Show Notes
Tony Barr has invented several computer programming languages, including the Statistical Analysis System, better known as SAS. He describes the concept of language, as “a mental model to think about the world.” Tony believes he has found the “concept of a concept” that will enable his new language, AMORE, to “redefine computer science.” As a child, he “lived inside his head,” and “never got any support from teachers.” Tony’s mother was “a brilliant person” and his grandfather held a patent on making ice cubes. His words of wisdom, “share the credit, but also take the credit.”
TRANSCRIPT:
Intro: 0:01
Inventors and their inventions. Welcome to Radio Cade a podcast from the Cade Museum for Creativity and Invention in Gainesville, Florida. The museum is named after James Robert Cade, who invented Gatorade in 1965. My name is Richard Miles. We'll introduce you to inventors and the things that motivate them, we'll learn about their personal stories, how their inventions work and how their ideas get from the laboratory to the marketplace.
Richard Miles: 0:37
Amore in Italian, it means love. And it's a title of many, a song it's also an acronym for a model of reality and you computer programming language. I'm your host Richard Miles today on Radio Cade. My guest is Tony Barr, a distinguished inventor of other programming languages, such as the Statistical Analysis Systems, SAS, and Acme, the automated classification of medical entities. He holds four U.S. patents and is considered one of the world's experts on programming languages. Welcome to the show.
Tony Barr: 1:05
Thank you.
Richard Miles: 1:06
So if we talk about everything that you know, and every technology you've developed, this would probably be a six to eight hour podcast series, and we'd have to order in lunch and sleeping bags and everything else. So we're going to start out at the ground, and that is, let's start about the concept of language itself, because you have developed programming languages for computers, but really programming languages , just another language, right? So why don't we start at the conceptual level about the function of language, and then we'll move from there to how you have interpreted those functions in your various inventions from SAS on up to a model of reality. Language, what is it for? What do we use it for? Why is it important? How do you think about language?
Tony Barr: 1:45
Language is the way we represent knowledge, the way we communicate knowledge, I actually go one step further beyond language. I think we have mental models in our mind that enable us to think about the world. We build models of organizations called enterprise models that drive the world. The government's got the biggest model in the world. Every business has a model. The museum has got its model for its customers and its exhibits and who their vendors are. We build models to think about the world, to understand the world. Language is a medium to communicate models. And it's my thesis. That the way we represent things in our mind is very similar to the way we represent things in the computer. And my hope is to unify the way we think about the mind, the way we think about the computer. So we have a uniform model, a simpler model for the world to live in. Wouldn't it be wonderful if you could, at the age of 12 learn model of reality. Inside it would be mathematics and logic. It would also be the way the computers work. It could give you the insight is the way you separate the concepts from the objects in the world. And this would give us a uniform framework to about the world. So my longterm plan, which I won't live to see is that we all live inside a model of reality, which we share.
Richard Miles: 3:07
So when we talk about a model of reality, it's relation to language, is it like for instance, a young child before they learn how to speak, right? Actually I'm thinking in particular right now of my granddaughter, who's eight weeks old. Right? And so she's learning about the world around her. And so when she first say, looks at a dog or a cat, she's going to know what that thing is before she even knows the word for it, right? I mean , she'll recognize these objects, a mother, father, animals, and so on. And so we already exist in a world in which we can know and identify things. And language is one way of representing them to other people, right? Because if I want to tell you, I just saw a dog walk down the street, I have to use a word. They can be in English. It can be in Chinese, it can be in German, but I'm using a tool to represent what was already there, this concept of dog or cat. So when you say language as a model of reality, it's to represent concepts that preexist language, right . Or exist apart from language, am I close?
Tony Barr: 4:03
Well, you're hinting at the right direction.
Richard Miles: 4:06
Ok alright. I'm hinting. I'm getting, I'm warmer. Okay.
Tony Barr: 4:08
Okay. Now I think the baby is a visual learner. As an audible learner. It doesn't have the concept of phonemes and syllables yet very rapidly. It learns mama and Papa. Very simple, simple sounds. And they build eventually a oral language, or you could call it a visual language. There is a school of thought in computer science about visual languages to represent things like programs, to represent things like mathematical molecules. So we build visual languages as well as we build verbal languages. The underlying representation, the structural representation in my mind is all the same. Whether you see it visually, or are you seeing it as a series of characters
Richard Miles: 4:52
And in visual language again, just so we make sure we're talking about the same concepts . If I'm driving down the street and I see a sign and the sign has a picture of a man walking across a pedestrian,
Tony Barr: 5:00
That's a visual language.
