From Newsgroup: comp.lang.fortran

Short documentary streamed in 1982 for the 25th anniversary of FORTRAN.

Documentary:
https://www.youtube.com/watch?v=CXvyjbWkTyU

John Backus and other FORTRAN creators speaking about the development of
the language and the compiler.

Preserved written history of FORTRAN:
https://softwarepreservation.computerhistory.org/FORTRAN/

Stolen from
https://www.reddit.com/r/fortran/

Lynn

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On 1/7/2026 9:52 PM, Lynn McGuire wrote:

Short documentary streamed in 1982 for the 25th anniversary of FORTRAN.

Documentary:
   https://www.youtube.com/watch?v=CXvyjbWkTyU

John Backus and other FORTRAN creators speaking about the development of
the language and the compiler.

Preserved written history of FORTRAN:
   https://softwarepreservation.computerhistory.org/FORTRAN/

Stolen from
   https://www.reddit.com/r/fortran/

Lynn

I started writing software in Fortran in 1975, thought it the best thing
since sliced bread. Then I learned Pascal in 1983 (Turbo Pascal). Then
I learned C in 1986 (Turbo C).

Lynn

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Fri, 9 Jan 2026 15:09:58 -0600, Lynn McGuire wrote:

I started writing software in Fortran in 1975, thought it the best
thing since sliced bread. Then I learned Pascal in 1983 (Turbo
Pascal). Then I learned C in 1986 (Turbo C).

1975 was probably about the time I first encountered actual lines of
Fortran code (though not a computer to run them on). This was in the Encyclopedia Britannica article on computers.

Then I got hold of a book called “The Compleat Cybernaut”, by Anna
Burke Harris. It was a pretty comprehensive introduction to Fortran. I
remember the bio mentioned that the author was living in a city where
she was the only practising blacksmith. I devoured that in a weekend.

Later, I found another book, called “Programming in POP-2”. The
concepts of non-numerical programming, lists, structures etc -- way
beyond anything Fortran had -- just blew my mind. Also it was
stack-based.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

Lynn McGuire <lynnmcguire5@gmail.com> schrieb:

Short documentary streamed in 1982 for the 25th anniversary of FORTRAN.

Documentary:
https://www.youtube.com/watch?v=CXvyjbWkTyU

Nice video - seeing the peole that I read about in "Abstracting Away
The Machine" (which I thoroughly recommend).

John Backus and other FORTRAN creators speaking about the development of
the language and the compiler.

The job they did was absolutely amazing - almost inventing
everything from scratch.
--
This USENET posting was made without artificial intelligence,
artificial impertinence, artificial arrogance, artificial stupidity,
artificial flavorings or artificial colorants.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On 1/9/2026 3:40 PM, Lawrence D’Oliveiro wrote:

On Fri, 9 Jan 2026 15:09:58 -0600, Lynn McGuire wrote:

I started writing software in Fortran in 1975, thought it the best
thing since sliced bread. Then I learned Pascal in 1983 (Turbo
Pascal). Then I learned C in 1986 (Turbo C).

1975 was probably about the time I first encountered actual lines of
Fortran code (though not a computer to run them on). This was in the Encyclopedia Britannica article on computers.

Then I got hold of a book called “The Compleat Cybernaut”, by Anna
Burke Harris. It was a pretty comprehensive introduction to Fortran. I remember the bio mentioned that the author was living in a city where
she was the only practising blacksmith. I devoured that in a weekend.

Later, I found another book, called “Programming in POP-2”. The
concepts of non-numerical programming, lists, structures etc -- way
beyond anything Fortran had -- just blew my mind. Also it was
stack-based.

My father started a chemical engineering software company in 1968. In
May of 1975, he asked what I was doing that summer between my freshman
and sophomore years in high school. I said watching tv and reading
scifi books. "Nope, you are going to work for one of my programmers",
my dad said.

