An exponential function, if you zoom in to your local spot, looks linear. This got Malthus. (Actually, Malthus's model was wrong but his conclusion wasn't far o

An exponential function, if you zoom in to your local spot, looks linear. This got Malthus. (Actually, Malthus's model was wrong but his conclusion wasn't far off. The world had been continually experiencing Malthusian catastrophes until about 1850, when the rate of economic growth overtook population growth. Colonialism was, to a large degree, Europe outsourcing its Malthusian catastrophes. For an intra-continental example, see: Irish potato famine.) He thought economic growth was linear, not exponential. OP's claim that 1913 to 1967 was an era of minimal (linear) change couldn't have been more wrong. While many nations returned to economic "trend line" there was a lot of, um, painful stuff in between. Even if you downplay World War I and II, you've still got the 1918 flu, the Great Depression, etc. Also, the US went from a regional power to a superpower in about 30 years, while Europe dealt with poverty (and lingering remnants of fascism) into the 1960s. That's a change. Only Americans (excluding those who served in the wars) had the privilege of saying that 1913 to 1967 was a smooth ride. Oddly enough, this time period (in which he understates progress) was written when world GDP growth was at its all-time (unmatched, even now) peak of 5.7% per year (we're around 4.5% now, which is better than we've been). The R&D efforts that didn't seem to be delivering (because flying cars and AI hadn't been invented yet) were already pushing the economy to grow faster than it ever has (and will have, as of 2014). It got a bad rap ("people in those cushy R&D labs aren't earning their keep") because it underperformed relative to unreasonable expectations, not because it wasn't working. (Oh, and there was that perceptron thing; the bit of high-school geometry needed to show that a single perceptron can't do XOR became "neural networks don't work" became "AI is a dud".) So we have a lot to learn from the 1960s, and as a society, we should consider dumping the Snapchat nonsense in order to some real R&D.