Index ¦ Archives ¦ RSS

Where Good Ideas Come From

There is no clearer sign of us entering a new era than reading a book from 2010. Not so long ago, wide-eyed journalists still described the Internet — note the capitalization1 — as a force for good, an incubator of ideas whose capacity to connect people will lead to an exponential growth of innovation and prosperity. If one was to build a case against journalisitc blindness to externalities, Where Good Ideas Come From would make for a good exhibit.

The case against journalistic ignorance of patent law, too. The author, Steven Johnson, describes patents as means of rewarding inventors — and surely we can find a better way to reward them than a device that restricts the all-important distribution of knowledge. Only, that is not why the patent system was introduced, as described clearly if not succinctly by Vannevar Bush. Why a science journalist would straw-man a crucial factor of western technological superiority before attacking it is beyond the scope of this brief review.

The third feather in Johnson’s cap of muddled thinking is his conflation of discoveries and inventions, putting both under the broad category of “ideas”. The problem with that is apparent in the last few chapters of the book, where a series of 2x2s of ideas distributed according to the number of people involved (individual versus networked) and whether the enterprise is commercial (market versus non-market) “proves” Johnson’s case that non-market networked operations are superior, and should be supported above others. After all, from the 1800s onwards, most of the dots have been in their quadrant!

But here is a random2 sample of the 54 ideas listed in the non-market/networked quadrant: electron, RNA splicing, chloroform, cell differentiation, EKG, cosmic rays, universe accelerating, genes and chromosomes, atoms form molecules, radiocarbon dating. Only one of those ten, EKG, is now a physical product being sold and used. Two more, RNA splicing and radiocarbon dating, are methods that could be commercialized. Those three I would describe as inventions, and all three have a rather limited scope of use. Everything else are discoveries, telling us things about how the world works but not directly improving our lives in any meaningful way, other than satisfying our thirst for knowledge.

Now here are 7 ideas randomly selected from the 35 listed in the market/networked quadrant: contact lenses, washing machine, plastic, elevator, steel, television, radio. If Johnson’s quadrants prove anything, it is that having market forces involved is strongly associated with invention. We can, of course, discuss the direction of that particular arrow, and whether markets co-opt inventions they deem useful rather than actually developing them. The distinct lack of inventions in parts of the world where market forces aren’t in play hints at my preferred answer.

Having said that, there are some good ideas in this book about good ideas. One of them, the adjacent possible, a mental model for both discovery and invention, is also the name Johnson now uses for his (rather … good) newsletter. That is the concept that made me remember the book fondly after my initial reading, 10 years ago, and it is the newsletter that made me re-read it. And a good thing too — because even though the book is the same, both I and the world have changed enough to make it irrelevant. It won’t be in my re-read list.


  1. Also note that the Chicago Manual of Style and the Associated Press both revised their stylization to lowercase in 2016, Year Zero of the New Era. This was not, of course, the year’s only notable event. 

  2. Randomization was performed by my hovering a pencil above the page and dropping it with eyes closed. If the pencil hit an empty space, the first idea straight down from the spot was chosen regardless of distance. 

© Miloš Miljković. Built using Pelican. Theme by Giulio Fidente on github.