Richard Miles: 5:01
I know it doesn't say stop because there's a guy about to cross the street. I just look at the picture and I know that's what it means. So it's communicating the exact same thing as if it were all.
Tony Barr: 5:11
Yeah, that's an international visual language. Our street signs are.
Richard Miles: 5:15
Untease us now. So when computers talk to each other, they're using this whole system of symbols, right? To communicate things for the computer to do. And when computers first started out, walk us through what the development conceptually was of early computers, I guess, and where we are now.
Tony Barr: 5:32
Well, I think from the very early on, people use simple languages to represent computer programs, to represent mathematical relationships at the basis of computers was the binary system where you use just the symbol zero and one, putting 32 zeros and ones together. You can make a pretty good sized number out of that. Using the binary, arithmetic. It's languages embedded inside of languages. We use 16 bits to represent a character now, and we can represent 65,000 characters with 16 bits. And we manipulate these characters, which are at the basis, representatives bits to represent, turn the switch on or drop the ball or whatever you might have commands like that and your computer language. Uh , so we convert from strings of characters, into some electrical instructions in the computer and it drops the ball. So it is a process of building bigger and bigger structures from littler structures. And we use in this case, a language based upon symbols, but you can use it language based upon pictures as well, which is like your street sign is the exact analogy there.
Richard Miles: 6:42
Is that part of the underlying basis of Amore a model of reality going towards more intuitive system, or is it something else?
Tony Barr: 6:50
It's meant to be a complete system. And it's a complete system starting off with the concept of a concept that defines itself and defines all other concepts. I think this is the most profound thought you can have a system has the definition of itself inside the system. This is actually something that I look for for 30 years of my life. And I found the concept of a concept. I used to call it the type of type, but better terminology is the concept of a concept. With the concept of concept, you can define all other concepts with the concept she can represent. A list of pictures would be a concept. An element of the list of pictures would be a character. So you build up bigger and bigger pictures. You see I'm really redefining computer science. So this is why we're having trouble communicating here.
Richard Miles: 7:38
So we've heard an awful lot about artificial intelligence and like sort of self-learning, but this isn't that.
Tony Barr: 7:44
No, it isn't that at all.
Richard Miles: 7:45
So it's not the ability of computers to program themselves.
Tony Barr: 7:48
No, it is not at all.
Richard Miles: 7:50
It's really a whole different way of understanding what computers are or what they do?
Tony Barr: 7:55
Exactly. Also, it's a coherent model that ties together the conceptual world and the world of objects. In fact, concepts in the heart are constructed of other little objects. So you're really using objects to describe objects, but you have to break the ideas down and package them so that the concept is the definition of an object,
Richard Miles: 8:18
This strikes me, is it in some way , similar to fractals and that in fact, those things like trees or snowflakes, right? The large thing is simply the smaller thing repeated over and over and over again. So that the same structure that you find in a leaf, it's essentially the same structure you find in a tree. So there's a definition of the tree so to speak in the leaf.
Tony Barr: 8:38
You are somewhat getting on there. A concept is essentially a list of connections and a connection can be a list, or it could be just an object or it could be an array. So that little framework there is applied over and over again, to build up the world. So you have concepts which are all defined by the concept of concept. It's very finite. That's the fractal idea, but you develop new concepts and concepts and inherit properties from their parent concept. So it grows.
Richard Miles: 9:12
Tony, what sort of applications do you think this would have a model of reality and how would this change the world in which we live in now?
Tony Barr: 9:20
Well presently, the programming model is programming languages and databases to store the objects. So when you're signing up for a class, you, you access some program on your screen, you enter in the things and they do database operations to build the new registration of you for some class, the teacher sees that the students in the class, the student sees that I'm in this class. So you're manipulating objects through a database mechanism. My idea is to have one level of formality so that the person is just in the model and all of these operations are done in a simple conceptual manner. So the separation of the types and the objects or the schema and the objects has disappeared, and those all become one together. So you see the thing in a uniform understandable way. It really is an enormous change. So right now, any application you have has databases behind it. And this system, you would go through programs and access the student records, whatever, but it'd be the same model used to store the objects as you would when you ran a simple little program. So it eliminates enormous amount of technology. And it's a whimsical idea, but I have the idea of a one world model where everybody live inside it. Now I could, well imagine that the country of China would have its own model reality and maybe the Western world would have its model reality. And actually, I , in reality, there'd be the Google versus the Yahoo thing. I mean, Yahoo's lost, but there would be that type of competition. But Google now is one to race.
Richard Miles: 11:10
Some people would argue that Google is our new reality, right? I mean, what can we do that? Doesn't involve some way Google, right?
