Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Episode #106: Jimmy Soni and Rob Goodman
176 Plays7 years ago
On this week’s episode of the podcast, I welcome two special guests, Jimmy Soni and Rob Goodman, authors of the new biography of Claude Shannon, the father of information theory. Jimmy Soni has served as an editor at the New...
The post Episode #106: Jimmy Soni and Rob Goodman appeared first on PolicyViz.
Recommended
Transcript
Introduction and Guest Overview
00:00:11
Speaker
Welcome back to the Policy This podcast. I'm your host, John Schwabisch. A very exciting pair of guests on this episode. As you may know, I am a big fan of John Gertner's book, The Bell Labs, a history of the AT&T Bell Labs in New Jersey. One of my favorite books really helped me start thinking about information technology and the way we communicate now.
Claude Shannon's Impact on Information Theory
00:00:32
Speaker
One of the central figures in that book was Claude Shannon.
00:00:35
Speaker
whose theory of information was sort of groundbreaking work for all the things that we think about and use to communicate today. And now we have, I believe, the first biography of Claude Shannon, a new book by Jimmy Sony and Rob Goodman, a mind at play how Claude Shannon invented the information age. And so I'm very excited to have both of them here on the podcast with me to chat about the book and to talk more about Claude Shannon and his theory. So Jimmy, Rob, welcome to the show. Thanks for coming on.
00:01:02
Speaker
Thank you so much for having us. Thanks a lot. It's great to be here. I'm really excited to talk to you guys about the book. It's a great biography and walks all the way through all the work that Shannon did. But before we get into the book itself, can I ask you guys maybe just to introduce yourselves and talk about your background and how you guys came together to write this
Writing Claude Shannon's Biography
00:01:20
Speaker
book? So maybe, Jimmy, we'll start with you.
00:01:22
Speaker
Sure. So my name is Jimmy Sony. I'm an author. I live in New York at the moment. I've known Rob, oh gosh, since our days as college students. So going on, you know, over 10 years now. And we were, funny enough, debate partners at Duke and then became co-authors on an earlier book, biography of an ancient Roman senator named Cato the Younger. And our latest book is the biography of
00:01:47
Speaker
of a 20th century mathematician named Claude Shannon, so we moved forward in time and moved around a little subject, but that's just a little bit about me. Yeah, and speaking for myself, I'm currently a PhD candidate at Columbia University. My day job is as a political theorist, so it's been nice to have a break from politics.
00:02:08
Speaker
to turn to this work on Claude Shannon. It's been really good to dive deep into a subject that is not necessarily related to my main field, but it has so much that's fascinating in its own right.
Shannon's Role at Bell Labs and Influence on Communication
00:02:21
Speaker
So can you guys talk a little bit about how you decided, hey Claude Shannon should be our next book. You've done something on ancient civilization. Where did this idea to do a book on Claude Shannon come from?
00:02:32
Speaker
Yeah you know group pretty organically i like you was really taken by the idea factory i thought it was just a wonderful book and in the idea factory which is a narrative history about labs which many regardless the most innovative twentieth century company in the united states and possibly in the world.
00:02:50
Speaker
Claude Shannon is a central figure. And I just assumed, perhaps naively, that someone had already written a biography of Claude Shannon. And so I went on Amazon to buy it, assuming there would be one. And then I couldn't find it. And so I decided to talk to Rob. And we sat down and talked about Claude Shannon and kept doing some digging. And then finally, one thing led to another. And we sort of embarked on the journey of writing his life.
00:03:16
Speaker
And so Shannon passed away, what, 25 years ago. So how did you start thinking about how are we going to write a biography about this figure who is fundamental to basically
Shannon's Achievements and Personality
00:03:28
Speaker
how we live and communicate today? Did you have a strategy in mind to write a biography this way? I'm sorry to jump in. I think he passed away in 2001.
