If "Next Time" Were "Tomorrow"
(Fri Jan 14, 2005) [/CodeCraft] #
This is just sad and embarrassing. I take no pleasure in reading about failed software projects. I've been there, and felt the anger that comes with putting time and energy into work that's never used. I've also been on successful projects, and felt the gratification of delivering software to the delight of its users. I just want to use my experience to help more projects succeed... because it feels good.
I won't speculate on how or why this project turned into "a train wreck in slow motion", but I'm guessing the final report won't be all that shocking. I'm most curious about the following quote:
One official said that "next time" the FBI would seek a modular system in which capabilities can be added or changed to the existing structure.
Without reading too much into the quote, it would seem the $170 million computer program fails on two fronts: it doesn't work and it can't be changed to work. Or at least changing it to work would cost more in terms of money and/or opportunity than scrapping the program altogether. If that's the case, then for all the bureaucratic and process failings we're bound to read in the report, the program is also flawed by bad code---it won't respond to change economically.
I hope the report has more to say about that quote, and I hope there is a next time. In the meantime, perhaps assuming that "next time" will be "tomorrow" helps put the inevitable change of software in perspective. Now, we don't know exactly how our code will need to change tomorrow, so we shouldn't risk speculating at great expense. But by assuming that someone will be in our code tomorrow, we can do inexpensive things to put that person in a better position to get the work done quickly.
