Quoted: The fact that it's dreamed up by nerds is the problem. I've watched feature after feature added to simple, already proven weapon system prototypes, by male engineers who are getting their ideas from video game combat, and female engineers who are just thinking about getting to daycare before 5pm, till a system that should take a few hours to train on winds up being a 250 page book, and a 2 week long class, with end users walking out still making basic mistakes due to the systems overbearing complexity.
|
Ed: I don't know what kind of systems you've been involved with, but in the Big Electronics Business it is
exactly the opposite situation.Even with a 100 page spec. the acquiring activity only has the roughest idea of what it
functionally wants. Because the resulting specification compliant system is a mere shell of what it needs to be there is a ton of "we should do this" and "wouldn't it be cool if" proposed by your male engineer prototype as described above.
But very little of it ever makes it in.Why doesn't it make it in? Because it means the original specification is wrong, which embarrasses the goverment program office and requirements guys. Because it means that the job, which was underbid to start with, now is going to go even further over budget and schedule than it already is, which is an even more deadly form of embarrassment for both government and contractor program management. In fact it can get the job cancelled. So we wind up delivering 100% specification compliant systems that only do half the job.
The sad thing is that this is acceptable to everyone except the grunts at both ends: the grunt engineer and the grunt user. It is acceptable because this is how you play the game. The federal budget is approved. We execute what we can for that year (or years). Too often what needs to get executed is too much to bite off for that much money, but if the services don't sell Congress the entire thing up front you don't get any money at all. Then when we fail, we do it all over again to get it right in the next budget cycle. If we demand that it be done right in the current budget cycle there is no money, the program gets cancelled (which means at best it will have to start all over again, probably with another contractor because "the first one screwed up") and the user gets nothing. And when I say "we" I mean BOTH government and industry.
It's a screwed up way to develop stuff because in the Big Systems world it is NOT a free market economy. You can't afford to go through a multi-million dollar development on speculation when you are only going to sell 10,000 of something. And from the service perspective they have to get Congress pregnant with each project and they have
so many projects, so inevitably some of them are going to be underfunded in the first go around.
What amazes me is how people allow themselves to be talked into playing these crazy games. But that's why government (and the military) are full of politicians. But that's no surprise. After all, I'm an engineer, the exact opposite of a politician
aa