00:03:36
Speaker
2001 so okay, so really recently so I mean you have Relatives you have many sections in the book where you've spoken to his daughter How do you think about planning out a biography about a figure where you can talk to some of his his peers and colleague? Of course, this is a totally different challenge than the work on the Akato book, right? The eyewitnesses have been dead for 2,000 years and can't call us out on it But in this book, I think we had to do two particular things or one we had to get a decent handle on
00:04:01
Speaker
on the technical aspects and what it was that Claude Chan did that was so revolutionary in the history of communication. What's the histories and the ideas that he came up with and why is he so revered within the field for being a pioneer of what it takes for us to communicate these days. The other thing, as you said, we needed to get a sense of who he was as a person, as a colleague, as a father, as a husband, and so on. So I think we sort of pursued these two things in tandem. We did a lot of archival research. We did a lot of
00:04:28
Speaker
Learning about the history of information as it were but also the interviews were just invaluable to get a sense of you know what a lovable decent if sometimes a standoffish person he was and finding.
00:04:43
Speaker
my favorite part i think was finding ways in which his personality found its way into his work i did the idea of curiosity and. Playfulness and questioning of and all these things that people identified in clodshan's personality shape the way he approached problems as well so whenever could draw those links about the kind of person would ask the kind of questions that clodshan asked. Those were moments of which i think we're doing our best and making the best contribution we could to understand
Legacy of Shannon's 1948 Paper
00:05:10
Speaker
his life.
00:05:10
Speaker
I want to get back to his personality and his curiosity, which comes in throughout the book, but can you give us a quick primer on Shannon's main work, the Theory of Information, for those who may not be familiar with it? Yeah, just to get the brief kind of layman's rundown summary.
00:05:28
Speaker
Elevator pitch version I like to give is that Shannon is essentially the person who events the bit. The bit is the fundamental measure of information. People have been groping towards a kind of objective way of understanding what it means to say there's X amount of information in a thing like a telegraph or another kind of message or a television broadcast or a letter or any kind of medium. Shannon is the one who comes along and explains how we can quantify this by thinking about information probabilistically.
00:05:56
Speaker
What I mean by that is he starts us thinking about the fact that messages don't just happen at random. Whenever we send a message because communication is all about following rules to make ourselves understood, every subsequent symbol we send, whether it's a word or a letter or an image or whatever it is, is related in probabilistic ways to the ones that we've already sent. And measuring these probabilities helps us understand how to quantify messages. But the more practical upshot of this is that
00:06:26
Speaker
Once we understand these probabilities, once we understand that a lot of the bits of information we send are redundant, they can kind of be done without, we can do things like compress messages and send them with a lot less bandwidth, or we can do things like
00:06:41
Speaker
protect our messages from noise and distortion by adding in redundancy that acts as a sort of shield against distortion. So the big practical upshot that Shannon proves in his Great Information Theory paper in 1948 is that it's essentially possible to send a message from point A to point B if you code it right with essentially perfect accuracy.
00:07:02
Speaker
And no one believe this was possible before shannon came along and prove that must be the case and this insight is essentially the inside behind the internet and all sorts of other high tech high volume communication systems that we depend on every day because she showed how you can use a coding. I take advantage of the probabilistic nature of our messages to make sure we send them accurately without accurate transmission you don't have the internet you have cloud storage you have.
00:07:27
Speaker
cell phones, you know, texting, you don't have anything that we rely on. All these things are derived from Shannon's insights into bits. So when Shannon's paper comes out in 1948, do people immediately recognize the genius and the importance of it and how it's going to fundamentally impact technology? I think they recognize the impact of the paper, but as a piece of theoretical thought, nobody could see, you know, in 1948
00:07:57
Speaker
the kind of world we live in in, you know, 2017 to 2018. But they could see that the way that Shannon had thought about information was a radical break from how people have thought about information up to that point. And so within a year, for example, the best evidence that we have for this is within a year.
00:08:17
Speaker
A paper that Shannon somewhat modestly titled a mathematical theory of communication when it is published in book form a year later is retitled the mathematical theory of communication, but more than that scholars in the field within that period and just after.
