COBOL

[kevin Abel]

I may have to look at some old COBOL for QA work.
The closest language to it that I used was called JOSS in 1979.

Has anyone here worked with these languages?

THanks,

Kevin

[Paul Clapham]

I worked with a COBOL program once. I may even have made changes which went into production but I can't remember for sure.

[Paul Clapham]

... This was back in the days when programmers were expected to pick up a new language with no notice. Of course I have to admit it was earlier to do that back then.

[Tim Holloway]

Picking up a new language is easy. Picking up its philosophy and infrastructure is what's harder. Philosopy was always there, LISP takes a lot different kind of thought than COBOL, for example, We think lambdas are cool as a new feature in Java, but LISP was built on lambdas.

JOSS is a military dialect of COBOL and was one of the myriad of custom languages that Ada was intended to replace.

The salient features of COBOL are:

1. Pretends to be English so that Pointy-Haired Bosses can think they understand the code.

2. As originally implemented, was a pure procedural language with the entire program generally being in a single source file (build systems weren't yet a thing, much less Structured Programming or OOP).

3. Can really stress your carpal tunnel syndrome. Instead of "A = B + 7" you code "ADD 7 TO B GIVING A". Verbosity, thy name is COBOL.

4. Interfacing to databases and online transactions systems is generally cumbersome. COBOL programs generally only did I/O using magnetic tape (150MB 9-track reels, for example), Disk a/k/a DASD (350 MB was a huge disk back then). And, of course, punched cards. LOTS of punched cards.

If you want to play with COBOL on a Linux machine, install the gnucobol standard package along with the gcc compiler. The gnucobol program translates COBOL into C and then gcc compiles the C code.

Or, if you want the full vintage experience, you could run the original IBM D COBOL compiler under DOS/VS os OS?VS on the Project Hercules mainframe emulator. But operating a mainframe console is a challenge in its own right.

[Paul Clapham]

Tim Holloway wrote:Pretends to be English so that Pointy-Haired Bosses can think they understand the code.

It didn't work then and even though many languages in the ensuing 60+ years have explicitly made that claim, it didn't work for them either.

[kevin Abel]

Paul,

I remember being in Highschool(1975 - 1979)and thinking computers would be fun.  We first had a machine that read in punch cards.  I don't remember the output.  The machine cost the school about $35,000 which was probably 250K now.  I didn't understand it well.

Then I saw Cobol with punch cards and I didn't find it interesting either but I'm glad I learned some of it because I ran into a project at DeVry University where I entered cards describing an electronics circuit using an application called Spice.  I understood enough of JCL and keeping my cards in order to get it to work.

I took some COBOL also in Highschool but I couldn't figure out why they declared variable types at the top of the code.  There were Letters and numbers that described the variable type and it all seemed cryptic and it annoyed me so much I went away from it.

Kevin

[kevin Abel]

Tim,

I had a machine called Pecos One by APF.   16K of RAM and 40K of ROM.

It used JOSS.   I liked how the syntax followed English with Subjects, objects, adjectives and verbs.   I thought it was just for my machine.  If I knew the military was using it I  would have tried to find work with it.
I notice JOSS and COBOL both use the keyword DISPLAY for showing text on the monitor.  

I wonder if JOSS is still being used anyplace.

Kevin

[Tim Holloway]

I don't think $35,000 would get much of a computer back then. In 1978 our PR1ME minicomputer was valued at about $100K. You may have simply had a remote terminal. Those were popular back then. At University we had an IBM 3780 remote workstation linked via leased line to a System/370 mainframe. It had a card reader and 1443 line printer, for example. No interactive consoles, just batch processing.

I was incorrect about JOSS. I confused it with JOVIAL. JOSS is more similar to BASIC.

Although if you dig hard enough, you can generally find an implementation of virtually any ancient language — and emulators for a lot of obsolete processors — I didn't find anything immediate on JOSS, and unfortunately, as an acronym JOSS has been recycled at least twice, making the search more challenging.

On the other hand, JOSS is an ancestor of FOCAL (https://github.com/maurymarkowitz/RetroFOCAL) and of MUMPS. FOCAL was very popular for a while, and MUMPS was actually a hirable skill for years. I would not be surprised if there are still shops running MUMPS in production to this day. http://www.cs.uni.edu/~okane/

You might find this illuminating, too: http://bitsavers.trailing-edge.com/pdf/rand/joss/P-3486_JOSS_Introduction_To_The_System_Implementation_Nov66.pdf

[Peter Rooke]

Arh COBOL   Way back at work, I was a C programmer, but the part-time college course was based around COBOL.  

candygrammar: n.
A programming-language grammar that is mostly syntactic sugar; the term is also a play on ‘candygram’. COBOL, Apple's Hypertalk language, and a lot of the so-called ‘4GL’ database languages share this property. The usual intent of such designs is that they be as English-like as possible, on the theory that they will then be easier for unskilled people to program. This intention comes to grief on the reality that syntax isn't what makes programming hard; it's the mental effort and organization required to specify an algorithm precisely that costs. Thus the invariable result is that ‘candygrammar’ languages are just as difficult to program in as terser ones, and far more painful for the experienced hacker.