Tony Barr: 11:16
Exactly.
Richard Miles: 11:17
You were, as I mentioned at the top of the show, the founder of Statistical Analysis System, SAS , which went on to huge commercial success, what was it like in the early days, starting a company, running a company. And then do you look back on that time with any regrets or you left relatively early in 1979, correct? Before SAS was this huge company that it is now, what was that like being part of a startup young company. And then what is it like to look on that now? What, 40 years later,
Tony Barr: 11:46
It was my second startup.
Richard Miles: 11:47
Second startup. Okay.
Tony Barr: 11:49
So, the first startup I left because I wanted to focus SAS because I thought that was the bigger fish I didn't do well in a organization run by committee. That's what I would say.
Richard Miles: 12:01
You also did some work for department of defense, for awhile.
Tony Barr: 12:04
That was an incredible experience I got in there right at the time, things were really changing, I guess, McNamara was in there and he was going to computerize everything. And I worked at the Pentagon. They hired me, they said, we're going to put all the military information into the computer. What a challenge. I just loved that place. And they had the self-defining file and that has become etched in my mind. Self-definition however, when it takes self-definition to the ultimate , uh , step so that if you have a question it's always inside the system, it's like having the Webster dictionary there. I mean, you always have the definition of everything that you need. And as you navigate through it, you have two pointers, one point or in the object, one point or in the concept. And as you navigate down, you get little or little or concepts. And so you never lose track of where you're at.
Richard Miles: 12:54
So this is probably a really bad analogy, but it was one that gentlemen, is this almost like a computer perpetual motion machine. And then once you've written it, you don't need to tinker with anymore. Cause it kinda just keeps self-defining yourself or providing its own answers, I suppose.
Tony Barr: 13:07
Not all, not at all.
Richard Miles: 13:10
Alright, we should probably move on from me trying to define this, but when you said self-defining file. So I was trying to grasp how that would actually work.
Tony Barr: 13:17
Well, the self-defining file just meant that they had the dictionary of all the elements on the front of of the file.
Richard Miles: 13:24
And this was in the late sixties?
Tony Barr: 13:26
That was 64.
Richard Miles: 13:29
Okay . All right . And how long were you there?
Tony Barr: 13:31
Just short of two years.
Richard Miles: 13:32
Prior to S AS, right? Well, before SAS.
Tony Barr: 13:34
Well it was Immediately before SAS. Right.
Richard Miles: 13:37
Tony, you said that as a child, you quote, lived inside your head and just created some problems in school because you're always planning projects rather than listening to your teachers.
Tony Barr: 13:48
That's right.
Richard Miles: 13:48
Which probably describes a lot of young boys in particular. Did your teacher finally come around at some point to realizing how bright you were or was it always an issue in your school career?
Tony Barr: 13:59
I never had any support from the teachers.
Richard Miles: 14:02
Really? Right to the bitter end.
Tony Barr: 14:03
Not till the bitter end.
Richard Miles: 14:05
That must've been frustrating. I imagine that because,
Tony Barr: 14:07
N`o I, I enjoyed my life .
: 14:08
You enjoyed it? Okay. But even in school, it wasn't frustrating because the teachers probably thought what that you were lazy or you just weren't focused. What did they think?
Tony Barr: 14:16
I'll give you this little thought. One time we took a test to see what was going to go into New York city to compete on this math test that they had. And so I took the test and she had these other guys, they were going to go. I said, I don't think I made a mistake on that test. And she went up and regraded my test and found out that I was the guy to go in on that bet. Now went into New York city at NYU and all these kids were there. And I believe I would've have gotten one of those scholarships, except they didn't teach anything about logic in my mathematics. And all these guys from Brooklyn Polytech had courses where they discussed logic, the ands and or's and not''s and whatever. If I just had that one little thing, I might've gotten that, but I'm just using this to illustrate that I got no respect.
Richard Miles: 15:05
You sound like Rodney Dangerfield. Were you always drawn to math? When was the first time you remember loving math?
Tony Barr: 15:11
Seventh grade.
Richard Miles: 15:12
Seventh grade. Okay.
Tony Barr: 15:13
When algebra came, then I loved it.
Richard Miles: 15:14
Did you just like the beauty of it? The elegance? Is that what attracted you?
Tony Barr: 15:17
It was the elegance, right, right, that's the power of it.
Richard Miles: 15:18
I remember the same thing happening. I was not a math major, but I remember really liking geometry because geometry operated according to these definite rules. And if you, if you worked through the steps, you got a certain answer and there's something very beautiful about that. Were you ever interested in things like biology or chemistry or was always so focused on numbers and math and satanical things.