00:08:34
Speaker
Take his paper as the jumping off point for other papers. So they write papers derived from it. They analyze sections of it. They go through it
Exploring Shannon's Broad Interests
00:08:42
Speaker
very closely and that continues to this day. It's over 90,000 citations of that paper at our last count and I think even more now.
00:08:50
Speaker
But to answer your question the technology that would connect to that information theory wasn't quite as clear because he was speaking about something that actually rose beyond technology that this conversation over Skype that the web page you loaded and that the phone call you made you know are all the same that they can all be reduced to bits and that they're all they all the same essential ingredients.
00:09:10
Speaker
on the other hand you know theoretically the paper made a big splash uh... and it it uh... it was in it was important at the time continues to be important today i want to talk about shannon the man because uh... what comes through in the book is his curiosity obviously his genius but also his curiosity uh... there's a sentence in in the book towards the end that i just love you both right his style of work was characterized by such lightness and levity in fact that we can sometimes forget the depth and difficulty of the problems he took on
00:09:39
Speaker
Can you give folks a sense of what he was like as a person and what you think his keys to success were? Yeah. Well, as a person, he was someone who pursued questions about the nature of information for the same reason that he pursued questions about the nature of juggling or the nature of how to pick stocks on a stock market or about how to game roulette wheels in Vegas. He did these things just because they were interesting to him.
00:10:06
Speaker
because he was a sort of person who love to take her to solve problems to write codes to invent just crazy outlandish gadgets that he had a special motorized line would take him down from the house to the lake shore nearest house he carved a tree into the shape of a pirate flagpole also has property he invented a fleet of customized unicycles in the calculator that worked in roman numerals.
00:10:29
Speaker
He invented an electronic maze solving mouse. He was just the sort of person who did these things for giggles, which didn't mean that he was always a fun, jolly person. I think people who weren't up to his level intellectually would find that he was a sort of withdrawn, sometimes standoffish sort of personality.
00:10:47
Speaker
If there's something that tickled his fancy and if he was intellectually engaged, people describe him just coming alive in a way that was just purely, just with a pure kind of absence of ego and just totally absorbed in the problems he was working on, totally absorbed in the puzzles that he was trying to solve with his colleagues.
00:11:05
Speaker
So I think it's hard to specify exactly what the secret to Claus Jan's success is, but I think part of it has to do with that idea of combining remarkable intellectual gifts with the lack of self-importance that kept him asking some silly questions, some odd questions that turned out to have remarkable results. He actually gave a speech in the early 1950s to his Bell lab colleagues about technical problem solving and about the kind of personality you need to solve problems. And one phrase that I really liked
00:11:34
Speaker
from that speech was the idea of cultivating a sort of what he called creative dissatisfaction or constructive dissatisfaction the idea that a great engineer or great scientist.
00:11:46
Speaker
knows when there's a problem and knows when something isn't quite right and knows how to poke around outside and inside that problem to figure out why it may or may not be interesting that something isn't quite right there. And the idea to be able to look at things and ask, why not this and why not that? That idea of cultivating that dissatisfaction, not in a kind of grumpy cynical way, but in a
00:12:06
Speaker
playful, curious way. I think that's a big part of what we can learn from Shannon. Obviously, we can't learn to copy the remarkable intellectual endowments that he was lucky enough to have that they were part of his gifts, but I think we can learn a little bit more from his disposition of what he did with those gifts.
00:12:23
Speaker
Yeah, he had interests, as you mentioned, you know, that spanned a lot of different areas, a lot of different disciplines. Let me put it this way. I've done over 100 podcast interviews, and basically every person that comes on the show that's working in the fields of data or data visualization comes to it from a different background.
