2) Phases of Y2k and Post-Y2k Effects:
As Y2k-effects developped in 4 phases, related experiences can be equally grouped:
Phase I: Preparation and Testing for Y2k Avoidance in all Sectors
Phase II: Roll-Over Experiences esp. in Critical Infrastructures
Phase III: Restart and Work in Daily Usage (starting January 3, 2000, and including February 29, 2000)
Phase IV: Medium and Long-term Effects
Phase I: Preparation and Testing for Y2k Avoidance in all Sectors Experiences and reports show that even in many of those systems which had been analysed and corrected, adverse unforseen effects have been observed during testing. Most of those effects have probably been corrected before rollover. In this phase, information about "problems" may have leaked because someone observed unforseen effects, or because experts were interested in the "information leak", e.g. to enforce Y2k-related activities in their organisation.
In this section, documents or links
refer to related Pre-Y2k experiences.
Phase II: Roll-Over Experiences esp. in Critical Infrastructures During and shortly after "Roll-Over", essentially those systems were on-line which had been well analysed and prepared. For some such systems, negative effects could have been observed directly (e.g. failure of electrical grids, telephone lines and traffic systems). For other systems, effects were almost unobservable (e.g. work of specific nuclear power plants, emergency systems of police etc). When some unforseen Y2k effect materialized, it was a priori very improbable that this would be reported soon. Only after some time, "information leakage" will give some hints what may have happened. Indeed, there are reports about Y2k-related failures during Roll-Over.
In this section, documents or links
refer to reports about Y2k-related effects during Roll-Over.
Phase III: Restart and Work in Daily UsageThis phase deals with applications in offices, industry and government which were switched off during roll-over period and were started before or when "normal work" began on Monday January 3 or later, esp. in offices and production. Reports about related experiences will surface only with significant delay. But some of these effects became visible in Internet webpages where the usual date procedures (often used in similar ways in client and server software) produced a manyfold of "bad dates", such as year "100" (arithmetically following on .., 98, 99) or "20,100" (improperly joined from "2000" and "100") and "3900" (wrong combination of "2000" and "1900") or "1900" (on some Microsoft websites :-) and others. Screenshots with related findings are available from different websites.
One special effect will appear on "February 29, 2000" , as Year 2000 is (different from other full centuries) a leap year. As media are interested in related effects, such reports are more likely than reports about "normal failures" in phase III.
In this section, documents or links
refer to Post-Y2k Experiences in Daily Applications.
Phase IV: Medium and Long-term Effects
In this phase, long-term effects may arise as some specific programs or date-related functions in database software are only very rarely invoked, e.g. quarterly or yearly. It is expected that related events may even appear after 2000, esp. including roll-over from December 31 at 23:59:59 hours to January 1, 2001 at 00:00:01 hours.
In this section, documents or links
refer long-term Post-Y2k experiences.
(Last updated: Monday, January 17, 2000)