we
got this idea of a personal computer of
computers of empowering device for children so
that's where it all started and then you can see
how that is woven through his career at
Xerox PARC later Atari
Apple Disney 8
HP and now as a
professor as an independent researcher
you know still working on computer
programming that educational programming and
while he was a you know a pioneer he's
still at it hasn't stopped and
sign of slowing down so
one of the things that I when I last
when I interviewed Alan for my book
said at the end of the little bit long interview I said so
are you pleased with the progress you've seen
in in 40 years and he said as
a utopian you know you're pleased to see
some progress along a path but also
as a utopian you're never satisfied so
I think that that is the way that you
know you want to approach life never satisfied always looking
the next thing that you can do at this great
I do
so I I've given a million talks
about the
Dynabook and children's
education and
ideas about changing the future by the way the
quote is actually the best way to predict the future is to invent
it and I think
the fact that it was misquoted in
the history museum as an important thing
to notice about history
many of us were asked to write histories
of various kinds by various organizations like
the ACM over the years I think we all
found that even trying
to tell the truth it doesn't come out right because
there are too many actual factors
and reality is messy and
the tendency is to try and turn it into a story and
the other thing is that I
think many of us have been at
least as influenced by things that we hated
in our field as those that we've
loved and you
can you can in a history you can reference something
that you love in a sentence and a
footnote but if you hate something people
expect you to justify that hate
and all of a sudden you've run out of
pages for what you started doing the thing
so so I think it's very very difficult
to actually explain
what happened in any reasonable
way but those of us who
are privileged to be the right age with
the right funders to a
person when asked about this
what actually happened we've
all said that the context was
what made us better
than we were we weren't any smarter
than anybody today but
we got a lot more done because
we were just bricks like any other bricks but
we just happened to be in the era where the arch was functioning
and in other areas the bricks
could not do more than make simple piles but
there was an organizing principle in that community
and some of the people who set up
actually understood what those organizing principles
was great about it wasn't completely accidental
so I thought to set
the stage here I would just show
you a few things that I noticed around
that time that were part of the context
leading up to this idea and
my title for this is it's the 40th
anniversary of the Dynabook idea so
the idea is actually I think
still a good idea because it wasn't with
particular reference to any kind of hardware
or software is basically a kind of a service idea and we'll see
what that that was and it's in a second
so I can tell
by the amount of reflection from
the audience from either people with white hair or
no hair
that there
are a lot of people who know what the 50s were like
by actually being there and
not even
tossed in the shape of things to come up
there because even though the movie was made
in the late 30s you could actually watch it on television
in the 50s and it featured flat
screen displays and one of the things
that was in the air in science fiction salon before the
50s but especially in the 50s
was that there were going to be flat-screen televisions
a whole host of other things
that were going to be part of what modern life was
about it was just part of the context
and then we had this great enemy
who was a bit more like us than anybody in
America wanted to admit namely
the Russians they liked gadgets too
and so this
set up this pretty good
for most of the time rivalry which
only way that funders can really respond to anything
so a lot of really interesting things
started happening in the 50s and especially around MIT
so a whirlwind which I think
there's a piece of whirlwind in this museum was
arguably of the
many early computers it was one of the most influential
of all times in so many ways including
its architecture it was rather like
a mini computer although it was large and
it had some of the earliest
display as you can see in
the original Earl the that circular tube
that got made into the sage console of which there's at
least one downstairs and the
light gun for pointing at things on
the display and as the processor
cycle through the display list the light gun would see
a flash of light and they could correlate it through what piece
of graphics was being drawn back then and
through a little more machinery figure
out what what you're actually pointing at so all of this
was part of this huge
complex that was called sage and
if you've ever seen a sage computer I never
saw the original q7
I've seen pictures of it but
I thought the cue thirty-two and it was enormous because
it wasn't just one computer the
sage computers were actually in pairs both
computing the same thing and when I first
was taken to see the queue 32 I thought it
was small because it looked like a pdp-10 machine
room and then I was informed that this was just the console
man could they build computers
then when they were really hard to build it actually
took days for the Q 32 to crash
it was set up to detect
errors and it
had subroutines that emulated every
instruction out of whatever instructions were left
so it had this enormous thing so it's constantly
figuring out what instructions were actually working
on the thing and then patching those in so
it took a long time to go down
so people who saw this and it
wasn't me I wasn't I was in high school when this was going on
but this made a big impression on
a lot of people and
while this gigantism was going on
people I carry
Husky U is one of the earliest computer pioneers
beloved of many of us one
of the nicest guys ever in computing as
early as 1955 he
had done the g15 and actually with
my mentor later on in graduate school dave evans
appendix and so this is the other impulse
the impulse to have a machine
that you could just sit at for hours and make
things make things happen so this impulse towards
what we call personal computers goes way back
and then
we're gonna celebrate something in
2008 it's got to be John
so this really pisses me off
that it's great they were celebrating
I'm the most trivial thing we're celebrating is this
in a couple of weeks we're going to celebrate the
anniversary of the mother of all demos that Engelbart did
up in San Francisco and that was something worth while celebrating
that was a hell of a show but 50
years ago and on either side
of it McCarthy kind of thought his way out of
pretty much every assumption
anybody had ever made about computing up to that point
and it's just shocking
when you actually
look into all the stuff that John did more deeply
and so
he looked at sage and also saw that oh
the future of computing has got to be interactive
he realized
instantly in a way that
hardly anybody has been able to realize since then
that we couldn't deal
with computer in computer terms so
right at the very dawn of this he said we have to have something like
he calls an advice taker we
have to be able to deal with computers in terms of our own common sense and
so he thought of it as an artificially
intelligent entity and we'll
see later there's another way of thinking about dealing with computers
in terms of common sense and then he
just sat down and invented the best single programming
language ever invented Lisp just
I think
it worries me about this is that Lisp is now 50
pound-for-pound it's
still truly impressive compared to almost anything
most programmers are programming in because they just
don't get it and then John
did something even better which he wanted to show that
wonderful new symbol manipulation language
was actually a Turing machine and
not just in a trivial way but an ax turning
machine that could realize itself in one shot and that's
what is in the air which I've always called Maxwell
equations of computing and the
philosophical leap there for people who
troubles to go through it is enormous because it gave
about programming and programming languages
and making programming language that had never been around
before etc etc so this is one
of the great years of all time in our in
