Enterprise

COBOL Today and Tomorrow, Part 1

It’s no secret that Linux at the server level is growing fast in the business world.

It’s stable, it’s infinitely customizable, it has the kind of support that an enterprise demands and its return on investment is amazing to people more accustomed to the costs of other approaches. So, you find it everywhere these days.

You discover something else everywhere, too: common business-oriented language, or COBOL, the computer’s original business language.

Not a Dying Language

The acronym COBOL has its share of humorous expansions (“completely obsolete, burdensome old language” is one) but if you think it’s dying, take another look.

A Gartner study found COBOL in about 75 percent of enterprise business processes, acting as the solid, transaction-focused foundation for the world’s core business processes.

People at the user level don’t see it — when they use these processes, it’s generally through interfaces written in C++, Java, .NET or some other newer language. However, it’s there — like the engine under the hood that powers the car.

The New Latin

A recent LinuxInsider article described COBOL as “the new Latin” because of its pervasive influence on computing. That’s a good description and a compelling reason for developers to get acquainted with it.

However, Gartner reported the COBOL code base is growing at more than 5 billion lines a year. Let me repeat myself: 5 billion. That’s not dead — it’s thriving.

Why COBOL Matters

OK, COBOL exists — so what?

Here’s what: COBOL does much more than merely exist. It is a basic fact of business life. Without it, global enterprises would look so different, they’d be unrecognizable.

Additionally, today’s mainstay COBOL developers are starting to look at retirement, and some have used that fact to push wholesale conversion projects on alarmed companies. However, it’s not really a crisis and, as industry makes known its need for replacement COBOL programmers, we need not expect a serious shortage to develop.

After all, someone has to maintain that 75 percent of business processes in COBOL and develop the 5-plus billion lines of new supporting code each year.

New Directions

In addition, COBOL is taking new directions to extend and amplify its legendary business effectiveness.

For example, Micro Focus has just introduced a suite that adapts COBOL processes to service oriented architectures (SOA). There’s no need to reinvent stable, debugged software in new languages or migrate to new hardware, absent other compelling reasons to take the hard way.

System migration is not a walk in the park, after all. It can take huge expenditures of man-hours to analyze, rewrite, debug, redeploy and then further debug core processes that had long since been thoroughly debugged and made stable in COBOL.

Why go through all of that if you don’t need to?

More COBOL Developers, Please

Still, the two trends I’ve noted — developers retiring and COBOL processes evolving and extending — make the growing call for COBOL developers inevitable.

If you’re proficient with Linux, know servers and can also support COBOL applications, you’re likely to pull a premium in the IT job market these days.

There’s an irony in the predicted shortage of COBOL developers, because COBOL is remarkably easy to learn. It’s a plain-English programming language that uses clear phrases to carry out real-world business data processes.

If you know C++ or Java, you can probably become productive with COBOL in four to six weeks. Put it on your resume and watch your opportunities grow.

The extra coins that will then drop into your pocket will track back to the misperception that COBOL is fading. Maybe that seemed true for awhile after the Y2K excitement, but not anymore.

COBOL is simply too fundamental and too essential to the world’s business to go away.

Where You Find COBOL

COBOL, perhaps the first high-level language, is pervasive because of its hard focus on essential business processes. It’s just about everywhere businesses conduct transactions, from finance to flexible phone rates to administrative systems for companies and governments.

Every time you go to an ATM, you touch COBOL somewhere along the way. Credit card billing, online banking, business forecasting, inventory control, government processes, even just calling your spouse — solid, invisible COBOL is there.

COBOL’s flexibility allows it to evolve with the times and meet the changing needs of commerce.

A computer language developed in the 1950s, in wide use by the early 1960s and one of the early computer standards by the early 1970s, COBOL can still operate the interlinked ATMs of the post-1990s and meet the business reporting and international data flow requirements of the 2000s.

Just as COBOL’s evolution continues — object-oriented COBOL was added to the COBOL standard in 2002, for example — so does the business world’s need for qualified developers to support it.

Grandma COBOL

It’s appropriate in a way to celebrate the vitality of COBOL now, having just observed the centennial of the birth of its chief architect,Grace Murray Hopper.

Born December 9, 1906, Hopper was a rarity during her time — she earned a Ph.D. in mathematics from Yale University and was a faculty member at both Vassar College and Harvard University. A rear admiral in the U.S. Navy by special congressional act, when she retired with 40 years of naval service she was widely, and fondly, known as “Grandma COBOL.”

Hopper was at her peak during the early days of computing. Working on the electromechanical Harvard Mark II computer at the university in 1947, she came up with — or at least popularized — the term “computer bug” after a moth got caught in a relay. The Smithsonian Institution still has the original moth, taped to a page in the Mark II log book.

Later, Hopper worked on the more advanced UNIVAC I (universal automatic computer), as well. About the time UNIVAC became a Remington Rand project in the early 1950s, she developed the first compiler, the A-0. That’s when she had the radical insight that programs didn’t have to be developed in machine or assembly language.

She thought programming could be done more usefully in something resembling English. That insight launched the progress toward COBOL.

Plain Language Is No Accident

Did I mention COBOL is easy to learn partly because it has so much in common with plain, spoken language? The similarity was no accident — it drove the design.

Other drivers also came into play by the time the first formal COBOL came out of committee in the late 1950s and early 1960s. In particular, COBOL’s flexibility and ability to evolve were structured indelibly into the language, so it has never fallen into stasis.

What’s Next?

As COBOL continues to evolve, it remains deeply enmeshed in the world’s business processes.

Supplying qualified replacement developers could become a problem, but probably won’t once people look realistically at COBOL’s staying power.

In Part 2, I’ll explore some options available if you want to take advantage of the opportunities COBOL offers in today’s — not yesterday’s — world.

COBOL Today and Tomorrow, Part 2


Irving Abraham is a Unix and Linux product director forMicro Focus.


Leave a Comment

Please sign in to post or reply to a comment. New users create a free account.

Related Stories

LinuxInsider Channels