August 6, 2014 • 42 mins
TechStuff salutes an incredibly influential (and yet relatively unknown) tech genius: Claude Shannon. What did he do?
Learn more about your ad-choices at https://www.iheartpodcastnetwork.com
See omnystudio.com/listener for privacy information.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:04):
Get in touch with technology with tex Stuff from how
stuff Works dot com. Hayley and welcome to text Stuff.
I'm Jonathan Strickland and I'm Lauren Folk Oban, and today
we wanted to talk about a an important figure in
tech who often I think is overlooked. H not on purpose,
(00:26):
it's just he himself was a very kind of conclusive
is probably the wrong word. They didn't seek the spotlight.
He became a very private person. And also the work
that he was doing was technical enough in nature that
I think it's a little bit less dynamic as explained
to the general public. Yeah, it's a little more tricky
than saying this person built this thing which changed the world.
(00:47):
This is the person who came up with the idea
that the things that were built that changed the world
were built upon Did that make sense, I'd have to
diagram the sentence. We're talking about Claude Shannon Shannon Folks,
the father of in anmation theory right, also known as
the father of the electronic communication age, and his full
(01:07):
name Claude Ellwood Shannon. Very important person he's been. He's
been compared to, you know, some some pretty impressive, big,
basic big people like Einstein, Yeah, Einstein being one of them,
and you might say, well, whoa you know Einstein, Like,
Einstein's name has become synonymous with just the concept of genius,
(01:29):
like to the point where we use it in phrases
where we're being you know, a little a little condescenating. Yeah,
way to go Einstein, that kind of thing. But as
you'll see when we go through this this episode and
explain what Claude Shannon did and his his contributions to technology,
as well as just kind of his wacky personality, you'll
really kind of see how that that applies. So exactly
(01:54):
who was he and what did he do? When was
this guy born? He was born in nineteen sixteen in Potaski. Yeah. Yeah,
his father was a probate judge and his mother was
a high school principle. He also did have some mildly
famous family. A very distant cousin of his kind of
(02:14):
made a name for himself, Yeah, for killing an elephant
with electricity, Thomas Edison. He did a few other things too, Yeah,
that's the requisite doing from the internet. Thomas Edison obviously
did many, many important things, some of them not remotely
involving putting an animal to death with electricity. Yeah, thet
(02:37):
the large majority of which so kill an elephant once. Yeah,
I know, you just sticks with you right. Well. As
a boy, Claude Shannon became interested in electronics and began
experimenting with different stuff. He was just curious about how
things work and how to build them himself. He built
a working model of an airplane. Pretty impressive. Think I
(02:58):
think he was born in nineteen sixteen. You didn't have
airplanes for very long. They were pretty new. Yeah, they
were brand new back in the early twentieth century. And
he also reportedly made a working telegraph system that they
set up between his bedroom and a friend's bedroom. His
friend lived half a mile away, and it was all
made out of fencing wire. Yeah, so he could all
(03:19):
but I mean the wire itself. Yeah, he could actually
end up sending messages to his friend have a mile away.
He was also really into radio circuits and built a
radio controlled model boat. Yeah, so very much that. Yeah. Yeah,
this is this is the growing world of radio technology
and the growing world of communications technology. So he was
interested in it as a kid. Now a little bit
(03:43):
later on, when he was a teenager, he got work
as a basic mechanic in a drug store, running a
fix it shop in a drug store, because that's that
was like the center of town. Yeah, where you go
and you go and get your your chocolate malt and
your your your fan fixed. You know, it's a one
stop shop. He attended an Arbor College, where he studied
(04:06):
mathematics and electrical engineering. He graduated an Arbor College in
nineteen thirty six and then went on to enroll in
graduate level study at the Massachusetts Institute of Technology. And
he decided upon m i T because he saw this
work study ad like pinned onto a physical bulletin board
(04:26):
on his college campus that was advertising for someone interested
in working on Vanavar Bush's differential analyzer, which was an
analog computer that used these physical mechanical connections to make calculations.
The deal here was that he would spend half his
time working towards his degree and the other half in
the lab with bush Um, who was then m i
(04:48):
t s vice president and also their dean of engineering.
So this was kind of sort of a big deal Um,
and this machine was huge. It was the system of
gears and pulleys and rods that calculated with an entire
range values that were based on the physical rotation of
the rods, and you could program it by physically rearranging
all of these mechanical bits to correspond with different equations
(05:11):
the control circuit. I mean that this is how early
this was in computing technology. The control circuit itself was
a system of some hundred electromagnetic switches. Yeah. This this
is a kind of the the evolution of what Charles
Babbage created way back in the day, the difference engine.
Uh so we've done the text us done episodes about
(05:34):
and a Lovelace, who was the first computer programmer she built.
She kind of saw that computers could be things that
could do more than just crunch numbers. They could analyze
any kind of data. Yeah, they could represent stuff that
isn't numbers as numbers, so that you could She had
this brilliant idea of, oh, a computer might be able
(05:56):
to represent something like a piece of music and be
able to create, you know, replicated in some way. Years
and years ahead of her time. And the computers of
those days were these giant analog actual machines. Yeah, sometimes manpowered.
Sometimes they had this electro mechanical element to it. So
we're predating the time of the electronic computer at this point,
(06:18):
so uh As Claude Shannon began to work on this machine,
you know now that he had had enrolled with M
I T. He noticed something interesting. He saw that the
switches corresponded with a concept he had started on studying
first as an undergraduate, and that was really focusing on,
which was symbolic logic. Now. I took symbolic logic in college.
(06:41):
I loved it because the basic idea of symbolic logic
is you reduce logical statements to mathematical statements. Actually, I
took a similar class. It was it was basically the
at least mathematical math class I could get away with
as an English major. Well, the neat thing about it
that if you could prove that it mathematically made sense,
(07:04):
then you could say that the statement is true right,
and if it does exactly so, you could you could
start to listen to your friends argue and sketch it out.
And then he said, look, here's where you went wrong.
But at any rate, while he was at M I T.
He started really studying the work of a thinker named
(07:25):
George Boole, who was from the nineteenth century and back
in eighteen fifty four, George Bull published an investigation of
the laws of thought on which are founded the mathematical
theories of logic and probabilities, sometimes known as the laws
of the We usually shorten that to just laws of thought.
So this discussion about the mathematical theories of logic had
(07:48):
Bull using algebraic equations to represent logical forms and syllogisms,
which is exactly what you know I experienced when I
was in college. In this work, he also said that
the only i'd impotent numbers, which are numbers that can
be put through a certain operation multiple times without changing
the result, are zero and one. For example, one times
(08:09):
one equals one, and no matter how many times you
will multiply one by one, it will always be one. Right,
So if you take the product of that of that
that equation and then multiplied by itself, you still stay
with one, same thing with zero, although also with zero
you can add and subtract and still end up with zero.
So zero zero, zero, zero, so bool use zero and
(08:31):
one for the values of the symbols. In his algebraic logic,
he said an argument held in logic if when reduced
to an algebraic equation, it held in common algebra with
the zero one restriction of the possible interpretations of the symbols,
meaning that if you could replace the symbols with a
zero or a one and it's still made sense, it
still worked, then it held true. So Claude Shannon looked
(08:53):
at this and he was thinking, this is a really
cool idea. I love this, this approach to logic. And hey,
you know a switch has two positions on and off,
so sort of like a one in zero. Yeah, I
mean what if we were to, you know, kind of
so play with that, that whole switch process, And that
(09:14):
became something that would percolated in the back of his
head for a while. In fact, it percolated so long
that people suspect that he had fully formed this whole
idea of applying boolean logic to electronic devices for years
before writing it down, and once he wrote it out
and presented it, well, we'll get there. We'll get there.
(09:35):
I also do want to note that around this time
Shannon became interested in juggling, I think originally for like
physical mathematical purposes. He showed up, he started showing up
at the m I T Juggling Club, Juggling Club I
see what you did there, and asking some of its
members if he could like measure their juggling, and and
(09:55):
thereby sort of got involved with them, and this would
be a lifelong in trist As we will get into
a little bit later on a little bit of trivia.