our field and
introduced the the sixties in many
ways was the decade of Marshall McLuhan he'd
written one great book just on one side of
1960 and understanding media was on the other side
and was loaded
with many kinds of insights and
McLuhan had this had
this notion that our biggest problem is
we are always thinking in terms of
what we know and therefore we're
completely locked in we're
basically not thinking at all so his idea was
if you're going to try and communicate with something in some way
some idea in some language that people know you
have to do it in a way that wakes them up so use something like Zen koans
but he called them pros so
this is one of them and
here's another nice image
steering only by looking
and most
can only see the present through the past that's because
we only use what we know to think about things
but artists can see
a little of the present as it is as a construction rather
than reality and once you can see something
as a construction you can see other constructions so artists
can see a little of the future and
his nobody could understand what this meant
medium is the message and so he wrote a book
called the medium as the massage
and that
came home to more people
because what he meant is when you're dealing with any
embodiment of any idea in any kind of system
it's what you have to become to
deal with that system that actually counts the
actual things are talked about in this system
are interesting but secondary but you
have to become a different kind of person in order to be a reader you have to become a
different kind of person in order to do science and
you have to become a different kind of person in order to watch
television for 24 hours a day
all of those things to him we're much
more important than the attempts to put content on
these media and another nice line
we shape our tools and then they turn around and reshape us and
Thoreau anticipated
this by saying we become the tools of our tools
children are the messages
we send to the future and children
are actually the future we send to the future so
started some
interesting thoughts from people who are actually looking at computers
is more than arithmetic engined pdp-1
appeared first in 1960
prototype is at BBN n its
immediate forbearer was a machine
called the tx2 which was on the TX 0 which is a
version a kind of a version of whirlwind
and highly
influential on a number of people especially this man
known as wick because
he was a psychologist who hung a rat hung
around BBN he was consulting there and he got fascinated and
because he was not a computer person or a technical
person he knew other things
like he understood symbiosis
and he realized right away that the
computer was more than a tool because of its
active nature it could act more like a like a
helpful Cynthia and
interesting thing that pretty much only biologists know you
may know we have a hundred trillion cells in
our body which is kind of a lot of cells
but you may not know that
forty percent of those cells forty trillion of
your hundred trillion cells are bacteria
think about that for just a second
turns out half of the 60 trillion that are left are red
blood cells and they only lived
for about a hundred days in your body
so just percolate on that for
a second and so liq
also had a beautiful personality people
like this man and
because of that and these
ideas that he started having in
a couple of years when the space program
has moved from DoD to NASA there was money left
over and according
to the stories there was literally one
of these meetings in the cosmos Club
in Washington probably with cigars and
certainly no women around the
agenda the old boys club and they
table and somebody said well hey let's give this money to lick
and that thus was born
ARPA i pto the information processing techniques office
and because of this man the
world is completely different today because
he was wise and we can talk about these
in the panel if people are interested many
people here knew him personally and
will never forget him
in the early 60s the johniac
downstairs one
of the worst computers to program ever invented if
you ever looked at the order code on it
very old and slow they
were going to get rid of it at Rand and cliff
Shaw of Newell Simon and Shaw he was also a
programmer and just a pretty nifty character
spoke about two words a year
he got them to give
throwing this machine away he said give it to me and let me
do something with it and so he did the first great end user
system and you can sum up what
Joss was like in this quote of says
the acceptance of an interactive computing system
depends on the little things the hundreds and hundreds of little things in
he's one of the few people in our history who
ever taken the pains to do every single one of those hundreds
of things so anybody who ever used Joss back
then Joss couldn't do a lot but
what it did it did it perfectly and in such
a way it made you feel good to be a human being and be privileged
to actually use a computer
then on the other end we
have this crazy genius Bob
Barton who
said why are people programming a machine
code they need to program and
higher-level languages and the answer was well that they
through slowly in Bart and said well let's make a machine to
run them fast so the B 5000
was the first really serious attempt
to make a higher-level architecture and
almost every idea in it is a classic
idea and these ideas
been used over and over again and reinvented over and over again
unfortunately mostly in software today because it's
very hard to buy any piece of silicon that has
an architecture anywhere near as interesting as the
B 5,000 was
the third machine I learned 1962
landmark and I can't
resist just showing a little bit
of sketchpad cuz I've been doing it for as long as I've
been able to show movies
lovely thing about this was there was no actual display
on the TX - Ivan actually programmed it
using the multiple program counter feature that
it had so the basic idea just sketched
something and then he indicated at the sides there
want them all to be mutually perpendicular and you just saw sketchpad
solve that problem and now
he's going to show that it can find
other solutions of course it could have constrained to the ratios
of size also but
here the constraints always wind up with a symmetrical
and wants one that looks more like a real rivet
of course one rivet is not
that interesting so what
you want to be able to do is make zillions of these and
so this system not only did
real time problem solving of constraints
and gave you
the first real working graphics system but it also
had this thing that people have called object-oriented
programming so there's one rivet
that he's going to stick
into that flange there
and then he's going
to show that he can make a few more rivets
and each of these rivets is
and then he realized whoops
I got that still got those crossbars I should make those invisible
so now he's making them invisible on the
master which we would call a class today and but
lo and behold the instances feel that
so here in one year and a PhD thesis you have kind
what I would call the romance of interactive computing
here's my vote for the first wheel personal computer
there's one downstairs Wes
hell of a guy I
love when
the ACM asked him to write a paper about
for the history of workstations
conference he was so modest his
title was the link was early and small
other thing that was great about this machine was it was
a study in parsimony and
that it got a hell of a lot of function
out of almost nothing and was
set up actually as a kit that all the first links were
by the people who are going to use them they had a kind
of a summer camp they came to Lincoln labs
and they were made really for biomedical
experimentation and so the people would come and they actually
built their own link and the everybody learned how to take
care of it and took it back to their labs and
many many links were made in the 60s
so pw1
phase two if we go up to 1962
Steve Russell he
was I think here tonight I know that he's here
at this but he was down there
he is my hero
30 percent for space war by putting it up here because this
is a very significant thing to happen because it got people
thinking and it was a spreadable thing
in the way that sketchpad wasn't thinking 7 70
 % Steve is my hero because here's
the guy that made the first list work and
going from McCarthy's way of thinking
about it to something would actually run
requires a very special kind of person as
special as John McCarthy but of a different kind so
