A Twitter thread by Ben Reinhardt.

1/ Did you know that Vannevar Bush (you know, the guy who helped enable everything from radar to the manhattan project, the NSF to memexes) wrote an autobiography?

Turns out that yes he did, it's been out of print since the 70's, and it's *excellent*


2/ Before you even start, it has one of the best forwards I've ever read* and like much of the book, doesn't feel dated at all.

"I have drawn on the wealth of the vocabulary of the youth of our times. Theirs is a pungent stock of words, and action marks most of them"

3/ Bush is the king of precision and nuance.

Studying history can be *both* good (learn from the past) and bad (it can devolve into mythologizing.)

4/ Sometimes it gets super meta. He doesn't do a chronological tale of his life at all - instead he cherry picks stories he thinks are the best to learn from.

5/ More meta - "yes, I realize there are a bajillion books on the Making of the Atomic Bomb and the Radar* so I will skip those"

*I recommend The Making of the Atomic Bomb and Tuxedo Park if you are interested

6/ What *is* dated is his oft-repeated concerns about overpopulation. It's a wonderful reminder of how many things we thought would bring about the end of civilization that are now (arguably) non-issues.

7/ It's also fascinating which breakthroughs he thought were just around the corner, and which were not even on his radar.

Why did he think general antivirals were imminent? Why weren't they? (Would certainly be nice to have right now)

8/ Individuals making the difference between success and failure of critical technologies and programs is a huge theme.

9/ Specifically, he repeats the theme of innovations only happening because individuals went against the system.

Which raises the uncomfortable question - in order to get more amazing sci-fi shit, how do you systematize going against the system?

10/ Tons of detail about how they structured organizations like the NRDC and *why*

Feels like he's basically giving blueprints for people who want to follow that are easy to understand and hard to execute.

11/ He describes the *massive* shift in how science and research was done in the US during WWII.

This is where our modern system was born - both the good parts and bad.

12/ Today, university administration is basically bloatware.

But it actually started as a valuable labor aggregator so researchers could focus on research.

13/ This is the moment that Universities began to depend on federal research dollars.

I think this started as a good thing, but the relationship has metastasized.

14/ From what I can tell, Bush was a *great* manager.

15/ Also, apparently F.D.R. was a good manager too. 🤷‍♂️

16/ They could review a project in a week and get work started on it *the next day.*

More fodder for @patrickc 's https://patrickcollison.com/fast

17/ Bush earns his reputation as the person who can go between Military/Government, Academia, and Industry.

He gets both how the structures of each of them make sense in their own context, but then clash when they interface.


19/ Trust in lines of communication is underrated as a goal to strive for.

20/ The consequences of these organizational structures today - NASA changes it's goals every 4 or 8 years ...

21/ I'm still not sure if Bush or Jewett was right ...

Deserves more digging!

22/ Many of the stories emphasize @SafiBahcall's point in Loonshots about the need to manage the transfer of technology from the people who create it to the people who use it.

Also hi Millikan! I love how random science heroes just drop into the story.

23/ "I am a hog on the ice - see how I slide!"

24/ Committees can actually be good: when there are two groups who think very differently but have a common goal.

25/ Made me think a lot about the value of coordinating efforts vs. letting a thousand flowers bloom.

Where I came down is that you should absolutely have parallel efforts but there is an optimum amount of coordination that isn't zero. Were that optimum lies an open question

26/ War may be one of the few situations where *massive* numbers of people all have a real stake in the outcome.

27/ Innovations appearing outside the organizations which find them useful is still a huge problem today.

28/ To be written of like this: LIFE GOALS

29/ There are so many things that fall into this category: the concept is straightforward so it's not 'novel' but it takes a ton of R&D and grinding to even get it to the point of a proof of concept.

30/ I think Bush was a truly kind person. This is the story of what he did for civilians who kept sending him ideas for inventions they thought were desperately important to the war effort.

31/ The duality of command - sometimes commanders need to be obeyed no questions asked, and sometimes they need to be challenged. I love the idea of having a literal physical signal for it.

Apparently businesspeople used to use their ties for this too. We've lost these signals

32/ I feel like we don't talk about inventing things anymore. We have entrepreneurs, hackers, researchers ... but nobody is an inventor.

33/ Some things never change ...

34/ A sobering analysis of why large companies with a single major product have no incentive to incorporate an improvement (even a large one) to a single part of that product.

35/ I feel like *accidental* blackouts don't happen anymore in developed countries. One of those subtle but big improvements.

36/ Nuance about commercialization. I feel like *commercialize all the things!* can be as dangerous as *commercialization taints your soul!*

37/ Now this was a 🤯🤨🤔 moment - the assertion that patents aren't actually there to reward an inventor. Instead it's to give a venture captialist incentive to *fund* the inventor.

Still mentally masticating this one.

38/ I'm *very* hesitant to use the B word. But if there was actually a good way of encoding what was new about an invention - what constraints it relaxed and which constraint it had, might that be a place for (furtively looks back and forth and whispers) Blockchain?

39/ The level of meta self-awareness here is off the charts.

Also a lesson for a lot of people talking about AI today ...

40/ NBD

41/ He's calling out that electric cars would be awesome ... in 1970. Problem was the batteries.

Imagine if we had a good way to encode which technologies would be amazing except for a precise constraint - how much awesome could we unlock?

42/ Did you know that Stirling (of Stirling Engine fame) was a clergyman? So much early science (see: Mendel) was done by clergymen because monasteries were one of the few places where you had a bunch of educated people with time and relative safety on their hands ...

43/ Advances in material science are the root cause for a mind-blowing number of technological advances. The field is still underappreciated.

44/ Air force needing oxygen for pilots -> research to make oxygen -> cheap oxygen -> cheaper steel making -> order of magnitude cheaper steel. 🤯

45/ Oh those youngersters

46/ Why don't we have great fleets of hydrofoils?

47/ Another sobering look at why single-product markets like cars and trains are hard to revolutionize.

48/ He also invented new art techniques?

49/ Nothing like a well designed and fabricated crankshaft

50/ More things that haven't changed since 1970 :-/

51/ "These aren't the nineties you are looking for..."

52/ Learning how math works by building a machine to do the math!

53/ More nuance. Instead of the debate over whether education is just skill training or signaling, maybe it's complex and many things.

54/ The soundtrack for this passage: https://www.youtube.com/watch?v=WySzEXKUSZw …

55/ The book has Biographical Notes on almost every person mentioned!! I've never seen this before and it is amazing.


There are lots of great pieces I skipped so if this intrigued you - read it! It's (used) on AMZ or on http://openlibrary.org .

This is part of my ongoing research into people/orgs that enabled quantum tech leaps.
http://www.benjaminreinhardt.com or http://ideamachinespodcast.com