In the midst of all of the celebration of the successful repair mission, I’m going to dash a little cold water here. While I criticized the O’Keefe decision to not do the Hubble repair, it wasn’t a criticism of the decision itself, but rather of the rationale for it. I’ve never supported the mission, because I think that there were much better uses for the money, even for astronomy. But cancelling it because of risk to the crew was a dumb reason to do so (particularly because it made it easy for Griffin to reverse it when he came in). As always in these cases, it’s the money, stupid. And while I’m sure that we’ll get spectacular results from the fix, I remain convinced that there were better uses for the money.

Anyway, Rocketman agrees. Though I would note that the cost of a Shuttle flight isn’t a billion dollars for the purpose of determining the cost of this mission. What’s important for this analysis is the marginal cost of the flight, which, ignoring the cost of the telescope upgrade hardware, was less than a couple hundred million, including crew training. But it was still a lousy deal.

Jon Goff has been doing yeoman’s work in digging into the ESAS appendices. I’ve done a lot of this kind of architecture trade, and it is indeed extremely sensitive to assumptions. And from just this excerpt, while I haven’t read it myself (I can’t find the time from my day job, and I’m glad that Jon did), it sure looks like the game was rigged (which isn’t at all shocking in the context of all of the problems that have arisen). I was simultaneously saddened and amused by this:

I know that Mike Griffin was claiming that part of the reason for doing Ares-I was to teach NASA how to design launch vehicles again, and I guess we have documented proof of the need.

Seriously though, this is a common rookie mistake. You don’t go basing decisions worth tens of billions of dollars on an unvalidated design tool. Now, this isn’t saying that INTROS is a useless tool, just that it obviously doesn’t capture all the state of the art in stage design, and until it does, its results ought to be taken with the appropriate sized (apparently multi-ton) grain of salt.

Well, apparently NASA does exactly that. Or at least it pretends to base them on it. I don’t think that they’re going to pull the wool over Norm’s eyes, though.

There’s an old saying attributed to Voltaire that “the perfect is the enemy of the good.” There are multiple interpretations of this aphorism, but one of them is that one must often accept the achievable, even if flawed, in preference to the perfect but unachievable solution.

Many defenders of Constellation use this argument, saying that if it’s not the best architecture, it’s good enough, and the only one politically feasible at this time, and that any other solution will only cause further delay. Implicit in this argument, of course, are the assumptions that further delay is unacceptable and that it is indeed good enough. Clark Lindsey points out an amusing analogy that Roger Pielke came up with in response to a similar argument by Paul Krugman that current plans for cap and trade are good enough — “Get on the bus.”

I don’t think that Ares is quite as bad as the idea of jumping the Grand Canyon with a Greyhound, but I do think it a vast waste of money, if our goal is space accomplishment rather than keeping parking lots full in Huntsville. From this space enthusiast’s perspective, it is not only imperfect, but it’s not even good enough. Even in the unlikely event that it is successful by NASA’s own cost and schedule criteria, it will be a disaster from the standpoint of making us space faring, flying far too little for far too much money per flight. The goals of the Aldridge Commission were good ones, and as currently planned, it contributes little to nothing to achieving them. As I wrote Sunday, if we need some additional delay to get it right, then we should do that, if it’s at all possible. And if it’s not possible, then we should just give up on having a useful and affordable space program, because this plan is so far from one that I certainly don’t want any more of my tax dollars spent on it.

[Update a few minutes later]

The goals of the Augustine review have been posted on the federal register:

The Committee should aim to identify and characterize a range of options that spans the reasonable possibilities for continuation of U.S. human space flight activities beyond retirement of the Space Shuttle. The identification and characterization of these options should address the following objectives: (a) Expediting a new U.S. capability to support utilization of the International Space Station (ISS); (b) supporting missions to the Moon and other destinations beyond low Earth orbit (LEO); (c) stimulating commercial space flight capability; and (d) fitting within the current budget profile for NASA exploration activities.

I can think of many architectures that would do a good job of those four goals, but Constellation isn’t one of them. It really only supports goal (b), and even that not very well. As Clark says, it will be both sad and amusing to watch NASA’s attempts to explain how it does.

You’ll be as shocked as I was to learn that NASA’s evaluation of DIRECT may not have been completely impartial:

“NASA’s October 2007 analysis of DIRECT, on the surface, appears to be a carefully executed analysis of the DIRECT architecture and its central launch vehicle, Jupiter,” notes the rebuttal document.

“However, a closer examination of the document reveals significant flaws in the evaluation of DIRECT that sets up a scenario where DIRECT would inevitably look inferior when compared to Ares.

“The errors are so numerous that the only conclusion possible is that this document was not a true analysis, but rather an attempt to discredit the DIRECT architecture.”

I’ve never been a proponent of that architecture, and haven’t even looked at it in any detail or given it much thought, other than that it’s almost certainly better than NASA’s current plans. But that’s a pretty low bar. I think that if we’re going to be investing billions in new launch systems, we should get some that actually, you know, reduce launch costs, and particularly marginal costs. But I know, that’s just crazy talk.

Taylor Dinerman reviews them in the context of the current NASA:

Law number XXIV would seem a particularly good one: “The only thing more costly that stretching the schedule of an established project is accelerating it, which is itself the most costly action known to man.” The urge of many politicians to spend more on NASA’s Constellation program so as to shrink the “Gap” is well known. The impact of such a decision on the rest of NASA, or on the future space exploration program, is obviously something that Mr. Augustine is going to have to look at very closely.

Another insight: “…we are attempting to develop major new systems with ten year technology, eight year programs, a five year plan, three year people, and one year dollars.” Constellation is trying to escape this dilemma by using existing technology: this may work, but it is dangerous since it assumes that the systems involved are already well understood, something that was so heavily critiqued in the 2003 Columbia Accident Investigation Board report.

As he notes, they still hold up a quarter of a century later. They explain well why a government space program is always doomed to disappoint its boosters.