Steve is one of that kind and
a little bit later is
another one of these special
there's gonna be a huge celebration
I'd like to see one
I say now computer
switching that will bring in a camera picture
from the camera mounted on his console such as the
camera mounted on - I built that's
great now we're connected audio you can see
my work you can point at it and I can see your face we could talk
so bill was about billa's down
in Menlo Park and the demo was up in in San
Francisco and I always ask computer
people today how did they get sub-second response and
all of this stuff on a computer whose total
memory was the size of that picture
192k picture and that's how much memory the
STS 940 had
it was a time-sharing machine it was a half MIT
24 bits and I've
only gotten gotten to correct the answer from one person over
the years who was a sophomore at UCLA and
the answer was they goddamn wanted to get sub-second response
want to get sub-second response you can get sub-second response and
the reason we don't have it today is because most
people who are making the systems don't
want it badly enough and most people who using the systems don't
realize that would help them
that's why it's this old stuff is really fun because
the people who who did this
stuff were just completely nutty as fruitcakes
in exactly the right way
you know who
another wonderful system so
the mouse was invented in 1964 but most people don't realize
the tablet the RAM tablet was also
move the connector out of the way so that we may draw a box in
its place
printing in the box is being used as commentary only
in this case oh if we
could only do with that well today isn't
it interesting so one of the simple things they did was just buffered
the recognition so you could write as
fast as you could write didn't make
you wait until it there was just almost
perfect user interface
and unlike the ankle Barb's which
had the feeling of a really beautiful tool
you have much more of a feeling
of being in a warm embrace with grail because
you're interacting with it tactically
in a way that you couldn't do
with a mouse so this is a very influential
system going on and
pdp-1 again it's kind of an epicenter for a
lot of things around 1964
Peter Deutsch who is maybe
16 at the time did an interactive
list on the PDP one that
had a number of wonderful points too it's a beautiful
machine code program but
he also realized that if you had a powerful
language like Lisp you don't need no operating system
that as
Dan Ingalls has pointed
out an operating system is just all that stuff that got left out
of the programming language you'd rather use
and you sort
confirming that if you start with the operating system so
starting with the programming language and just keep it
alive during the whole life of the of the
system and also
in that time there are just lots and
lots of things that I would
computers but there were certainly single user computers
they lack the service idea
of a personal computer but the
16:20 of course is the early 60s control de Deus
160 a I wrote some programs for that I
learned much more than I ever wanted to
know about how the 1130 worked when I was doing
my thesis project so
this idea of the utility of single user
computing was around and
65 is
when Gordon Moore wrote his day donation article the
original prediction was this green guy right here
which is a factor of two
doubling and then later
at God have banded to a factor of two doubling every 18 months
here every year every
18 months and so the reality for processors
was kind of on the doubling
every 24 months and
memories dynamic
RAM memories are kind of in between those
curves and the prediction went out to just 1995
so if you're willing to
believe this and a lot of people didn't believe
think it was relevant because his predictions were actually
for MOS process it was
really slow most people who are using
scale integrated circuits we're using bipolar because they were fast
bipolar we had a lot of layers and was difficult to predict
mos was simple it was slow but
it also was essentially pumping capacitors
electrons and so it was easy to calculate how
fast things would run if you could make them smaller
university of utah started thinking
yes this is
thing that Dave this I would say was the perfect way
of characterizing the whole Arco community
ittsan Carl Hewitt down there nodding
basically nobody cared how
crazy anybody was as long as they could sing
1966 I saw this article in
the ACM about
Simula and had a stroke of
little flash of lightning because it was
programming language that was kind of like sketchpad
and it reminded me a
little bit of biology and I'll take that story
up in a little bit so
in 1967 there are rumors of a 512
bit wrong from national
semiconductor it's hard to explain to
anybody who was young today how exciting
but I
because it was really
I mean we went the Apollo guidance
system was a core rope computer
magnetic cores that actually
had wires strung through them and it looked like a basketball-sized
of mess and that
was basically to hold
the controls for the instruction
sequences so the idea of a 512 bit
wrong that you could just use for storing
various kind of controlling information was very exciting
and this guy ed Cheadle wanted to do a little
machine as he called it and so the two
of us started started working on it and this
is its self-portrait on its own display
but what it actually looked like was this as a hewlett packard
display on top of in 1130
with a whole bunch of the hardware and software that we had done
so that was very instructive
and right around that time
Butler Lampson was in full
flight doing amazing things
and it's hard to believe that in
1965 or 66 about the Lampson there's
only 23 years old because he from
my way of looking at Butler
he was always a senior guy and
a couple of years
later a guy
whose hair is now whiter than this Chuck
Thacker joined this project
and as he said he never looked back
so this is a very serious attempt
to provide a service as
a time sharing system but it was attempt to actually provide
real services to real users real early
and Minsky
huge influence back
then and for me because like Minsky was
loved Papp Ritz work
and talk about Pafford all the time and it was through Minsky
that I decided to go visit tapper
then look lighter and now Bob Taylor
as he looked back then Bob
was the third of the funders at ARPA wrote
this highly influential paper thinking
about what the computer is not as a computer but as
a medium and that fit
in very well with the way Marshall
McLuhan thought of media and
the way I was starting to think about media and
liq did
not want this network that he wanted to have worldwide
be small so he always
call it the intergalactic Network back
then because he realized that you know
engineers are human and
so if they're given a
task of a certain size they will try and meet that task
so he wanted to give them a task that
was too large for them to me in
the hopes that the result might actually scale
and so as always
talked about in these terms and liquid had this wonderful
a psychologist he knew he was not qualified
to pass judgment on goals
even make up goals so he didn't make up goals he just made up visions
and he trusted the technical people
something about that visions and oh if the funders were just like that
today because aslak
said you can't have a decent idea inside
the beltway in Washington DC so
you need to do is to find people outside the
Beltway to do things so
this way of thinking about things I've
been painting a picture gradually here of what
I think of as one of the dominant ways that
ARPA community thought about things is that they gradually started
thinking about things in terms of no centers
so in order to scale the networks
were interested in couldn't have a center and
the programming languages that
sketchpad didn't have a center and
the object stuff that I was thinking about didn't have a center and
object stuff that Carl Hewitt sorry think about I didn't have a center all
of these things were like these
networks of things and like biological
cells no Center so we lose
many many cells each
year and we lose every atom in our body every
7 years gets replaced even the ones in our bones
Barry Wesler is here and
was working in the ARPA eye PTO office
at that time as Taylor's
deputy and he
was also very interested in the ARPANET and making it happen it
hadn't happened yet in 68 and Barry
the youngest guy there and there was an idea that the
people who actually did the work namely the graduate students
