Hoff --

Doing some ruminating on the Debate Thread. I'm very dense on these two paragraphs:

"Finally, we need to consider errors due to system replacement. Implementations introduce errors at a much higher rate than normal remediation. Jones estimates that 5 errors per function point are introduced through new development. To be fair, not all errors are equal, just as many types of Y2k errors can be lived with. But to be ultra-conservative, let's assume only 15% of delivered errors in software implementations are comparable to Y2k errors. That leaves 75% of the function points in replacement systems containing errors on par with Y2k errors.

In our universe of non-compliant systems, only 33% were left untouched, leaving 66%. 50% of these were replaced, as opposed to remediated, leaving 33% of function points in non-compliant applications being replaced. From above, 75% of these will contain errors on par with Y2k errors, or 24.75% of function points."

I'm going to say, "duh", once I parse it, but could you tell me take me through the 15%, 75% and 24.75% calculations?

I think I may have some comments, questions, rejoinders once I get that through my thick skull.

Thanks.

-- BigDog (BigDog@duffer.com), September 17, 1999

Answers

Well, I'll give it a shot.

Jones cites studies that show as installed, software contains 5 errors per function point.

These errors run the gamut from trivial to catastrophic, with obviously many more "trivial" errors; the assumption being many "catatstrophic" errors are caught prior to implementation.

Y2k errors also run the gamut from "catastrophic" to "trivial", but my guess is the distribution would be more "normalized", since we are talking about errors in non-remediated systems, and as such assuming no testing of Y2k had occurred.

To account for this, and add a level of conservatism, I made the assumption that only 15% of the errors introduced through implementations are "on par" with the full range of Y2k errors. Taking 15% of the 5 errors per function point left .75 errors per function point introduced through implementations at the same level as the full range of Y2k errors.

Part of the original assumptions was that 66% of the "non-compliant" function points were either remediated or replaced, with the other assumption being that 50% of those were replaced by new systems, and 50% remediated in the old systems. So 50% of the 66% of non-compliant function points were actually replaced, giving 33% of the non-compliant function points replaced. From above, 75% of those would have errors "on par" with the full range of Y2k errors, leaving 24.75% of the starting non-compliant function points with errrors "on par" with the full range of Y2k errors.

-- Hoffmeister (hoff_meister@my-deja.com), September 17, 1999.