RESULTS of my "very unusual poll" & the ANSWER to question #2greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread
To poll participants:
After sifting through the responses to my three questions, I found that your degree of optimism/pessimism did NOT correlate with your answer to question 2, but there appeared to be a strong correlation with question 3. For example, the "very pessimistic" group was highly consistent in their belief that the Federal government must have >75% of ALL its systems compliant in order to avoid major problems, but the "pessimistic" group was much less consistent and required the Fed to have between 50-75% of their systems operational. Unfortunately (from a statistical viewpoint), only a few participants were either neutral or optimistic, but when I lumped these two remaining groups together, it appeared that they required only about 50% of all Fed systems to be compliant. With these results in mind it's instructive to see what Congressman Horn had to say on this subject. In his March 18, 1998, "Opening Remarks", Representative Horn told his Y2k Oversight Subcommittee the following:
"Roughly speaking there are 8,000 mission-critical systems plus 60,000 second and third tier systems. In addition, there are thousands of data exchanges with foreign, state and local government, businesses, and citizens. There are telecommunications systems, biomedical devices, and uncounted millions of embedded computer chips. We cannot allow all these so-called non-mission-critical systems to fail...The collective confusion of tens of thousands of secondary systems failing could be catastrophic."
As many of you know, since that date, the Federal government has repeatedly whittled down the number of mission-critical systems. (Why do you think they keep doing that?) I believe the most recent figure in 6399. Therefore, the MOST conservative estimate of the "percent mission-critical systems" would be based on the ratio 6399/(8000+60,000) which gives 9.4%. However, this percentage is unacceptably high to me because compliant Fed-based data exchanges, and embedded chips MUST be considered essential if THE MISSION OF THE FEDERAL GOVERNMENT in all essential areas has any chance of operational reliability. Coupling this requirement with the millions of desk top computers that are part of this matrix and which have NOT been counted as part of the Fed's mission-critical "systems", and I think you come to the horrific conclusion that not only is the actual number well below 5%, but realistically speaking, Horn and virtually every other government spokesperson, have been playing a giant shell game with the American people. Folks, the emperor has no clothes, all you have to do is open your eyes and look.
-- Dr. Roger Altman (firstname.lastname@example.org), March 20, 1999
Good work and report.
-- Watchful (email@example.com), March 20, 1999.
So very roughly speaking, there would appear to be a positive correlation between mistrust of the government, and assessment of the importance of government in our lives. This makes good sense -- the larger the role we think government plays, the more concerned we are with that role being handled appropriately. Conversely, those who believe most government activities could vanish and almost nobody would notice a thing, aren't too concerned about their noncompliance.
-- Flint (firstname.lastname@example.org), March 20, 1999.
I received the following email:
Hi, Dr. Altman.
Can I ask you a question regarding your poll result posting? You stated as follows:
"Therefore, the MOST conservative estimate of the "percent mission-critical systems" would be based on the ratio 6399/(8000+60,000) which gives 9.4%. However, this percentage is unacceptably high to me..."
Did you mean to say "unacceptably low"? Meaning that truthfully, the number of mission critical systems have been underestimated.
Or am I misunderstanding the context?
Sylvia raises an excellent point which brings up the basic semantic definition of "mission-critical" systems. If we accept the government's number of 6399 mission-critical systems as absolute and unequivocal than ALL OTHER SYSTEMS must be, by definition, "non-mission critical". Therefore, newly "discovered" system could only be accounted for by assuming it is "non-mission" critical by adding it to the denominator. Thus, my comment that the 9.4% was probably way too high.
However, common sense tells us the there are far more than 6399 mission-critical systems that, indeed, could and should be added to the numerator. Unfortunately, if they weren't counted as part of the TOTAL NUMBER OF SYSTEMS in the first place (as in the case of PC's data exchanges, and embedded chips) then we would have to add them to the denominator as well. Since the resulting percentage could wind up confusing just about everybody, I chose the former logic, however illogical it may appear at first.
-- Dr. Roger Altman (email@example.com), March 20, 1999.