I guess the reason I am so strongly advocating single-purpose hardware here is because of the fact that I don’t think that any government would be willing to fund the long-term development needed for a viable space-truck. That said, NACA and its descendant programs within NASA have existed since the 1930s, right? The idea of theoretical, government-funded aerospace research is there in the X-program, it just must have looked too expensive to have been funded without an express purpose.

The silver-bullet answer to that challenge was “reusability and high flight rates.” Both cost money up front in order to save over time: reusability meant wings, complex TPS, more robust airframe and engines (all of which cut into payload ratio, which sucks to begin with for rockets); and high flight rates meant more infrastructure and big turnaround teams. It was supposed to be done on an Apollo-ish schedule, on a budget that in the end was about 40% of Apollo (constant dollars.)

That was like greeting Roald Amundsen on his return from the South Pole: “Nice job — now we’d like you to build a railroad from the coast to the pole, two trains a week, suitable for passengers, cargo, scientific and military needs. Here’s 40% of what you spent before. Good luck.” That was not going to happen, Anthony, despite all the Monday-morning quarterbacks asserting since 1981 that their alternate STS approach would have been 1/10th the cost and flown 10x as often with a ground crew of 3.

Having been a space fan since 1957, and science writer through the 70s and early 1980s, knowing the STS principals, I reject the neat and convenient “lies NASA and the Administration told.” We all drank the Kool-Aid; we were all on a post-Apollo, we-can-do-anything high, and chose not to look too closely at the profound difference between the challenge of a closed-end moon race and the challenge of getting into space frequently, affordably, and flexibly.

In hindsight, in a perfect world, we would have said not “We’ll create an operational space truck in 6-9 years,” but “We’ll start a series of limited, overlapping X-craft programs, each with narrowly defined obectives: one for propulsion, one for flyback stages, one for TPS, one for fast turnaround procedures, etc. None will deliver any payload to speak of; candidly, we can’t say how long before the results add up to a space truck, and we can’t say how much the total cost will be, so keep the checkbook open.”

Good luck selling that to any government, ever, not just Nixon’s in 1972.

Given the technology available c. 1970 and those reasonably expectable over the next few years, there was no way NASA (or anybody else) was going to achieve the promised flight rates and costs within a decade. They (and OMB and Congress and the public) pretended otherwise, and proceeded, and declared the result “operational” after four flights (!).

Given the Shuttle we actually had in 1984, there was no way that a multi-hundred-ton space station (or space commercialization, or SDI, or anything else dependent on much lower $/kg than Apollo) was going to happen on a reasonable schedule and budget. We pretended otherwise, and proceeded…

And 22 years later, say “Wha’ happen? Gosh, I guess the manned-spaceflight side of NASA got stupid after Apollo.” What we should be saying is: “We all told ourselves a string of pretty lies about keeping up the Apollo pace of ‘milestones’ while dropping NASA spending from ~5% of the federal budget to less than 1%, and reality bit us.”

I don’t have a silver-bullet answer to CATS, which I believe to be a much tougher engineering and economic challenge than Apollo was. I don’t believe the newer pretty lies — that a different STS design would have been dramatically better, or that Unleashed Free Enterprise will now do quickly what Big Dumb Government screwed up over 35 years.

What I do have is a conviction that any mission or program conceived as if we’d solved CATS, when we haven’t, is bound to be a disappointment.

I think it’s probably much simpler to run a mission-driven robotic program because the costs are smaller, and there are fewer hands in pot, so to speak. But nevertheless, this doesn’t justify why NASA has totally failed to accomplish its stated goals with either the Shuttle or ISS programs.

Thanks to Moore’s Law and prohramming advances, we get a lot more function today from a kg of circuitry in a planetary probe or comm or sensing satellite) than we did decades ago. But the hardware required to enclose, support — and bring back — human beings hasn’t gotten much lighter at all.

So I trace the ISS’ shortcomings not to some inherent managerial dysfunction on the manned side, but to the underlying delusion — Reagan 1984 and ever since — that we had a cost-effective “space truck” to build and support a multi-hundred-ton station… when we didn’t then, and still don’t.

Try this exercise: imagine that Freedom/Alpha/ISS had been planned as fully automated, unmanned, from the beginning — but the same total mass. Would it have been completed on schedule, on budget, and be doing lots of good science and/or free-fall industrial research?

“That’s absurd,” you say. “An unmanned station wouldn’t need anything like the same habitable volume [i.e. payload mass]. An unmanned station wouldn’t need anything like the same level of resupply [i.e. payload mass.] An unmanned station would have been built from components delivered by ELVs with payloads that were all payload, not payloads less ~100 tons of vehicle to return to eartrh…” Are we seeing a pattern here?

We don’t yet have the ability to do anything massive in space cost-effectively; that applies just as much to SPsats or SDI battle stations as it does to ISS. If you asked the “robotic side” to soft-land a 50-ton package on Mars instead of a Sojourner or Rover, the results would be waaaay slow and waaaay expensive. IOW, “heavy vs. light” explains a very large part of what you pose as “mission vs. program” or “functional vs. dysfunctional.”