There is a computer scientist that is

super wellknown, but a lot of the things

that he did during his career are well,

they're unknown. I have a good feeling

like you know him by his shortest path.

Yes, today we are talking about Edgar

Dystra, of course, creator of Dystra's

shortest path algorithm. By the way, you

don't love him at his shortest path, you

don't deserve him at his longest. And

today we're going to be celebrating

Edgar Dystra by giving him the highest

known achievement any programmer can

possibly receive in their lifetime. The

programmer of the month at the startup

touring award, Nobel Peace Prize,

programmer of the month. The thing about

Dystra is he didn't just do the shortest

path, which by the way, the way he comes

up with the shortest path is just out of

control, but he actually comes up with

many more things that we use to run

operating system and distributed

systems. He also did a lot of

foundational work that ends up appearing

in a lot of the modern languages that

you use today. And he prevented one of

the greatest disasters in all of

programming. And to top it all off, the

man has some oneliners that are just

So today we celebrate Edgar Dystra,

programmer of the month. Now, before we

begin, a quick word from the sponsor.

>> She's seen the latest database CEO

monthly.

>> Whoa, nice uptime. I'd love to get my

data in that.

>> You mean you haven't even connected yet?

We can use my Planet Scale account.

>> It's called Planet Scale and it's really

rad. Those database queries are super

fast. Branches, migrations, query

insights too. My SQL and Postgress are

easy to use. Yeah, go planet scale. Get

data.

>> It's Postgress.

>> Planet Scale. The easiest MySQL or

Postgress database to use. Your

developers or parents help you hook it

up. Data sold separately.

>> Welcome back. All right. Before we get

to know Dystra, the legendary computer

scientist, let's learn a little bit

about his early years. Dystra was born

in Roderdam in 1930. He described his

father at one time the president of the

Dutch Chemical Society as an excellent

chemist and his mother, a brilliant

mathematician who had no job. Huh,

sounds like any mathematician I know. In

1951, Dystra's father suggested he

attend a 3-week course on programming in

Cambridge. It turned out to be an idea

with farreaching consequences. Yes, that

little tiny suggestion by his father.

Those three weeks changed the course of

computing history by getting Dyster to

get off the track of physics and onto

the real and pure science of computing.

Now, the early part of Dyra's career

started off in the center of

mathematics. This is where he studied

computing and defended his PhD thesis.

He was there from 1952 to 1962. Now, he

received his PhD in 1959, but he

actually created the shortest path

algorithm in 1956,

and the story for it is just so dang

good. One morning, I was shopping in

Amsterdam with my young fiance and

tired, and we sat down on the cafe

terrace to drink a cup of coffee, and I

was just thinking about whether I could

do this. Of course, this being the

shortest path from Rotterdam to

Gronigen, Grroning, Gronigan, the n the

name is Gronagen. So he was already

thinking about the shortest path and

then I designed the algorithm for the

shortest path and as I said it was about

a 20inute invention. In fact it was

published in 1959 3 years later. The

publication is still readable. It is in

fact quite a nice one. Was really just

patting himself on the back by the way.

That entire paragraph mogged everybody

several times over. He was just off

having a nice cup of coffee. A little

tired from shopping. H man. I just not

you know it's kind of difficult here.

Let me just do a quick invention that's

going to blow the minds of the world,

make gaming super awesome, and also make

networks possible. Oh yeah, he's also

going to do it in 20 minutes without a

whiteboard, just his brain. You know,

sometimes I like to pretend myself a

smart man, but I this this is just

mogging, okay? There's nothing there's

nothing I can do in the light of his

intellect, okay? [laughter]