our pair should actually have a conference of their
own it shouldn't just be the the contractors and the first
one was held at the University of Illinois in the
summer of 1968 and
whole bunch of interesting people showed up there
but we had a tour of the
University of Illinois who saw that thing remember Barry so
this is Don bit sirs first working
plasma panel display for the Plato terminal
and suddenly something that
was 50 science fiction was but yet only
an inch square was actually working and could
display the University
of Illinois on it so it
was going to happen we started talking about gee
maybe a little desktop computer like the Flex machine could go on
the back of one of these lasting displays someday
so here's the
the slide that involves me my
background was in bio math I did
a lot of theater and music and
basically I got
reacting to things so I saw a sketchpad and Simula
and because of the bio and the
math I thought objects
and I saw Engelbart and the
Grail system and
came up with this thing that didn't have a mouse but a tablet
on it that was kind of a desktop
personal computer that had cheadle and I did
and then right
about 40 years ago today maybe
40 years ago last week I went to
visit Seymour papper because I'd
started visiting people who
could be users
of a desktop computer so I was interested in
working with people who are not professional
computer users and Seymour was working the children
and I saw the following
interesting thing that
Seymour is a mathematician and
he had studied with Piaget he
should be here tonight I think many of you know the
tragic story that a couple of years
ago Seymour was in Hanoi and got run over by a motorcycle
had to have part of his brain
removed to save his life but he's not recovered
from that so this is the this
the evening that is really dedicated to this man
who has meant so much to us in every phase
of our of our lives
and our ideas in
both the professional and educational levels
and Seymour was just
a complete genius at coming up with insights
and one of
the ideas he got was children are
egocentric but in
a charming way they do everything relative
to them so if they were a coordinate system there would be an
inertial coordinate system and an
inertial coordinate system is the differential
geometry of Gauss and
if you keep track of this in the right way you actually
get the differential geometry of vectors
and in fact a child is one
of those vectors and so is a turtle and
he thought boy this is
so close to the way children already think I wonder what they
would do if we put some formalism on it and allowed them
to treat it as mathematics and so
you get a little kid and say can
you draw a circle with your body and the little kid would go like this
and you'd say to the little kid what
are you doing and the kid would say well I'm going a little
and turning a little over and over and going
a little and logo is forward and turning a
LITTLEST turn and over and over is repeat
in so if you just tell the
turtle to do that by
golly you get a circle and since those numbers
whatever numbers you throw in there you're going to get circles of
different kinds and all of a sudden you've
differential equation here that is infinitely
simpler than trying to do differential equations
of geometry and analytic geometry and a little sudden you've
got something that children can so this completely blew my mind
this is the turning point on this
idea because once you've
got something it is incredibly powerful that a child can learn
you've no longer got an
you actually have something like the printing press that's
one of the great 500-year inventions
in human history because if children can
learn these powerful ideas then you actually have a chance
of having them not just increment on what's already known they
will actually help over several
generations and bend something new and so so
that combined with just seeing
the splat panel display and all
these wise words and the clue ins in my mind on the
plane ride back to Utah from that
meeting with Seymour I drew this little cartoon
with kids out in the grass because if
make a personal computer for kids don't put it on a desk
that isn't them and so
it immediately took the fun
idea of making a flat screen computer and made it paramount
that if you're actually going to do something serious
here with regard to children you had to make
a computer that was in every way made for
a child to take
with them to do with them just as they would take a book away from
adults to learn by themselves and
because our Poe was
thinking about packet radio not just doing
a packet network back then I thought okay well
absolutely it's gonna have Wireless and
it'll have a stylus and it
will have a touchscreen and it has to have a
keyboard because the one thing we learned from Grail is either if
you have perfect character recognition it's still not fast
enough for doing bulk typing and so
all of those things coalesced in
somebody helped me had
this where I could find it so
I after I got back to Utah I
found a cardboard box and
started making a model of this thing that I'd
drawn and the original version of this for quite
a few years did not have this cover his cover was put
on at Xerox it was not called the Dynabook until
Xerox but
basically the ideas that had to have a removable memory
which I would have killed for one megabyte of
removable file storage back then
has a stylist
many interesting
problems on it and probably the most interesting thing we
did is you could load it up with lead shotgun pellets
make it heavier and heavier
and heavier we could even make it as heavy as some of
the portable computers today
back then because we weren't interested
in coping we're interested in actually what is a max
weight and the definition of
portability I came up with was well you're always
carrying something else too so something
that's portable if you can carry it and something else
at the same time and so
if you combine a thing full of lead pellets
with other things you quickly realize that two pounds is if
there's two pounds of 1968
and it's two pounds in 2000 a
but of course as a service
idea you don't have to do it like this
so Ivan at that very
time was working on the head-mounted display just moved from Harvard
to Utah and we thought that
actually a head mounted display with liquid crystal
would beat a large panel because of the
contamination problems that you
their area dependent on large panels Thomas nobody
was cared about these the Japanese cared
a lot about making flat screen televisions so they put
enormous amount of effort into getting yield out of those
larger displays in a saner world
the head-mounted glasses would have come the
Negro pot II had this idea that basically
the whole world is going to be networked up and
so all you need is a thing on your wrist that identifies
you to the room that you walk into and knows where you're
pointing and they actually did a pretty scarily neat version
of that in the 70s later
those ideas got attributed
to somebody else and given a different name but there's actually
originally negroponte zayed in
the seventies and again if you think of this as a service idea
right away what you're interested in asking
is what what are the actual services
so then we're almost
to the end of this little context here
les Ernest is in the crowd where
are you les yay my favorite
besides being
nice guy he was willing to put up with me for a year at
Stanford when I was supposed to be working on
the hand/eye project
there but actually spent a lot of time thinking
about what it would mean to do the end user experiments
because the problem is it's not the technology
stupid basically that if you're gonna do
one of these things you have to have some understanding of the end-users
and so what I really wanted was some
portable little computers like
this or maybe like this was the kind of a thing
I sketched out at Xerox PARC and
in between Xerox PARC happened and when
I asked tailor what I should do he just said follow your instincts
and I never asked him another question
again and we just started working on it and
here's the thing in a nutshell and
we can talk later about why there isn't a Dynabook
today if you want but basically it came down
to four things that besides
all the media stuff that were obvious and simple
to do in emulating regular books and doing
television and animation and all that stuff there
are four four things we have to learn about
what powerful ideas were and how
they could be learned we're interested
in whether computers could aid as Patrick