So I was the keypuncher for Sun Fu, a Chemical Engineer PhD. Sun Fu
would write an algorithm on the back on a piece or two of printout
paper. I would keypunch the algorithm in Fortran 66, get the subroutine
to compile cleanly on the time share Univac 1108, and deliver the card
deck to Sun Fu. Dad paid me $2.00 per hour. I should have paid him.

I knew Sun Fu well. He lived with us for a year in Houston in 1973
after he got his PhD at the University Of Oklahoma where my Dad had
taught Chemical Engineering from 1963 to 1968. Sun Fu was one of my
Dad's best grad students, an incredibly smart guy from mainland China.

Lynn

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Fri, 9 Jan 2026 19:33:27 -0600, Lynn McGuire wrote:

So I was the keypuncher for Sun Fu, a Chemical Engineer PhD. Sun Fu
would write an algorithm on the back on a piece or two of printout
paper. I would keypunch the algorithm in Fortran 66, get the
subroutine to compile cleanly on the time share Univac 1108, and
deliver the card deck to Sun Fu.

In the summer job I had at the end of my first year, I was writing
Fortran code to model sawing up a log for planks or peeling it for
plywood. I started by submitting coding forms to the data entry
operators, as per recommended practice, but the cards came back full
of typos (“1” for “I”, “O” for “0” etc).

There was a 129 keypunch in the room where you submitted jobs, which
was meant to be used by ordinary users for quick corrections only. I
started using it to punch all my own cards. Once the operators
realized I could do my program entry as fast as they could (and more accurately), we came to an agreement that I would give up my place
every few minutes and go round again, to give others in the queue a
chance.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

Lawrence D’Oliveiro <ldo@nz.invalid> writes:

Then I got hold of a book called “The Compleat Cybernaut”, by Anna
Burke Harris.

That's a great title. Amazon listing prices make it look like a real collector's item.

Later, I found another book, called “Programming in POP-2”.

That looks interesting, too. It looks like it's still available in the
form of POP-11/Poplog.
--
David Meyer
Takarazuka, Japan
papa@sdf.org
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

Lynn McGuire <lynnmcguire5@gmail.com> schrieb:

On 1/7/2026 9:52 PM, Lynn McGuire wrote:

Short documentary streamed in 1982 for the 25th anniversary of FORTRAN.

Documentary:
   https://www.youtube.com/watch?v=CXvyjbWkTyU

John Backus and other FORTRAN creators speaking about the development of
the language and the compiler.

Preserved written history of FORTRAN:
   https://softwarepreservation.computerhistory.org/FORTRAN/

Stolen from
   https://www.reddit.com/r/fortran/

Lynn

I started writing software in Fortran in 1975, thought it the best thing since sliced bread. Then I learned Pascal in 1983 (Turbo Pascal). Then
I learned C in 1986 (Turbo C).

I started out on my father's programmable calculator (Casio,
38 steps). Enough for factorization. Then, a better programmable
calculator, also Casio, the FX 602-P. In parallel, a C64 (starting
1983?) using Basic and a bit of assembler. Then, Turbo Pascal and
some Modula-2 on the Atari ST, plus Fortran as a student, starting
around 1987 (on IBM-compatible mainframes, later vector computers).
Around 1989, I got access to UNIX workstations and also worked in
C and various script languages, especially Perl.
--
This USENET posting was made without artificial intelligence,
artificial impertinence, artificial arrogance, artificial stupidity,
artificial flavorings or artificial colorants.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Sat, 10 Jan 2026 21:20:47 -0000 (UTC), Thomas Koenig wrote:

Around 1989, I got access to UNIX workstations and also worked in C
and various script languages, especially Perl.

What else was there in terms of “scripting” at the time?

In those days, “scripting” meant “interpreted”, which in turn meant “slow”. Perl managed to break the mould by compiling to an
intermediate byte-code form, which made it fast enough to be a real productivity booster for many common quick-and-dirty programming
tasks.