A certain podcaster by the name of Jonathan Strickland was
a founding member of the University of Georgia Juggling Club. So, uh,
that's the only thing I really share in common with
claud I loved symbolic logic and I enjoyed juggling. They're
(10:19):
the comparison ends for he was far more intelligent than
I can ever hope to aspire. But yeah, you have
to agree with no, It's it's fine. I I have
come to grips with it. Okay. If you told me, hey, Jonathan,
you're never going to be as smart as say Claude
Shannon or Albert Einstein, it's alright. Most people won't be,
(10:39):
so I guess. Ninety eight, Claude Shannon writes a thesis
applying Bulls approach to circuitry by equating the zero one
restriction as the off and on positions of a switch
within a circuit. He was twenty two years old. This,
this had never been done. This has never been the
first time anyone had ever said this, certainly out loud,
(11:01):
and other thinkers have said that it would have taken
decades for anyone else to have come to this kind
of conclusion. Right, we could have been sort of groping
around with other approaches for years before someone had come
up with this particular version. And not only did he
come up with this idea, but the way he he
presented it in his thesis, it was very elegant, and
(11:24):
he would he would expand upon it a little bit later,
to the point where people said, this is this is
why he gets compared to Einstein. It's like Einstein saying
not just I figured out this one component to how
the universe works, but being able to express it elegantly
and have a whole picture right. Like it's like it's
not just a fact, it's a hill host of facts
(11:45):
that are all support one another. And it's like they say,
it's it's like you come up with a fundamental theory
of science and unfold it all at once. It's just so.
His thesis also laid out how logical functions such as
and or and not could be implemented within a physical circuit,
so building of logic gates. Now keep in mind this
(12:07):
is all in a hypothetical slash theoretical approach, right, It's
not like he was. He wasn't building this mechanically or
or electronically. That's the case. Maybe exactly, yeah, he was.
He was. He was laying out how this could be possible,
not actually building them himself. Claude Shannon leaves m I
T after earning a doctorate in mathematics to teach for
(12:27):
one year at Princeton Um. And here's the story. Has
a couple of different who has some alternate endings. We
will present you with the two that we know of.
But the story goes that he was teaching at Princeton
and while he was teaching a class he was holding
a lecture. Albert Einstein himself opened the door and stepped inside,
(12:48):
and Claude Shannon kept going on with a lecture, but
obviously was very much impressed with the fact that this
genius has walked into his classroom. He sees I'm Stein
bend over and whisper something to one of the students
in the back. He sees that the student replies and
then he sees that Einstein quietly leaves the room. He
(13:08):
continues on with his lecture. At the end of the lecture,
he holds the student back and with great anticipation, asks
the student, what did this brilliant man have to say
about my lecture? And my version of the story was
that Einstein had very quietly asked the student where are
they currently serving tea? I've heard that he asked where
(13:31):
the men's room was, so it maybe there's where are
they currently allowing you to peet? Could possibly been at
any rate. Apparently that became one of Claude Shannon's favorite stories.
He would love to tell the story about how Albert
Einstein walked into his classroom and asked something completely not
connected with what he had to say, and that made
(13:51):
him like just tickled in it tickled it, And I thought,
well that that also tells you a lot about his
his personality that he did not take himself seriously. Yeah. Uh.
In nineteen forty one, he joined a company famous for
its research and development, Bell Telephone Labs, and his work
mostly focused on things that had to do with the
(14:13):
war effort. In this ninety one is World War two,
and it included anti aircraft devices that could calculate and
target counter missiles, which came pretty seriously in handy during
the German blitz on England. Yeah. Yeah, it turns out
if if your enemy is blasting you with missiles, counter
missiles are a high priority. He also got to work
(14:35):
in cryptography, so here's something where he's got a you know,
a connection with people like Alan Turing who was working
on cracking the Enigma machine back over in England. He
was now Claude Shannon was designed devices used by Allied
powers to send messages back and forth, so he was
looking at keeping Allied messages safe rather than cracking German
messages or access power messages. He later wrote a paper
(14:58):
about communication theory of secrecy systems, which, according to M. I.
T is generally credited with transforming cryptography from an art
to a science. UM it was a mathematical proof that
an encryption scheme called the one time pad or the
Vernon cipher is is unbreakable. And it's the that cipher
(15:19):
is the basic idea of encoding a message with a
random series of digits a key, as we have talked
about on the show before UM which both parties communicating
have a copy of But you know, this is a
very simple concept in cryptography, but having the mathematical proof
that it is in fact unbreakable if the system is,
(15:40):
then that's really awesome. And when we talked about the
Enigma machine, that was one of those systems that could
have been unbreakable had people actually been able to follow
the rules properly. But because there were two things that
really fell apart for the Enigma machine. And I know
this is a bit of a tangent, but it relates
to this. Yeah, those two things were. One, the Enigma
(16:01):
machine was designed so that no matter what the letter
you pressed would never light up as the same the
same letter would never light up as the letter that
you had pressed, So knowing that meant that you could
remove one variable from all the possible outcomes. Secondly, people
were not as careful with their log books, with their
code books as they needed to be um and that
(16:22):
that led to the code being broken. But everyone seems
to agree that had every had the Germans, had the
access powers, been incredibly careful, then that would have been
an unbreakable code. Of course, times of war, you can't
really do share in human error being what it is. Yeah,
I mean it's it's that's the difference between the ideal
(16:44):
and reality. Meanwhile, uh, Claude Shannon began to develop theories
on how to apply his ideas about bully and logic
and circuitry to telephone switching lines. Because of course he's
working at Bell Labs in something else not involved of
in Claude Shannon happened that Bell Labs the development of
the transistor. Now, the transistor was a huge breakthrough. It
(17:08):
meant that the world of electronics could move away from
things like vacuum tubes and allow this other device to
take its place, essentially, which ultimately lead to the manatorization
of electronics. But it wouldn't be until Claude Shannon um
published his concepts about information theory that would let that
(17:30):
be a functional item in the way that it became. Yeah. Yeah,
it was really this idea of digitizing information that Shannon
had that made this a a practical device beyond just
especially that early transistor. It's enormous if you ever see
a picture of it, I think compared to the If
you think that billions of transistors can now fit on
(17:52):
a microprocessor chip, and then you look at the first
one it's it's enormous difference. Obviously. Now, this idea of
digitizing information was pretty much what would allow the transistor
to become useful. And also it's what would lead to
things like encoding information onto storage media like uh, like
a compact disc. This is what would make not just uh,
(18:17):
processing data possible, but storing it. Yeah, and right, it's
it's kind of a really beautiful coincidence that both of
these technologies were being developed at Bell Labs within a
year of each other. As it turns out, because in
that is when claudean and actually published his paper Mathematical
Theory of Communication. Yes, and that's available in PDF form.
(18:39):
Will will share the link because you can actually read
his paper on information theory. And this is the one
that I said earlier that you know, people, people who
were information theory experts, they say like, this is this
is like Einstein coming out with the theories of relativity.
This idea of a complete picture, not just an idea,
but a complete picture of an approach that laid the
(19:01):
groundwork for digitizing information so it can be transmitted and stored. Now, again,
he was a theorist. He did not build this. He
explained how it is mathematically possible, right, and so it
left it up to engineers and computer scientists to figure out, Okay,
if this is theoretically possible, how do we make it real?
(19:22):
What do we do to actually put this stuff into
into reality and have it work for us? Uh? Now
was when it was published, But there are people who
have looked into Claude Shannon's life who say that he
may have had this fully formed as early as ninety three,
and he thought that it was a really cool idea,
but just didn't think, you know, no one else is
(19:43):
going to care about this. I would, I would argue.
I mean, from from what I've read, it sounded to
me more like he kind of had it brewing and
just didn't want to present it until it was done.
He did seem like the kind of person who he
wanted to make sure that he had as complete a
picture of an idea as possible before presenting it to
(20:03):
anyone else. He did not want to have the experience
of coming forward with just half an idea. So yeah,
he's kind of a perfectionist in that sense. And it
really is a challenge to explain just to an average
person exactly how important this theory was, but you know,
in a in a practical sense at the time that
(20:25):
he was coming up with this, it was necessary to
create a better telephone system. So in the old analog
telephone system, you've got some pretty big limitations, some some
barriers you've got to get across due to signal loss
or noise, and analog telephone signal gets weaker the longer
that the telephone line it's traveling along is. Yeah, so
in order to get around that, engineers would place amplifiers
(20:48):
along a telephone line to boost the signal. So you
get a weak signal coming in, it goes through the amplifier,
the signals boosted, it's stronger going out. But unfortunately, um
the along with the signal that you want to get staid,
all of the noise that's on the line also gets boosted.
So eventually you run out I mean, I mean just
the noise takes over. Yeah, Yeah, you lose the signal
in the noise. So that would be you know, if
(21:09):
you've ever heard like one of those those telephone conversations
that goes on in an old movie where it's just
like all you hear is cracked, like yeah, just imagine
that if you're far enough away that all you would
get was the stack. You would not get any voice
at all. So, uh, the interesting thing was that by
switching from analog signals to digital signals, they didn't have
(21:32):
to worry about the signal boosting problem. Instead of a
continuous signal like a sign wave, which is, you know,
an acoustic wave, is what you would get with an
analog telephone line, digital signals are sent in a series
of bits, and a bit is either a zero or
a one. That's all based off of Claude Shannon's application
of Boolean algebra to electronics, and it worked so you
(21:54):
could do this with telephones, which was great, but it
meant you could also do it with just about any
other kind of nation transfer from radio to telegraph, telephones, everything.
