|
Y2K: Where's the fix?
|
 |
May 3, 1999: 3:03 p.m. ET
Most millennium-bug issues are management, not technical, concerns
|
NEW YORK (CNNfn) - Given how long everyone's known about the Year 2000 bug -- and how close we are to the new millennium -- two questions come to mind: What's taking everyone so long to solve this Year 2000 bug? Also, if it's a standard problem, why isn't there a standard solution?
The problem itself is simple enough to understand. Many computers are programmed to recognize only the last two digits of the year because programmers in the early '60s needed to save memory wherever possible. Using 99, for example, to represent 1999 was a more economical solution.
We now know, of course, that when computers make the transition to 2000, they may recognize the year as 1900 or could cease to function altogether.
If that's the common problem, why is it taking so long -- and why is so much money being spent -- to fix the problem.
The short answer is that there are more than 5,000 different programming languages. It's impossible to write a universal code to solve the problem across the different languages, many of which are proprietary systems for firms and organizations.
Also, even if a computer's hardware is Y2K ready, that doesn't mean all the applications installed on it are ready.
The task of finding a solution, however, runs deeper than repairing faulty programming code.
Hide and seek
Market research firm Gartner Group has estimated that solving the Y2K problem will cost between $1 trillion and $2 trillion worldwide. Of that total, $300 billion to $600 billion -- 30 percent -- will be spent on resolving technological issues.
"From the start, the problems has always been more of a managerial challenge than a technical challenge," said Jack Gribben, a spokesman for the President's Council on Year 2000 Conversion.
In fact, more spending occurs outside corporate information technology (IT) budgets to fix the Y2K problem.
"As of January 1999, companies are spending an equal amount in IT and outside, for things like risk assessment on their business partners and contingency plans," said Lou Marcoccio, Y2K research director at Gartner Group. "By the end of the year, companies will spend two-and-a-half to three times more outside IT."
A large part of the reason a solution is slow in coming is that companies and governments won't be aware of many Y2K issues until the problems occur.
The issue lies with what Marcoccio calls "date-forward processing" -- that is, transactions that were set to occur weeks, months or years earlier, such as 30-year mortgages.
"Many organizations have failures sitting in their systems, but they won't show up until they run the transactions," Marcoccio said.
In other words, it's difficult to solve a problem that, as far as you're aware, doesn't exist. Organizations may be under the impression that all their programming code is ready for the new millennium, but that may not be the case.
Although most people expect the worst to happen when the clock strikes midnight on New Year's Eve, Gartner has determined that only 10 percent of the anticipated Y2K-related failures will occur within the two weeks surrounding the new year. Rather, a sizable chunk of millennium-bug failures will happen in July and October this year.
"In July, we'll have 80 times more six-month date-forward processing than in any period before 2000," Marcoccio said. Those transactions include parts ordering from manufacturers and diagnostic health data.
"In October, that figure goes up to 800 times," he added. "There is more first-quarter date-forward processing transactions -- things like financials, accounts receivable, parts procurement, shipping."
Furthermore, things will only get worse in 2000. Marcoccio estimated that 55 percent of all Y2K-related failures will occur in the first three quarters of the new year.
"We'll have millions of transactions running later in 2000 that have never been run before," Marcoccio said.
-- by staff writer John Frederick Moore
|
|
|
|
|
 |
|