I would point to Perl as the start of the wave of “metaprogramming” languages (by which he meant “very-high-level” languages) that Fred
Brooks predicted in the final edition of his classic book “The
Mythical Man-Month”.

His own proposed example, AppleScript, is probably best forgotten.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

Lawrence D’Oliveiro <ldo@nz.invalid> schrieb:

On Sat, 10 Jan 2026 21:20:47 -0000 (UTC), Thomas Koenig wrote:

Around 1989, I got access to UNIX workstations and also worked in C
and various script languages, especially Perl.

What else was there in terms of “scripting” at the time?

I actually started out with awk, which preceded Perl (I did
not mention that because I mainly switched to Perl once it was
available, but I will use awk '/foo/ { print $1,$3 }' style of
scripts every now and then.

awk actually has the same logic as RPG, but is many orders of
magnitudes saner.

In those days, “scripting” meant “interpreted”, which in turn meant “slow”. Perl managed to break the mould by compiling to an
intermediate byte-code form, which made it fast enough to be a real productivity booster for many common quick-and-dirty programming
tasks.

I think the same applies to awk.
--
This USENET posting was made without artificial intelligence,
artificial impertinence, artificial arrogance, artificial stupidity,
artificial flavorings or artificial colorants.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Sat, 10 Jan 2026 21:47:15 -0000 (UTC), Thomas Koenig wrote:

Lawrence D’Oliveiro <ldo@nz.invalid> schrieb:

On Sat, 10 Jan 2026 21:20:47 -0000 (UTC), Thomas Koenig wrote:

Around 1989, I got access to UNIX workstations and also worked in
C and various script languages, especially Perl.

What else was there in terms of “scripting” at the time?

I actually started out with awk, which preceded Perl ...

Yeah, but Perl does everything awk does, just as concisely, and more.

So yes, at the time people were pushing awk as some kind of
general-purpose high-level language, we already had Perl.

Tcl might have been another possibility, but I think Perl had less bad
syntax. ;)
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

Lawrence D?Oliveiro <ldo@nz.invalid> wrote:

On Sat, 10 Jan 2026 21:20:47 -0000 (UTC), Thomas Koenig wrote:

Around 1989, I got access to UNIX workstations and also worked in C
and various script languages, especially Perl.

What else was there in terms of ?scripting? at the time?

awk and sed, called from shell scripts.

Of course if you blend shell, sed and awk, you get perl 4.
--
pa at panix dot com
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Fri, 9 Jan 2026 15:09:58 -0600, Lynn McGuire
<lynnmcguire5@gmail.com> wrote:

On 1/7/2026 9:52 PM, Lynn McGuire wrote:

Short documentary streamed in 1982 for the 25th anniversary of FORTRAN.

Documentary:
�� https://www.youtube.com/watch?v=CXvyjbWkTyU

John Backus and other FORTRAN creators speaking about the development of
the language and the compiler.

Preserved written history of FORTRAN:
�� https://softwarepreservation.computerhistory.org/FORTRAN/

Stolen from
�� https://www.reddit.com/r/fortran/

Lynn

Thank you very much for this nice, no-fluff video.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Fri, 9 Jan 2026 15:09:58 -0600, Lynn McGuire
<lynnmcguire5@gmail.com> wrote:

I started writing software in Fortran in 1975, thought it the best thing >since sliced bread. Then I learned Pascal in 1983 (Turbo Pascal). Then
I learned C in 1986 (Turbo C).

Lynn

I first came across Fortran (IV) in 1975 at the Indian Institute of
Technolgy (Bombay). I would say that we were were taught
"Theoretical Fortran" as we were taught the sytax in class and
our exercises, homework and tests consisted of code written
on paper and submitted for evaluation. No actual running of
the programmes as the IBM 360 was doing much more
important work! It was very much like learning maths, physics
or chemistry (but without any lab work!).

