Job postings for mainframe experts are up 15 percent over the last 18 months, despite a stall in the sales of systems such as the IBM Z Series. Companies that invested millions of dollars on large systems find themselves needing experts who are no longer employed by the company. The alternative is to find them on the open job market.
This is the kind of development that HP 3000 managers believe has a forthcoming echo in your community.
Under the heading "HP3000 Workers Chances Improving?" Tracy Johnson noted a story in tech Web site slashdot, reporting on the short term folly of companies releasing experts who knew stable, reliable, but aging systems.
"I presume the hidden cost of letting HP 3000 workers go, way back, may also apply here," Johnson said. A few others with success at selling their 3000 savvy have agreed.
Craig Lalley of EchoTech pointed out that COBOL, which turned 50 this summer, remains a valuable skill. "I know some very valuable COBOL programmers," he said. "Laugh if you want, but it still a required skill set in some places."
The slashdot article notes that "Businesses that cut experienced mainframe administrators in an effort to cut costs inadvertently created a skills shortage that is coming back to bite them. If you do a total cost of ownership, the mainframe comes out cheaper, but since the costs of a mainframe are immediately obvious, it is hard to get it past the bean-counters of an organization."
Donna Hofmeister at Allegro Consultants, who's using both her MPE and Unix skills to support HP customers, pointed out that large-scale systems like the 3000 have cost advantages that are not apparent to less-experienced management. There's the sticky value of mature apps, too.
"I said essentially the same thing [about cost of ownership] to someone at HP about MPE," Hofmeister said. "The recipient’s eyes flew open like I had said the most amazing thing ever heard. I’m not a bean counter, but it was so obvious that this was the case -- and yet this had never occurred to this very well-educated HP’er.
"But anyhow, 'tis true -- it can be very very hard to replace legacy apps (for a multitude of reasons), and some people just can’t/won’t get that."
Since the 3000 has 30 years of marketplace history, a couple of 3000 managers and developers mentioned the downside of the MPE model, shortfalls that led companies away from 3000 investment. "Managers could not get a needed hardware upgrade approved," said Chuck Ryan, "because the upgrade fees demanded by their various third party software vendors vastly overshadowed the cost of the hardware. The licensing model for software on the 3000 is what killed the machine."
Pete Eggers, whose name has been mentioned for the OpenMPE board of directors, also took note of what MPE would need to remain a large-scale solution with small scale flexibility. He summed up a steep technical hill to climb.
"An updated MPE could easily have been created to sit atop a stripped down and MPE optimized Linux kernel," Eggers said on the 3000 newsgroup, "supporting all manner of server hardware and new operating system technologies. But as the late Wirt Atmar once told me, the chance for that happening passed in 2001. So now HP has regulated the HP 3000 and MPE users to hospice care, making them as comfortable and happy as possible until they pass on."
Neither view is absolute in the 3000 community, but companies are shedding 3000 staff before shuttling the system out of production use. Companies make a practice of cutting these experts out of shrinking IT budgets, so now there’s a lot of bareback computing going on in this community. Meanwhile, the 3000s continue to run. But how long, before a software failure? And who will be left on staff to understand these apps when the work begins on migrations?