“The Forty-Year Programmer”

[Lynn McGuire]

“The Forty-Year Programmer”
https://codefol.io/posts/the-forty-year-programmer/

“Software Development is Young”

“The computer language Fortran dates back to 1957. There are other
languages about the same age (LISP and Algol: 1958, COBOL: 1959) and a
few weird contenders for being older (Konrad Zuse’s Plankalkül, 1942-ish.)”

“Let’s say there have been programmers since roughly 1957. 65 years.
I’ve been a full-time paid programmer since 1998 (24 years) and a
programmer at all since 1984 (38 years.) I’m pretty experienced. Alan
Kay, inventor of the Smalltalk programming language, starting somewhere
around 1963 — 59 years. He’s one of the longest-active I’ve found.”

I started writing Fortran in 1975 at age 15. I wrote some Basic in 1982
on our new IBM PCs to help with my engineering calculations at the power
plant I was working at. I bought my own PC in 1983 and bought the new
Turbo Pascal compiler which was the most amazing thing that I had even
seen, using an interactive development environment. I bought the Turbo
C compiler in 1987 and moved on to that. I took an engineering software
job in 1989 writing C on DecWindows. I moved to a Smalltalk and C
product in 1992. I moved on to C++ around 2002. I write engineering
software in F77 and C++ nowadays.

Lynn

[Robin Vowels]

On Tuesday, September 6, 2022 at 2:26:38 PM UTC+10, Lynn McGuire wrote:
> “The Forty-Year Programmer”
> https://codefol.io/posts/the-forty-year-programmer/
>
> “Software Development is Young”
>
> “The computer language Fortran dates back to 1957. There are other
> languages about the same age (LISP and Algol: 1958, COBOL: 1959) and a
> few weird contenders for being older (Konrad Zuse’s Plankalkül, 1942-ish.)”

.
Some date FORTRAN a little earlier, to 1956. GEORGE came in 1956.
.

> “Let’s say there have been programmers since roughly 1957.

.
Programmers have been around since the mid-1940s.

[Thomas Koenig]

Lynn McGuire <lynnmc...@gmail.com> schrieb:

> “The Forty-Year Programmer”
> https://codefol.io/posts/the-forty-year-programmer/
>
> “Software Development is Young”
>
> “The computer language Fortran dates back to 1957. There are other
> languages about the same age (LISP and Algol: 1958, COBOL: 1959) and a
> few weird contenders for being older (Konrad Zuse’s Plankalkül, 1942-ish.)”
>
> “Let’s say there have been programmers since roughly 1957. 65 years.
> I’ve been a full-time paid programmer since 1998 (24 years) and a
> programmer at all since 1984 (38 years.) I’m pretty experienced. Alan
> Kay, inventor of the Smalltalk programming language, starting somewhere
> around 1963 — 59 years. He’s one of the longest-active I’ve found.”
>
> I started writing Fortran in 1975 at age 15.

What computer did you have available to run it on? At the age of 15,
I was lucky that my father had a programmable calculator.

[Lynn McGuire]

A Univac 1108 with 64K words of memory (32K words data, 32K words
program code). My father had an engineering software company and I
start working for one of his programmers.

Lynn

[John McCue]

Lynn McGuire <lynnmc...@gmail.com> wrote:
> ?The Forty-Year Programmer?
> https://codefol.io/posts/the-forty-year-programmer/
>
<snip>

>
> I started writing Fortran in 1975 at age 15.

<snip>

You and be both :)

It was my first paying job (I was at college) and I messed up
royally, they did not renew the contract.

But in a couple if years I got back into it and been there ever
since.

John

[gah4]

On Tuesday, September 6, 2022 at 9:47:37 AM UTC-7, Thomas Koenig wrote:

(snip)

> What computer did you have available to run it on? At the age of 15,
> I was lucky that my father had a programmable calculator.

When I was 14, my father borrowed a brand new HP 9810A programmable
desktop calculator. (While its owner was away.)

