5 years ago today, I slipped away from a Xmas party at a friend’s house and snuck upstairs and watched a Falcon 9 first stage return to land at the Cape after boosting 11 Orbcomm satellites toward orbit. I posted this reaction and promised, “…thoughts later.”

I think 5 years is long enough to now share my thoughts. A thread follows.

When engineering students ask me, “Isn’t it cool when Space-X lands their Falcon first stage?” I nod in vigorous agreement, but then follow-up with a hint of condescension, “You have *no* *idea* just how cool…”

For a decade, I sat through stacks of vu-foils AIAA meetings, and another decade of PowerPoint after that, listening people argue that SSTO was a bridge too far for reusability, and that two-stage reusability was the key. Spreadsheets, renders, endless slide decks. No hardware.

Then, in the mid-90s, things got serious. In Seattle, in addition to having our leggings tucked into Doc Martens beneath oversized flannel shirts, everyone and their cousin were discussing two-stage-two-orbit with a reusable first stage and plans to reach a reusable second stage.

The catalyst: Seattle-based Craig McCaw, founder of McCaw Cellular, floated a concept called Teledesic. Originally pitched, Teledesic would have been a constellation of 840 commercial satellites to provide internet worldwide. Think Starlink, 25 years too early.

The race was on to develop a way to deliver the Teledesic birds, and two-stage with a reusable first stage and ambitions for reuse of the second stage was in the drinking water. This zeitgeist led to the creation of Kistler Aerospace, by Walter Kistler and Bob Citron.

This early analysis of the Kistler TSTO approach by Jason Andrews—who was at UW-Seattle at the same time I was in mid-1990s—with its “powered bounceback” return of the first stage, is interesting to compare to Space-X’s current Falcon-9 fly-back technique: https://doi.org/10.2514/6.1996-2772

Those of us hanging out on Usenet in those days also remember William Mook’s GreenSpace project: Two-stage-to-orbit with full reusability, which also presaged Space-X’s approach in many ways, right down to pintle-fed engines.

Heck, even Bill Gates was getting in on the reusable TSTO action. Check out this 1994 patent on a two-stage vehicle with recovery of both stages at the launch site:
https://patents.google.com/patent/US5568901A/en

Yes, that really is *the* Bill Gates. With this much talent and money involved, almost everyone paying attention was expecting to see a reusable TSTO on the pad within a decade. But we had to wait.

The collapse of Teledesic and the fact that Iridium, Globalstar, and Orbcomm all went bankrupt around the same time that the dot-com bubble popped (~2000) meant these launch vehicle development programs went into hiatus, although their DNA can be found throughout NewSpace today.

So, how *did* Space-X do it when these earlier efforts never got to the pad?

Skipping Falcon 5 and going directly to F 9 was an unbelievably risky move, but the extra capacity of F 9 v1.1 and upgraded Merlin engines gave the margins (i.e., excess propellant) to experiment with powered recovery for landing, after earlier attempts with parachutes failed.

These tests were done under the NASA Commercial Resupply Service program, Orbcomm, and other commercial satellite launches in 2014, all paying customers.

In short: Get other people to pay for your development program for first stage reusability—since the customer won’t care what you do with the booster once it’s done its job—and even make money in the process.

𝗧𝗵𝗶𝘀 is the really amazing thing that leaves me in awe of Space-X’s achievement. Not that they thought of flying back the first stage booster (people had been talking about it for decades), but that they 𝗱𝗶𝗱 it while 𝗺𝗮𝗸𝗶𝗻𝗴 𝗺𝗼𝗻𝗲𝘆, even when it 𝗱𝗶𝗱𝗻’𝘁 work.

And five years later, this is where we are at... https://twitter.com/sciguyspace/status/1341382374496583680