Commenting on COBOL

As we approach its big anniversary, attention is turning towards the most resilient of all technologies: COBOL. And the comments are already flying...

I recently penned an article in which I described the COBOL computer language as facing an existential challenge. The challenge wasn’t because it was old or outdated (as some might have you believe), but that it just isn’t talked about any longer, jeopardising future investment decisions.

Ahead of the September anniversary date, I have been asking experts, users and commentators to offer their perspectives on the secrets to COBOL’s unrivalled success in the business computing world. While we will save the full list of comments for another day, we can already get a sense of the general tone.

The following are some of COBOL’s foundational strengths, as testified by people who use it or are experts in the field. I withheld names and roles for this article but will publish the full details of all the comments received in the future. Safe to say they all know what the comment line is in COBOL.

COBOL: be sure to include comments

So, what of the commentary so far? Did it match recent commentary about the COBOL language? Here’s what we were told so far.

COBOL’s syntax is famously readable, resembling English language sentences. And, it holds, the easier it is to read, the easier it is to learn, and maintain. Supporting that accepted wisdom, one expert claimed, “COBOL is so easy to understand, I could teach it to you in an afternoon." Another comment, talking about their own experience in learning the language, confessed, “the real heroes were the COBOL language designers who had enabled my lightning-fast understanding of COBOL with no prior programming experience”.

Back in the day, there were no IBM Z mainframes, and the hardware world was in its infancy. COBOL’s designers sensibly realised it would need to work ‘wherever’ the market demanded it. And so was born COBOL’s often-overlooked portability: mainframes, Linux, UNIX, Mid-Range, Windows, Cloud – literally hundreds of platforms can run COBOL. As one expert proudly testified, “I have worked, and continue to work, in both the mainframe world and the distributed world continually leveraging previous skills. COBOL allows me to do that”.

Similarly, no-one in the early 60’s could reasonably predict the meteoric change in the world of technology that lay ahead of COBOL. Its designers insisted that it could adapt and change with the times. One expert’s perspective was effusive, “In today’s world of computing, COBOL utilises modern development processes and environments, and happily integrates with other cutting-edge platforms technologies. It’s as modern as everything else”. Another remark echoed the same sentiment, “From architecting and performing migrations, to modernizations and upgrades, [COBOL] is the only coding language that continues to stay modern”.

COBOL was developed as a business-level language, to enable the growth of the embryonic technology industry, and to train ostensibly non-technical people how to use it. It was designed with business in mind, handling large volumes of structured data, and the associated complex calculations against it, making it ideal for large-scale commerce. As one commentator put it, “Many critical IT systems, covering billions of lines of code and across many industries, dating back as far as the sixties, rely on the power and stability of COBOL”. A more specific account confirmed – “We have been counting on COBOL since 1970. We currently have more than 125 million lines of active COBOL code running in production,” while another former user took the context in the other direction, with this showstopping remark, “The second most valuable asset in the United States – after oil – is hundreds of billions of lines of COBOL." 

A Technological Gem?

The above comments are just a teaser, of course. We are receiving a bountiful harvest of individual perspectives, from commentators and practitioners, including – to be fair – at least one or two concerns and criticisms. In fact, COBOL has had its detractors since its inception. Infamously, somebody remarked that it wouldn’t last the decade, and that was back in 1960. One comment we received asserted that COBOL’s success was little more than “an accident of history,” and that “if PL/I had come first, it would be more popular.” Yet here we are, heading towards COBOL’s 65th birthday.

Whatever the case, the debate will unfold again, around COBOL’s sapphire anniversary. COBOL was, after all, created through collaborative effort, never owned by a single entity. Everyone is entitled to an opinion, and I look forward to sharing those I’ve received nearer the time.

What’s your view?

Feel free to reply, or DM me, if you want your say.

Further reading

COBOL, the incessant number cruncher - Planet Mainframe

HELLO, WORLD! (REMEMBER ME?) - Elnion