In 1976, I got into actual programming on an IBM 370 at
Indiana University. My bible was, and remained, Daniel
McCracken's " A Guide to Fortran IV Programming" till
F77 came by. Then, work took a turn and I got more
and more involved with PV Wave as the integrated
graphics was what was needed on the jo -- which
also meant dealing with Matlab. But, there were quite
a few problems that were still Fortran based till the end
of my hardcore, numberica, programming days around 2010.



--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Sun, 11 Jan 2026 20:24:21 -0000 (UTC), pa@see.signature.invalid
(Pierre Asselin) wrote:

Sorry to use this NG for a private message, but I would like say
hello to a long lost contact.



Pierre, Great to see your name again after many, many years.

I met you in Santa Barbara (around 1980, I guess) when you
were in AMC, working with Vince Kelly. I was a friend of Bob Gray
and Victor del Prado. I was doing my PhD at UCSB at that time
and we exchanged some email after I graduated and joined
Xerox. Victor and I sometimes talk about you and wonder
what you are up to.

Hope all is well with you. Happy New Year!

No, the email will not work, but do let me know if you would
like mine .... and the real name!





--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Tue, 13 Jan 2026 11:59:54 -0500, Charlie Roberts wrote:

Then, work took a turn and I got more and more involved with PV Wave
as the integrated graphics was what was needed on the jo -- which
also meant dealing with Matlab. But, there were quite a few problems
that were still Fortran based till the end of my hardcore,
numberica, programming days around 2010.

MATLAB is, or was, written in Fortran. Out of curiosity, I looked up
PV-Wave -- looks like the kind of thing you would do in NumPy these
days. Similarly in place of MATLAB you could use NumPy with
Matplotlib.

I did a GUI programming job in MATLAB once. Trying to fit GUI
functionality on top of a language primarily designed to work with arrays/matrices was ... an experience I would not like to repeat.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Tue, 13 Jan 2026 19:42:26 -0000 (UTC), Lawrence D�Oliveiro
<ldo@nz.invalid> wrote:

On Tue, 13 Jan 2026 11:59:54 -0500, Charlie Roberts wrote:

MATLAB is, or was, written in Fortran. Out of curiosity, I looked up
PV-Wave -- looks like the kind of thing you would do in NumPy these
days. Similarly in place of MATLAB you could use NumPy with
Matplotlib.

I believe that IDL/PV Wave also have their genesis in Fortran. While
the integrated graphics was the primary push for the partial
migration to PV Wave, the vector nature won a lot of engineers
towards that platform. Still, as I said, there were problems that
were for the Fortran (and later C++) people.

I did a GUI programming job in MATLAB once. Trying to fit GUI
functionality on top of a language primarily designed to work with >arrays/matrices was ... an experience I would not like to repeat.

Agreed, if you really want a very smooth GUI. The last real job
I did was for a real time data acquistion scanner and I had to
learn LabView. Though I ended up writing real code for the
most part, I did do some GUI. Much better than Matlab.

But, we are a long way from where this thread started. If I
had to write a quick programme now, I guess I will tend to
head towards Fortran!

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Tue, 13 Jan 2026 15:08:38 -0500, Charlie Roberts wrote:

If I had to write a quick programme now, I guess I will tend to head
towards Fortran!

If you want something good for “scratchpad” programming (maximum
results with just a few quick lines sof code), I would recommend
Jupyter notebooks. Built on Python, together with toolkits like NumPy,
Pandas, Matplotlib and a bunch of others.

You know how, back in the day, BASIC was famed for its immediacy? Boot
your PC straight into the BASIC ROM: type a line without a line number
and it executes at once. Put a line number on the front and it gets
added to your program in memory.

Jupyter gives you that same immediacy, only in a more advanced form --
each cell can hold multiple lines of code, and you don’t need line
numbers. And you have rich output -- graphics, sound, video -- and
also some basic interactive widgets with minimal effort.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

Around 1989, I got access to UNIX workstations and also worked in C
and various script languages, especially Perl.

What else was there in terms of “scripting” at the time?

One had OS-dependent scripting at least as far as early 1980s.

