As an industry, IT is not exactly famed for finding uses for older technologies. Typically, it opts to replace them with newer and seemingly more exciting innovations whether or not the business value of these investments has been proven.
In few areas of IT is this tendency more prominent than in legacy systems, whose "legacy" label alone implies an inherent need for upgrading and replacing.
"IT eats the weak and also the end-weary," said Charles King, an analyst with Pund-IT Research. "As long as I have been an analyst, people have been speaking about the death of the mainframe. This is not a case that can be made, or at least any time soon."
IBM officials have been making that case for years, and until recently, the numbers for their mainframe business have been strong. However, they dropped in the third quarter. Revenues from System z server products dropped 31 percent over the same period last year, while the total delivery of mainframe computing power—measured in mips (million instructions per second)—fell 21 percent.
Still, IBM is pushing the mainframes as the perfect platform for server consolidation and virtualization and has rolled out models aimed at the midmarket and customers in developing countries.
IBM wants the mainframe to be an IT destination. Click here to read more.
The majority of discussions around mainframe skills in the IT workplace address either a looming shortage of workers with legacy programming knowledge or the lack of a future in a mainframe career. But those with knowledge of mainframes suggest that if the latter issue were put to rest, the former one would follow.
"The fact is that if you talk to huge percentages of Fortune 500 companies, they'll tell you that the cost of migrating to a different platform or different software is significant," King said. "As long as the mainframe is doing the job that it was designed to do, those companies have no reason to change this. From a practical standpoint, this means that mainframe skills will continue to be in demand."
However, the message that mainframes are not the dead-end career path they once were made out to be has not yet reached the IT masses. Few are more aware of this than Kristine Harper, program manager with zNextGen, a Share and IBM user community for recent graduates and older IT professionals who have a common interest in the mainframe platform.
Harper ended up on a mainframe career path because both of her parents were mainframe developers who showed her the ropes.
"When I go to career fairs, half the students don't know what the mainframe is and the other half didn't know that they still exist today," Harper said. "They're sleeker and smaller—they don't take up a whole room anymore—and there's financial security in this path."
She said she has to explain to computer science students the viability of the mainframe career path.
"Seventy percent of the world's data is on the mainframe," Harper said. "They're airlines, banks, stores, insurance companies. … They're great places to work."
Harper said she understands that the mainframe career path may not have the flash or trendy appeal of working in the newest technology.
"It's not a sexy career at all," she said. "The work is behind the scenes. You really have to be a self-motivator and a hard worker. The draw for me was the sheer power of assembler code, the reliability and capability, and to know that your code is running in the background of all of the big corporations around the world, versus building a game in Java."
Some analysts say this lack of appeal is the mainframe's chief obstacle to attracting new blood.
"The dream of working for a hot startup and making money in an IPO is more along the lines of playing baseball in high school," Pund-IT's King said. "It's a reality for a small percentage of IT professionals."
Page 2: Big Iron Remains Career Option
While the mainframe career path may seem less glamorous to college students, it offers the chance to work with large com-panies that have the money to spend on keeping their systems maintained and keeping those who maintain them employed and happy.
"Mainframes offer professionals some options that don't exist in other skills sets, [such as] the ability to work with very big companies at the top end of the IT food chain," King said. "It's like getting out of law school and putting a shingle out with your buddies versus working at a Wall Street firm in the city's center."
Much of the ongoing viability of the mainframe career path is credited to IBM, which has worked to get more than 300 universities to support mainframe education. In 2000, IBM made a commitment to train 20,000 college students on System z mainframes by 2010, a number it surpassed in early 2007.
"What's interesting is that the mainframe recruiters now go to those universities when they're looking to hire," King said. "They know they're there. Along the arc of an IT professional's success, attending a school that specializes in the mainframe skills is not a bad way to go."
Adding to this effort, IBM announced in 2003 the Mainframe Charter, outlining a commitment to customers that it would manage the System z costs and dramatically simplify and innovate on the platform.
The difficulties getting university students enthralled with a platform they view as old and unexciting is part of a larger issue of the practical application of professional skills in business settings, something sorely missing from university educations, King said.
"There are no IT classes on how to make a living, but there are people making livings in IT," he said. "There are people making a living with mainframe IT skills. The universities IBM works with and supports take a practical, common-sense approach to learning and making a living."
In IT, there have always been trends that follow certain skills, platforms and applications, and when something is in a popularity upswing, hundreds of students acquire the same skills simultaneously.
"What this means is that you might have great skills, but you're diving into a market where thousands of others share the same skills, when it might help you more to have differentiated skills," King said.
The difficulties overcoming student disinterest in legacy computing echo the larger problem of getting younger people to pursue any math or science career.
"Trying to convince students to study computers, science and math today—[much less] mainframes—is like pulling teeth," said Michael Stack, a product developer at Neon Enterprise Software who was a Northern Illinois University applied systems programming professor for 40 years. "The first goal is to get them interested again, whether they want to focus on legacy or modern computing systems."
Meanwhile, the demand for mainframe specialists is not diminishing, as evidenced by Stack's difficulty retiring.
"Having said all that, having mainframe software as the framework of your computer education is one way to utilize it, even if you are worried that there will be fewer mainframe jobs in 20 years," Stack said. "But I have no reason to believe this is the case—I am trying to quit working, and I keep getting solicited."
Harper echoed this sentiment, witnessing firsthand the inevitable shortage of IT professionals with legacy computer skills.
"I am one of the youngest people where I work," she said. "None of my co-workers are going to be here in 20 years, and all of the stores, airlines, banks and more … will continue to use our software and rely on the mainframe. We need people to take the reins." ´
Check out eWEEK.com's Careers Center for the latest news, analysis and commentary on careers for IT professionals.