michael_nielsen

I've read it. Very underwhelmed. The result of the paper is that in some fields it's taking more people to achieve the same rate of growth (unsurprising). They define "new ideas" as rate of growth per researcher. This definition makes little sense, as far as I can see.

One amusing consequence of their definition: suppose I have an idea, I, that produces an increment of growth, Delta in, say, transistor density. If instead that idea had been invented 10 years later, no reason it would have produced the same increment,

In other words, it makes their evaluation of ideas very strangely time-dependent. Of course, that's not the main point: the main point is that they don't make any serious argument for their definition.

The first paragraph could be directly taken from the physicist community in the late 19th, early 20th century

That was my first thought, not even due to the "end of science" phrase, but just in general, due to the ever-changing ratios of number and scale of research efforts, society & economics, & culture. This is a cycle, albeit perhaps for different combinations of reasons each time.

Interesting take that computers can just be used to obtain more bits of precision. It is interesting to contemplate when computers have lead to surprise.

Fun fact: quantum gate teleportation was discovered, in part, by numerical search. My memory - Ike may differ - is that Ike wrote a little program to brute force answer a question. He showed the solution to me, & I said "Oh, I can explain what's going on using teleportation..."