And again this was one of those things that could
not immediately be implemented. The engineers had to build the
technology sporting. But once they did, they realized, we can
build out a nationwide telephone, even a global telephone system
(22:18):
that doesn't require amplifiers every x number of miles because
you're never going to lose that that signal clarity, all right,
Like hypothetically, you can do this with literally zero loss
in quality. So so long as you don't mind taking
the necessary amount of time for each bit to be transferred,
really the transfer speed is the only cap that you're
(22:39):
working with at this junction, exactly. And Claude Shannon he
kind of came up with that too. He said, uh,
you know, if if we have an infinite amount of time,
you'll have zero signal laws. But that any medium of
transmission is going to have ultimately a cap of how
much data it can care y at any given within
(23:01):
a given amount of time. So it was interesting because
that was one of those things that ended up becoming
a challenge to engineers. He said, look, for whatever medium
you choose, it's and it's specific to each medium. You're
going to have this limit that you're going to hit
and you can't go beyond it. And the engineer said,
all right, we agree, there's no way we can go
(23:23):
beyond that limit. So what our goal is is to
get as close to that limit as we possibly can.
And and this also led into some really interesting side
concepts about digital compression and error. Yeah exactly, Yeah, you
had to. You could end up compressing data into smaller
data packages, which helps you get around that bandwidth cap
(23:46):
But in order to do that, you also have to
have that that error correction software, that those algorithms that
are able to detect and and fix any errors that
come across while you're transmitting this information. These were all
laid out his ideas, and and that that error correction
concept also ties back into the idea that, uh, you know,
(24:07):
if you scratch a c D, you can still it
can still be read. Yeah, yeah, because you have these
extra bits that are built into the data itself, these
bits that otherwise would seem superfluous. They're not necessary for
you to have the full message, but those extra bits
actually allow some redundancy. So if there is some damage
to the physical medium, you can still end up using it.
(24:30):
And it's not like you get a smudge on your
your your disk and now you can't use it. Right.
So the concept of a disc also being new because
that was something that he laid out in here, saying
that this is a method for possible storage, not just transmission,
but also storage. Yeah, so so big big ideas. Uh.
At any rate, moving on with his life, I mean
(24:51):
he's so he's already gotten to the point where he's
laid out everything that's going to lead to things like
JPEG's m P three's ZIP files. UH, data transmission a
ross cable across telephone lines. All of this stuff is
possible because of the ideas he came up with. His
life continues on and in nineteen forty nine he marries
Mary Elizabeth Moore Betty Betty. She was a new miracle
(25:16):
analyst at Bell Labs, and they would go on to
have two children together. And he also, during his time
off from changing the world UH, decided to build a
simple computer to play chess, and he wrote a paper
about programming computers and computer chess algorithms. A lot of
computer like chess playing computers are still based upon the
(25:38):
foundations that he laid out while he was working on
this UH. You find that the Claude Shannon in his
spirit time often did things that that most of us
would be like, well, you could have a full time
job doing that. He's like, no, I just want you know,
I'd like to keep my hand in. Around that time,
engineers at Bell Labs at that time being ninety nine,
(25:58):
began to actually create the technolog The implemented Shannon's ideas,
and they built something called a regenerative repeater and the
idea was that a bit could be regenerated perfectly and
repeatedly as long as the bits weren't quote unquote too small.
So as long as the messages weren't too small, they
could consistently regenerate a message. Uh and that would mean
(26:21):
that you would again have no signal loss, You wouldn't
lose any data in the process because you could just
just as quickly as it was coming into the regenerative
regenerative repeater, it would send out a copy the same
data message back out again. Um. Also to around this time,
as the engineers at Bell Labs were creating that that
(26:42):
physical technology to incorporate Shannon's ideas, he started to introduce
the idea of bandwidth limits. Yeah, this is what I
was talking about when he said, it doesn't matter what
medium you're using, eventually you're going to hit that capacity.
And eventually they started calling this the Shannon capacity or
Shannon limit. So it was again a very important idea
(27:03):
that ended up being playing a huge role in the
telecommunications industry as well as just electronics and computing in general.
Uh So, this is what gives engineers that goal, This
is where they want to hit as close to that
number as they possibly can to maximize the amount of
data they can shove through any particular medium at top speed. So,
you know, we often talk about data transmission speeds, but
(27:26):
speed is really kind of a deceptive term because it's
not just how fast something gets from point A to
point B. Usually we're talking about speeds that are approaching
the speed of light. That's really fast. What we're what
we're really concerned with is throughput, which is the amount
of data that can travel at that speed to get
from point A to point B. Because if you're dividing
(27:48):
that data up into lots of of bits like a
long string, yes, each individual bit is moving at the
speed of light, but you still got to get that
whole string through. Yeah. Yeah, it's it's the you know,
getting the campus through at the end. Yeah. Yeah, it's
the idea of if the if we hear that there's
pizza in the kitchen, uh, and we're all invited to
go and eat it. Then the problem isn't that we
(28:09):
have a bunch of slow people on staff. We're all
very very fast. The problem is the doors only so wide,
and eventually four or five of us while just try
and cram through it, at the same time. So that's
the difference between just speed and throughput. Now, sept ones
and zeroes don't usually elbow you in the face, that's true,
but we have no such restriction, as we have demonstrated
upon multiple occasions. Now, at this time, engineers were also
(28:33):
trying to find on ways to take on other elements
of this theory, like the compression and redundancy ideas, and
build working devices and algorithms that turned that theory into reality,
actually making products that could take advantage of the ideas
that Shannon had produced. And uh. Meanwhile, Shannon received a
very special present at Christmas of from his wife this year,
(28:58):
a unicycle, and stories say that he frequently rode through
the halls of Bell Labs at night on this unicycle
while juggling. He is my hero because of why not? Now, See,
if my wife gave me a unicycle for Christmas, I
would imagine she was plotting my demise and perhaps had
put taken out yet another life insurance policy on me
because she knows my my lack of balance. But but
(29:23):
I I have nothing but respect for someone who is
transforming information theory while writing a unicycle and juggling juggling. Yeah,
so because because it Meanwhile he was looking into machine
intelligence and memory. Yeah, he was really branching out, you know,
he was he was very much interested in exploring all
(29:44):
these different ideas. Now, by nineteen fifties six, he decides
to leave Bell Labs, though he continues on as a
consultant and he goes back to M I. T. To teach.
He also wrote a paper he was called the Bandwagon,
and uh, that's when he said he didn't really like
how the words information theory were being thrown around. So
(30:06):
essentially what he was saying was that they were losing
their value. Information theory as a concept was losing its
value because companies were using it to describe things that
didn't really fall within the umbrella of information. Yeah. It
was a really popular and pop culture almost term in
the scientific community at the time. And I mean people
were publishing papers that had information theory and the title
(30:27):
just because they thought it sounded cool, when in fact, right,
it had nothing to do with that. So it was
kind of like how virtual reality became this buzzword that
began to lose meaning, particularly when the public started to
see what the reality of the field was as compared
to the Hollywood depiction of what virtual reality was back
in the early nineties. Sure, sure like artificial intelligence or
(30:51):
I read an essay recently from the guy who coined
the term manic Pixie dreamgirls saying that he just wished
he had never done that thing. I would like to
apologize to the world. Yeah. So this was one of
those interesting things were the paper wasn't so much about
advancing the concept, but just saying, let's use our words
carefully and correctly. He said that perhaps the term had
(31:13):
quote ballooned to an importance beyond its actual accomplishments end quote.
I think that's a little bit modest on his part, Honestly,
I think so too, considering that again, without that theory,
computers and electronics would not work the way they do today. Yeah,
but at any rate, this kind of marked the beginning
of Shannon's disappearance from the research and technology scene. He
(31:36):
he really didn't want to be a celebrity, I think,
and he had this huge push from the media and
the government and science in general to be made into one,
and it it kind of pulled him away from from
both research and public education, right and he was it
wasn't that he was cold from why, I understand whenever
he gave talks they were really great, and whenever he
(31:58):
wrote papers they were really great. He was constantly being
pressured to do that, and it was starting to become
more of something that would cause him anxiety as opposed
to something that he would enjoy doing well. In nineteen
seventy three, the Information Theory Society, which is part of
the I Triple E or I, created an annual Shannon
(32:18):
lecture that became the Shannon Award UH And in nineteen
seventy eight, Claude Shannon officially retired from m T, although
he had not really been actively working there for some
years before. Certainly UH and in nineteen eight seven, Claude
Shannon gave his last interview to Omni Magazine. Now, by
the late eighties, Claude Shannon began to suffer from Alzheimer's
(32:40):
and withdrew from the public eye entirely. His wife would
go and attend events instead in his place, and in
February two thousand one, at the age of eighty four,
he would pass away. Yes, there are some very UH
inspiring and moving tributes to Claude Shannon that were published.