shown us we
needed to understand how
this could work with human mentors teachers and parents and
I was most interested in
whether you could make computer mentors to
do this because my confidence
in adults was very low back then
and it still is basically
the biggest bottleneck in all forms of education
today and especially all forms of educational reform
are the adults that are in the system
and many in the third
world it's the lack of adults
but in our world is almost the lack
of adults almost no elementary school teachers understand
anything about math and science and so in
a way things might even be better if they weren't there
because the children would not be getting
misinformation about it so these
were the four parts and the idea that whatever
it is it's like reading and writing so it has to succeed for
over 90% of the children not for the 10% who
are set up by nature to be good at it and
of course the problem with technologists doing is we were all set up by
nature to be in that 10% that were good at all of
us learn to program in a week I'll bet you anything
it's not that hard if you
almost know what it is and if you don't almost
know what it is it could be really daunting for people
and this is why computer people generally design
terrible computer interfaces because they
are so they're not only willing to
cope with something bad they're pleased to
because it's a little challenge for
okay so at Xerox
we had this HP 35 calculator
which I only had for two weeks because it was
tolen cost a lot
of money and after
crying about it for a day I was happy because I realized
this is the first time I've ever seen a computer that
somebody could actually steal
so Butler
instigated and Bill English and others
instigated a thing called the old character
generator that
to try and understand what it would be like to make
readable things on a computer display and by
character generator we actually made a bitmap
memory that
allowed painting and other kinds of animations and
be done and so this was kind of the first bitmap
memory and what it looked like and
so one of the things that happened in 1972
is a lot of things that were
being thought about got simulated using
very expensive equipment like this old character generator data
general Nova computers and so forth so
a lot was known about what animation
was going to be like and what music was going
to be like and and so forth
and now we're just about to the end of the story here
because we're starting to take
that was an idea and is still an idea and
in a way ideas on account
for a little in computing because you
kind of have to implement the stuff this is
part of the story that really makes me plush up my throat
because every time you implement something you five
years go away and you
do learn something so but
the important part here is that the
kind of the super basketball team the
super sports team that's a real Research Center has people of every different
kind of temperament and when
they can work together you get this enormous synergy
of talents and we have a
couple here tonight that chuck was
kind of the master hardware
architect at Parc made
these ideas and ideas of his ideas
of Butler's real in such
a way that we could actually try things out and Dan
made a lot of these software ideas real
in such a way that we could actually try things out and under understand
them and so I'll end this
little context by
just pointing out that whenever
we honor somebody in our community the
community we're actually honoring
every single one of us because none
of those ideas would have happened without this
incredible intellectual environment
and practical
environment to have ideas and to
bring them to life thank you very much
thanks very much Ellen
and that 15 minutes did stretch beyond here
yeah yeah I was warned about this
effect you warned me about it so
just very quickly some quick introductions if that
Chuck Thacker why don't you just
Chuck you
saw Chuck in the and the slide show in his earlier days but
he hasn't aged that much
Chuck a close
colleague of Alan's at Xerox
PARC of the project leader of the alto
computer the first personal computer
of Ethernet and it just it just it just
to the laser printer laser
later worked at digital's Research Center
and now works for
Microsoft Research where he helped himself as
a Cambridge lab he's now back in Pell in
Mountain View effect just across the street here and
he was the hardware designer of
Microsoft's tablet PC
which was the realization of all those years of
dreaming of pen computing and finally
getting one that got in the market and actually sold
well and proved that that idea was
more than just a concept it was actually something could be achieved and you
know when I asked Alan would
look back on the 40 years and we've gone
with this the decision of Dynabook what's the computer
that's closest to your vision didn't hesitate for a second he
said the exo computer from the one left her child
organization and Mary Lou was
who designed that computers who looks great the cheese on stage
here now she's a scientist an entrepreneur
and inventor she's concentrated on advanced
display technologies but her interests
and skills go way beyond that
zone for work with the OLPC co-founder
of the One Laptop Per child project along with with
Nicholas Negroponte and the really
there are lots of remarkable things about that machine but
I'll call out is the fact that they needed to have
a machine that was very versatile very flexible
could be and could be used in all sorts of situations lighting
situations including outdoors and
she achieved that with a display on the on the
EXO computer so we're gonna have to
little discussion here I'm gonna ask some questions hopefully
they'll be kind of like a conversation but on
your seats there are some cards we'll
time at the end for some questions from the audience and so
please pass them up to the people who
collect them and we'll get some of those answers as well so
very good
all right Chuck
that's your chance you know one
of one of the great things I thought about the the contextual
setting that how long did
was it just showed all those threads that
that kind of came together over those couple
of decades that the that
fed into his ideas invited a lot of people ideas
and there was this incredible fertile ground and
I think you know a lot of people know that it was a Xerox PARC in
the 60s it's so much of this came together we
came from you know concepts
into some technologies that were really truly personal
yeah seventy-five years yeah
anyway so what
was it about that time and that place that
made that possible well I think it was a it was
combination of things number one there was the
arbor community and our community I
mean you can't quite understand how how it was
in those days because it was actually possible
ssentially everybody in the field because
it wasn't all that big and ARPA
had had
generated a style of doing research
that was
really quite effective and when
Park was founded
they were fortunate us to bring in Bob Taylor to
run the computer science lab and he
knew everybody so he could pick
who he wanted and he did pick
a lot of us so it's a pity that he isn't here
tonight but he he's insist on staying upon
his skill in Woodside but
I think that yeah
it was a it was a pretty collegial enterprise
yeah there's almost kind
of a middle ground between an
academic kind of research and product developers
well no we were much much more like academics than
like product developers we really did not think
of these things as products
what we were doing was essentially the way I characterize
it we were spending a lot of money to simulate the future
and that's not the way you build a product right
you're at the opposite end of the spectrum we're
taking these decades of technology
and advancement all the tremendous work done by
the Japanese on displays and trying to take
that some of these machines we were looking before downstairs
there was the note taker the side next to it
cost fifty thousand dollars to make this this one device
and I think Doug said that it really wasn't quite
that much but it still a tremendous amount a
expensive computer very expensive portable computer the
first one and you've made one that where
the material cost I think is $180 so
if you and and
but in a very collaborative kind
environment so if you would just talk about how you
could have you guys worked kind
of the other end of that almost like an open source
project hardware and software right mostly
some things are closed but well Allen came
