“If this is true, it simply will not work in time. No way, no how.”

I stated a few posts ago regarding this paragraph discussing the healthcare.gov website…

– One person familiar with the project says it’s only about 70 percent of the way there, and has heard estimates of somewhere between two weeks to two months to fix it. As a programmer I know points out, “two weeks to two months” is the programming equivalent of “40 days and 40 nights”: “A long time, but I have no way of knowing how long.” When I used to hear estimates like that, I used to assume it would be coming in on the late end of that range, earliest.

Well, it turns out I AM not alone with this line of thinking. Lane Core Jr. commenting on the Facebook “Best of the Web” page states…

As an old IT guy myself (I was a coder for 20 years), this paragraph jumped out at me:

“One person familiar with the system’s development said that the project was now roughly 70 percent of the way toward operating properly, but that predictions varied on when the remaining 30 percent would be done. ‘I’ve heard as little as two weeks or as much as a couple of months,’ that person said. Others warned that the fixes themselves were creating new problems, and said that the full extent of the problems might not be known because so many consumers had been stymied at the first step in the application process.

Here is my interpretation: 70 percent is a number somebody pulled out of his butt. There is no realistic way to quantify percentage complete in a situation like this: multiple projects …spread across multiple contractors, especially when the systems are already in production and horrifically buggy. It is difficult to do such estimates with an in-house project. Frederick Brooks, in his book “The Mythical Man-Month”, observed that coding is 90% finished for half of the total coding time: anybody relying on such like estimates to

An estimate of anywhere from 2 weeks to 2 months means they actually have no idea how long it will take. Because they have no idea how much needs to be done and how much needs to be fixed.

And fixes do tend to introduce more problems. Brooks estimated that fixing a defect has a 20-50 percent chance of introducing another defect.

That’s my $0.02.