If you stacked every line of COBOL code remaining in the world, it would easily reach the moon.
It’s an open secret that a supposedly dead programming language is still lurking in thousands of data centres around the world, often running their most critical applications. While there’s a general consensus that COBOL is going to have to be replaced at some point, it’s not clear how long that will take, or exactly how it should be accomplished
But when that time comes for you, get ready to reach for your wallet.
The cost of moving
As Phil Murphy, principal analyst at Forrester Research puts it, companies that help you migrate off the mainframe are like real estate agents: they make money when you move. And this is especially true when it comes to your precious COBOL. There are few major players in the market with enterprise-grade COBOL compilers, and those that do have one are calling the shots.
“The world needs a COBOL compiler vendor other than, obviously IBM and mainframe, and really the two biggies are Micro Focus and Clerity, which Dell now owns,” says Murphy. (In April, Dell acquired Clerity Solutions, which specializes in COBOL re-hosting, as well as Make Technologies, a Canadian company that focuses on re-engineering COBOL code).
Here’s the result: there are now just a few big vendors you can turn to to keep your COBOL applications alive. The majority of companies either have them re-hosted by by Clerity or Micro Focus, or they simply stick with their mainframes. None of these options are going to be cheap.
There’s a whole other debate about whether it makes more sense, financially, to get off your mainframe in the first place, but let’s assume that moving to x86 platforms is the right move for you. You’ll find it’s a bit of an uncomfortable place to be, essentially stuck between a few goliath vendors. And don’t look to HP and Oracle for relief, says Murphy.
“They’ve got the same value prop, if you think about it. But what they don’t have is a COBOL compiler to do it with, unless they buddy up to Micro Focus, and historically, I’d be lying if I told you those were warm and fuzzy relationships,” he says.
If Dell doesn’t sell Clerity, he adds, it would force hardware vendors who want to run COBOL apps to deal with Micro Focus, giving them a “virtual monopoly.”
“That lack of choice was already an issue for COBOL compilers,” he says. “Some of Micro Focus’ customers were complaining about increased system maintenance and pricing, and contract negotiations, so surely this will only fuel that fire.”
COBOL and mainframes (circa 2032)
But Mary Nugent, vice-president of solutions consulting at Micro Focus, says her company’s history is what has propelled it to its position as a market leader. Micro Focus’ name is synonymous with COBOL, having worked consistently with the language for decades, even while the rest of the world was trying to forget it.
“You have to go back to the roots,” says Nugent. “When I was at Texas A&M [university] in the late ‘70s and early ‘80s, when you had to learn COBOL, because every business major had to—I was in a business school—you learned it on the Micro Focus IDE.”
“We are COBOL,” she says.
With the experience and customer base the company has, the market will look favourably upon Micro Focus in the future, she adds. “It’s going to keep us busy for 20 or 30 more years.”
But it remains to be seen what kind of competition Dell will provide over that time period. The company has made it clear that application modernization, including COBOL re-hosting, will be a big part of its strategy to move into the IT services realm.
Hans Wolf, managing director for Dell’s application services business unit, says the company recognizes application modernization as a definite “growth market.”
“We’re in the ITO outsourcing business and we’re starting to see more and more that people have really kind of improved their data centre processes and efficiencies to the point where there’s not a lot of low-hanging fruit left,” says Wolf.
Thus, more companies will be looking to save money the only remaining place they can: by getting off their legacy infrastructure, he says. And Dell has no “legacy biases we’re trying to protect or keep,” he adds, perhaps alluding to Dell’s competitor, IBM, which has made great efforts to convince the market that mainframes aren’t “dinosaurs”
that are soon going to be extinct.
Certainly, the company recognized that buying Clerity and Make, with their rare and valuable COBOL expertise, was the linchpin of the strategy to put Dell in the application modernization game.
“We can deliver the processing power,” says Wolf. “The issue is how we get those legacy apps moved off the mainframe, or AS400, or HP3000 — whatever the platform happens to be.”
Murphy says Dell now has all its bases covered. Clerity caters to enterprises that want straightforward COBOL re-hosts, while Make digs a bit deeper, parsing and re-writing COBOL code. Dell can now say “‘come to my hardware, come to the cloud, come wherever — I can move you quickly via re-host and I can re-engineer you to fit the new platform if that’s what you need,'” he says.
How big this need will eventually grow remains a matter of speculation. IBM, naturally, would have you keep your mainframe, while Dell and Micro Focus insist that COBOL applications will hum along just fine on x86 hardware. It’s hard to predict the future, but the hundreds of billions of lines of COBOL code are a fact we can’t ignore. And if the market remains in its current state, you can expect to pay top dollar for every line that’s disturbed.
(UPDATE: Don Fitzgerald pointed out in the comments that there are plenty of other companies with COBOL compilers out there. I didn’t mean to imply that there were literally only three choices, just for most practical purposes. I’ve changed the article accordingly to avoid giving the wrong impression. I’m also trying to reach the analyst as well to get some figures, hopefully, on how much of a market share these other companies with COBOL compilers have.)