Excerpts from Y2K IEEE white paper

greenspun.com : LUSENET : Grassroots Information Coordination Center (GICC) : One Thread

With all of the arguments going back and forth, I decided it was time to present an INFORMED and UNBIASED report on what Y2K means. This is from the IEEE white paper on Y2k published last year (which has gone largely ignored. The full paperexplains WHY there was no Y2K bomb going off in the infrastrucure on Jan 1st. It also highlights the real threat and problem with Y2K. (one of the biggest threats listed was complacency caused by no major problems during the rollover).

What you are starting to see now in the posted reports is backing up their predictions.

If the right steps are taken, we'll see a short slowdown. If they aren't, expect a worldwide recession/depression.

This is a long posting. Go to their website at http://www.ieee.org/organizations/tab/Y2Kfocus.html click on Year 2000 technical information statement for the full document. [note: document is pdf format]

. . . we can say with relative certainty that the situation is not as bad as the worst-case scenarios of overly alarmist Chicken Littles, who do not understand the intrinsic discipline of engineering. In contradistinction to the serially coupled probabilities, given the innate reliability engineered into life-critical impact areas (like the robustness evolved into all natural systems) there is a relatively LOW probability of ALL things going WRONG at once.

It is sobering to realize that many Y2K problems, when analyzed quantitatively with even the most optimistic approximations and assumptions, are simply not amenable to any conventional process of repair  which is why remediation must be expanded to include alternatives. Note again, in this context, why the situation is deemed not as harmless as the wishful-thinking views of crisis-minimizingPollyannas, who do not understand the complexity of the problem:

A superficially trivial issue, the omission of the century digits from dates, has created computational ambiguities that first corrupt individual systems and then propagate to endanger interrelated systems and entire organizations. To discover, identify, isolate and correct every date-related operation within each individual system is challenging in itself; to integrate, coordinate, synchronize and validate all changes across the web of interconnected systems and organizations is the most daunting task our community has ever faced. Without a coordinated international effort to investigate, review, disseminate and promulgate practical, shareable, standardized problem descriptions and solutions, it is arguable that the current interconnective complexity of the global electronic infrastructure is at severe risk of collapse.

The purpose of this document is to provide an engineering perspective to a problem that spans an extremely wide range of technologies, organizations, geographers and human activities. The scope of this work is therefore not restricted to any one industry, technology or phase of the problem but seeks to encompass the full dimensions of the whole. Its goal is to support informed decisions and to encourage responsible action along productive courses carefully balanced between apathy and alarm. The name of the problem and its association with a point in time notwithstanding, the Y2K problem will impact many parts of the global information system infrastructure for years into the new century. This document is the first step in an ongoing process undertaken by the Institute of Electrical and Electronic Engineers to facilitate effective knowledge and information sharing by people who must address the problem both before and after its manifestations.

The implied instantaneousness of the term Millennium Bomb, all of the celebrated count-down clocks, the anxiety surrounding the rollover at midnight of December 1999  and even the term Y2K itself  all relate to the relatively minor clock issues, more solvable problems of hardware and low-level software. These issues, although important, are dwarfed by the Y2K problems inherent in the vast accumulated base of data in files and databases all over the world that must be processed by application software, software that may not know how to resolve the ambiguity caused by the missing century digits in year representations in that data.

For every non-PC computer clock involved, there are millions of data records. For the one-time rollover of system clocks there are astronomically countless times data records with potentially ambiguous century representations must be properly processed. While the rollover effect will happen in a moment and dissipate relatively quickly, data processing errors are happening and accelerating now and will go on for years after the rollover. Impacts may peak in the early weeks and months of 2000, when todays date is on the new side of the boundary, but they are not by any means confined to that period.

 The Good News is that the situation is not as bad as feared by the Chicken Little alarmists who dont quite understand the innate long-term robustness of naturally evolved systems or the discipline of innate, redundant and fail-safe reliability engineered into life-critical systems;  The Bad News is that the overall prospects for minimal impact are not as good as hoped by the Pollyanna optimists who dont quite understand softwares complexities and uncertainties, let alone its historical dismally low probability of successful remediation on the first attempt.

Figure 11 attempts to depict the notion of the cumulative effect of time on unresolved Y2K issues after the advent of the Year 2000. As time goes by, the aggregation of unremediated systems will tend to hasten the urgency of the problem. As these issues tend to cascade upon one another, the central tendency could migrate into more socially critical sectors as time passes. This leads to a central conclusion of this Technical Information Statement. The longer lasting are technical failures, and even more important, the longer lasting are the impacts or effects that are allowed to escape a given organization or control entity into the external world, the more cascading and ultimately more destructive the Y2K problem will be.

-- Sam Walker (swalker2000@earthlink.net), January 06, 2000


You are correct: nobody has been paying attention to this.

I have used IEEE material in What's Wrong with the Way the World Thinks about Y2K Part II and Part III.

-- Lane Core Jr. (elcore@sgi.net), January 07, 2000.

Moderation questions? read the FAQ