Tony Barr: 15:38
Mechanical things .
Richard Miles: 15:39
Mechanical things, okay.
Tony Barr: 15:39
More than anything else.
Richard Miles: 15:42
You grew up in New Jersey, correct?
Tony Barr: 15:43
Right.
Richard Miles: 15:43
Neither of your parents was a engineer or mathematician or had that background? No . Did you have any other influences? Do you think this is just, you were born this way? Or did you have particular teachers or other folks who got you curious about this?
Tony Barr: 15:57
I want to straighten that out. My mother was an incredibly brilliant person. I think my father was as well, but he didn't finish high school. He was from England. His father was a refrigeration engineer who had a patent on how to make ice cubes on a passenger line.
Richard Miles: 16:12
This is the English UK patent? Have you ever found that patent?
Tony Barr: 16:16
My sisters found it and I asked her to send it the reference to me, but I will , I will get that reference. Now, my grandfather was an exceptional individual who was a head of the school board, state representative owned half a bank, owned a car dealership and was wiped out by the depression. So he had the biggest house in town, still the biggest house in that town and two maids. And my mother never talked about them. I had to go visit my uncle to find the history of that because we never were living that life.
Richard Miles: 16:46
Did you go into New York city a lot? The reason I ask , I just interviewed somebody a few days ago and they attribute a lot of their intellectual curiosity to the fact that they went into the city. They went to these great planetariums and museums as a young child.
Tony Barr: 16:59
I love that. I did the same stuff. Also. We would visit in Philadelphia, the Franklin Institute. That was a big, big thing for me as well.
Richard Miles: 17:07
And those were in the days. I imagine you just went by yourself, right? Just hopped on a subway or a bus or something.
Tony Barr: 17:12
I was too young, no, I went to New York city on my own, but in Philadelphia I went with my parents.
Richard Miles: 17:18
Right, right. Tony, we always offer everyone on the show, a chance to dispense wisdom. If you were meeting a younger version of yourself, are there things that you would say, Tony, stay away from this or Tony definitely do that. Anything that you would have done differently than if you knew what you know, now?
Tony Barr: 17:34
This is not exactly Jermaine , but I watched the picture about Steve jobs. And there was one thing in there that was very dramatic, which was Steve jobs was reviewing what they're doing with the MAC. And he where's the typeface support? And program manager said, we're not going to do it because we can't meet the deadline. And Steve said, well, what? You can't make the deadline. No, we can't meet the deadline. And Steve said, we're going to meet the deadline. And the guy said, no, we can't meet the deadline. And Steve said, you're out of here. Well, I had an incident in my company that was rather tragic where I had a team of five people, one engineer, four programmers, working on a phone system that I thought was an easy way to make money. And I had some pretty good ideas on how to do it. I actually thought I could do 6,000 phones on a PC. So I was go into these meetings and the guy was a brilliant guy. I had as the manager there, he would argue in front of the whole group there. That is non-viable . And I'm thinking we spent $80,000 for our phone system. This is viable. And then I go to a trade show and somebody comes over and says, have you seen this guy's booth? He's over there with the tutorial for windows 95. So he got the whole team working on that. Not on my work. I would felt insulted time after time having to defend this product in front of the whole team. I see if he came to me one-on-one right, that would have been fine. But that was a big mistake for me to tolerate dissent in front of the team. He took the whole team with him .
Richard Miles: 19:09
I see . But there were supposed to be working for you.
Tony Barr: 19:12
Right! Right. So that's just an easy thing to talk about in this little thing. So perhaps a word of wisdom is that you need to be a self promoter and make sure that people know who's getting what done and share the credit, but also take the credit. That is a word of wisdomI would say.
Richard Miles: 19:29
Tony, we could talk all morning. It would take you at least another two hours to get me up to speed and what I need to know about computer programming language, but it's been a fascinating conversation. You have an incredible mind and incredible career, and we're just very pleased that you could join us today on Radio Cade.
Tony Barr: 19:43
Well, I'm very pleased to be here. Thank you.
Richard Miles: 19:45
I'm Richard Miles.
Outro: 19:48
Radio Cade Would like to thank the following people for their help and support Liz Gist of the Cade Museum for coordinating and vendor interviews. Bob McPeak of Heartwood Soundstage in downtown Gainesville, Florida for recording, editing and production of the podcasts and music theme. Tracy Collins for the composition and performance of the Radio Cade theme song, featuring violinist, Jacob Lawson and special thanks to the Cade Museum for Creativity and Invention located in Gainesville, Florida.