I worked for some time to write a prime number program, doing indirect
addressing to try dividing by all primes found so far. Much fun, and
complicated in its loops.

Then, the last two weeks of 8th grade, our teacher got some of us two
weeks (about an hour a day) learning Fortran on the school district's
NCR Century 100. That got through about a table of square roots.

My 8th grade graduation present was the IBM Fortran manual, just
before a month long family road trip across the country. I brought it
along, but didn't read so much of it.

But after the road trip, and before school started in September, I did enough
Fortran programming to mostly understand it. I would read IBM reference
manuals from cover to cover. (I knew lots of strange things, along with the
more usual ones.)

My high school had keypunches, which we were supposed to use to run
programs on the Century 100, but I punched them and had them run
on the IBM S/360 instead.

Among the features of the OS/360 compilers, is one to print out the
generated assembly code. I also had the assembly source for the
Fortran library on microfilm. Mostly with those, and some other manuals,
I started S/360 assembly programming when I was 16.

[jfh]

I may be one of the last people to have got a PhD in theoretical fluid mechanics from a university that already had a computer without ever using it: thesis submitted 1963, degree awarded 1964. I first used Fortran II in 1963 on an IBM 1620 in a different university to solve an equation of the form integral_0^infinity p(x,y)*f(y) dy + c*f(x) = g(x) where f is the unknown function, p and g are known functions and c is a constant. A few months later that machine was replaced by an Elliott 503 (much faster) and I changed to Algol 60 (much nicer) and translated my program manually because I could not get FEAT (Fortran to Elliott Algol Translator) to work. More recently I hit the same integral equation with a different right-hand side function g. My original Fortran code no longer existed and I was now using Fortran 95 (after deviations through Pascal, Algol 68, and Fortran 66 and 77) and translated my old Algol 60 code; a useful test was getting the old answer with the old function g. The result: on seeing what the solution f(x) looked like with the new function g(x) I could see how to prove that analytically.

[LC's No-Spam Newsreading account]

On Tue, 6 Sep 2022, Lynn McGuire wrote:

> On 9/6/2022 11:47 AM, Thomas Koenig wrote:

>> What computer did you have available to run it on? At the age of 15,
>> I was lucky that my father had a programmable calculator.
>
> A Univac 1108

When I was in the 4th or 5th year of high school (so 17-18), a
schoolmate brought in a brand new pocket
https://en.wikipedia.org/wiki/HP-35 which his father (a scientific
journalist) had in vision, and we played with it for a while (he did
not buy it after all).

I started programming (Fortran of course) on the University Univac 1108
during the 2nd year of university. Among my first (user-chosen)
programs there were: a program to print the decimal values of all
characters (Univac had 36-bit word of 6 6-bit bytes); a program to
compute with full precision (using an heuristic algorithm) Sissa's
problem https://en.wikipedia.org/wiki/Wheat_and_chessboard_problem (was
beyond the precision of 36-bit integers); a finite-difference solution
of the two body problem (on the first run Earth leaped away on a
straight line because of an incorrect implicit variable typing) :-)

[gah4]

On Saturday, September 17, 2022 at 7:43:19 AM UTC-7, LC's No-Spam Newsreading account wrote:

(snip)

> I started programming (Fortran of course) on the University Univac 1108
> during the 2nd year of university. Among my first (user-chosen)
> programs there were: a program to print the decimal values of all
> characters (Univac had 36-bit word of 6 6-bit bytes);

In high school, we had access to an IBM 370 running Call/OS,
and Teletype ASR33 terminals.

I wanted to find the EBCDIC to ASCII conversion table, so wrote
a Fortran program to print out all 256 characters, with
a loop and A1 format.

Partway through running it, the system crashed.
Oh well, try again the next day.

Partway thought running it, the system crashed.
I didn't try running it again. It seems that there is a
character that crashes the system when printed.

So, the opposite of your program.