greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

Have you noticed that of the all the 39 questions that were provided by Mike Adams and Y2K Newswire that question 11: Why has no Fortune 1000 firm yet conducted full, end-to- end testing of the Y2K compliance of their computer systems is proving to be the one that isn't really getting answered. Why hasn't CHARLIE P REUBEN been able to debunk this particular question? Does he agree with the Weatherman that "the cost is too prohibitive"? Just when does the cost of this type of testing come into the picture(Can you say about 21 days)? So come on CHARLIE P, I'm waiting for your answer.

-- headwacko (isout@duhbunkie.site), December 11, 1999


I agree, headwacko, although I have yet to see any real answers to the questions.

Mike the Rat's answers would be fun to debate (since there's virtually no substance to them), but unfortunately he doesn't seem to like interactive discussions.

If anyone knows of any fact based answers to those questions, please let us know.

-- TA (sea_spur@yahoo.com), December 11, 1999.

Maybe no one answers you because they don't want to waste their time.

-- weare (laughing@you.com), December 11, 1999.


There is, like it or not, a point of diminishing returns to testing. In theory, you reach that point when the cost of administering the next test exceeds the cost you'd incur if you didn't test and that test would have failed. In a nutshell, it's silly to spend a dollar to test a dime.

In practice, you know how much the test will cost, but can make only an estimate of what the problems you're testing for might cost if they happen. The accuracy of this estimate can vary fairly widely. But if you've tested each link in a chain, and testing the whole chain would cost more than the chain is worth, why bother?

Beyond this, it's common knowledge that no organization is an island -- they depend on customers, suppliers, and infrastructure. They can only test what's within their jurisdiction to test (although they can, and in cases have simulated power and phone failures).

The point I'm making is that while you're correct that TRUE end to end testing cannot take place until actual rollover, Adams' question is rhetorical -- Why haven't these people done the impossible? Adams KNOWS this is impossible, so why ask? To spread FUD, most likely. Adams can't ask why big companies haven't done *sufficient* testing, because that's simple -- they HAVE!

-- Flint (flintc@mindspring.com), December 11, 1999.

Didn't MasterCard engage in end-to-end testing with all the clients interested? I sure got that impression from Arnold's TIME MACHINE reports.

Check out the last ones...at bottom of list

Not every company has an Arnold that is willing to document and publicly declare every stage of progress made down to the last minute, but if one wants to extrapolate, one must remember that there are positive experiences from which to extrapolate also.

-- Anita (notgiving@anymore.com), December 11, 1999.

Hey wacked-out, why not go to the Debunking Boards and check out the Y2K Weatherman's answers to your question. Each of the 39 were addressed and undoubtedly, you won't like the answers.

-- Bad Company (johnny@shootingstar.com), December 11, 1999.

Moderation questions? read the FAQ