00:12:39
Speaker
I don't think I've had people come to it from the same place. So that wide interest, being able to think about different areas and different disciplines, is that important to success today? I mean, it's clearly something that Shannon had. I guess I'm just wondering whether that broad Renaissance perspective on the world is something that
00:12:59
Speaker
you think is important to success and something that people should cultivate in an era where things are increasingly technical. You can't really take your car to the garage anymore in the corner because it's essentially a computer on wheels. Well, I think the risk of giving a broad answer to your question would say, yes. I would say broadly, yes. But to bring it back to Shannon's life, I think one of the lessons of Shannon's life is that it's okay to
00:13:28
Speaker
dabble, that it's okay to have promiscuous curiosity, that it's okay to pick up a project and finish it to your satisfaction and then set it aside and know that some lingering element of that project is going to find its way into other work you do. I think there's a real premium today on focus and that's not altogether a bad thing.
00:13:48
Speaker
There's a real premium on seeing projects through to their ultimate conclusion, whether that's making sure to finish a book that you're midway through or whether that's taking the company public. I think those are all good things. We want people to focus. We want people to finish things, but Shannon's Life does offer a few powerful counter arguments.
00:14:08
Speaker
He picks up something like stock picking, and he doesn't necessarily want to design a hedge fund. He just becomes interested in it and writes papers about it and gets other people interested in it. He picks up wanting to build a wearable device and builds one to try to play the tables at Vegas, and he doesn't want to then go and commercialize the product. He tries it out. He and Thor build it together. It works. They have some fun, and it is set aside, and now, I believe, is in a museum.
00:14:33
Speaker
And they also get worried about the mafia, right? I love that part of the story. They kind of assume that two somewhat bookish mathematicians are not going to be able to weasel their way out of talking to the mob. And so, again, you can argue for or against that style of work. I think what you cannot argue
00:14:53
Speaker
is the significance of Claude Shannon's work and his body of work. And I do think that part of the reason it was significant is because he didn't have one input. He wasn't just an engineer. He was an engineer and a mathematician who also studied logic, which is what led to one of his earliest breakthroughs. He was someone who played around with codes when he was a boy and then as an adult participated in the war effort by breaking codes and doing cryptography. And so that cryptography finds the way in information theory.
00:15:20
Speaker
And I think in an era in which we, again, justifiably worry about distraction, Claude Shannon's life is just a little bit of a kind of pump the brakes. It says it's okay to dabble.
Environment at Bell Labs and MIT
00:15:30
Speaker
And there's a difference, I think, between dabbling and being distracted.
00:15:34
Speaker
What about Shannon's work with others? I mean, starting his professional career at the Bell Labs, working at MIT, was there something about the teams and the people that he was surrounded by that contributes to that and that sense of dabbling? I mean, I think that's part of the message of Gertner's book is this team of brilliant scientists and engineers working for a phone company, but basically being able to work on anything they wanted to work on.
00:16:00
Speaker
Yeah i think that's certainly true i think an important part of shannon's story that i hope we tried to tell is the notion of the networks and the communities in which is embedded that help make what he did possible. Even starting at mit a big influence on his later career was the neva bush who is his first mentor as a graduate supervisor.
00:16:20
Speaker
who was a major proponent of what Jimmy called dabbling of being really ecumenical in one's scientific interests. He had a talk that we found in a biography of him discussing the need for what he called grad students in the sciences to get out of what he called their modern monastic cells or their little disciplines of little interest. So he was always pushing Shannon to take his talent in a direction that might not have seemed natural to him. And a great example of this is early in Shannon's life,
00:16:50
Speaker
After he completed his famous study of switching and binary coding that is one of the bases of the modern computer, which should have been enough for most advisors, Bush asked him to go off and do his PhD research on theoretical genetics. Shannon had never been a biologist, never had any experience in genetics. He said when he took up the problem, he didn't even know what the words meant, and he had to kind of educate himself from a kind of basic undergraduate level to even get up to the level at which he could write a dissertation. But he did it.
