Re: [vcf-midatlantic] COBOL
Something I've noticed being overlooked whenever this topic comes up (what, several times a year): the industry *has* been trying to migrate off COBOL for as long as I've been in the game (30+ years). At one point, the "4GL" suites were going to let you capture all that business logic in a 'better' language. Most are all long forgotten and barely lamented. PowerBuilder, anyone? Then IBM decided the answer to COBOLness was, of all things, Smalltalk. They dumped all kinds of resources into retreading COBOL programmers into object-oriented quiche-eaters. A glorious experiment; I think pretty much the only legacy is Object COBOL and a lot of programmers that really, really hate the whole OO paradigm. Then about a decade ago I guess, Java was going to put the final stake in COBOLs heart. Jury still out there, I suppose. I do see a *lot* of business apps for conservative, disciplined, COBOL-esque IT shops being done in Java. Not sure they're replacements, tho. I've even heard of real projects to migrate COBOL to Ada (reasonable), Visual BASIC (icky) and C (kill it with fire). I'm sure over time all these 'alternatives' eroded away the *number* of COBOL apps in production; after all, if you've got some smallish, auxiliary app why not make it easier for more people to be competent to maintain. And there's an entire ecosystem of products that let you put a nice UI in front of what is on the back-end a block-oriented, 3270 terminal abstraction. All programmable in VB, Javascript, etc. What seems clear, however, is that little migration of business critical, high volume, high reliability transaction processing applications that are COBOL/Mainframe core competencies is being seriously undertaken. P.S. - If you're really interested in learning COBOL academically, take a peek at Columbus State University TSYS School of CS (https://goo.gl/0cx6qH). You can get a COBOL/Mainframe oriented certificate or BS degree. It's a joint venture of sorts between the University System of Georgia, IBM and TSYS, who is one of the largest credit card processors in the world. This is a tangible answer to "not enough new COBOL programmers". P.P.S. - I'm currently an IT executive at the 5th largest bank in the US. Yes we use COBOL. No, it's not going away any time soon. They're hiring, last I checked. KJ
What seems clear, however, is that little migration of business critical, high volume, high reliability transaction processing applications that are COBOL/Mainframe core competencies is being seriously undertaken. Not surprising. I assume most of those systems are too critical to a) risk replacing with something that may or may not be as reliable, and b) to even shut off for the few seconds/minutes/hours it would take to switch to the new system.
On Tue, May 23, 2017 at 5:00 PM, Kenneth Seefried via vcf-midatlantic < vcf-midatlantic@lists.vintagecomputerfederation.org> wrote:
Something I've noticed being overlooked whenever this topic comes up (what, several times a year): the industry *has* been trying to migrate off COBOL for as long as I've been in the game (30+ years).
At one point, the "4GL" suites were going to let you capture all that business logic in a 'better' language. Most are all long forgotten and barely lamented. PowerBuilder, anyone?
Then IBM decided the answer to COBOLness was, of all things, Smalltalk. They dumped all kinds of resources into retreading COBOL programmers into object-oriented quiche-eaters. A glorious experiment; I think pretty much the only legacy is Object COBOL and a lot of programmers that really, really hate the whole OO paradigm.
Then about a decade ago I guess, Java was going to put the final stake in COBOLs heart. Jury still out there, I suppose. I do see a *lot* of business apps for conservative, disciplined, COBOL-esque IT shops being done in Java. Not sure they're replacements, tho.
I've even heard of real projects to migrate COBOL to Ada (reasonable), Visual BASIC (icky) and C (kill it with fire). I'm sure over time all these 'alternatives' eroded away the *number* of COBOL apps in production; after all, if you've got some smallish, auxiliary app why not make it easier for more people to be competent to maintain. And there's an entire ecosystem of products that let you put a nice UI in front of what is on the back-end a block-oriented, 3270 terminal abstraction. All programmable in VB, Javascript, etc.
What seems clear, however, is that little migration of business critical, high volume, high reliability transaction processing applications that are COBOL/Mainframe core competencies is being seriously undertaken.
P.S. - If you're really interested in learning COBOL academically, take a peek at Columbus State University TSYS School of CS (https://goo.gl/0cx6qH). You can get a COBOL/Mainframe oriented certificate or BS degree. It's a joint venture of sorts between the University System of Georgia, IBM and TSYS, who is one of the largest credit card processors in the world. This is a tangible answer to "not enough new COBOL programmers".
P.P.S. - I'm currently an IT executive at the 5th largest bank in the US. Yes we use COBOL. No, it's not going away any time soon. They're hiring, last I checked.
KJ
participants (2)
-
Drew Notarnicola -
Kenneth Seefried