APPA President Joe Spoonemore is director of special projects, facilities management, at Washington State University.
At this point in time I am convinced that most facilities managers are familiar with the potential for computer system failure attributed to the inability of computer hardware or software to properly interpret a "00" in the year-date field. In reality, the problem lies in the potential inability of the CPU or BIOS to respond to the information delivered by the RTC(Real-Time-Clock). I will also assume that a lengthy discussion of the relationships of the motherboard, RAM, ROM, CMOS, BIOS, RTC, and CPU will be unnecessary. What we need to recognize and deal with is that in the US alone there are some 25-billion embedded chips similar to the old Z-80 type that now cost in the order of $1 and can wipe out an entire enterprise. Whether they are in PCs or PLCs (Programmable Logic Controllers) or whether the associated software cannot cope with the double-zero issue the task is to identify them, test them and develop a compliance strategy for mission critical services. The cost of world-wide compliance is estimated to be $600 billion and climbing. Less than six months ago it was assumed that 50% of this sum would be required to test the existing systems. This figure has increased and experts on the subject expect an excess of 60% of the total cost will be required just for testing.
The immensity of the problem, at first consideration, is almost overwhelming. So too are the consequences of not familiarizing yourself with the location and function of these little denizens of computer architecture. Unlike some pundits, I am convinced that with investment in preparation and readiness our industry can avoid the collapse of our modern computer driven society. Some skeptics are convinced that the domino effect will cause failure of electrical generation and distribution systems which will lead to the complete breakdown of our transportation, telecommunication, utilities and financial systems. Recent predictions from the electrical industry, however, indicate that somewhere between 1% and 2% of the national grid may actually fail. The lesson here is that we must recognize the potential impact of the loss of purchased power and be ready to react.
The word "ready" is the key word. Some would argue that we must be 100% compliant to avoid any possibility of loss of essential services. Realistically, 100% compliance cannot be guaranteed. No matter how diligent, we cannot absolutely guarantee the functionality of any system that requires the support of elements that are not under our control. As individual consumers we cannot guarantee the delivery of electricity or other critical services and materials to our site. Nor can we give assurances that all elements of a computer network will function flawlessly. The failure of a single piece of hardware or software could result in the loss of the entire network. Again, the challenge is to be ready ! We simply must not fail to deliver essential services to the campus community. Regardless of the depth of coverage you have in terms of written covenants from manufacturers or the potential for subsequent recovery via litigation the bottom line is that the campus community will find little comfort in these pieces of paper if they lose power and water. Speaking of litigation, it is my understanding that 40 states have passed legislation that holds computer and software vendors harmless should their product fail to be compliant as the world moves into the year 2000. Some protection may be afforded by complying with the "Good Samaritan Law" signed by President Clinton on October 19, 1998. The Year 2000 Information and Readiness Disclosure Act covers the time period from January 1, 1996 thru July 14, 2001. The retroactive period (prior to October19, 1998) must be addressed prior to December 3, 1998 to be effective.
Non-networked, non-IT, or smaller non-administrative computing systems require different strategies to address their specific needs. One approach has been taken by BOMA in their "Meeting the Year 2000 Challenge," General Motors in their "Year 2000 Test Procedures" or the University of Oklahoma as managed by Reginald Tempelmeyer. The difference between our approach at WSU and those previously mentioned is that (1)we are concentrating on only those systems that we perceive to be critical to life/safety, service delivery or research and (2)we have been given the task of searching beyond the Physical Plant purview and into the academic/research community.
At the outset, we considered the more-or-less classic approach of identifying and cataloguing all potential locations of embedded systems. Realizing that, with a very limited staff, we were considering a stupendous task we quickly refocused on only those systems that had a potential impact on critical services. The procedure is much the same as the other approaches in that once the potential impact on service delivery is determined a triage/prioritization is performed, testing is pursued, estimates for correction are prepared, funding requests are organized and finally, compliance projects are scheduled. What has been discovered to date is that the single most pervasive non-compliant networks are our Central Control and Monitoring System (CCMS/EMS) and the Safety Division Communication network. Initial estimates to bring the Landis&Staeffa CCMS/EMS system into compliance by updating the MBC's and SCU's is $250,000. The Safety Division has discovered that their Motorola Centralink, Silent Knight, Intesis paging system and 911-Dictaphone recorder are all non-compliant and will require a total of some $35,000 worth of upgrade effort.
Other plant infrastructure systems that have been examined and determined to be either compliant or inconsequential in terms of potential failure are as follows:
Unlike most Facilities organizations that I am aware of we, at WSU, have been asked to go beyond the infrastructure systems and examine the entire campus. For a very small staff this could be an overwhelming task if the rigorous approach were to be assumed. We, therefore, have deviated somewhat and simply ask the question, "will the failure of a given piece of hardware, research equipment or software result in life/safety trauma or significantly impact our ability to deliver service?"
To systematically assess the impact of embedded, time/date sensitive systems a Y2K Embedded System Checklist and an attendant Y2K Embedded Chip Risk Assessment Matrix has been organized. These forms are appended to this article and can be used to initiate thinking about how to begin a program. Between these two instruments the intent is to isolate those situations that would have the maximum impact on our ability to deliver service. A combination of Information Technology analysts, departmental computer technologists and the Y2K project administration has enabled WSU staff to begin a site visitation and assessment process. Once findings are recorded, compliance-cost estimates completed, and non-compliance consequences reviewed, contingency plans will be prepared as justification for further funding. The request for plant compliance funding, which appears to be the most overarching issue, has been submitted.
What have we discovered that is worthy of passing on to our fellow techno-victims?
In summary, it is not likely that the year 2000 will cause society to collapse. It is clear, however, that serious consequences await those that do not make the effort to root-out embedded systems that can impact the delivery of essential services. The following advice is offered to reduce the failure of computerized systems as they make the transition into the next century:
Appendicies:
