I like your analogy to programming a computer and I would take it a step further. In "The Mythical Man-Month," the author points out that computer programmers have a tendency to believe that the bug they just fixed is the last bug. i.e. "This time it will surely run." In other words, they are overly optimistic.

So too with self-improvement. It's easy to fool yourself into thinking that you've finally figured it out and that the diet you started is the sure path to permanent weight loss or whatever.

To make matters worse, it's very tempting to evangelize your new-found self-improvement technique. To proudly announce to the world that you've got the True Path to weight loss, quitting smoking, or whatever. And urge people to follow your plan.

As a consequence, the internet is full of diet and other self-improvement advice that doesn't work all that well. Not to mention that it's condescending and annoying.

A reputable company would not ship untested software, and I would propose a similar rule for any kind of self-improvement plan or advice:

1. Do not evangelize it until you've used it successfully for at least a year, preferably a year involving normal human experiences.

2. If you feel you must push your ideas before that, make it clear that you are doing so and commit to following up a year or two down the road to honestly let people know if things are still working.

3. In any event, commit at the beginning to honestly following up at the 2, 3, 4, and 5 year marks.