I am like a caveman. I feel like the guy

on idiocracy trying to put the square

block into the round hole. Now, right

around when he defended his PhD thesis,

he started to work on the Algo 60

compiler. Now, this was an impressive

feat. First off, Alol is a computing

language that's going to be a highlevel

language. This is one of the first ones

kind of really entering the scene. And

second, it's not just that you're

building a compiler which had relatively

low amounts of study comparison to what

we have today, but it's also the machine

that it ran on. The most obvious problem

the pair faced, by the way, the pair

would be Mr. uh Dykes run along with it,

Dr. Zonavald, was that the machine

designated to run the software, the

Dutch Electrologica X1 computer, had a

memory of only 4,096

words. Now, you may not be familiar with

the term word. That's just going to be a

chunk of memory. Specifically for this

machine, it was 27 bits per word. I know

this was I mean, during this time,

computing was so out of control, powers

of two weren't even the standard. 27

bits. What? What are we going to do with

27 bits? I bet he would kill a bite for

some more memory. By comparison, the

memory of a present-day laptop is larger

by a factor of a million. But towards

the end of his 8-month stint with Algo

60, he separated ways. Algo 60 was

designed by an international committee.

And although Dyster attended several

meetings during the design process, his

name does not appear among the 13

editors of the final report. Apparently,

he disagreed with a number of majority

opinions and withdrew from the

committee. This was perhaps the first

public sign of his fiercely held

independence. In 1962, he ends up

leaving the mathematical center and

heads on over to the Einhovven

University of Technology where a lot of

his work, the things that you use, the

things that you rely on happened and you

probably don't even know they happened.

What? You just a jerk maneuver right

there. Now when Dyster goes to the

university, he starts working on the

multi-programming operating system where

his love of parallel programming was

fostered and he ended up coming up with

a lot of really cool concepts. This work

was taken from EWD123.

Hey, you're probably wondering what the

heck is an EWD123.

Well, it's actually pretty simple. EWD

is just Edgar's initials, but the 123 is

the 123rd post in his little kind of

blogging series kind of handwritten

letters that he would send out to other

academics. He effectively was a YouTuber

in his day. He's just like me except for

he was smart and probably pretty

good-looking. An extensive 87page report

titled cooperating sequential processes.

Now, you're probably like thinking like,

"Hey, Prime Prime, is that is that the

Go thing? Is that Goof Funks?" No, no,

no, no, no, no, no. This is kind of the

first version of it. Later on, Tony

with some inspiration from Edgar

Dystra ended up writing communicating

sequential processes which became the

foundation of Go Funk. Now, in the same

report, he introduced the first known

synchronization primitive which he

termed semaphore. Yes, he created the

semaphore which was the basic and the

first kind of version of a mutex. A

semaphore of with one that's just a

mutex. Obviously not modern sweet

amazing mutexes with the threads in

recursion and all these kind of things

but the very first standard version just

a semaphore. Now you may not be familiar

with what a semaphore is. It's pretty

simple. Let's imagine you have this

thing right here which is a lock and you

can put a number on it. Let's just say

we have a number of one and you have a

bunch of threads that are all going into

this trying to say reach or touch some

sort of value or mutate some sort of

value. Well, you can't have all these

extremely and well-drawn straight lines

go in and mutate the same value at the

same time. That creates a data race. So,

what you do is you have a little lock

and whichever one arrives first takes

this number and down like decrements it

atomically. Oh no, it's at zero. Nobody

else can go in. So, this guy goes in,

does whatever, you know, maneuvers he

needs to and gets on out. Next person

can go in, decrement, do whatever they

need to, and go out. The next person

goes in, does whatever it needs to, and

leaves. Each time they leave, you can

imagine they increment the number, and

the next person decrement, increment,

decrements, increments. You get the

idea. This effectively allows only so

many things through at the same time.

Now, there's probably a bit more

complication that's going on, but this

is just like the simplest way to look at

it. He also identified the deadlock

problem, which he named the deadly

embrace. Now, you're thinking,

what is this, a Greek tragedy? Is this

is this this is Odysius going out

somewhere? Yes, the deadly embrace,

which by the way, amazing name. Also,

pretty sure it's a slay the spire card

if I'm not mistaken. If it was up to me,

I think we should continue to call it

the deadly embrace and not a deadlock.

To give you the simplest explanation,

let's pretend we had some sort of

database resource and it was like a

mutex. You can only have one person

owning that resource at a time. And

let's pretend that there's two different

threads running. We have A and we have

B. Now, these guys can run

independently. They don't rely on each

other. Everything's fantastic. And A

goes, "Hey, I want to use the database."

