One of the mantras we have in software engineering is
premature optimization is the root of all evil. The reasoning behind that is if you start optimizing parts of an application before you know it's going to increase your performance, you risk wasting a bunch of time without much (if any) performance benefit. Smart engineers gather data on the performance characteristics of the application and attempt to optimize the parts of the application where the most time is spent. That way you can get more benefit from your effort, since small optimizations can yield big improvements.
You can use the same principle when discussing the Federal Government's budget. Our country's budget has a performance problem: it budgets more than it takes in taxes. The stupid thing is to take this problem and cut things based on ideology (ie.
get rid of the NEA and you'll get rid of the budget problem). The smart thing is to look at what you're spending the most money on and cut based on that data. That way, you can make smaller (percentage-wise), easier cuts and make big differences.
It's why I love looking at this illustration of the 2009 budget. It's data that we can use to see where our cuts can have the biggest impact. So if, for example, you suspected that cutting the National Endowment of the Arts would make our budget problems go away, you could quickly see that this would only reduce the total discretionary budget by .01%. So while it still might be a good idea to take a look at how that money's spent, it wouldn't be the first thing to look at. On the other hand, if we eliminated the Global War on Terror, we would reduce the discretionary budget by 16%. And if we didn't want to eliminate either of those things and instead had a goal of reducing the total budget by, say, $100 million, it'd probably be easier to reduce the budget for the Global War on Terror by .06% than it would be to reduce the budget for the NEA by 78%.
So when I read about McCain complaining about spending $3 million to study bear DNA while exempting defense spending from any possible spending freeze he'd put on most of the rest of the budget, I get the sense he's more concerned with cutting based on ideology and not on actual data. It may be that $3 million to study bear DNA is excessive, but I get the feeling he didn't do any in-depth analysis on that. What I do know is taking a budget item that's 60% of the total discretionary budget and saying
this is exempt from any spending freeze that might occur is not a decision that can be based in rational thought. Indeed, anyone who looked at the budget objectively would spend a lot of time and energy looking at the defense budget since it's by far the department with the largest budget.
Not that I'm saying we need to put a spending freeze on defense (or any aspect of the budget, for that matter). It may be that the budget allocates money in the exact amounts we need to spend for exactly everything we need. And if that's the case you need to come up with a way to increase revenue. My suspicion is that this isn't the case, that there are budget items that could be decreased or eliminated; but I don't know for certain because I don't have enough information on the subject. I just found it interesting that, of the few things McCain didn't want to freeze spending on, the biggest item on the budget is on that list.