and he sort of said you know here's
here's how it's got to work you know
how do you make a hundred
dollar laptop well if you look at the cost of the screen it's more
than a hundred dollars and so you can start
there and some Nicholas had started with a kind
of a science project projector and I looked
at that and I thought we got a ship the
kids are still kids were there it's about sin
so that really bounded the project and so we
just were able to bring in all sorts of people from all
you know this was the chance we had
a chance to everybody
came to the project had a chance to use their skills
in a way not just to make another product but
this product if we could pull it off could change the
world and so you think you work hard on any
product that you make but this product you don't stay
up just one night two nights three nights you know just
don't sleep for two years and I think that was the
the effort of a large number of
extraordinary individuals that that came
together to make it happen a lot of lots of different
ideas yeah but but a focus when we got to
ship this it's got actually work no no
yeah and I knew well you you
clearly had a goal in terms of that kind of a hardware
spec kind of goal and price kind of goal but
the but the but the main thing that was driving you was this desire
to make an affordable and flexible computer
for children I have a dramatic impact on that
certainly that was what was been driving
Alan all these years how did that
kind of inform your engineering and
design decisions well
you know a lot
of people thought it was impossible but if you looked
at the cost of you know super low cost CPUs they
were available a lot of people
came to the project AMD came to the project and and
really put forward a
lot of companies put forward extraordinary price points
and if you looked at you
know the cost of Wi-Fi and come-come way down stuck
with a screen problem of that being a hundred bucks so
we had I designed a new screen in convince
manufacturers to
but really the way we did it
was to not start with the engineering to
really just think about it a long time
a lot of thought a lot of going out talking to
the countries convincing them that
that they they believed us that we could
low-cost laptop and they said wow if he can do that why
don't you come and talk to the press I'll buy two million
of them and the president Lula da Silva did that first in
Brazil and we went actually back to the
US after that and got
calls from every other head of state in Latin America saying you
know we were thinking you know when a
lot of it was sort of getting
the groundswell in the the
different heads of state throughout the world
to say they would buy these laptops if they existed where
when the manufacturers heard that they could fill
their factories with this kind of volume they
were interested and so we thought we'd
pick a country in Latin America a country in Africa a
Middle East a country in Southeast Asia and so forth
but when we took Brazil on as
our country in Latin America
every other head of state in Latin America called us
and said how can Brazil possibly represent us
don't even speak Spanish
that's a good point
that's right they do rot in and so
that just shut I think almost after the UN summit
on the digital divide where Kofi Annan unveiled
our functional tethered prototype
Allen was there and convinced
us to actually have a motherboard underneath the table
I think every head of state except for Miramar
North Korea and Belgium had essentially
signed on to the project that was enough to
convince the manufacturers to work with us so they
so the the XO was the closest thing to your
vision of the Dynabook but the
data book is the you know the the computer
built the idea that was never built so why was
it never built in spite of all these people
really trying very hard to make
it happen well I think
you know in spite of the fact that I
have a cardboard model and
drew pictures of it as
I said to me it was a service idea
and the four
things that I put up in one of those
slides near the end so we have to learn what
our powerful ideas are and how to learn them how computers
could help learn them what kind
of mentoring can be done by humans and
to do my computer
those are the four things and
just working on the first two
with our 90%
success threshold for children led
to 25 years of
failure because we
were paying for this research ourself
nobody would actually fund the children's
research because we did these long projects
and we
didn't believe in most forms of the testing that are reported
in educational literature and stuff so we wanted
to actually convince ourselves that children were
getting fluent and that 90% of them were getting fluid
so by those criteria it was just one failure
after another and but
basically after each one of these which would last
five to ten years depending on the
effort we've learned something and about 10
years ago one of the systems that we did
started touching many more
children with adult health in
a much stronger way and so
I think it's safe to say after 40 years the first
two ideas
and a little bit about what kind of human mentoring
you need have been solved of the three things
but boy when Nicholas started
up the OLPC project my heart sunk
as I supported it because
mentoring if it's tough
to get good human mentors in the United States it is really
tough out in the third world and so
for instance there doesn't exist today
even the simplest
program like
interface that's in common use today can even find
out what its user is and what its user
knows and what it can do or just the simplest things
it can't find out what level of reading the
user do and help them do the next level of reading
so if you think about the the
common sense version of we did of McCarthy's idea
of what you needed for a common sense was the
common sense of using the world so we
made a relatively we made a world populated
with objects that reacted to what humans did
but they didn't interact very strongly whereas
that isn't enough pure discovery
learning took us a hundred thousand years to get to
science and so you actually need
learning that is actually facilitated
and if you can't make a
thousand good teachers in a year to save
yourself you have to have a user interface that can do that
to me that is and actually Danny Barbaro is
here somewhere Danny was
at parking known Danny for a thousand
years and you guys were supposed
to be doing this Danny what what happened
you can't don't evolve it
invent it no but so the
this this dream of having a user
interface that could facilitate is
as old as artificial intelligence that's
why I brought it up it is artificial
but you know you could say robot
yeah but what I'm saying is is that the
to me that was the thing we
would have worked on it if we'd had even one good idea but
I've never had a good AI idea in my entire career
and so it's a different temperament
a different kind of person and my thought was eventually
they'll catch up at the past and we'll
be able to actually make up for no teachers and
make up for bad teachers at least we'll probably never do as
well as a good teacher but we should
be able to do something and so when the OLPC project
started thinking oh my god we
are lacking the one piece of the technology
like if we could just ship that
with a simple program that would teach the children to learn how
to in their native language that would be the killer
app on that machine and wouldn't have to worry about anything
else for a number of years but that
technology doesn't really exist yes
I know everybody has it's like
it's like in the old days the
rule was everybody can make a one-inch
square flat panel display and nobody can make a 5 inch
square one still
be so I'm just saying that it's it's
that gap I believe that has to be bridged in
order to fulfill the educational goals that the Dynabook has if
you have to have some way of getting around the adults in
the system that make educational reform difficult
happening I mean here's the here's the day that prude
didn't believe that the kids could learn how to use the laptops
themselves they did a couple pilots and Peru
measured themselves on a World Economic Forum study
of primary school education throughout the world
142 countries participated in a study Peru
came 140 second dead
last but they said well you know at least we have a baseline
there are more than 200 countries in
world and so you know we're not that bad so they got
the laptops they're in primary
school education 15 1 5
of the children could pass reading comprehension tests
