NJ seeking help from COBOL programmers in coronavirus fight

[corbe]

NJ seeking help from COBOL programmers in coronavirus fight

By Kaelan Deese - 04/07/20 10:36 AM EDT

The state of New Jersey is seeking volunteers with knowledge of how to code COBOL to aid in the coronavirus outbreak, according to the governor's request on Monday.

COBOL is a much older programming language than what most people are familiar with today. The coding language was developed in the 1950s alongside efforts with the Department of Defense, according to CNBC.

Most of the programmers today are familiar with more contemporary languages such as Java, HTML, or CSS, but COBOL is still used in financial applications for larger government agencies or business purposes.

<..snip..>

https://thehill.com/homenews/news/491526-nj-seeking-help-from-cobol-programmers-in-coronavirus-fight

[Cyber Liberty]

Good luck with that.  The last time there was an "urgent" demand for COBOL coders was in Y2K, and they were hard to find then.  Now it's two decades down the road.

[Polly Ticks]

IF user_COBOL_language_flag = 'Y'
     THEN move "NJ State Employee" to user_job_title
          ELSE perform 999-NJ-Is-Screwed
END-IF.

[PeteS in CA]

As dinosaurian computer languages go, COBOL is probably from the Triassic era. But it's also an example of, "If it's not broke, don't fix it." COBOL became so useful and ubiquitous that users kept using their existing programs instead of reinventing the wheel.

Thing is, at some point, probably well over a decade ago, users should have rewritten, verified, and alpha-tested functions written in COBOL before user volume or other stresses "broke" the COBOL program, resulting in an emergency situation in which haste would almost certain cause waste, if not chaos. I seriously doubt NJ is the only state or institution that let this major update effort slide.

As I posted long ago on a techie discussion site, by the time the can everybody kept kicking down the road hits the wall, the people who kept kicking it down the road are in higher offices or retired, where they won't have to deal with the problem of the can.

[PeteS in CA]

https://www.cnn.com/2020/04/08/business/coronavirus-cobol-programmers-new-jersey-trnd/index.html

Wanted urgently: People who know a half century-old computer language so states can process unemployment claims
By Alicia Lee, CNN
Updated 4:00 PM ET, Wed April 8, 2020

(CNN) On top of ventilators, face masks and health care workers, you can now add COBOL programmers to the list of what several states urgently need as they battle the coronavirus pandemic.

In New Jersey, Gov. Phil Murphy has put out a call for volunteers who know how to code the decades-old computer programming language called COBOL because many of the state's systems still run on older mainframes.

In Kansas, Gov. Laura Kelly said the state's Departments of Labor was in the process of modernizing from COBOL but then the virus interfered. "So they're operating on really old stuff," she said.

Connecticut has also admitted that it's struggling to process the large volume of unemployment claims with its "40-year-old system comprised of a COBOL mainframe and four other separate systems."
...
For instance, with more than 362,000 New Jersey residents filing for unemployment in the past two weeks, the 40-year-old mainframes that process those claims are being overloaded.

Coders have moved away from the aging language

COBOL, which stands for Common Business Oriented Language, is a computer programming language that was developed back in 1959, according to the National Museum of American History.

"It's a programming language that was used to create a very significant percentage of business systems over the period of the 60s, 70s and even into the 80s," Joseph Steinberg, an expert on cybersecurity, told CNN.

It's not just the State of New Jersey that is running into resource limitations in their systems running COBOL applications.

[Cyber Liberty]

https://www.cnn.com/2020/04/08/business/coronavirus-cobol-programmers-new-jersey-trnd/index.html

Wanted urgently: People who know a half century-old computer language so states can process unemployment claims
By Alicia Lee, CNN
Updated 4:00 PM ET, Wed April 8, 2020
It's not just the State of New Jersey that is running into resource limitations in their systems running COBOL applications.

One would think that after Y2K, where COBOL was the source of much of the angst, this stuff would have been re-written.

[roamer_1]

I am too young at it for COBOL

But one day, the hue and cry will be raised for PASCAL,
And I shall rise like a phoenix from the ashes!

[Cyber Liberty]

I am too young at it for COBOL

But one day, the hue and cry will be raised for PASCAL,
And I shall rise like a phoenix from the ashes!

I like compiled programs for industrial use.  Interpreted code, like Basic, always bugged me because it can be so...damned......slow.

I used C and an interpreted in-house script code, and I really like the way the C tester ran, compared to the other.  Rocky Mountain Basic was popular for test equipment, but I never liked it.

[roamer_1]

I like compiled programs for industrial use.  Interpreted code, like Basic, always bugged me because it can be so...damned......slow.

I used C and an interpreted in-house script code, and I really like the way the C tester ran, compared to the other.  Rocky Mountain Basic was popular for test equipment, but I never liked it.