But unfortunately, B already has a lock

on the database. So A is just going to

sit there and wait. You know, it's just

going to be a happy little boy hanging

out, having a good time. Well, what

happened if B is like, "Hey, the result

from A." What's going to happen? Well,

A's waiting for B, but B's waiting for

A. Deadly embrace. They've embraced each

other. deadly like [laughter]

this is. See, this is why I'm not the

smart guy. This is why I don't have my

own crime 123. The mutual exclusion

problem along with deadlock detection

and prevention are now mandatory topics

in courses on operating systems and

parallel programming. Yes, Dystra is

effectively the father of parallel

computing. And you didn't even know

that, did you? Kids these days. In 1968,

Dystra ended up penning a nice little

article here called go-to statement is

considered harmful. Now, the thing is is

that originally it was not called that

in the book we programmers, which I

might have written a forward to. You

learn that he originally titled it a

case against the go-to statement in

which Nicholas Worth, the editor,

renamed it to Go-To Statements

Considered Harmful, which became such a

popular paper, you probably even knew

about that paper whether or not you read

it. Now, here's the thing. Back in those

days, in 1968, the largest possible

program was some IBM OS totaling right

around 1.7 million lines of code. I just

want you to think about that for a

second. There were 1.7 million line code

projects

with just go-to statements, hundreds if

not tens of thousands of them all over

the place. Like programming had to be

just insane during those days. And Dyrup

had to go and argue, which took a decade

to convince people, hey yo, go to that's

not good. That's not good at all. If you

don't know what a go-to is, you can

imagine you just have like a bunch of

lines of code anywhere in here. You can

just be like go to and then you can say

some sort of offset, some sort of label

into another location and it will just

go over there whatever there is. Now at

one point this made sense before we had

any sort of highle language. If

statements are just gotos, right? If

this condition is true, continue else go

to the end of the statement. That

actually makes a lot of sense. So you

can imagine why gotos were around. Now

if this war didn't happen and Dystra

didn't really lead the charge against

the go-to statements and people were

kind of just okay with second best, we

could have had JavaScript with goto

statements. I you think React is bad

now? Imagine server components with

gotos and you have to be like use client

go to jumping into like server code. It

would have just been nuts. But honestly,

there's just like this small part of me

that thinks JavaScript it deserves a

go-to. And this kind of war against the

go-to became a larger war in which

Dystra argued that we need to be using

if statements and for loops and higher

level constructs to easier understand

the program that we're writing. This

became structured programming. It became

the like mantra of his that we need to

have easily comprehendable programming

languages. Thank you, Dystra. I mean, I

still know it resulted in JavaScript,

but you get the idea. Thank you, Dystra.

And still at Einhovven in 1972, he did

receive the ACM touring award. Now,

remember touring award, Nobel Peace

Prize, programmer of the month. Don't

forget that. Widely considered the

second most important prize in computing

science, he was recognized for

fundamental contributions to programming

as a high intellectual challenge for

eloquent insistence and practical

demonstration that programs should be

composed correctly, not just debugged

into correctness. for illuminating

perceptions of problems at the

foundation of programming design. This

statement right here is actually just

like I mean in today's LLM world, this

is kind of like this almost hilarious

view. It's almost like we have inverted

what Dystro would consider good cuz I I

always often wondered would Dystra like

LLMs doggy. Hi, what do you want from

me? I'm trying TO DO A YOUTUBE VIDEO.

Silly dog. You know, I often wonder what

would Dystra think of today's LLM world.