On HP RTE-6 VM there were "transfer files" with some two-letter control statements. On IBM VM/CMS one could write scripts in EXEC, then EXEC2,
and later REXX (REXX is still kicking and alive). On VAX VMS "command
files" where written in DCL (Digital Command Language).

I was exposed to all of them (and of course a lot of Fortran).
Cannot remember what was used on PDP 11, I had a limited exposure.

At least in astronomy, dedicated scripting languages were implemented in
some data analysis environents, usually mimicking those of the OS under
which they ran.

So e.g. ESO IHAP had something mimicking HP TR files, ESO MIDAS
mimicking VAX DCL, amd NOAO IRAF (running on VMS and Unix) a "cl" of its
own.
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Sun, 18 Jan 2026 17:06:35 +0100, LC's No-Spam Newsreading account
wrote:

One had OS-dependent scripting at least as far as early 1980s.

In those days, “scripting” meant “interpreted”, which in turn meant “slow”. Perl managed to break the mould by compiling to an
intermediate byte-code form, which made it fast enough to be a real productivity booster for many common quick-and-dirty programming
tasks.

I would point to Perl as the start of the wave of “metaprogramming” languages (by which he meant “very-high-level” languages) that Fred
Brooks predicted in the final edition of his classic book “The
Mythical Man-Month”.

His own proposed example, AppleScript, is probably best forgotten.
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On 1/18/2026 3:00 PM, Lawrence D’Oliveiro wrote:

On Sun, 18 Jan 2026 17:06:35 +0100, LC's No-Spam Newsreading account
wrote:

One had OS-dependent scripting at least as far as early 1980s.

In those days, “scripting” meant “interpreted”, which in turn meant “slow”. Perl managed to break the mould by compiling to an
intermediate byte-code form, which made it fast enough to be a real productivity booster for many common quick-and-dirty programming
tasks.

I would point to Perl as the start of the wave of “metaprogramming” languages (by which he meant “very-high-level” languages) that Fred Brooks predicted in the final edition of his classic book “The
Mythical Man-Month”.

His own proposed example, AppleScript, is probably best forgotten.

REXX could also be compiled, if desired.
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Sun, 18 Jan 2026 16:31:51 -0600, Gary Scott wrote:

REXX could also be compiled, if desired.

From what I can gather, REXX does everything with strings. I would put
it on a par with TCL, perhaps.

Perl was more advanced than that.
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On 1/10/2026 3:27 PM, Lawrence D’Oliveiro wrote:

On Sat, 10 Jan 2026 21:20:47 -0000 (UTC), Thomas Koenig wrote:

Around 1989, I got access to UNIX workstations and also worked in C
and various script languages, especially Perl.

What else was there in terms of “scripting” at the time?...

...

I used to play Lunar Lander on the teletype master console of the Univac
1108 that my Dad used back in 1970. Lunar Lander was written in BASIC.
https://en.wikipedia.org/wiki/Lunar_Lander_(video_game_genre)

Lynn

--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

LC's No-Spam Newsreading account <nospam@home.invalid> schrieb:

Around 1989, I got access to UNIX workstations and also worked in C
and various script languages, especially Perl.

What else was there in terms of “scripting” at the time?

One had OS-dependent scripting at least as far as early 1980s.

On HP RTE-6 VM there were "transfer files" with some two-letter control statements. On IBM VM/CMS one could write scripts in EXEC, then EXEC2,
and later REXX (REXX is still kicking and alive). On VAX VMS "command
files" where written in DCL (Digital Command Language).

JCL is also a script language, of sorts (the worst prorgramming
language of all times, according to Brooks).
--
This USENET posting was made without artificial intelligence,
artificial impertinence, artificial arrogance, artificial stupidity,
artificial flavorings or artificial colorants.
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On Sun, 25 Jan 2026, Thomas Koenig wrote:

One had OS-dependent scripting at least as far as early 1980s.

