The End is Nigh!

Zack Urlocker
Wednesday, September 01, 1999

Whenever a new technology emerges in the software industry, programmers are quick to claim their early involvement. Those cigar-chomping, coffee-swilling managers puff out their chests and claim, "Why we were writing object-oriented code back in the '60s with Fortran IV using arrays of Holerith data." Younger programmers look with awe upon the graybeards and wonder if Herman Holerith was the guy who wrote "Last Train to Clarkesville."

Y2K or Not 2K

The newest hot technology to get on the scenes today addresses the so-called year-2000 problem, or Y2K to the MTV generation. The source of the problem is that historically, many dates have been stored as two digits, for example 97 for 1997. In the early days of mainframe computing and the Commodore 64, memory and disk space were expensive. By compressing the data format into two bytes, programmers were able to conserve memory to use for business-critical applications like printing Snoopy calendars. When IT management tells programmers to fix the problems or elseIyou've got to wonder who they think built those systems in the first place.

The year-2000 problem is a large one and a problem with potentially far-reaching consequences. The Gartner Quadrant predicts that the Y2K problem will require six billion dollars to correct. And that doesn't include taxes, tips, or dealer fees. Of course, this estimate could be off by quite a bit: It might have been calculated on a Pentium.

Millenial Machinations 

So what are the options?

Wait it out: Maybe this is like that old Michelangelo virus a lot of doom and gloom and then nothing. In fact I'm positive that if you just ignore the whole Y2K issue, nothing will happen. No payroll, no payables, no invoices, no receivables...

Hire more programmers: It's well known that adding more programmers to a project slows it down. More programmers mean more code, more Bugs, and ultimately, more work. Hire enough programmers, and you'll eventually reach a point where time will not only slow down, but will run backwards. When the clock gets back to 1986, buy Microsoft stock so you won't have to face this problem again.

Be proactive: If you're proactive in eradicating year-2000 bugs, you'll find the work easier to manage. Get involved with a task force to define the effects of Y2K in your organization. Then choose a strategic, high-profile area, say screensavers, and focus on it. If you work hard and stay focused, you should be able to get it completely under control and on schedule. While others complain about tackling accounts receivable or payroll, you'll feel confident that with another few years' testing, you'll have done your part.

Fight fire with fire: just like the way Jeff Goldblum fought the aliens with a computer virus in the film adaptation of Hemingway's novel Independence Day. For example, with the right programmers, you can combat the year-2000 problem with disciplined insertion of off-by-one errors. If you're lucky, you'll not face your year-2000 problem until a year later. Or earlier?

Start a trend: With six billion dollars being spent on the Y2K problem, I wonder if there isn't a long-term entrepreneurial opportunity here. Why not "Money for bugs"? Certainly, we can find other causes equally worthy of budget dollars in the new millennium. We owe it to ourselves to raise other bugs to the same level as Y2K. The key is to make sure we have good marketing names. After all, if you want to get project funding, you'll have to come up with a better name than "heap leaks." It sounds like a personal-hygiene problem-but who's to say what's a feature and what's a bug? And if you can work the Internet or Java in, then all the better. Maybe something like, "the Web-2000 Dateline Initiative for Java." If you can't beat 'era... And finally, the best way to deal with the year-2000 problem: Get a job at Gartner.

About the author
Zack Urlocker is a pseudonym used to keep his boss from finding out who's been running up those wireless ISP charges.