at grade level and in the schools we
put the OLPC xo1 to zero
percent of the children could pass reading comprehension
tests at grade level and they have
they have scores three two one or zero this
three is passing to is almost passing one
and zero are self explanatory we
went into schools where zero percent of the children
could pass at grade three and they were all zeros
or ones six months with the laptops
30 percent of the children got
grade passed at the reading comprehension at
grade level of at three forty percent
of the got scores of two and the
rest were heroes one that was incredible and
that wasn't perfect they could learn how to read and
it was a testing data we
didn't actually expect this as a result
in fact we were sort of saying all while testing teaches the
test let's not do that but that was enough to convince Peru
to buy four hundred thousand laptops and really deploy
them across the country so it's working
somehow we don't understand how but
you know when Nicolas started this project I said to
him if nothing else it's likely
to make this a topic of conversation
now to actually bring the idea of a children's computer
topic and if you can get a bunch of other companies trying
to compete with you in doing it that
will be a wonderful result of
the oil PC because that kind
of competition is very
often healthy for the genre and
I think there are at least seven
vendors now they're making machines
that they claim or aimed at children one way or
another and are less expensive and
the stuff that powers education
is the software that can ride on these machines
and so just like laptops became a category after
awhile I think more and more people are
thinking about this who maybe not have
children's best interests in mind but just think of it as another marketing
area that draw it can drive some good things
and so I believe this whole thing
is a step forward and
I still think there's a ways to go before
the thing I haven't done is escalated
the goals of the Dynabook from
back then because the of course
the the goodness of displays and
the size of the memory and the speed of the processors is far far
far beyond anything I postulated
back then but doesn't actually matter to
that makes some of these goals a little easier but
the problem is these goals are actually things
require invention and so they're only facilitated
by technology not they're not brought
forth by it automatically you
you know Xerox PARC days
through most of your career very
you know you're an academic type researcher
yet it was you and other
people bert healy by feeling
and I think also Lampson at
Microsoft Research a little bit he who made
who made the tablet happen and you know this is an incredible thing you know early
90s John doar or some of these others
who kind of whipped this excitement up around that the pen
computing also Apple I'm involved in it
really collapsed at
that time because partly because handwriting
just wasn't good enough but you guys came back
a decade later and made
it happen so so what's the thing that what
was different about a decade later that that could wasn't
just the technology resembled it's Moore's
law and that's all it all that it is and that in
1999 it was possible to build
a processor that dissipated less than five watts
it was not the geode that's in the oil
PC it was a transmitter processor and
with that and a modest modest
sized battery you could get you know essentially all day all
day life and we actually showed this machine
to Alan I took it down to Glendale when
we got the first prototype and I had
loaded squeak which was the system alan alluded to upon
it and Dan Ingalls grabbed it out of my hand
started running a benchmark and he said this is a fast computer
and I said yeah that's a pretty fast computer
that prototype is nicer than any
of this commercial versions
yes handwriting recognition is
bad but you can do exactly the trick that that Alan alluded
to in the talk you don't do the recognition initially
you do it in a background thread and and you fire
off strokes as you get them and it all works just
fine I have built that system unfortunately you
can't quite get it on on
the current tablet PCs but I think
we're getting there and there are some fine
new pieces of silicon out there in the world
ideas are so easy because
they have no mass and they're not affected by gravity
right exactly
have several questions here talked
about here question was what hardware limitations
still remains whether
the battery life that would you most like to see overcome
well the the problem that
I see in these devices for children is
that you would that you would really think of very much
its ruggedness and I guarantee
you that if you drop one of these things
it will break
well I actually know a
small company in in Palo Alto that
I like I got to design a thing that was we've
actually looked at this idea too about how to build low-cost
computers for children and and yes
we came up with a number 185 dollars to
the I was very worried about the ruggedness
of the liquid crystal displays now you could make them out of plastic
but that's you know that's a tough thing to do right
now manufacturers much rather make
of glass and the problem is when they make it they build
incredible stresses into it and so if you drop
it and it means a little bit then you got a stress
crack propagating right through the cover plate and I've
I've broken laptops displays and
probably many of you have too and we could not
tolerate that in a machine that's going to be used by children because
children use their their
school objects for many different purposes including
as weapons fact
one of the things we did it at Utah with
the cardboard because when I was a kid we used to
hit waiting for the school bus we used to hit baseballs with our
schoolbooks yep exactly so 700
G's we tested the exo
for that and five foot shocks
well I I was
interested in this question and
so I went and I Ivan I'm not
a mechanical engineer so I I went and hired one and
they built one for me that
was they used
some secret sauce but they called me up a couple of days later
and they said Chuck were not going
to be able to get the drop tests finished
time we said and I said what's the matter and he said well
the idea here was this machine unlike the XO has
a case that is an integral shock
control thing the case is actually made out of an elastomer
and the
display is supported in such a way that it can't bend
or without adding hundreds of G's
they said well the problem is every time we drop it from a
nut from an altitude of a foot higher we have to
take it all apart because it gotta live it up and we got to look at the display
carefully magnifiers to see if it's broken and
he said and we're up to ten feet and we're out of a light out
of ladders so if you need
a highly ruggedized laptop I know someone
who knows how to make it my
research the first think pad was
actually a tablet and it was made for
with the idea of field service you know
technicians and things like this and also insurance
adjusters oh yeah when we got into the
tablet business we knew immediately that we were going to
get 200,000 units a year and that was quite
predictable because all the previous tablets
had the resistive digitizers would go over to a more standard
thing and they all have this Atlantic
thought that they so they I think it was nation
something the adjustments would take you things on the field and
the first thing they would do when they wanted
of kneel down and write something as they would put they
would actually put the computer down and kneel on it
they busted a bunch of screens that way you
know I wanted to ask me you know you're you're doing a lot of work with
is place now that's really I understand the focus
of pixel Qi so what's your
vision of what you've wait where you can take the the work you
with the XO and One Laptop show what were you taking
that now it goes back to
the question that was just asked neglecting power consumption
that is the thing that we care about is battery life and
it feeds into everything
it feeds into robustness it feeds into use K feeds
into the fan noise everything and so in the
limit I just
think why is there a motherboard why is there a CPU
why isn't the laptop just a screen with it with some kind
wireless and cloud computing so physically all the circuitry
is in the screen your time or in the
so what what
we're focused on is really
trying the screen is by far
the most expensive part of a portable
and the most power hungry and the second most
