May 07, 1998
Harvard
University Gazette

 

Full contents
Notes
Newsmakers
Police Log
Gazette Home
Gazette Archives
News Office
Feedback

SEARCH THE GAZETTE

 

Harvard Seeks to Swat Year 2000 Computer Bug

By Alvin Powell

Contributing Writer

The year 2000 is coming and Harvard's computer specialists are preparing to meet it.

The preparation is needed because, along with the new millennium comes the "millennium bug," a worldwide glitch in computer software caused by programmers saving space by noting dates with just two digits.

That means that computers, which take everything literally, will assume the year "00" is 1900. If untreated, this seemingly innocuous problem could cause trouble ranging from fax machines marking documents with the date Jan. 1, 1900, to wide-scale disruption of financial networks and power grids.

Exactly how large the problem will be is subject to debate. Some doomsayers predict a wave of bankruptcies and a global recession. Others say it will be noticed hardly at all. Most agree, though, that it is a problem worth paying attention to and that early assessment and action can minimize the potential impact.

Harvard has been working for more than a year to minimize any disruption. The University has surveyed its many schools and departments, quizzed vendors on their software's "Year 2000 compliance," contacted outside institutions with which it exchanges information electronically, and put a University-wide structure in place with the goal of gaining near-total compliance by January 1999.

"We are prepared and getting better prepared," said Liz Eagan, a project manager with University Information Systems, who is helping coordinate the University-wide effort to prepare for the approach of the year 2000.

Eagan and Scott Bradner, another coordinator of the effort, say the University's major computer systems, systems that handle financial, admissions, and other critical data, are either ready or being made ready for the century's end.

But Harvard's decentralized structure makes it difficult for a few people to oversee University-wide efforts. That's why it is critical that departments, individual laboratories, and other locations take responsibility for ensuring their own computers and other equipment won't fizzle when the calendar flips over to 2000.

"The University is so large, diverse, and decentralized that it's pretty likely there's critical systems that the core will never know about," Bradner said.

Harvard has already selected a small army of about 60 coordinators at different schools and departments who will be responsible for making sure efforts are under way to identify potential problems and take whatever steps are needed to prepare the major systems.

But it's what the University doesn't know that can hurt it, coordinators say.

The critical, highly visible systems -- University-wide and at the various schools -- are being taken care of. Bradner and Eagan's main concern is small, custom-made or freestanding systems that were purchased long ago and that are chugging away in a corner, nearly forgotten, running or monitoring a critical experiment.

"The things people really need to concentrate on are mid-range, mid-criticality things that people may not be paying attention to," Eagan said.

Another challenge for Harvard computer staffers is that of "embedded chips."

With computers increasingly saturating modern life, computer chips are no longer just found in computers and other items readily identified as computerized. Chips, some of which may not recognize the new millennium, are in everything from microwave ovens to elevators, from automobiles to telephones.

Harvard's University Operations Services (UOS) is tackling the problem of embedded chips in critical systems like elevators and fire alarm systems.

Director for Applied Technologies Paul Malagrifa said UOS is still assessing different systems and equipment, but the results after six months have been encouraging.

"We have yet to uncover a system, or a set of systems, that pose a considerable risk to the safety and well-being of the University from a facilities operations perspective," Malagrifa said. "This is not to say that one does not exist, but after six months of focused analysis, the situation appears stable."

But much work needs to be done, Malagrifa said. And, though UOS has responsibility for interbuilding systems, like the campus fire loop and energy management systems that end at a central point, individual schools have responsibility for the systems within their buildings and need to check those themselves.

UOS is checking most critical systems first and working its way down to the least critical. Staff are contacting manufacturers to ask about chip compliance as well as testing systems and equipment to ensure they won't malfunction. For systems and equipment that are not compliant, the equipment can usually be reprogrammed or the chip replaced, Malagrifa said.

Computers Helping Computers

When looking for help with computer problems, where else would you turn but to a computer?

The University has set up an extensive Website stocked with contacts, forms, links to other sites, and other information on the Year 2000 computer bug. The address is www.uis.harvard.

edu/year2000.

The site includes a list of coordinators for different areas and a list of equipment that could potentially be affected by the Year 2000 bug. Included on the list are copiers, door locks, card access systems, buses, cars, ticketing and signaling systems, and telephone switching and data switching equipment.

Eagan and Bradner encourage people to visit the site. Those with questions or information about potential problems should contact their local coordinator or, if they don't know or can't find their coordinator, contact Eagan and Bradner.

With computers pervasive in today's society -- and in the University -- it's unlikely Harvard will escape without a single glitch, Bradner said. The goal, though, is to make those glitches as minor as possible.

"Our aim is that we don't have anything that closes down a part of the University or endangers anyone or endangers ongoing research," Bradner said. "The aim is to make sure the impact is local and inconsequential."

 


Copyright 1998 President and Fellows of Harvard College