Taken from --> Cat B CandyGramma

[kevin Abel]

Tim,

I learned JOVIAL about 35 years ago.   It seemed pretty good for its time.   The acronym stands for Josh's own version in a language.  Maybe it's not Josh and another name.

I enjoyed JOSS.   I think it was before BASIC I think.

Kevin

[Campbell Ritchie]

Tim Holloway wrote:I don't think $35,000 would get much of a computer back then. . .  You may have simply had a remote terminal. Those were popular back then. . . .

I remember them well. But surely a remote terminal would have cost well under $35,000.

[Tim Holloway]

Campbell Ritchie wrote:

Tim Holloway wrote:I don't think $35,000 would get much of a computer back then. . .  You may have simply had a remote terminal. Those were popular back then. . . .

I remember them well. But surely a remote terminal would have cost well under $35,000.

I would not count on it. The IBM 3780 Binary Synchronous Remote Data Terminal System was comprised of up to 3 optional units, each about the size of an office desk: Card reader, card punch,, and IBM's 1443 typebar printer, capable of 425 line-per-minute output. At school we had the reader/printer setup with a control unit in the reader module.

There wasn't a lot of empty space in there and a lot of it was heavy-duty precision mechanical parts. Plus, if it's like a lot of similar IBM devices, the control unit was probably an IBM 1400-series small mainframe using their unique Solid Logic (transistor) Technology. Microprocessors were yet to be invented.

And, most critically, back then, a computer or computer peripheral wasn't priced based on the cost of its components or labour, but on whatever IBM could gouge you for. Which they would freely do. I suspect that even the keypunch machines ran over $10,000 1970's dollars each. So likely a minimum of £5000 newly-decimalised pounds.

The CRT terminal system units were practically brain-dead. All the smarts were in their shared control units. I remember being all excited [i]circa[i] 1983 when the first terminal systems using Intel 8080 and Z-80 processors were installed at work.

We're very spoiled when it comes to computer equipment pricing these days. I just saw a 12 TERAbyte SSD advertised fot USD $99. It's hard to believe that back when a 5 MEGAbyte disk storage was state-of-the-art that it would cost about the same as a small-to-medium house.

And then there's my Raspberry Pi collection. Most of them were $35 each and likely every one of them more powerful than IBM's top-of-the-line System 370 Model 165 with the water chiller cooling system that everyone salivated over in the early 1980s. And the Pi comes pre-configured with vastly more RAM.  An IBM System/370 Model 145 which was about the midrange for that series could run from $705,775 to $1,783,000 and maxed out at 512KB of RAM. So you see, no one ever really needed more that 640 KB.

[Campbell Ritchie]

Tim Holloway wrote:. . . wasn't priced based on the cost of its components or labour, but on whatever [XXX] could gouge you for. . . .

Hahahahahahahahahahahaha!

The CRT terminal system units were practically brain-dead. . . .

I remember some of them, too.

. . . back when a 5 MEGAbyte disk storage . . . would cost about the same as a small-to-medium house. . . .

In those days, you could have bought a small‑to‑medium street round here for that.

[Peter Rooke]

Before my time I started when the "green screens" of VT200 Unix terminals had mostly started to replace the older mainframes.
Of course, it's been my experience that if you dig deep enough you often find a mainframe somewhere with COBOL, VAX/VMS, and who knows what else!

“But you're out of your mind,” they said with a shrug.
“The customer's happy; what's one little bug?”

But he was determined. The others went home.
He spread out the program, deserted, alone.

The cleaning men came. The whole room was cluttered
With memory-dumps, punch cards. “I'm close,” he muttered.

The mumbling got louder, simple deduction,
“I've got it, it's right, just change one instruction.”

It still wasn't perfect, as year followed year,
And strangers would comment, “Is that guy still here?”

He died at the console, of hunger and thirst.
Next day he was buried, face down, nine-edge first.

And the last bug in sight, an ant passing by,
Saluted his tombstone, and whispered, “Nice try.”

Taken from GNU humor - credited to Lou Ellen Davis.

[Tim Holloway]

The last large-enterprise employer I had had zero in-house mainframes of any brand. They did tap into 2 external mainframe service systems, one of which was with a company I worked for for well over a decade, in fact. One of my jobs was to maintain a data import system that did ETL services to our local servers (originally Solaris, later Red Hat). Alas, independent ETL tools were not yet available, so we had to do everything the hard way. The fact that I was fluent in EBCDIC and packed-decimal data formats was an asset.

Of course, any respectable corporation knows what to do with assets — you liquidate them. So the next 3-year fire-everyone round I was out on the street, again. And that's when I swore off working for Big Business.

[Liutauras Vilda]

kevin Abel wrote:in 1979

At that time I had no business on this planet.

[kevin Abel]

At that time I had no business on this planet.

We are Glad you eventually showed up.