The Importance of “Legacy” in Colleges and Universities
By Cameron Seay, guest blogger
The global economy runs on mainframes. Depending on who you speak to, between 70 and 80 percent of all business transactions are still running through a mainframe. Even greater than that are ATM and credit card transactions, where 90 percent still rely on these legacy systems. Yet, when we discuss “legacy” in the technology space, people rarely understand what we’re talking about. When I say that I work with mainframe computers, people envision a room full of massive devices that cannot be moved and require specialized knowledge to use. In reality, IBM has been coming out with an upgraded IBM z system every two years, for as long as I can remember. They run efficiently and they are secure. But the myth about needing specific skills to run these systems? Unfortunately, that may be more accurate than we would like to admit—or than it should be.
As a professor of programming, I have had the pleasure of teaching at five different colleges and universities across the U.S.: North Carolina Central University, NC Agricultural and Technical State University, Alcorn State University, Tennessee State University and East Carolina University. Most of these are historically black colleges/universities (HBCUs). In each position, I have encountered similar feelings and beliefs about perceived “legacy” technologies. Most faculty are oblivious to mainframe technology and its potential value to their students. Those who are aware of mainframes and have learned to program on them feel so far removed from these systems that they would be unable to teach a class on the subject. This has created a major disconnect between what the industry needs and what programmers are taught. Academic departments have little incentive to teach something that they haven’t traditionally taught, especially when there is no money attached to it. This has been the issue with mainframe computers and legacy technology. Most schools don’t teach it and don’t want to, yet recent events have shown that this viewpoint will become a bigger issue, especially with COBOL programming.
COBOL programming has been around for more than 60 years, and many consider this a “legacy” language, even though it is still used by governments and the financial services sector. COBOL codes and programs that were created 30 or 40 years ago are still being used on these systems. And while moving the 220 billion lines of COBOL code that exist to a modern language is technically doable, it makes no sense from a financial or risk perspective. Although there is no current shortage of COBOL programmers, older coders are now retiring; in fact, the majority of the mainframe workforce is starting to retire. There are many companies, such as Rocket Software, who have very strong youth-oriented recruitment programs, but this won’t be enough. If we want COBOL and mainframes to remain viable options, and if we want our governments and financial sector to remain operational, the feasible solution is to teach COBOL in colleges.
I’ve met many students who have never encountered mainframes, or may have outdated perspectives about them. Many of them subscribe to the belief that these systems are rarely used. But no matter the school, when I arrive at a college to teach these programs, the sequence is largely the same: once you clearly explain what mainframes are, what they do, and why they are vital to the global economy, classes fill up. It may take a semester or two, but it always inevitably happens. It’s easy to see why mainframe classes are a hit at HBCUs. Many of the students there went to not-so-great high schools, and usually did not get the technology exposure that students who went to better high schools received. Accordingly, they look for whatever advantage they can find in their job search. The mainframe has provided this in abundance.
Recent government events that have resulted in urgent calls for COBOL programmers have proven that the outdated belief that COBOL doesn’t need to be taught is simply not true. Colleges and universities need to follow the lead of HBCUs and begin teaching students about legacy technology, COBOL and their importance in modern society. Making these topics appeal to students is simple: mainframe jobs pay more and have more stability. Colleges and universities, on the other hand, may be more difficult to convince. Many schools believe that there is no financial viability to teaching this, and will only consider it when companies financially incentivize these courses. But legacy does not mean replaceable. COBOL programs can take advantage of improvements on the hardware that allow them to run faster and more securely. Legacy means secure. There is no doubt in my mind that COBOL will be around for many years to come. We just need to convince colleges and universities of that.
Cameron Seay is an evangelist for the mainframe. Currently serving as Adjunct Faculty in the College of Engineering and Technology at East Carolina University, He has been an academic in the mainframe area for the past 14 years and was in IT for 21 years before that. His focus is an “enterprise-centric” perspective on computing, including the cloud, which encompasses every aspect of a heterogeneous infrastructure. He holds a doctorate in educational psychology, and master’s degrees in business, information systems, and economics.