Really beautiful things. You can certainly go online and read
(33:03):
a lot of those those tributes that were written the
week and month following his passing. And we have a
collection of interesting little trivia that we didn't really want
to fit into the overall episode. But it didn't really
fit into the timeline. But so much of I mean,
if it wasn't charming enough, I mean, if charming is
(33:23):
the correct word, actually charming is totally the correct word.
According to me, I find it downright charming that he wrote,
you know, papers that mathematically proved the computers can exist.
But but but but other than that, there's just a
lot of little just so. So one of those things
is that, you know, we just said he he was
(33:44):
not big on on pursuing the limelight. He didn't he
didn't go after that at all, and and often he
would reluctantly take the stage, but as time went on,
he did that even less frequently. He wouldn't go out
very much at all to address the public, and according
to M I. T. Technology Review, he even had a
(34:05):
file labeled letters I've procrastinated too long on So if
he got something from colleagues or government officials or scientific
institutions and had just been sitting around for a really
long while. He would just put this in a file, saying, well,
that's too that's too late, and that's never gonna happen.
So I'm just gonna put that in this file. Um. He,
(34:26):
like we said, love to build stuff, to engineer stuff.
You know that whole telegraph line stories one of my favorites. Um. Now,
as a parent, he built a chairlift that would take
his kids from his house to a nearby lake so
they didn't have to walk the whole way to the lake.
He also, from what I understand, designed a hidden panel
in his office that didn't lead anywhere at all. He
(34:48):
just he just felt like building one. He just needed it.
It made me think of a Mitchell and Web sketch
where this wall must rotate both here and not here. Look, Mite,
that's a load bearing wool. But anyway, he just decided
he wanted to make one. He also built a life
sized electric mouse named Theseus, after the Greek mythology figure
(35:12):
that's the one who was stuck in the labyrinth that
had to find his way out in the minotaur or minotar,
depending upon your preferred pronunciations, after him. So this mouse,
what it would do is it would explore a maze
and quote unquote remember where it comes from. It was
it was going after some little metal cheese bits. I think.
So the the way this mouse would go through the
(35:33):
maze is it would go down a pathway and whenever
the pathway would branch, it would start to rotate. Yeah,
so it would take one and then it would, uh,
it could backtrack if it went down an incorrect route, right,
and then it could take the path it had not
taken as opposed to you know, if this were just
an electronic mouse that had some collision detection, it wouldn't
(35:53):
It could potentially just go back and forth down the
same little pathway forever. Yeah, but this was branching that
this one knew. Okay, well I already took the path
that's on the right, so I have to take the
path that's on the left. So it's pretty cool that
he built this thing, you know, just for the fun
of it. He built it also probably my my favorite
(36:14):
robotic piece of his eight juggling robot, a bounce juggling
robot to be precise, bounce juggling robot that like w
C Fields to be even more precise. Yeah, it was
like having a like, imagine a drumhead, right, and the
drumhead allows things that are dropped on it, like a
ball bearing to be bounced on it. And then two
(36:34):
little uh angled platforms that are serving his hands that
are bouncing this again, these little these balls. Yeah, and
it just kept it going in a in a bounced
juggling pattern perfectly. And he basically made it out of
like erector set pieces. Yeah, you know, just like you do.
And then he wrote a paper on the dynamics of
keeping multiple objects in the air simultaneously. It's pretty famous
(36:55):
within the juggling community. I tried to read it what
I actually wrote, how Juggling works for how stuff works
dot com. In fact, if you go to that that
article on how stuff works and you look up how
juggling works, there's a video of me juggling in that article.
I still I still say it because I juggle a
little bit. I still say that we really need to
do a video of all Right, I juggled torches in mine.
(37:17):
You're ready to pick those up? Okay, well, well we'll
start small. Uh. He also made a robot that could
solve a Rubic's cube, which is pretty amazing. I mean,
obviously that needs I can't either. I know there are
algorithms for how to solve it the most efficiently, and
I've seen people who are really good at who just
like it's like it's like magic. You know. The way
(37:40):
I saw a Rubik's cube is by peeling the stickers
off and then replacing them properly. I cheat, but yeah, no.
He he created a robot that could follow these algorithms
and also just recognize what the pattern was on any
given side, so it could, you know, create the rules
that needed to solve it. UM and he made a
calculator that worked with Roman numerals. It was called throwback,
(38:02):
which stood for a thrifty Roman numerical backward looking computer.
UM also rocket powered Frisbees and motorized poco sticks. Yes,
the motorized pogo stick. I was thinking, like, again, that
sounds terrible. If the unicycle hadn't killed me already, that
certainly would. He built the ultimate machine. My favorite machine
(38:23):
of all time is the ultimate machine. All right, tell
us about it, Jonathan. All right. Now, imagine you have
before you a box, and on that box you can
see the outline of a trap door, and the only
other really interesting feature on this box is a simple
switch has switched to off, and you push the switch
to on. The trap door opens and a hand emerges
(38:46):
from beneath the trap door and hits the switch back
to the off position, with draws back inside and trap
door closes. That's it. That's it. You hit the switch
and the harm comes back out yet the switch, the
arm comes back out. Uh. I want to share this
video too. There's a video of a brilliant variation of
the Ultimate Machine that is hysterically funny. It doesn't just
(39:10):
do that like, it starts to do it so um.
It ends up at first looking like it's a variation
on the Ultimate Machine, like oh, that's cute, But then
it starts doing other things too, because this particular box
had wheels on it and can move autll the way,
so it's starting to avoid the person who's trying to
hit the switch, or it would playback prerecorded messages saying
like hey, hands off, buddy, that kind of stuff and
(39:32):
was really really entertaining. So we'll share that one as well.
But you have to remember that that particular very entertaining
machine is based off this thing that Claude Shannon built
for no reason other than it tickled him just because
he could. Um. He also had a collection of exotic unicycles,
including some that were because he he was wondering, how
small could you make a unicycle before someone would be
(39:53):
unable to write it? Uh? For me, that's any size.
But but I think me too, that would be any size.
Assuming that you are capable of writing a unicycle, how
small could you go before you could no longer maintain
your balance? In fact, he had a couple that I've
heard are essentially unwriteable. Uh. He also lectured on using
information theory as an application to playing the stock market,
(40:16):
though he never really published any work on this. He
did do a lecture, but he didn't write a paper.
He also did really well in the stock market himself,
although he wasn't necessarily employing information theory to do so.
He was investing in companies that friends of his. Yeah,
he made some very savvy stock purchases based on amazing
work that his friends were doing. These are These are
(40:38):
the people who were inventing like the basic components of
computers and electronics, going on to form their own companies,
and he would invest in those and then they ended
up being these these enormous companies we know today. So
he did quite well. Um, and there's no Nobel Prize
for mathematics, which is why Claude Shanna never won one, right,
but he certainly did win a number, I mean, probably
(41:01):
way too numerous to mention here awards, but but one
that we wanted to mention it is the very first
Kyoto Prize, which was created in Japan to award honors
to contributions in mathematics. Essentially, it was supposed to be
the Nobel Prize for mathematics, right, and this was all
the way in the nineteen eighties, and this came into invention. Yeah,
the very first one went to Claude Shannon, and from
(41:21):
what I understand, it actually came with an even larger
cash prize than the Nobel Prize does. So so if
you if you feel like he was he was snubbed
because Nobel Prizes don't recognize mathematics, fear not, the Kyoto
Prize had him covered. So I hope you guys, uh,
if you had not ever heard of Claude Shannon before,
I hope you learned something in this episode, because he
(41:43):
really did seem to be a remarkable person. In multiple ways.
I mean, this guy seems like the kind of professor
I would have absolutely adored um. But then you know,
I like all of my professors who had lots of personality,
and we're unafraid of coming across as a little unusual
or yeah those are my favorite. Yeah, so awesome. And
(42:05):
I hope if there are any other really important figures
in technology that you would love to hear us cover,
you should let us know. Send us a message you
can in says an email that addresses tech stuff at
how stuff works dot com, or drop us a line
on Tumbler, Twitter or Facebook, or handle it. All three
is text stuff hs W, and we will talk to
(42:26):
you again really soon for more on this and thousands
of other topics. Because it has to works dot Com
Get in touch with technology with tex Stuff from how
stuff Works dot com. Hayley and welcome to text Stuff.
