I hold a private pilot’s licence, I’ve written a lot of science fiction, and I spent the spring of 2026 doing something neither of those prepared me for: reading every published cislunar roadmap I could find and working out where they agree. The answer surprised me. They agree almost completely about the architecture. DARPA, the National Space Society, the Aerospace Corporation, AIAA, the Chinese Academy of Space Technology: a phased buildout from government-anchored exploration, through a propellant economy, to manufacturing in cislunar space for cislunar customers, with the Lagrange points as the connective tissue.
What they don’t agree about, and what the commercial discourse is getting wrong, is the timing and the sequencing. That’s the argument of The Long Bootstrap, a nineteen-thousand-word roadmap running from 2026 to 2120. Its thesis in a sentence: the cislunar economy is correctly hyped as a destination and profoundly underhyped in timing and architecture, and the keystone of the whole thing, a depot at Earth–Moon L1, is the piece nobody is funding.
The numbers that matter are small and specific. From L1 it costs about 2.5 km/s to reach the lunar surface, 1.7 to GEO, 3.8 to LEO, and station-keeping is roughly 10 m/s a year. Those three numbers decide the shape of the economy more than any launch-cost curve does.
This section will take other things too. The regenerative rocket engine over in Interactive is here because I wanted to see the thing I was writing about, cut open and firing. A game about the July crisis of 1914 is on the drawing board; it isn’t space, but it’s the same habit of mind, which is wanting to understand how large systems actually move.