This article was published in CAUSE/EFFECT journal, Volume 22 Number 3 1999. The copyright is shared by EDUCAUSE and the author. See for additional copyright information.

Y2K--No Time for Panic
by Daniel A. Updegrove

We�ve been updating or replacing enterprise systems, testing data centers and networks, advising depart- ments on local systems and PCs, and consulting with facilities and health care personnel on devices with embedded chips. We�ve attended numerous task force meetings, briefed boards, and presided over community panel discussions on �the millennium bug.� What have we learned from this experience, and what should we focus on between now and January 1, 2000?

What we have learned

First, as most technologists know, there are really two problems with the year 2000 (Y2K): the device problem and the data problem. The former has received too much attention and the latter too little. In the case of devices--elevators, sensors, laboratory and medical apparatus, computers--concerns were raised that the inability of the hardware to recognize the year 2000 (would it be 00 or 1900 or 1980?) would cause widespread failures on January 1. Fortunately most non-computer devices either don�t know the date or won�t fail catastrophically, even if the date displayed after December 31, 1999, is misleading. As for computers, the robust economy and falling hardware prices have led to wholesale replacement, and even many obsolete PCs can have their dates reset once in the new year. Thus Saturday, January 1, figures to be a much calmer day than the Doomsday crowd has predicted.

The data problem is more complicated since calculations and displays involving calendar dates permeate millions of different computer programs, some of which lack source code or programming staff. Moreover, output of some systems provides input for others. Fortunately we�ve had a much longer time to solve this problem, as student information systems had to be year 2000 aware by 1996, and most financial systems needed to handle fiscal year 2000 starting this summer. Unfortunately we can�t possibly test all of these systems, and we won�t observe all possible impacts of erroneous calculations and data feeds for several months into 2000.

Second, we�ve learned that there are an astonishing number of computers and computer-related devices and systems in our institutions, and a large number of (but never enough) very talented and hardworking people maintaining them. Creating an inventory of systems and a network of information technology (IT) personnel has been a worthwhile exercise in its own right.

Third, we�ve been forced to confront our institutions� dependence on our IT system and to update (or create!) contingency and business continuity plans. Stuff does happen, and recently tested contingency plans help us all sleep better.

Fourth, we�ve come to understand the interdependence of our institutions and our communities. Y2K compliance may be only an �academic� exercise for a college or university lacking water or electricity or 911 service. Conversely, a campus with its own co-generation and heating plant may have a special responsibility to its community during a mid-winter emergency, whether it be caused by Y2K or an ice storm. Many of us have found Y2K to be a good reason to get to know IT leaders in other institutions, city government, and nearby industries; these contacts will serve us well in the future.

Priorities for the months ahead

Testing and contingency planning have occupied most IT shops this year, and these efforts will continue until the end of the year. While much progress has been made on remediation, some systems administrators and computer users elected to wait for the �final� update from vendors, so we can anticipate late nights as December 31 approaches.

Another focus will be communications, both within the institution and with our many external constituencies. Communication is a key element of contingency planning, of course, and failure of any electronic media will have serious consequences that must be thought through well in advance. In some disasters, campus e-mail systems and Web sites have been critical news and communication resources for entire communities. Does your Web server have a hot backup?

Communication may be equally crucial if �nothing happens� since overreaction and panic could pose greater problems than Y2K-related system failures. Hoarding, susceptibility to scams, and other adverse reactions might be minimized if individuals had confidence in key institutions, such as colleges and universities. Keeping the community informed of the Y2K readiness of the institution will assist faculty, staff, students, and neighbors in making their personal preparations for the new millennium.

Daniel A. Updegrove ([email protected]) is university director of Information Technology Services at Yale University. the table of contents