OK, first of all let’s get this straight cost $70,000,000.00 not $634,000,000.00. The latter near astronomical number is the total amount of contracts awarded to CGI, the Canadian Development Company that built the POS.

As for the “tech surge” nonsense, and the notion that they can fix this pork barrel programming project by December. The simple answer is NO. The reason I can state this so unequivocally is because I know from experience that rebuilding a large scale functional program is not the same as doubling down on the battlefield. More boots on the ground won’t fix the problem anymore than a bunch of cooks can fix that Christmas ham you left in the oven way too long.
If just a few of any of the myriad issues that have already come to light are true:

  • they used pirated open source code
  • it represents largest Oracle 11gR2 Identity and Access Management deployment in the world
  • it’s is over 500 Million lines of code
  • it wasn’t tested until the week before the rollout
  • it’s unsecure
  • and obviously just plain doesn’t work

then this pig ain’t gonna fly…

A key indicator of just how bad this program is is the sheer size of the code. Contrary to popular opinion really, really big programs are not the key to successful development projects. In fact, the opposite is true. Small is better because the odds of outright failure are much lower when each task to be accomplished in a much larger project are completed tested and assembled into a larger functioning whole.

500 Million lines of code is enourmous. To put this in perspective this is 10 times the size of the Vista operating system, a dysfunctional operating system that was also know for coding bloat.

This in and of itself, outside of the fact that the site simply doesn’t work, is a strongest indicator of a seriously dysfunctional development process where lines of code were used as a metric for progress. The precedent of counting lines of code to justify big ticket budgets is not new. IBM had a group that would count K-LOCs, to define a budget. In the overly simplistic world of K-LOC’s, a 20K-LOC project would be twice the budget of a 10K-LOC budget. Now consider the implication of a “clever” programmer who can do the same work, better, with only 4K of code. That guy just cut the budget of the 20K project to a fifth. So much for hiring clever programing ideas.

If I’m correct in my assumption about the overall quality of the code (and believe me, I am.) Then there is no option other than starting from a clean slate. Any thought that you can fix this pig by combing through the myriad recesses of it 500 million lines of code is absurd and even if you did take on this Sisyphus-ian task, the program’s poor architecture would ensure that it still would not work. This program should be gutted and we need to seriously re-evaluate the entire concept of these insurance exchanges where major insurers can pull out them preemptively in the first place.

Unfortunately, this is not the way things are done. So what is going to happen?
1) They will be forced to push the date of requiring people to be signed up with an insurance provider by a few months.

2)They will throw a bunch of money at the project, it will never work, but many people, to avoid the penalty, will still sign up off exchange, in order to avoid the penalties. And this will be described as a qualified success.

3) One good thing that will actually happen is that many people who could not afford insurance will get on some state subsidized program that they didn’t qualify for before due to lowered requirements.

4) Those in the middle will find that their coverage will be more expensive and cover less than what they could have had before ObamaCare.

In the end, the number of people who file for bankruptcy due to medical debt who had insurance will actually go up, because the fundamental premise that someone can afford to continue to pay high insurance premiums when they have a long term debilitating illness is simply absurd.

Leave a Comment

Your email address will not be published. Required fields are marked *