Interview: Machine Pix Weekly with Kane Hsieh
by Jason Crawford · October 23, 2021 · 1 min read
Kane Hsieh, who runs the popular @MachinePix Twitter account, interviewed me for his blog, Machine Pix Weekly. An excerpt:
At the highest level, economic growth throughout all of human history is like a super-exponential curve. In specific segments, it’s maybe exponential: certainly since about the beginning of the industrial revolution. But even in the segments before that there is some growth, some progress, so it wasn’t totally flat. If it was, the world of 1700 would have looked exactly like the world of 10,000 BC—which is obviously not the case.
So growth was extremely slow in the hunter-gatherer era, right? Major innovations that come along, maybe once a century or something, or less than that: maybe once a millennium. And then growth probably kicked up into a higher rate, but still extremely slow by modern standards when we got the agricultural revolution. I mean the first agricultural revolution back in 8000 BC or so when you got permanent agriculture and settled societies and so forth, and then growth was sort of slow for another 10,000 years or so—then kicked into high gear a few hundred years ago.
At the broadest level, some fundamental technology or societal shift happens, like agriculture or settled societies, or the printing press or the industrial revolution—and each of these things provide some fundamental infrastructure or capability, which not only helps the economy and human living, but also feeds back into the process of progress itself. That’s why exponential growth is even possible: these things work on self-reinforcing feedback loops.
And on the cathedral projects that used to take multiple generations:
On the one hand, you know, we can look back on those projects and marvel at the longevity of those projects; on the other end, construction productivity was super low if it takes you a hundred years to build. Rivaled only by the New York subway lines.
Comment on Reddit
