White Elephant

There's a dispiriting article in the Washington Post this week entitled "Space Shuttles Bound to Technologies of the Past". It's not dispiriting for what it says - there are good points, and quotes from knowledgeable people - but rather for a certain kind of attitude towards technology that shows the science reporter doesn't really get it.

The article is correct in diagnosing the problem - instead of being a reliable, if unglamorous, space truck, the shuttle as it now exists is a fabulously expensive vintage roadster. It isn't just that there are no spare parts being made - we don't even have the machines to make the spare parts. The production lines are gone, and in many cases the expertise to build the production lines is gone, frittered away through decades of attrition. It would cost more to build a new shuttle now than it did to develop the original program.

What's troubling is the slant of the article. Early on, we hear from one critic of the program, a University of Maryland engineer who was asked by NASA to evaluate the shuttle's robotic arm:

Pecht found that the arm was still in good working order. But while the engineer answered NASA's questions, the space agency never answered his. "Why are we using this old technology?" he asked repeatedly. "Why don't we change the ways we buy and design so we can always be updating, so we can always be putting in the latest technology? I could never get the clearest answer on that."

From the rest of the article, it's clear that the author shares Pecht's attitude. But the answer to Pecht's question is ironic: the Shuttle program has to use old technology because it relied too much on new technology at its inception.

What Pecht should be asking in his evaluation of the robotic arm is the same question that we should ask about the Space Shuttle program: "Is the technology appropriate for the job at hand"? From an engineering point of view, it's the only question that matters. No one complains that their bathroom fixtures use 1850's technology, because nineteenth-century plumbing turns out to be a great solution to the problem of waste disposal. And maybe the robotic arm is actually a good tool for the job, with the added bonus of twenty years of testing and debugging to get the kinks out.

We know that no one asked that question of the Shuttle program because it's clear that the "job at hand" for the Shuttle was never defined. The program was sold to Congress as a reliable and cheap way to deliver people and cargo to low Earth orbit (never mind why that was necessary). But it was an impossible goal from the outset. Rather than abandon the project, NASA administrators just swapped the dog for the tail, and made the Shuttle's mission contingent on what the technology could realistically achieve. There was no way to point out that the design goals were not being met, because the design goals were being redefined on the fly, to fit the limitations of the technology.

The Shuttle program became the equivalent of the client who says "I want a website that uses Flash and Java, with a talking avatar", without defining what need the website is supposed to serve. Think Boo.com, with rockets attached.

There was nothing but new technology in the Shuttle. In addition to being the first reusable spacecraft, it was the fastest airplane ever built, and one of the first 'lifting body' airframe designs. It had not one, not two, but three of the world's most powerful rocket engines, all of them designed for multiple firings (another first). To lift the whole contraption into space, the designers needed to tack on a pair of the largest solid rocket motors ever built, and make them reusable, too. And for a crowning touch, the entire orbiter was encased in the now-infamous heat protection system, with its tens of thousands of frangible tiles. There was nothing like it in the world.

And there still isn't. The hodgepodge of new technologies made the shuttle an unmaintainable mess. The design was so complex that it could not be modified in the slightest without grave risks to safety. But the Washington Post reporter, proceeding from the assumption that new technology is always better technology, misses this point. She misses it most spectacularly in choosing to quote an astronomer critical of the Shuttle's onboard computer system:

"They have these ancient computers that are really pathetic," said Jonathan McDowell, an astronomer at the Smithsonian Astrophysical Observatory in Cambridge, Mass., and a space program analyst. "They are many years out of date." Indeed, to run high-speed science experiments, McDowell said, astronauts have to carry and plug in laptops. "It's a strange mix of very robust but very old computers that will absolutely work, and a bunch of notebooks that are running the latest version of Windows," he said.

And so the best thing about the Shuttle - its flight control software, probably the only bug-free complex software program ever written, comes off as a liability. To the reporter's credit, she cites a second source who explains why the computers have not been upgraded:

The computers haven't changed a lot since the advent of the vehicle," agreed Jeff Carr, a spokesman at United Space Alliance, a Houston company that runs the shuttle fleet. "It's one of those things that are very adequate for the job and have always been very adequate. They don't need to be faster. . . . There has never been any impetus or need to change them."

From a safety perspective, having astronauts with laptops is ideal, because you never have to risk introducing bugs into the core computer system. Just imagine the nightmare of having to upgrade a Shuttle's entire computer system every two years, or discovering subtle hardware bugs in the new processor during a space mission. The hardware is ancient, and God bless it.

But the tone of the article suggests that these kinds of archaisms are both an embarrassment and a liability - that somehow the Shuttle is an old beat-up jalopy, and that we have not kept it up as well as we should. When the reality is that the original design, by adopting too much at once, effectively froze the project in place.

The Shuttle design is too brittle and too complex to allow for the kind of incremental improvements that are the cornerstone of safety. Boeing can continuously improve and rejuvenate the 747, originally built in 1969, because the underlying design is simple enough to understand and tweak. By contrast, the Shuttle is almost too complex to touch. Hell, it's almost too complex to fly.

To really see the difference between high and low technology, compare the Columbia disaster with the nearly fatal re-entry of Soyuz-5, a Soviet space capsule, in 1969. During that flight, the equipment module failed to separate from the command module on re-entry, and cosmonaut Boris Volynov found himself in a falling spacecraft pointing exactly the wrong way:

As the spaceship fell back into the atmosphere, [Volynov] heard grinding as the deceleration stresses built up. The ship was slowly tumbling end over end, exposing all of its surface to the growing fireball. Then it stabilized with its nose forward, which was exactly the wrong orientation possible because that part of the capsule's skin was the thinnest. In the top area, there was only an inch of insulation, compared to the 6 inches along the bottom, and during a normal reentry three inches of that was expected to burn away

[At this point the better half remarks "I hope Russian cosmonaut suits are brown!" They are!]

Volynov survived because the equipment module finally burned off during reentry, and as soon as it got free, his command module flipped around to face the right way, heat shield forward. It flipped around because of simple, passive design - it was shaped to point in the right direction. Put a tennis ball in a sock and throw it, and you have the technology that saved Volynov.

On the shuttle, a single pencil-thin puncture on the leading edge is enough to doom the spacecraft, and there's no way to fix it.

(There's a full account of Volynov's flight in this article, which wins extra bonus points for non-ironic use of the phrase "fiery death".)