I'm Jonathan Strickland and I'm Lauren Folk Oban, and today
we wanted to talk about a an important figure in
tech who often I think is overlooked. H not on purpose,
(00:26):
it's just he himself was a very kind of conclusive
is probably the wrong word. They didn't seek the spotlight.
He became a very private person. And also the work
that he was doing was technical enough in nature that
I think it's a little bit less dynamic as explained
to the general public. Yeah, it's a little more tricky
than saying this person built this thing which changed the world.
(00:47):
This is the person who came up with the idea
that the things that were built that changed the world
were built upon Did that make sense, I'd have to
diagram the sentence. We're talking about Claude Shannon Shannon Folks,
the father of in anmation theory right, also known as
the father of the electronic communication age, and his full
(01:07):
name Claude Ellwood Shannon. Very important person he's been. He's
been compared to, you know, some some pretty impressive, big,
basic big people like Einstein, Yeah, Einstein being one of them,
and you might say, well, whoa you know Einstein, Like,
Einstein's name has become synonymous with just the concept of genius,
(01:29):
like to the point where we use it in phrases
where we're being you know, a little a little condescenating. Yeah,
way to go Einstein, that kind of thing. But as
you'll see when we go through this this episode and
explain what Claude Shannon did and his his contributions to technology,
as well as just kind of his wacky personality, you'll
really kind of see how that that applies. So exactly
(01:54):
who was he and what did he do? When was
this guy born? He was born in nineteen sixteen in Potaski. Yeah. Yeah,
his father was a probate judge and his mother was
a high school principle. He also did have some mildly
famous family. A very distant cousin of his kind of
(02:14):
made a name for himself, Yeah, for killing an elephant
with electricity, Thomas Edison. He did a few other things too, Yeah,
that's the requisite doing from the internet. Thomas Edison obviously
did many, many important things, some of them not remotely
involving putting an animal to death with electricity. Yeah, thet
(02:37):
the large majority of which so kill an elephant once. Yeah,
I know, you just sticks with you right. Well. As
a boy, Claude Shannon became interested in electronics and began
experimenting with different stuff. He was just curious about how
things work and how to build them himself. He built
a working model of an airplane. Pretty impressive. Think I
(02:58):
think he was born in nineteen sixteen. You didn't have
airplanes for very long. They were pretty new. Yeah, they
were brand new back in the early twentieth century. And
he also reportedly made a working telegraph system that they
set up between his bedroom and a friend's bedroom. His
friend lived half a mile away, and it was all
made out of fencing wire. Yeah, so he could all
(03:19):
but I mean the wire itself. Yeah, he could actually
end up sending messages to his friend have a mile away.
He was also really into radio circuits and built a
radio controlled model boat. Yeah, so very much that. Yeah. Yeah,
this is this is the growing world of radio technology
and the growing world of communications technology. So he was
interested in it as a kid. Now a little bit
(03:43):
later on, when he was a teenager, he got work
as a basic mechanic in a drug store, running a
fix it shop in a drug store, because that's that
was like the center of town. Yeah, where you go
and you go and get your your chocolate malt and
your your your fan fixed. You know, it's a one
stop shop. He attended an Arbor College, where he studied
(04:06):
mathematics and electrical engineering. He graduated an Arbor College in
nineteen thirty six and then went on to enroll in
graduate level study at the Massachusetts Institute of Technology. And
he decided upon m i T because he saw this
work study ad like pinned onto a physical bulletin board
(04:26):
on his college campus that was advertising for someone interested
in working on Vanavar Bush's differential analyzer, which was an
analog computer that used these physical mechanical connections to make calculations.
The deal here was that he would spend half his
time working towards his degree and the other half in
the lab with bush Um, who was then m i
(04:48):
t s vice president and also their dean of engineering.
So this was kind of sort of a big deal Um,
and this machine was huge. It was the system of
gears and pulleys and rods that calculated with an entire
range values that were based on the physical rotation of
the rods, and you could program it by physically rearranging
all of these mechanical bits to correspond with different equations
(05:11):
the control circuit. I mean that this is how early
this was in computing technology. The control circuit itself was
a system of some hundred electromagnetic switches. Yeah. This this
is a kind of the the evolution of what Charles
Babbage created way back in the day, the difference engine.
Uh so we've done the text us done episodes about
(05:34):
and a Lovelace, who was the first computer programmer she built.
She kind of saw that computers could be things that
could do more than just crunch numbers. They could analyze
any kind of data. Yeah, they could represent stuff that
isn't numbers as numbers, so that you could She had
this brilliant idea of, oh, a computer might be able
(05:56):
to represent something like a piece of music and be
able to create, you know, replicated in some way. Years
and years ahead of her time. And the computers of
those days were these giant analog actual machines. Yeah, sometimes manpowered.
Sometimes they had this electro mechanical element to it. So
we're predating the time of the electronic computer at this point,
(06:18):
so uh As Claude Shannon began to work on this machine,
you know now that he had had enrolled with M
I T. He noticed something interesting. He saw that the
switches corresponded with a concept he had started on studying
first as an undergraduate, and that was really focusing on,
which was symbolic logic. Now. I took symbolic logic in college.
(06:41):
I loved it because the basic idea of symbolic logic
is you reduce logical statements to mathematical statements. Actually, I
took a similar class. It was it was basically the
at least mathematical math class I could get away with
as an English major. Well, the neat thing about it
that if you could prove that it mathematically made sense,
(07:04):
then you could say that the statement is true right,
and if it does exactly so, you could you could
start to listen to your friends argue and sketch it out.
And then he said, look, here's where you went wrong.
But at any rate, while he was at M I T.
He started really studying the work of a thinker named
(07:25):
George Boole, who was from the nineteenth century and back
in eighteen fifty four, George Bull published an investigation of
the laws of thought on which are founded the mathematical
theories of logic and probabilities, sometimes known as the laws
of the We usually shorten that to just laws of thought.
So this discussion about the mathematical theories of logic had
(07:48):
Bull using algebraic equations to represent logical forms and syllogisms,
which is exactly what you know I experienced when I
was in college. In this work, he also said that
the only i'd impotent numbers, which are numbers that can
be put through a certain operation multiple times without changing
the result, are zero and one. For example, one times
(08:09):
one equals one, and no matter how many times you
will multiply one by one, it will always be one. Right,
So if you take the product of that of that
that equation and then multiplied by itself, you still stay
with one, same thing with zero, although also with zero
you can add and subtract and still end up with zero.
So zero zero, zero, zero, so bool use zero and
(08:31):
one for the values of the symbols. In his algebraic logic,
he said an argument held in logic if when reduced
to an algebraic equation, it held in common algebra with
the zero one restriction of the possible interpretations of the symbols,
meaning that if you could replace the symbols with a
zero or a one and it's still made sense, it
still worked, then it held true. So Claude Shannon looked
(08:53):
at this and he was thinking, this is a really
cool idea. I love this, this approach to logic. And hey,
you know a switch has two positions on and off,
so sort of like a one in zero. Yeah, I
mean what if we were to, you know, kind of
so play with that, that whole switch process, And that
(09:14):
became something that would percolated in the back of his
head for a while. In fact, it percolated so long
that people suspect that he had fully formed this whole
idea of applying boolean logic to electronic devices for years
before writing it down, and once he wrote it out
and presented it, well, we'll get there. We'll get there.
(09:35):
I also do want to note that around this time
Shannon became interested in juggling, I think originally for like
physical mathematical purposes. He showed up, he started showing up
at the m I T Juggling Club, Juggling Club I
see what you did there, and asking some of its
members if he could like measure their juggling, and and
(09:55):
thereby sort of got involved with them, and this would
be a lifelong in trist As we will get into
a little bit later on a little bit of trivia.
A certain podcaster by the name of Jonathan Strickland was
a founding member of the University of Georgia Juggling Club. So, uh,
that's the only thing I really share in common with
claud I loved symbolic logic and I enjoyed juggling. They're
(10:19):
the comparison ends for he was far more intelligent than
I can ever hope to aspire. But yeah, you have
to agree with no, It's it's fine. I I have
come to grips with it. Okay. If you told me, hey, Jonathan,
you're never going to be as smart as say Claude
Shannon or Albert Einstein, it's alright. Most people won't be,
(10:39):
so I guess. Ninety eight, Claude Shannon writes a thesis
applying Bulls approach to circuitry by equating the zero one
restriction as the off and on positions of a switch
within a circuit. He was twenty two years old. This,
this had never been done. This has never been the
first time anyone had ever said this, certainly out loud,
(11:01):
and other thinkers have said that it would have taken
decades for anyone else to have come to this kind
of conclusion. Right, we could have been sort of groping
around with other approaches for years before someone had come
up with this particular version. And not only did he
come up with this idea, but the way he he
presented it in his thesis, it was very elegant, and
(11:24):
he would he would expand upon it a little bit later,
to the point where people said, this is this is
why he gets compared to Einstein. It's like Einstein saying
not just I figured out this one component to how
the universe works, but being able to express it elegantly
and have a whole picture right. Like it's like it's
not just a fact, it's a hill host of facts
(11:45):
that are all support one another. And it's like they say,
it's it's like you come up with a fundamental theory
of science and unfold it all at once. It's just so.