00:17:20
Speaker
Even though he didn't publish in the field, I think what he got from those early experiences was the value of getting outside the comfort zone, the value of being an amateur in a lot of fields at the same time, and what might come of it. Later on at Bell Labs, he's in an entire environment that encourages that exact mindset.
00:17:39
Speaker
Bell Labs was extremely fortunate in that it benefited from a federal monopoly on telecommunications, which meant that it had a lot of money to throw around at R&D and hiring people like Shannon who were on the payroll to do basic research, research that might not pay off for decades. But there are a number of comments that we got both in other works and in our own interviews.
00:18:01
Speaker
from people at Bell Labs talking about what a great environment this was for doing original research. One of Shannon's colleagues we spoke to said that there's an attitude at Bell Labs that it doesn't matter if what you're doing pays off in 10 or 20 to 30 years, we'll still be around then. Shannon himself said that he doesn't remember outside of his time when he was sort of conscripted as part of the war effort. Outside of that, he doesn't really ever remember at Bell Labs being told what to work on.
00:18:30
Speaker
Some people sank in this environment, some people couldn't handle that degree of freedom, but other people swam tremendously well. Shannon was one of them who was able to step back and think not about how do I improve this particular telegraph system or how do I improve this particular long distance phone line system, but to think about what do we mean when we talk about information at all? How do you actually step back and measure it?
Shannon's Humility and Intellectual Honesty
00:18:55
Speaker
These things didn't actually pay technical dividends, practical dividends, for a number of decades. It was intellectually a bomb, as one of Shannon's colleagues called it. But practically speaking, it was a dud, at least for a couple of decades, at which point it started to revolutionize the world. But I don't think there are a lot of other companies then and certainly now that would have given someone leeway to do basic research on that scale.
00:19:18
Speaker
Is there interest that Shannon had that each of you identified with or found the most endearing? You know, I did think that among other things, the juggling was actually one of the more endearing parts for a couple of reasons. One is that we know of Shannon the Juggler, the origin story for the more serious interest in juggling goes to his
00:19:42
Speaker
daughter who read about the MIT Juggling Club in a newspaper and asked her father to take her. And as a new father of a daughter myself, I could just see in the future being dragged to things that I wasn't necessarily keen on, but then be developing a really serious interest in it. Now granted, my daughter is two, so that hasn't happened yet. The keen interest right now is the Watermelon Seed, which is a book about an alligator who eats a watermelon seed. But I could sit for sea in the future
00:20:11
Speaker
being taken in different directions that I wasn't necessarily interested in. The other thing that's endearing with the juggling story is just his interest in it again goes beyond what an average person's curiosity would be. He invites a juggling club over to his house. He measures jugglers and what they do. He becomes a juggler himself and tries to sort of get past four balls, which is a pretty high target to hit and he's not able to hit it and makes him upset. He's also one of the first people to write a serious mathematical paper about juggling.
00:20:41
Speaker
and it shows off just the range of things that he reads because it has citations on everything from Socrates.
00:20:48
Speaker
uh... to scientific to science fiction novels and so part of what the juggling illustrates is this point that we've been making throughout that became naturally and it was a real calling card to have a lot of interest in to be reading very widely and then to distill those wide pools of knowledge and learning into papers like struggling paper so to me that's that's one of the stories and that i'll say the other it's not an interest but it's a story that's pretty endearing
00:21:12
Speaker
is the moment when the shannons are on their back porch and they're hosting kind of a barbecue or an event. And Shannon's there and his daughter is there and his daughter Peggy is in charge of the toothpicks. And she runs up the stairs, this sort of porch or patio that they have. And while running, she trips and the toothpicks, they fall into a big pile in front of Shannon, the guests and Peggy. And you know, Peggy's embarrassed and she looks at her father and her father kind of stands there for a second and he looks at the pile of toothpicks
00:21:40
Speaker
And he says, did you know that you could estimate pi with that? And there's a gather, a formula, or a calculation that allows you to take a randomly dumped set of needles of that kind and calculate pi. And what Peggy remembers most is both the kind of absurdity of that moment, but then also the fact that her dad wasn't angry. And I found that to be endearing. Yeah. Rob, what about you?
