Kelly: More than My Share of It All comes by way of Paul Graham and I see why Graham likes it: Kelly is the sort of person who barely exists anymore. Kelly worked on numerous important aerospace engineering projects from World War II into the 1970s, and he oversaw vital projects like the SR-71 Blackbird and the U-2 Reconnaissance plane—both of which were major innovations. In an era of the totally fucked up F-35 and numerous similar systems, it’s shocking to read about genuinely innovative projects completed on time and sometimes even under budget. It’s shocking too to read about someone who sounds like a person rather than a bureaucrat, and who argues for responsibility instead of buzzwords:

There is a tendency today, which I hate to see, toward design by committee—reviews and recommendations, conferences and consultants, by those not directly doing the job. Nothing very stupid will result, but nothing brilliant either. And it’s in the brilliant concept that a major advance is achieved.

At the time Kelly worked, large aerospace and related companies acted like Google or Apple do today—perhaps because their founders still ran them. Kelly writes about how he once “telephoned Walter Baird [of the Baird Atomic Company] personally since he and I had worked together on a number of other Skunk Works projects. He immediately agreed to pick up his end of the log.” A direct call to a decision maker is often an improvement over hundreds of hours of committee bullshit. Many people know this intuitively but many systems, in universities and business, have run to committee. Kelly writes, “I fear that the way I like to design and build airplanes one day may no longer be possible.” In that sentence I think he should have “may” before “one,” but the important point remains: that day has arrived.

One could profitably read this book next to Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future. Musk is a Kelly-like figure, and, together with Zero to One, Kelly tells a story about how large swaths of the nominal technology sector have become sclerotic. Every single person working on the F-35 should be required to read Kelly. But the people at the top of fat and lax military-industrial companies are already getting theirs.

The sense of differences extends to Kelly’s university experience, where he writes that “The professors were broadminded people, with interests and contacts outside the university. They took a personal as well as professional interest in their students.” Today “a personal as well as professional interest” leaves professors open to university politics and sexual harassment claims. Money is tighter; he writes of how the University of Michigan built a wind tunnel and let Kelly operate it part time. He says, “The money didn’t mean anything to the university; renting the tunnel afforded them a chance to see what the students could do.” Money today means a lot to universities, and the faculty seem powerless to reverse that trend.

Kelly and Kelly are about taking reasonable risks with good cost-to-payoff ratios. Risk does imply the possibility of failure, though, and reasonable failure isn’t tolerated in many big institutions. That may help explain why startups are so important, why much innovation happens on the Internet, and why Amazon.com’s ebook systems are so important for writers. Much of Kelly is not written particularly well, but the larger points it makes make it a fascinating historical document anyway, and a reminder of what can be accomplished by determined people in systems that let them succeed.