His thesis also laid out how logical functions such as
and or and not could be implemented within a physical circuit,
so building of logic gates. Now keep in mind this
(12:07):
is all in a hypothetical slash theoretical approach, right, It's
not like he was. He wasn't building this mechanically or
or electronically. That's the case. Maybe exactly, yeah, he was.
He was. He was laying out how this could be possible,
not actually building them himself. Claude Shannon leaves m I
T after earning a doctorate in mathematics to teach for
(12:27):
one year at Princeton Um. And here's the story. Has
a couple of different who has some alternate endings. We
will present you with the two that we know of.
But the story goes that he was teaching at Princeton
and while he was teaching a class he was holding
a lecture. Albert Einstein himself opened the door and stepped inside,
(12:48):
and Claude Shannon kept going on with a lecture, but
obviously was very much impressed with the fact that this
genius has walked into his classroom. He sees I'm Stein
bend over and whisper something to one of the students
in the back. He sees that the student replies and
then he sees that Einstein quietly leaves the room. He
(13:08):
continues on with his lecture. At the end of the lecture,
he holds the student back and with great anticipation, asks
the student, what did this brilliant man have to say
about my lecture? And my version of the story was
that Einstein had very quietly asked the student where are
they currently serving tea? I've heard that he asked where
(13:31):
the men's room was, so it maybe there's where are
they currently allowing you to peet? Could possibly been at
any rate. Apparently that became one of Claude Shannon's favorite stories.
He would love to tell the story about how Albert
Einstein walked into his classroom and asked something completely not
connected with what he had to say, and that made
(13:51):
him like just tickled in it tickled it, And I thought,
well that that also tells you a lot about his
his personality that he did not take himself seriously. Yeah. Uh.
In nineteen forty one, he joined a company famous for
its research and development, Bell Telephone Labs, and his work
mostly focused on things that had to do with the
(14:13):
war effort. In this ninety one is World War two,
and it included anti aircraft devices that could calculate and
target counter missiles, which came pretty seriously in handy during
the German blitz on England. Yeah. Yeah, it turns out
if if your enemy is blasting you with missiles, counter
missiles are a high priority. He also got to work
(14:35):
in cryptography, so here's something where he's got a you know,
a connection with people like Alan Turing who was working
on cracking the Enigma machine back over in England. He
was now Claude Shannon was designed devices used by Allied
powers to send messages back and forth, so he was
looking at keeping Allied messages safe rather than cracking German
messages or access power messages. He later wrote a paper
(14:58):
about communication theory of secrecy systems, which, according to M. I.
T is generally credited with transforming cryptography from an art
to a science. UM it was a mathematical proof that
an encryption scheme called the one time pad or the
Vernon cipher is is unbreakable. And it's the that cipher
(15:19):
is the basic idea of encoding a message with a
random series of digits a key, as we have talked
about on the show before UM which both parties communicating
have a copy of But you know, this is a
very simple concept in cryptography, but having the mathematical proof
that it is in fact unbreakable if the system is,
(15:40):
then that's really awesome. And when we talked about the
Enigma machine, that was one of those systems that could
have been unbreakable had people actually been able to follow
the rules properly. But because there were two things that
really fell apart for the Enigma machine. And I know
this is a bit of a tangent, but it relates
to this. Yeah, those two things were. One, the Enigma
(16:01):
machine was designed so that no matter what the letter
you pressed would never light up as the same the
same letter would never light up as the letter that
you had pressed, So knowing that meant that you could
remove one variable from all the possible outcomes. Secondly, people
were not as careful with their log books, with their
code books as they needed to be um and that
(16:22):
that led to the code being broken. But everyone seems
to agree that had every had the Germans, had the
access powers, been incredibly careful, then that would have been
an unbreakable code. Of course, times of war, you can't
really do share in human error being what it is. Yeah,
I mean it's it's that's the difference between the ideal
(16:44):
and reality. Meanwhile, uh, Claude Shannon began to develop theories
on how to apply his ideas about bully and logic
and circuitry to telephone switching lines. Because of course he's
working at Bell Labs in something else not involved of
in Claude Shannon happened that Bell Labs the development of
the transistor. Now, the transistor was a huge breakthrough. It
(17:08):
meant that the world of electronics could move away from
things like vacuum tubes and allow this other device to
take its place, essentially, which ultimately lead to the manatorization
of electronics. But it wouldn't be until Claude Shannon um
published his concepts about information theory that would let that
(17:30):
be a functional item in the way that it became. Yeah. Yeah,
it was really this idea of digitizing information that Shannon
had that made this a a practical device beyond just
especially that early transistor. It's enormous if you ever see
a picture of it, I think compared to the If
you think that billions of transistors can now fit on
(17:52):
a microprocessor chip, and then you look at the first
one it's it's enormous difference. Obviously. Now, this idea of
digitizing information was pretty much what would allow the transistor
to become useful. And also it's what would lead to
things like encoding information onto storage media like uh, like
a compact disc. This is what would make not just uh,
(18:17):
processing data possible, but storing it. Yeah, and right, it's
it's kind of a really beautiful coincidence that both of
these technologies were being developed at Bell Labs within a
year of each other. As it turns out, because in
that is when claudean and actually published his paper Mathematical
Theory of Communication. Yes, and that's available in PDF form.
(18:39):
Will will share the link because you can actually read
his paper on information theory. And this is the one
that I said earlier that you know, people, people who
were information theory experts, they say like, this is this
is like Einstein coming out with the theories of relativity.
This idea of a complete picture, not just an idea,
but a complete picture of an approach that laid the
(19:01):
groundwork for digitizing information so it can be transmitted and stored. Now, again,
he was a theorist. He did not build this. He
explained how it is mathematically possible, right, and so it
left it up to engineers and computer scientists to figure out, Okay,
if this is theoretically possible, how do we make it real?
(19:22):
What do we do to actually put this stuff into
into reality and have it work for us? Uh? Now
was when it was published, But there are people who
have looked into Claude Shannon's life who say that he
may have had this fully formed as early as ninety three,
and he thought that it was a really cool idea,
but just didn't think, you know, no one else is
(19:43):
going to care about this. I would, I would argue.
I mean, from from what I've read, it sounded to
me more like he kind of had it brewing and
just didn't want to present it until it was done.
He did seem like the kind of person who he
wanted to make sure that he had as complete a
picture of an idea as possible before presenting it to
(20:03):
anyone else. He did not want to have the experience
of coming forward with just half an idea. So yeah,
he's kind of a perfectionist in that sense. And it
really is a challenge to explain just to an average
person exactly how important this theory was, but you know,
in a in a practical sense at the time that
(20:25):
he was coming up with this, it was necessary to
create a better telephone system. So in the old analog
telephone system, you've got some pretty big limitations, some some
barriers you've got to get across due to signal loss
or noise, and analog telephone signal gets weaker the longer
that the telephone line it's traveling along is. Yeah, so
in order to get around that, engineers would place amplifiers
(20:48):
along a telephone line to boost the signal. So you
get a weak signal coming in, it goes through the amplifier,
the signals boosted, it's stronger going out. But unfortunately, um
the along with the signal that you want to get staid,
all of the noise that's on the line also gets boosted.
So eventually you run out I mean, I mean just
the noise takes over. Yeah, Yeah, you lose the signal
in the noise. So that would be you know, if
(21:09):
you've ever heard like one of those those telephone conversations
that goes on in an old movie where it's just
like all you hear is cracked, like yeah, just imagine
that if you're far enough away that all you would
get was the stack. You would not get any voice
at all. So, uh, the interesting thing was that by
switching from analog signals to digital signals, they didn't have
(21:32):
to worry about the signal boosting problem. Instead of a
continuous signal like a sign wave, which is, you know,
an acoustic wave, is what you would get with an
analog telephone line, digital signals are sent in a series
of bits, and a bit is either a zero or
a one. That's all based off of Claude Shannon's application
of Boolean algebra to electronics, and it worked so you
(21:54):
could do this with telephones, which was great, but it
meant you could also do it with just about any
other kind of nation transfer from radio to telegraph, telephones, everything.