On HP RTE-6 VM there were "transfer files" with some two-letter control
statements. On IBM VM/CMS one could write scripts in EXEC, then EXEC2,
and later REXX (REXX is still kicking and alive). On VAX VMS "command
files" where written in DCL (Digital Command Language).

JCL is also a script language, of sorts

I had some exposure to JCL too on IBM mainframes (and to the equivalent "languag" on Univac 1100 ones), but I did not consider them script
languages (no interactivity, no obvious way of "passing parameters",
did it have branching ?),
My exposure to JCL occurred in the UK where MSSL used the computing
centre at RAL ... to solve the parametere passage issue they had an
additional layer called Electric written over JCL jobs, which allowed to change some parameters at submission time.
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

LC's No-Spam Newsreading account <nospam@home.invalid> schrieb:

On Sun, 25 Jan 2026, Thomas Koenig wrote:

One had OS-dependent scripting at least as far as early 1980s.

On HP RTE-6 VM there were "transfer files" with some two-letter control
statements. On IBM VM/CMS one could write scripts in EXEC, then EXEC2,
and later REXX (REXX is still kicking and alive). On VAX VMS "command
files" where written in DCL (Digital Command Language).

JCL is also a script language, of sorts

I had some exposure to JCL too on IBM mainframes (and to the equivalent "languag" on Univac 1100 ones), but I did not consider them script
languages (no interactivity, no obvious way of "passing parameters",
did it have branching ?),

You can write procedures in JCL, with parameters. You can also
skip steps based on the condition code of previous return codes,
but the logic is horribly backwards. And you can submit new
jobs from JCL to an "internal reader" via SYSOUT=(A,INTRDR), so
looping is possible, in principle.

Fred Brooks, who should know, called it "the worst computer
programming language ever devised by anybody, anywhere", adding
that as the /360 project lead, he was responsible, but that there
was more than enough blame to go around.

My exposure to JCL occurred in the UK where MSSL used the computing
centre at RAL ... to solve the parametere passage issue they had an additional layer called Electric written over JCL jobs, which allowed to change some parameters at submission time.

I once wrote a JCL script which would test-compile a Fortran program
on an IBM 3090, then cross-submit a job to a Fujitsu vector computer
if the compilation succeeded, to save costly CPU time on the vector
computer in case of syntax errors. I think I also included
some output processing back on the IBM 3090, submitting a job
backwards.
--
This USENET posting was made without artificial intelligence,
artificial impertinence, artificial arrogance, artificial stupidity,
artificial flavorings or artificial colorants.
--- Synchronet 3.21b-Linux NewsLink 1.2

From Newsgroup: comp.lang.fortran

On 1/27/2026 11:43 AM, Thomas Koenig wrote:

LC's No-Spam Newsreading account <nospam@home.invalid> schrieb:

On Sun, 25 Jan 2026, Thomas Koenig wrote:

One had OS-dependent scripting at least as far as early 1980s.

On HP RTE-6 VM there were "transfer files" with some two-letter control >>>> statements. On IBM VM/CMS one could write scripts in EXEC, then EXEC2, >>>> and later REXX (REXX is still kicking and alive). On VAX VMS "command
files" where written in DCL (Digital Command Language).

JCL is also a script language, of sorts

I had some exposure to JCL too on IBM mainframes (and to the equivalent
"languag" on Univac 1100 ones), but I did not consider them script
languages (no interactivity, no obvious way of "passing parameters",
did it have branching ?),

You can write procedures in JCL, with parameters. You can also
skip steps based on the condition code of previous return codes,
but the logic is horribly backwards. And you can submit new
jobs from JCL to an "internal reader" via SYSOUT=(A,INTRDR), so
looping is possible, in principle.

Fred Brooks, who should know, called it "the worst computer
programming language ever devised by anybody, anywhere", adding
that as the /360 project lead, he was responsible, but that there
was more than enough blame to go around.
snip

That would be my opinion of JCL. Hated it, used it only where
absolutely necessary.
--- Synchronet 3.21b-Linux NewsLink 1.2