The reason people, not just hackers, have opinions on macroeconomics is that unlike many other branches of science it hasn't come up with a lot of definitive an

The reason people, not just hackers, have opinions on macroeconomics is that unlike many other branches of science it hasn't come up with a lot of definitive answers. As the article notes experimentation, a basic ingredient in science, is very hard in macroeconomics. This means that there are all sorts of ideas floating around on how economies actually work. Some reputable, some not. Newtons laws of motion don't get discussed much because experimentation and time give overwhelming evidence that they're probably right. Thus there's not much point in discussing them, they're accepted as a foundation on which you can build, which is exactly what science does. Standing on the shoulders of giants. Economics is different. Partly because of the inability to test a thesis, partly because everything is closely linked making it hard to distinguish cause from effect. I actually believe macro economics is currently in it's early stages, with professional practitioners still trying to get a grasp on the basics. Even among Nobel laureates in economics there's still disagreement about whether Keynes, one of the most heralded economists ever, was right or wrong. And he wrote his main work "The General Theory of Employment, Interest, and Money" in 1936. Likewise, if you look at the wikipedia entry on the great depression there's still substantial debate as to what caused it, and how we got out of it. I also think that this discussion is good. Yes, there are some ridicolous theories on youtube, but the serious amateur probably knows enough to discard these as the crap they are. A public discussion of macroeconomics will, however, be very fruitful as people from other fields bring their expertise to the subject. Mandelbrot and his knowledge of fractals and mathematics has interesting points on economy, Kahnemann uses psychology to explain economic phenomena, and hackers, of course, can set up interesting computer models, which is probably the closest we come to experimentation in this field.