Maybe it takes longer and longer for new minds to reach the frontier
"In green is the relative number of researchers compared to the number in 1930. This isn’t the number of researchers, its a measure of effective researchers, found by dividing total R and D spending by the nominal wage of high-skilled workers. Relative to 1930, we have almost 25 times as many (effective) researchers at work today. This leads to the author’s title question - “Are ideas getting harder to find?”. And the author’s answer is “Yes”."
Thus looking at individual industries/lines will help us sort out what is happening. The first example is transistors, and the count of transistors per square inch on a chip. Famously, this appears to obey Moore’s Law, which is that the number of transistors per square inch doubles every 18 months. If we count the number of transistors per square inch as “ideas”, then this translates to a growth rate of 35% per year. At the same time, the effective number of researchers working at chip companies (Intel, AMD, etc..) has grown by a factor of about 80 since 1970. Plotting this together yields their next figure.
Maybe discovery is only partly connected to total effort All you get with more research mind hours Is redundancy and tight interval between rival teams in the same races Maybe large teams slow the rate of discovery Much to ponder Even if global pop growthslows to zero Of course as available free labor ratios rise Thru increased productivity for necessary labor The research effort can increase