And again this was one of those things that could
not immediately be implemented. The engineers had to build the
technology sporting. But once they did, they realized, we can
build out a nationwide telephone, even a global telephone system
(22:18):
that doesn't require amplifiers every x number of miles because
you're never going to lose that that signal clarity, all right,
Like hypothetically, you can do this with literally zero loss
in quality. So so long as you don't mind taking
the necessary amount of time for each bit to be transferred,
really the transfer speed is the only cap that you're
(22:39):
working with at this junction, exactly. And Claude Shannon he
kind of came up with that too. He said, uh,
you know, if if we have an infinite amount of time,
you'll have zero signal laws. But that any medium of
transmission is going to have ultimately a cap of how
much data it can care y at any given within
(23:01):
a given amount of time. So it was interesting because
that was one of those things that ended up becoming
a challenge to engineers. He said, look, for whatever medium
you choose, it's and it's specific to each medium. You're
going to have this limit that you're going to hit
and you can't go beyond it. And the engineer said,
all right, we agree, there's no way we can go
(23:23):
beyond that limit. So what our goal is is to
get as close to that limit as we possibly can.
And and this also led into some really interesting side
concepts about digital compression and error. Yeah exactly, Yeah, you
had to. You could end up compressing data into smaller
data packages, which helps you get around that bandwidth cap
(23:46):
But in order to do that, you also have to
have that that error correction software, that those algorithms that
are able to detect and and fix any errors that
come across while you're transmitting this information. These were all
laid out his ideas, and and that that error correction
concept also ties back into the idea that, uh, you know,
(24:07):
if you scratch a c D, you can still it
can still be read. Yeah, yeah, because you have these
extra bits that are built into the data itself, these
bits that otherwise would seem superfluous. They're not necessary for
you to have the full message, but those extra bits
actually allow some redundancy. So if there is some damage
to the physical medium, you can still end up using it.
(24:30):
And it's not like you get a smudge on your
your your disk and now you can't use it. Right.
So the concept of a disc also being new because
that was something that he laid out in here, saying
that this is a method for possible storage, not just transmission,
but also storage. Yeah, so so big big ideas. Uh.
At any rate, moving on with his life, I mean
(24:51):
he's so he's already gotten to the point where he's
laid out everything that's going to lead to things like
JPEG's m P three's ZIP files. UH, data transmission a
ross cable across telephone lines. All of this stuff is
possible because of the ideas he came up with. His
life continues on and in nineteen forty nine he marries
Mary Elizabeth Moore Betty Betty. She was a new miracle
(25:16):
analyst at Bell Labs, and they would go on to
have two children together. And he also, during his time
off from changing the world UH, decided to build a
simple computer to play chess, and he wrote a paper
about programming computers and computer chess algorithms. A lot of
computer like chess playing computers are still based upon the
(25:38):
foundations that he laid out while he was working on
this UH. You find that the Claude Shannon in his
spirit time often did things that that most of us
would be like, well, you could have a full time
job doing that. He's like, no, I just want you know,
I'd like to keep my hand in. Around that time,
engineers at Bell Labs at that time being ninety nine,
(25:58):
began to actually create the technolog The implemented Shannon's ideas,
and they built something called a regenerative repeater and the
idea was that a bit could be regenerated perfectly and
repeatedly as long as the bits weren't quote unquote too small.
So as long as the messages weren't too small, they
could consistently regenerate a message. Uh and that would mean
(26:21):
that you would again have no signal loss, You wouldn't
lose any data in the process because you could just
just as quickly as it was coming into the regenerative
regenerative repeater, it would send out a copy the same
data message back out again. Um. Also to around this time,
as the engineers at Bell Labs were creating that that
(26:42):
physical technology to incorporate Shannon's ideas, he started to introduce
the idea of bandwidth limits. Yeah, this is what I
was talking about when he said, it doesn't matter what
medium you're using, eventually you're going to hit that capacity.
And eventually they started calling this the Shannon capacity or
Shannon limit. So it was again a very important idea
(27:03):
that ended up being playing a huge role in the
telecommunications industry as well as just electronics and computing in general.
Uh So, this is what gives engineers that goal, This
is where they want to hit as close to that
number as they possibly can to maximize the amount of
data they can shove through any particular medium at top speed. So,
you know, we often talk about data transmission speeds, but
(27:26):
speed is really kind of a deceptive term because it's
not just how fast something gets from point A to
point B. Usually we're talking about speeds that are approaching
the speed of light. That's really fast. What we're what
we're really concerned with is throughput, which is the amount
of data that can travel at that speed to get
from point A to point B. Because if you're dividing
(27:48):
that data up into lots of of bits like a
long string, yes, each individual bit is moving at the
speed of light, but you still got to get that
whole string through. Yeah. Yeah, it's it's the you know,
getting the campus through at the end. Yeah. Yeah, it's
the idea of if the if we hear that there's
pizza in the kitchen, uh, and we're all invited to
go and eat it. Then the problem isn't that we
(28:09):
have a bunch of slow people on staff. We're all
very very fast. The problem is the doors only so wide,
and eventually four or five of us while just try
and cram through it, at the same time. So that's
the difference between just speed and throughput. Now, sept ones
and zeroes don't usually elbow you in the face, that's true,
but we have no such restriction, as we have demonstrated
upon multiple occasions. Now, at this time, engineers were also
(28:33):
trying to find on ways to take on other elements
of this theory, like the compression and redundancy ideas, and
build working devices and algorithms that turned that theory into reality,
actually making products that could take advantage of the ideas
that Shannon had produced. And uh. Meanwhile, Shannon received a
very special present at Christmas of from his wife this year,
(28:58):
a unicycle, and stories say that he frequently rode through
the halls of Bell Labs at night on this unicycle
while juggling. He is my hero because of why not? Now, See,
if my wife gave me a unicycle for Christmas, I
would imagine she was plotting my demise and perhaps had
put taken out yet another life insurance policy on me
because she knows my my lack of balance. But but
(29:23):
I I have nothing but respect for someone who is
transforming information theory while writing a unicycle and juggling juggling. Yeah,
so because because it Meanwhile he was looking into machine
intelligence and memory. Yeah, he was really branching out, you know,
he was he was very much interested in exploring all
(29:44):
these different ideas. Now, by nineteen fifties six, he decides
to leave Bell Labs, though he continues on as a
consultant and he goes back to M I. T. To teach.
He also wrote a paper he was called the Bandwagon,
and uh, that's when he said he didn't really like
how the words information theory were being thrown around. So
(30:06):
essentially what he was saying was that they were losing
their value. Information theory as a concept was losing its
value because companies were using it to describe things that
didn't really fall within the umbrella of information. Yeah. It
was a really popular and pop culture almost term in
the scientific community at the time. And I mean people
were publishing papers that had information theory and the title
(30:27):
just because they thought it sounded cool, when in fact, right,
it had nothing to do with that. So it was
kind of like how virtual reality became this buzzword that
began to lose meaning, particularly when the public started to
see what the reality of the field was as compared
to the Hollywood depiction of what virtual reality was back
in the early nineties. Sure, sure like artificial intelligence or
(30:51):
I read an essay recently from the guy who coined
the term manic Pixie dreamgirls saying that he just wished
he had never done that thing. I would like to
apologize to the world. Yeah. So this was one of
those interesting things were the paper wasn't so much about
advancing the concept, but just saying, let's use our words
carefully and correctly. He said that perhaps the term had
(31:13):
quote ballooned to an importance beyond its actual accomplishments end quote.
I think that's a little bit modest on his part, Honestly,
I think so too, considering that again, without that theory,
computers and electronics would not work the way they do today. Yeah,
but at any rate, this kind of marked the beginning
of Shannon's disappearance from the research and technology scene. He
(31:36):
he really didn't want to be a celebrity, I think,
and he had this huge push from the media and
the government and science in general to be made into one,
and it it kind of pulled him away from from
both research and public education, right and he was it
wasn't that he was cold from why, I understand whenever
he gave talks they were really great, and whenever he
(31:58):
wrote papers they were really great. He was constantly being
pressured to do that, and it was starting to become
more of something that would cause him anxiety as opposed
to something that he would enjoy doing well. In nineteen
seventy three, the Information Theory Society, which is part of
the I Triple E or I, created an annual Shannon
(32:18):
lecture that became the Shannon Award UH And in nineteen
seventy eight, Claude Shannon officially retired from m T, although
he had not really been actively working there for some
years before. Certainly UH and in nineteen eight seven, Claude
Shannon gave his last interview to Omni Magazine. Now, by
the late eighties, Claude Shannon began to suffer from Alzheimer's
(32:40):
and withdrew from the public eye entirely. His wife would
go and attend events instead in his place, and in
February two thousand one, at the age of eighty four,
he would pass away. Yes, there are some very UH
inspiring and moving tributes to Claude Shannon that were published.