I confess, my love is in-line scripting.
That's probably why I am such a batch queen. The freaky stuff I can do with simple batch on a cmd line probably spoiled me for serious programming. Even as a Delphi freak, I tend to hate object oriented programming - Sooper fat, and sooper slow. I like the tight code I can make line by line.

I like basic alright, but I was drawn away into pascal early on, pascal being more friendly to me, hence my winding up in the dead end that is now Delphi.

I should have learned assembly. I have just enough of it to do what I need in batch (writing debug on the fly)... Now there;s tight code. Those tiny little 64k programs that are soooo powerful.

[Cyber Liberty]

I confess, my love is in-line scripting.
That's probably why I am such a batch queen. The freaky stuff I can do with simple batch on a cmd line probably spoiled me for serious programming. Even as a Delphi freak, I tend to hate object oriented programming - Sooper fat, and sooper slow. I like the tight code I can make line by line.

I like basic alright, but I was drawn away into pascal early on, pascal being more friendly to me, hence my winding up in the dead end that is now Delphi.

I should have learned assembly. I have just enough of it to do what I need in batch (writing debug on the fly)... Now there;s tight code. Those tiny little 64k programs that are soooo powerful.

PASCAL encourages linear programming.  I knew test equipment programmers who were "subroutine happy" in BASIC.  They branch off to subroutines because only three lines of code repeat three times in the whole bloody program.

As a line tech for a testing floor, I got to know the programmers without having to look at the comments section.  They had "fingerprints."

[roamer_1]

PASCAL encourages linear programming.  I knew test equipment programmers who were "subroutine happy" in BASIC.  They branch off to subroutines because only three lines of code repeat three times in the whole bloody program.

As a line tech for a testing floor, I got to know the programmers without having to look at the comments section.  They had "fingerprints."

Yeah... Throw in a 50mg module for four lines of code... I HATE that. HATE it. And now it's just part of the furniture. 

[Gefn]

What you’re talking about makes as much sense to me as the Latin I attempted to learn a few years back.

I’m proud to be a Luddite!

[Cyber Liberty]

Yeah... Throw in a 50mg module for four lines of code... I HATE that. HATE it. And now it's just part of the furniture. 

I learned to deal with it.  I knew that when it was my turn to code, I'd do it differently.  My boss loved it.  Remember ISO9000?  It was a breeze for me because my boss insisted on my in-code commenting and a good header with all revision/pedigree information.  Everything traced.  In fact, I underwent several code audits for ISO, and I never even knew about it.  Nobody had to ask me anything because it was all right there embedded in the source code.

[roamer_1]

I learned to deal with it.  I knew that when it was my turn to code, I'd do it differently.  My boss loved it.  Remember ISO9000?  It was a breeze for me because my boss insisted on my in-code commenting and a good header with all revision/pedigree information.  Everything traced.  In fact, I underwent several code audits for ISO, and I never even knew about it.  Nobody had to ask me anything because it was all right there embedded in the source code.

YEP me too. I comment my code exhaustively too - for my own benefit, generally, as I will be going back in sooner or later, and notes are invaluable. Always wrote a readme and a help file too. Even for tiny little cmdline stuff.

[DB]

I confess, my love is in-line scripting.
That's probably why I am such a batch queen. The freaky stuff I can do with simple batch on a cmd line probably spoiled me for serious programming. Even as a Delphi freak, I tend to hate object oriented programming - Sooper fat, and sooper slow. I like the tight code I can make line by line.

I like basic alright, but I was drawn away into pascal early on, pascal being more friendly to me, hence my winding up in the dead end that is now Delphi.

I should have learned assembly. I have just enough of it to do what I need in batch (writing debug on the fly)... Now there;s tight code. Those tiny little 64k programs that are soooo powerful.

I've done a lot of assembly language programming. From controllers and CPUs to DSPs. Even written my own assemblers way back when. When the math coprocessor first became available in the IBM PC-1 I wrote all the matrix math functions using the 8087 in assembly language for a linear circuit analysis program. There wasn't any higher level support at the time. It ran about 7 times faster than what could be done on an 8086 alone. Sold the program to others for awhile. It is what launched my current business.

Assembly is fast, tight and small but very difficult to follow... Heavily pipelined instructions sets made DSP particularly challenging.

Outside of assembly language my experience is pretty limited. My primary experience is in Quick Basic, Forth, Visual Basic (up to version 6.0), FreeBASIC (something that compiles to a single exe and is free to use and runs on about anything. It also executes fast) and C for embedded systems. Delphi was much superior to Visual Basic compiling to an exe without tons of DLLs requiring a complicated installation process.

I program a lot in VHDL these day but it isn't a general purpose programming language. It is used to implement complex high speed logic in FPGAs.

[240B]

Never had time for comments or documentation.
Every thing was such a frenzy with constantly changing requirements and impossible deadlines.
And as soon as one project was nearing the end, they would throw two more at you.
It really was a fustercluck, with no time to think. It was always DO IT NOW! GET IT DONE!
And management did not care about comments. The coders just had to figure it out. That's your job.

