By Joel Warner
By Michael Roberts
By Alan Prendergast
By Michael Roberts
By Michael Roberts
By Amber Taufen
By Patricia Calhoun
By William Breathes
"If you have your top technical people doing this, you're taking $200-an-hour employees and applying them to a menial task," he says. "You can take a $30,000-a-year guy who's just out of college, and with our software guiding him through the code, he can fix it just as easily as your top person."
Its Y2K product line now accounts for roughly 80 percent of Accelr8's revenues. The company has landed contracts with Ford, Corning, Unisys and several other major companies that don't have any time to waste in upgrading their midrange systems. "There aren't enough people in the world to solve the Year 2000 problem," says Harry Fleury, Accelr8's president, "so the idea is to automate the task as much as possible."
Y2K investors tend to regard service companies like Ciber as more attractive than product vendors such as Accelr8, since the service companies often enjoy longstanding client relationships that will continue after the Y2K fix is completed. But that doesn't mean the handicappers at Hanifen Imhoff aren't bullish about Accelr8. "If they're successful with their product, they're going to generate a lot of cash over the next two or three years to invest in future acquisitions or other products," notes Welty.
Geimer plans to be in business well past the year 2000. Accelr8 has already weathered one attack of Y2K-related hype on the Internet, which, over a four-day period last year, drove its stock from the low teens to thirty and then back to sixteen dollars a share. After a year of steady gains and increasingly promising earnings reports, it now trades at around $25--with even better news ahead, its CEO insists.
"The Year 2000 problem isn't going to be solved by January 1, 2000," Geimer notes. "You'll probably see a two- to three-year carryover. Secondly, the migration business is still the core of our technology, and we intend to build on that, from VAX to Unix to the Windows NT platform. We got this Y2K problem because people became complacent about their old legacy code. Rather than facing up to the fact ten years ago that it needed to be modernized, or budgeting for it five years ago, they just let it go.
"That decision has created this incredible opportunity, not only for Year 2000 products, but to sell them the next generation of technology--which they haven't taken advantage of for fifteen years."
Geimer says his company's greatest challenge isn't cracking the Y2K problem but persuading manufacturers to move beyond the slow-but-familiar VAX technology. "They get into what I call the oil-and-grease mode," he says. "They don't invest in the information age; they use every piece of equipment until the last widget has been produced. The heavy industries in this country have not, because of competitive pressure, spent money on hardware and software."
By now most major corporations have at least some notion of the kind of problems the millennium could pose for their mainframe accounting and payroll systems, as well as the workhorse midrange computers and client-server networks that keep their businesses running. But business systems aren't the only "mission-critical" operations that rely on computer chips. One of the most critical areas also happens to be one of the least examined: the so-called embedded systems that run the physical plant itself.
Embedded systems encompass everything from the devices that control office-building elevators, air conditioning and security systems to the chips that operate the dashboard clock and automatically adjust the seat in a new luxury car. They're also essential to a wide range of industries--from mining and manufacturing to energy and public utility companies--in the form of "plant control systems," which regulate processing and emissions, monitor production and inventory, churn out required EPA reports and shut down assembly lines when alarm sensors go off.
For every ten business systems in a major manufacturing concern, there are probably two hundred such components on a factory floor. Until recently, though, few companies had bothered to examine their control systems for Y2K problems--and what they're finding is disturbing.
"The recognition that there are problems at the embedded-system level is pretty late in developing," says John Jenkins, CEO and chairman of TAVA Technologies, a company that specializes in providing consulting and software for plant control system integration. "Even our own people, until about nine months ago, tended to think that date-related issues aren't part of process control. The reality is that dates have a lot to do with it."
Compared with mainframe computers, many of the control systems in factories are relatively simple gizmos. But the vast number of them and the degree to which they rely on each other to keep the whole operation running pose a formidable challenge for Y2K-busters. "Think of the business system as a narrow, deep silo of millions of lines of computer code," says Jenkins. "On the factory floor you have a bunch of buckets, each with a few thousand lines of code. You have to search out each one; you can't strip out your code, send it to India and have it come back clean."
TAVA was formed last year out of the merger of five similar companies, including Denver-based Topro Systems Integration. The company, which now has 350 employees and offices in ten cities, reported revenues of $36 million in fiscal 1997. Jenkins expects double-digit growth in the coming year--total new orders in the first half of 1998 are expected to exceed $28 million--as the Y2K-related business continues to expand. TAVA already has nearly three dozen Fortune 100 clients seeking help with their Y2K problems, including some with hundreds of plants.