Really beautiful things. You can certainly go online and read
(33:03):
a lot of those those tributes that were written the
week and month following his passing. And we have a
collection of interesting little trivia that we didn't really want
to fit into the overall episode. But it didn't really
fit into the timeline. But so much of I mean,
if it wasn't charming enough, I mean, if charming is
(33:23):
the correct word, actually charming is totally the correct word.
According to me, I find it downright charming that he wrote,
you know, papers that mathematically proved the computers can exist.
But but but but other than that, there's just a
lot of little just so. So one of those things
is that, you know, we just said he he was
(33:44):
not big on on pursuing the limelight. He didn't he
didn't go after that at all, and and often he
would reluctantly take the stage, but as time went on,
he did that even less frequently. He wouldn't go out
very much at all to address the public, and according
to M I. T. Technology Review, he even had a
(34:05):
file labeled letters I've procrastinated too long on So if
he got something from colleagues or government officials or scientific
institutions and had just been sitting around for a really
long while. He would just put this in a file, saying, well,
that's too that's too late, and that's never gonna happen.
So I'm just gonna put that in this file. Um. He,
(34:26):
like we said, love to build stuff, to engineer stuff.
You know that whole telegraph line stories one of my favorites. Um. Now,
as a parent, he built a chairlift that would take
his kids from his house to a nearby lake so
they didn't have to walk the whole way to the lake.
He also, from what I understand, designed a hidden panel
in his office that didn't lead anywhere at all. He
(34:48):
just he just felt like building one. He just needed it.
It made me think of a Mitchell and Web sketch
where this wall must rotate both here and not here. Look, Mite,
that's a load bearing wool. But anyway, he just decided
he wanted to make one. He also built a life
sized electric mouse named Theseus, after the Greek mythology figure
(35:12):
that's the one who was stuck in the labyrinth that
had to find his way out in the minotaur or minotar,
depending upon your preferred pronunciations, after him. So this mouse,
what it would do is it would explore a maze
and quote unquote remember where it comes from. It was
it was going after some little metal cheese bits. I think.
So the the way this mouse would go through the
(35:33):
maze is it would go down a pathway and whenever
the pathway would branch, it would start to rotate. Yeah,
so it would take one and then it would, uh,
it could backtrack if it went down an incorrect route, right,
and then it could take the path it had not
taken as opposed to you know, if this were just
an electronic mouse that had some collision detection, it wouldn't
(35:53):
It could potentially just go back and forth down the
same little pathway forever. Yeah, but this was branching that
this one knew. Okay, well I already took the path
that's on the right, so I have to take the
path that's on the left. So it's pretty cool that
he built this thing, you know, just for the fun
of it. He built it also probably my my favorite
(36:14):
robotic piece of his eight juggling robot, a bounce juggling
robot to be precise, bounce juggling robot that like w
C Fields to be even more precise. Yeah, it was
like having a like, imagine a drumhead, right, and the
drumhead allows things that are dropped on it, like a
ball bearing to be bounced on it. And then two
(36:34):
little uh angled platforms that are serving his hands that
are bouncing this again, these little these balls. Yeah, and
it just kept it going in a in a bounced
juggling pattern perfectly. And he basically made it out of
like erector set pieces. Yeah, you know, just like you do.
And then he wrote a paper on the dynamics of
keeping multiple objects in the air simultaneously. It's pretty famous
(36:55):
within the juggling community. I tried to read it what
I actually wrote, how Juggling works for how stuff works
dot com. In fact, if you go to that that
article on how stuff works and you look up how
juggling works, there's a video of me juggling in that article.
I still I still say it because I juggle a
little bit. I still say that we really need to
do a video of all Right, I juggled torches in mine.
(37:17):
You're ready to pick those up? Okay, well, well we'll
start small. Uh. He also made a robot that could
solve a Rubic's cube, which is pretty amazing. I mean,
obviously that needs I can't either. I know there are
algorithms for how to solve it the most efficiently, and
I've seen people who are really good at who just
like it's like it's like magic. You know. The way
(37:40):
I saw a Rubik's cube is by peeling the stickers
off and then replacing them properly. I cheat, but yeah, no.
He he created a robot that could follow these algorithms
and also just recognize what the pattern was on any
given side, so it could, you know, create the rules
that needed to solve it. UM and he made a
calculator that worked with Roman numerals. It was called throwback,
(38:02):
which stood for a thrifty Roman numerical backward looking computer.
UM also rocket powered Frisbees and motorized poco sticks. Yes,
the motorized pogo stick. I was thinking, like, again, that
sounds terrible. If the unicycle hadn't killed me already, that
certainly would. He built the ultimate machine. My favorite machine
(38:23):
of all time is the ultimate machine. All right, tell
us about it, Jonathan. All right. Now, imagine you have
before you a box, and on that box you can
see the outline of a trap door, and the only
other really interesting feature on this box is a simple
switch has switched to off, and you push the switch
to on. The trap door opens and a hand emerges
(38:46):
from beneath the trap door and hits the switch back
to the off position, with draws back inside and trap
door closes. That's it. That's it. You hit the switch
and the harm comes back out yet the switch, the
arm comes back out. Uh. I want to share this
video too. There's a video of a brilliant variation of
the Ultimate Machine that is hysterically funny. It doesn't just
(39:10):
do that like, it starts to do it so um.
It ends up at first looking like it's a variation
on the Ultimate Machine, like oh, that's cute, But then
it starts doing other things too, because this particular box
had wheels on it and can move autll the way,
so it's starting to avoid the person who's trying to
hit the switch, or it would playback prerecorded messages saying
like hey, hands off, buddy, that kind of stuff and
(39:32):
was really really entertaining. So we'll share that one as well.
But you have to remember that that particular very entertaining
machine is based off this thing that Claude Shannon built
for no reason other than it tickled him just because
he could. Um. He also had a collection of exotic unicycles,
including some that were because he he was wondering, how
small could you make a unicycle before someone would be
(39:53):
unable to write it? Uh? For me, that's any size.
But but I think me too, that would be any size.
Assuming that you are capable of writing a unicycle, how
small could you go before you could no longer maintain
your balance? In fact, he had a couple that I've
heard are essentially unwriteable. Uh. He also lectured on using
information theory as an application to playing the stock market,
(40:16):
though he never really published any work on this. He
did do a lecture, but he didn't write a paper.
He also did really well in the stock market himself,
although he wasn't necessarily employing information theory to do so.
He was investing in companies that friends of his. Yeah,
he made some very savvy stock purchases based on amazing
work that his friends were doing. These are These are
(40:38):
the people who were inventing like the basic components of
computers and electronics, going on to form their own companies,
and he would invest in those and then they ended
up being these these enormous companies we know today. So
he did quite well. Um, and there's no Nobel Prize
for mathematics, which is why Claude Shanna never won one, right,
but he certainly did win a number, I mean, probably
(41:01):
way too numerous to mention here awards, but but one
that we wanted to mention it is the very first
Kyoto Prize, which was created in Japan to award honors
to contributions in mathematics. Essentially, it was supposed to be
the Nobel Prize for mathematics, right, and this was all
the way in the nineteen eighties, and this came into invention. Yeah,
the very first one went to Claude Shannon, and from
(41:21):
what I understand, it actually came with an even larger
cash prize than the Nobel Prize does. So so if
you if you feel like he was he was snubbed
because Nobel Prizes don't recognize mathematics, fear not, the Kyoto
Prize had him covered. So I hope you guys, uh,
if you had not ever heard of Claude Shannon before,
I hope you learned something in this episode, because he
(41:43):
really did seem to be a remarkable person. In multiple ways.
I mean, this guy seems like the kind of professor
I would have absolutely adored um. But then you know,
I like all of my professors who had lots of personality,
and we're unafraid of coming across as a little unusual
or yeah those are my favorite. Yeah, so awesome. And
(42:05):
I hope if there are any other really important figures
in technology that you would love to hear us cover,
you should let us know. Send us a message you
can in says an email that addresses tech stuff at
how stuff works dot com, or drop us a line
on Tumbler, Twitter or Facebook, or handle it. All three
is text stuff hs W, and we will talk to
(42:26):
you again really soon for more on this and thousands
of other topics. Because it has to works dot Com
TechStuff News
Hosts And Creators
Oz Woloshyn
Karah Preiss
Popular Podcasts
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by audiochuck Media Company.
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.