[Cyber Liberty]

I've done a lot of assembly language programming. From controllers and CPUs to DSPs. Even written my own assemblers way back when. When the math coprocessor first became available in the IBM PC-1 I wrote all the matrix math functions using the 8087 in assembly language for a linear circuit analysis program. There wasn't any higher level support at the time. It ran about 7 times faster than what could be done on an 8086 alone. Sold the program to others for awhile. It is what launched my current business.

Assembly is fast, tight and small but very difficult to follow... Heavily pipelined instructions sets made DSP particularly challenging.

Outside of assembly language my experience is pretty limited. My primary experience is in Quick Basic, Forth, Visual Basic (up to version 6.0), FreeBASIC (something that compiles to a single exe and is free to use and runs on about anything. It also executes fast) and C for embedded systems. Delphi was much superior to Visual Basic compiling to an exe without tons of DLLs requiring a complicated installation process.

I program a lot in VHDL these day but it isn't a general purpose programming language. It is used to implement complex high speed logic in FPGAs.

I cut my teeth on 8080/Z80 machine code in 1980 Tech School (Assembly was for beginners).  I was good at it, but never used it again.  My first programming class was Basic, in 1976.  We had a PDP10 mainframe (time-shared) to work with.  It took up a whole room, with several stacks of 1Mb hard drives.  Heady stuff!

[Cyber Liberty]

Never had time for comments or documentation.
Every thing was such a frenzy with constantly changing requirements and impossible deadlines.
And as soon as one project was nearing the end, they would throw two more at you.
It really was a fustercluck, with no time to think. It was always DO IT NOW! GET IT DONE!
And management did not care about comments. The coders just had to figure it out. That's your job.

And that was why I was glad to have a boss who insisted on in-code comments.  He didn't bitch about the extra time, and it kept my ass away from prying auditors.

[Weird Tolkienish Figure]

I've done a lot of assembly language programming. From controllers and CPUs to DSPs. Even written my own assemblers way back when. When the math coprocessor first became available in the IBM PC-1 I wrote all the matrix math functions using the 8087 in assembly language for a linear circuit analysis program. There wasn't any higher level support at the time. It ran about 7 times faster than what could be done on an 8086 alone. Sold the program to others for awhile. It is what launched my current business.

Assembly is fast, tight and small but very difficult to follow... Heavily pipelined instructions sets made DSP particularly challenging.

Outside of assembly language my experience is pretty limited. My primary experience is in Quick Basic, Forth, Visual Basic (up to version 6.0), FreeBASIC (something that compiles to a single exe and is free to use and runs on about anything. It also executes fast) and C for embedded systems. Delphi was much superior to Visual Basic compiling to an exe without tons of DLLs requiring a complicated installation process.

I program a lot in VHDL these day but it isn't a general purpose programming language. It is used to implement complex high speed logic in FPGAs.

Coding is assembly is prone to spaghetti code devices like goto's and such, there's no upper level language contructs like branching, and stack addressing. And a lot of assembly programmers basically coded their own versions of these meaning that even assembly coders preferred upper level coding, just their own version of it. Some assemblers had macros that, again, turned assembly into upper level language.

It's useful to know for reverse engineering.

[DB]

I cut my teeth on 8080/Z80 machine code in 1980 Tech School (Assembly was for beginners).  I was good at it, but never used it again.  My first programming class was Basic, in 1976.  We had a PDP10 mainframe (time-shared) to work with.  It took up a whole room, with several stacks of 1Mb hard drives.  Heady stuff!

My first experience with programming was also a Z80. Late 70's I think. I wire wrapped a Z80 CPU to 128 bytes of RAM and fuse link ROM... I think I still have it... The code was written on a large piece of graph paper with all the Z80 mnemonics converted to their hexadecimal values. The jump offsets were calculated by hand too... The good old days... It worked...

[DB]

Coding is assembly is prone to spaghetti code devices like goto's and such, there's no upper level language contructs like branching, and stack addressing. And a lot of assembly programmers basically coded their own versions of these meaning that even assembly coders preferred upper level coding, just their own version of it. Some assemblers had macros that, again, turned assembly into upper level language.

It's useful to know for reverse engineering.

My first macro assembler was for the IBM PC 8086 (really an 8088).

[Polly Ticks]

Never had time for comments or documentation.
Every thing was such a frenzy with constantly changing requirements and impossible deadlines.
And as soon as one project was nearing the end, they would throw two more at you.
It really was a fustercluck, with no time to think. It was always DO IT NOW! GET IT DONE!
And management did not care about comments. The coders just had to figure it out. That's your job.

True story.  And it made for 'interesting' times when what little documentation actually existed was destroyed when the towers came down in New York.  There was a LOT of time and effort that went into trying to reverse-engineer some of that old spaghetti code.