Stanley Chodorow
Provost

John Fry
Executive Vice President

A Short Story

The calendar reckoning that will give us the Year 2000 problem goes back to a monk in the 6th
century A.D. named Dionysius Exiguus ("Dennis the Short") who calculated the date of the birth of Jesus and proposed dating before and after that date. Though the system only began to come into common use in the eighth century, the very first person to use it appears to have been Dionysius' friend Cassiodorus, another learned monk writing in the year 562.

Scholars now date the birth to about 4 B.C., which means that if Dionysius had got it right, this year would be the year 2000 and we would have had that much less time to get ready. As it is we hope to have all key systems at Penn "fixed" and ready by December 1998.

--J.O'D.

Will Your Computer Know What Year It is?

by James J. O'Donnell

Every millennium begets its own terrors. While few of us expect the world to end in the Year 2000, we are already living with our own apocalyptic predictions: newspapers, magazines, radio and other media issue a daily Year 2000 warning. Will our computers stop working? Will our elevators and security systems suddenly freeze up? Or is it all media hype?

This fear is in many ways very real. Penn, like any other institution or enterprise, relies heavily on computer systems and devices that control the operation of equipment for crucial research and administrative activities. Many of these systems and devices will not function correctly in the Year 2000 unless we prepare them. (It's simple: if like many systems over the last decades, a system is programmed to record dates by a two-digit number, then the move from 97 to 98 to 99 is easy enough, but the move to 00 means that you've suddenly moved back in time 99 years: all sorts of calculations can go awry if this happens.) But at the same time, we need to take a balanced approach. The appropriate mentality is that of "risk management." If you know you will buy a new desktop system before January 1, 2000, then you can breathe a little easy: new machines are already 2000-compliant. But if your system is mission-critical and will not be substantially upgraded in the next 26 months, you should take a serious look at it to see what risks it contains, then think about the costs of remediation.

Many of Penn's academic and administrative offices are working aggressively to resolve their problems. Some changes have already been made (in time to cope with the arrival of the class of 2000, for example), but much more remains to be done. Our goal is not just to fix code but to assure that the essential business of the University continues uninterrupted. A working group under the sponsorship of the Provost and EVP is being formed to bring together and coordinate the activities already underway in many schools and centers. Our WWW site (http://www.upenn.edu/computing/year2000) should be consulted frequently for progress across the University and for up-to-date information from external agencies such as NIH concerned about the Year 2000 problem. The site will include names of individuals in each school and center who are leading their school/center efforts, as well as information about specific hardware and software releases that are and are not 2000-compliant.

But centralized and coordinated activity cannot do the whole job. Many researchers are believed to be at risk for systems built up over time, collecting valuable data, and continuing to play an important part in their work. Every computing system and program must be presumed Year-2000 vulnerable until proven otherwise. The key to success in bringing all systems safely into the Year 2000 will be communication and information sharing at all levels within the University. A survey to be conducted shortly by each school and center will assess current progress and the magnitude of the remaining effort. Please respond fully and quickly when we seek this vital information.

ISC, Penn's central computing organization, stands ready to work with the IT professionals in schools and administrative centers to advise and consult and make appropriate reference for individuals and organizations with year 2000 needs that may yet be unaddressed. Please write or call directly to me (jod@isc.upenn.edu, 898-1787) or to Robin Beck, associate vice president (beck @isc.upenn.edu, 898-7581).

Dr. O'Donnell is Professor of Classical Studies in SAS and is the University'sVice Provost for Information Systems and Computing.