expensive part of a portable is the battery
and so if you want to
to work better portables I think
somebody's got
and I think most
people when they're designing hardware don't even consider
this screen really part of the hardware
anymore so anyway I know a lot about screens
and I've convinced the LCD manufacturers
to work with me we're working with about 30% of the worldwide
LCD capacity sort of saying you know it's
kind of like CMOS TFT
LCD if you want there's all these sort of visions
of the future and display as well and
very few of them get into mass production
when they even starts a sample
that the patents have expired and I think it's
like 20-30 years ago face it CMOS one forget
the silicon and Sapphire gallium arsenide very good for
niche stuff but if you want to
make stuff that actually ships you
know there are six masks on top of an amorphous
is amorphous silicon you can make things so we're
focused on that making sunlight-readable splays displays
from reading things like that but
the XO is the first part of that it's
the only sunlight readable screen in a laptop right
now it's also a 1 watt laptop the
laptop battery lasts for 20 hours and
we made it for yeah poor kids
in the developing world but you guys can all buy
one starting November 17th and they give one get one program
sorry I had to plug it but um
you know you can get one too if
you if you buy two you have to give one to a kid in a
refugee camp but you can get one it's left
is the
computer as a as a empowering device for children
we don't have too many children in the audience today
but there must be but there must have been
at least one because this one
says though you may not have noticed there is representation
from the younger generation in the audience where
oh you left okay
all right so they went to bed so here's the question having
seen the whole spectrum of computing so far what is your
message to the younger generation
learn everything you can
forget it and just smell the perfume
that you have
to learn everything you can but
you can't take it first order seriously
in order to think you
has it has to be there where you can draw on it rather than it running
you around I think that the secret
of getting educated is
being able to use the knowledge rather than
having the knowledge use you that's that's the
and one of the biggest problems with a lot
of the youngsters doing computing today it's they turned
it into more like a pop culture where
and I've said to them on many occasions
is you could very well be smarter than the smartest person
back in the 60s and the 70s but you're not smarter
than a hundred of them and the 16 the 7
is very interesting because that was where aggregates
of very smart people in 20s 30s
and a hundreds did things
that no single smart person could
o and they did things of a different kind than most
of the stuff that's being done in computing today which
i think is hugely incremental and
there are some even some good reasons for the incremental
ISM but a lot of the incrementalism is just
because people are more interested in novelty
actually in getting above
some important threshold and the way to get beyond
novelty like Cicero said
the person who knows only his own generation remains
forever a child so
the way to get out out of that and I I would
say one of the things that was interesting about that time Chuck
can comment but cool thing was you couldn't get an
undergraduate degree in computing back then it was
so good you had to learn
something and
so everybody in
the ARPA community had gotten an undergraduate degree or master's
degree or some of them even PhDs or or not
but basically they had mastered something that was an
existing discipline that was really hard where the thresholds
were very strict and didn't matter where there was physics like with
Butler electrical engineering or mathematics and
everybody had a I think a really good
detector for what was
just playing what was just a or guitar in computing
hero machines including in Microsoft Research
I notice now yes yeah oh yeah this is one
outside very close to your office could be alright
look like you were about to say something what's your message well
young people I
outside I'm not sure exactly I would give this man
one of the things that I've that I'm very
worried about is that there
are a couple things that are happening in our society
this is this is an interesting
thing I've been trying to fact out my wife told me this that
had read somewhere that there was a study done of the
vocabulary of graduates she fought from middle
but it might have been from grammar school in
about 2008
now and
20 years ago and the answer was
that the vocabularies of kids have dropped
substantially more
so than then I think you could you
could account
for you know people just aren't getting are
you know we're not putting anything in the water but
we also see you know there's a rise with things
like attention deficit disorder and
a lot of that I think comes from
things like television
we did a lot of study on
how to help use computers to
help kids learn how to read and we
finally decided that the only thing that we
could possibly do that might work and
and this is in the u.s.
now this is not in Peru where they maybe want
in the US where they would much rather watch television the
only thing that we could possibly figure that would
rag the kid away from the television when their parents weren't
there because that's the norm now
and it wasn't 40 or 50 years ago is use
videogames and so we tried to develop a
system that combined both good pedagogy and
an experience that the kid would actually enjoy we
failed but when
I told Alan that we were doing this kind of thing he said he
you're gonna try to teach kids to read and I said yeah and he said I'd
never do that it's way too hard and he was right
you know I one
of the messages I would give to young people is
on occasion and maybe frequently
leave your computer behind and just go for a walk and
and don't plug your iPhone in your ears either just
like start to have some some sensory experiences
with the world around you how
many of you are without your iPod
right now
this is an audience filled of filled
with the elderly and only I don't
have them I don't have one like and to
think you used to be one day
was only here for about one year 1968
last word
it reminds me of this is very short story but
I was with Alan and Nicholas and in Tunisia northern
Africa a couple years ago and after
a grueling week we went to a Roman
ruins and dukkha and Alan
pointed to this rot in the road
that had been made by the carriage is being pulled over
the road for centuries and said he'd
buy lunch for anybody that could say
what the rut in the road was and I
was I'm tired I don't know it give up
but what's the name of the
rut in the Rhode Island the rut
in the road which is actually not worn by chariots for centuries
but was cut there for the chariots
is curriculum
is where our modern and you can look at
it as a bad thing and is a good thing you
know basically it meant the way towards
where you're trying to go the path that you take
to get there that's what it was and it came from
that rut in the road which by the way was
interestingly reversed and became
if you measure some of the ones that error then roman
ruins you'll find the gauges of some of the railroads
in Europe are exactly that with
to take that story I'm suggesting
just a few quick things to close up
Ellen I got a text from Cicero
he said you didn't quite have that quote right but
we're going to have it checked and
in the interest of protecting the
museum's reputation at the expense of my own let
me just say I blew the quote it's right downstairs
and I just want to
Cicero quote I was trying to
modernize it and remove gender from it
only his own generation
remains forever a child that's great
that's fantastic let's
let me just the Steve Allen had the very good suggestion
that everybody who was involved with the Dynamo project at
ARPA Park everybody who was involved in ARPA and
Park in the audience okay please
stand up so we can recognize you
wonderful thank
Steve's book is going to be on sale in the
lobby at the front desk afterwards so please
check that out I want to invite you on Thursday
November 20th if you can make it to one of our lunch
hail to the historian series Steve
Blank talking about secrets of Silicon Valley so
if you can make that we'd really love to have you and please join me in thanking
everyone who is involved tonight Alan
shock Mary Lewis Lee thank you very much
ave a safe trip home everyone