Well, we do have a little bit of a hint

the not just debugged into correctness

part of his touring award because when

he left Einhovven University in 1984 and

went to Austin, Texas up until his

retirement in 1999, he spent the

majority of his work working through

proofs and formal proofs for

programming. He believed that all

programs should be provably correct, not

just, hey, I think this is correct. not,

hey, I believe we have something that

works. We have enough unit tests to

prove that all the bugs we've ever found

haven't happened again. But he wanted to

believe that all things could be always

correct. Would he look on to this time

and think, what the hell's happening? Or

would he be stoked? because it is such a

uniquely magical period of time in which

we can speak English and out comes

computer program like that is it is

whether you like them or not that is

just absolutely magical. That is

something I never thought I could ever

see. In fact, I I have largely operated

under the assumption that a program

can't tell if it's going to halt. And

now we have things that appear to know

when a program is going to halt. It's

honestly kind of a a magical experience

that we live in. So, would he look at

this with disdain and say that we're all

bunch of slop cannon coding bums? Or

would he actually look at this and go,

"This is amazing. What a new and

incredible time period." Well, this is

where we get to a lot of his quotes. And

I think when you read through his

quotes, you probably get a little bit

better of an idea of what he might or

might not think during this time. It is

practically impossible to teach good

programming to students that have had

prior exposure to basic. As potential

programmers, they are mentally mutilated

beyond hope of regeneration. Damn, dude.

I started on Q Basic.

Dude, I'm getting second hand destroyed

right now. The use of cobalt cripples

the mind. Its teaching should therefore

be regarded as a criminal offense. Hey,

Admiral Grace Hopper, still love you.

You you might get yourself a little

programmer of the month. Okay. APL is a

mistake carried through to perfection.

It is a language of the future for the

programming techniques of the past. It

creates a new generation of coding bums.

Program testing can be used to show the

presence of bugs, but never show their

absence. It's actually an incredible

insight for this, especially in just our

world of, hey, just put the guy put the

guidelines on the slop cannon and bro,

you got yourself something amazing on

the other side, which is really this,

like, can you know all the outcomes? Do

you can you actually control for all

these things or are you accidentally

letting sinister bugs run loose in your

system? This is one of my personal

favorites. If in physics there's

something you don't understand, you can

always hide behind uncharted depths of

nature. You can always blame God. You

didn't make it so complex yourself. But

if your program doesn't work, there is

no one to hide behind. You cannot hide

behind an obstinate nature. If it

doesn't work, you've messed up. Well, I

mean, now we have Claude. Okay, Claudius

Maximus. We can we can blame Claude,

stupid Claude, for not supporting

openweight models. My other big favorite

one. Object-oriented programming is an

exceptionally bad idea which could have

only originated in California.

I mean something so beautiful. I mean it

just makes me so stoked. Also this one

reminds me of Lacun Lunquin. Laquan Lun

the guy the used to be guy from

Facebook. The effort of using machines

to mimic the human mind has always

struck me as rather silly. I'd rather

use them to mimic something better.

Which is totally right. Right. Because

like Stockfish is better than all of

humanity at chess. Like [laughter]

we as a collective whole couldn't beat

Stockfish. So having like human

intelligence, I don't know, it's

probably not all that good. And

obviously before we go, I do have to

give you a couple little fun facts. He

also never purchased a computer. Yes,

the man who carried computer science on

his shoulders never purchased a

computer. I married a Dutch marriage

rights require you to state your

profession and I stated I was a

programmer. The municipal authorities of

the town of Amsterdam did not accept it

on the grounds that there was no such

profession. Bro, imagine, dude, we

already have a hard enough time with

Arch Linux users, but imagine getting

rejected because they're like, "Arch

Linux, that's not even a real operating

system, bro. You can't get married. You

No. No. No marriage for you. You and

your fake profession of programming.

Okay, buster." Yeah, that's a that's a

real one. Well, this is to you, Edgar

Dystra, and we appreciate and celebrate

your contributions to all of the field

of computing. And this is why you win

this month's programmer of the month

from the startup, multi-billion, of

course. And hopefully this title makes

more sense. The man who carried computer

science on his shoulders. We could

literally have go-to statements in

JavaScript right now. So, you better

thank him. I want to see some thank yous

in the comments. By the way, just

special shout out to Kristoff Apt. Uh,

amazing article. I left out so many

amazing details. I strongly recommend

you go and read the entire dang thing.

The name is the shortest pathogen.