00:22:03
Speaker
One little moment that I respect as an academic was when Claude Shannon went to MIT later in his career after Bell Labs and became a professor. He was sort of revered as the center of the electrical engineering community there.
00:22:18
Speaker
Um, but, uh, you know, he's never especially keen on supervising lots of students doing sort of the grunt work that a less renowned professor was going to have to do. Um, I think at some point, uh, someone asked him why he didn't advise more students and he said, Oh, I couldn't possibly, I don't feel entitled to advise anyone. I don't have any advice to give, which was a great way. Uh, people like me are always looking for ways to get out of our, uh, our, our less exciting commitments. And I think that I have to remember that line one day to see if I can get away with deploying it. Yeah.
00:22:48
Speaker
There was another great quote about and we put this in the photo section with a great photo of a grad student, I think, shot from behind. We couldn't tell it was discussing with Shannon across his desk. There was a quote from one of Shannon's other students who said, you know, in those days you had to have quite an ego to even bother asking Shannon to advise you.
00:23:07
Speaker
So I think that was probably another thing that protected him from a lot of the work, simply because not that he was unapproachable, but that you had to think very highly of yourself to even bother approaching him in the first place was an idea. But what's interesting about that is what comes through throughout the book is his modesty. And you write towards the end of his career, and he's essentially traveling the globe receiving awards and accolades.
00:23:31
Speaker
He really didn't like to do that. So that modesty really comes through. So it's interesting in this sort of tension, as it were, between people feeling not up to asking him for things at the same time. He seemed like a very modest human being. He very much was. And part of that was, you know, there are a number of sources of that. One was, I think, just a kind of natural inclination to avoid the limelight, to focus on the work itself, to focus on the research.
00:24:01
Speaker
In the 1950s, he basically has the option to be famous and publicly really well known. The figure of the scientist is reaching a kind of cultural height. So he's got the chance to get in on that, and he's included on lists of the most important American scientists and all that. And then just as soon as that happens, he basically kind of backs out and goes and pursues his own research and work. And so I do think modesty was a part of what he did. The other moment that this happens is when information theory picks up and is taking off,
00:24:32
Speaker
he's recognizing something which is that people are treating this theory that he wrote for engineers to solve engineering problems as that era is like solution to everything, right? And so people put it on par with the theory of relativity and he's not as willing to go out and trumpet this theory. Now, that's an incredible thing unto itself because he's a pretty young man still. He's got a lot of self-interest in making this the theory of everything. I mean, he could have been on the lecture circuit and been on TV and all the rest. He could have been like Einstein.
00:25:01
Speaker
But he's willing to be intellectually honest and rigorous enough, and I also think modest enough, to write a quick 300-plus word editorial called The Bandwagon, where he essentially tells the rest of the world, like, hey, cool it. This is for engineers doing engineering. Please don't pretend that it's bigger than it is. And the thing for us to do right now is to take this work and to continue to advance it. So that was a modest
Conclusion and Book Promotion
00:25:25
Speaker
moment. I think it was also an intellectually honest moment.
00:25:29
Speaker
Well, it's a great book. I really enjoyed getting to know both the theory and his work and, of course, the man himself. So, Jimmy, Rob, thanks so much for coming on the show and thanks for writing a great book. Of course, thanks so much for having us. It's really good to hear that you enjoyed it. Thank you so much, John.
00:25:43
Speaker
And thanks everyone for tuning into this week's episode. I hope you enjoyed it. I hope you will check out the new book by Jimmy Sony and Rob Goodman, A Mind at Play, How Claude Shannon Invented the Information Age. You should also check out John Gertner's book on the Bell Labs. I'll put links to both books on the show notes page. So feel free to reach out, let me know what you think of the book. That's all we have for this week. So thanks for tuning into the Policy of This podcast, and I'll see you next week.