It's a bit unfair to say SpaceX spent only 800M to build the Dragon. A lot of the technology it uses was developed by and for NASA with NASA's money.
Not to say that NASA has been doing design-by-politics for a very long time. When practicality gives way to what politicians want in order to increase their own constituency, the technological endeavor is pretty much doomed.
A lot of the technology it uses was developed by and for NASA with NASA's money.
While Dragon may be standing on the shoulders of giants, isn't that equally true of the Orion program? I mean, sure, SpaceX has benefited from past lessons, but so does the current NASA development program. Those giant numbers quoted for Orion/Constellation surely don't include moneys spent on Apollo, Space Shuttle, etc.
If you read the following paragraph (I should have edited it, but when I realized, it was too late) you would see that NASA is a politician-driven endeavor. It also has a mandate to generate new technology, while SpaceX only needs to apply it.
Obviously, Orion/Constellation is not money well spent. At least it teaches that basing a project in common parts with an existing system and then specifying them in a way they are no longer common is appallingly stupid. Ares V may look like a shuttle tank with shuttle SRBs attached, but it's not.
0.8 billion for an unmanned rocket program is not exactly cheap. If anything it suggests that building rockets that work is expensive and private companies are going to spend similar amounts of money for similar results.
PS: That's not to say that improvements can't be made. A scaled up version of Scaled Composites White Knight that can lift a rocket capable of boosting 10 tons into to LEO would change the economics of space travel. http://en.wikipedia.org/wiki/Scaled_Composites_White_Knight_... However, it's an old idea and implementing it would be risky and expensive.
It's not entirely an unmanned rocket program. True, SpaceX hasn't launched anyone. However the most recent launch was a test of their Dragon crew module. Sort of a spin around the block to make sure the wheels don't fall off if you will.
There's a big potential bang-for-buck with cheaper launch technologies for unmanned and bulk cargoes. Engineer a payload like a Bigelow style inflatable space station, fire it into orbit out of a huge gas gun or ram accelerator, and you've a game changer for orbital R&D. Do the same for thin-film solar panels and constructor robots, and you've solved the world's energy problem.
Right but NASA also gets to make use of their previous work when
developing Constellation or whatever the new thing is called. We
wouldn't have gotten this far without NASA. At the same time however,
NASA's space craft development process needs some sharpening. Whether
that's due to bureaucracy or congressional pork, something needs to change.
NASA proved it was possible to spend money and gain practically nothing of scientific value (ISS & the entire Shuttle program). Of course, the main reasons it got those money was to spend it in favorable political endeavors.
There is a lot of value in learning how not to do something. The shuttle is a valuable (albeit costly) lesson on why you should keep committees out of a design process, on focusing on doing one thing well instead of many things badly, on why graceful degradation is important and so on. Likewise, I don't think the ISS is a complete failure - it teaches us every day how you can (and can't) build huge structures in LEO.
And it's not that the ISS is horridly expensive - it's shipping its parts to its final location that's expensive. I wonder how much would it cost if we had the Falcon-9 Heavy, a couple Saturn V's or a shuttle-derived expendable cargo pod available in 1980.
Could the money be better spent? Could politics be less of a problem? Probably. It's a shame we still rely on chemical rockets and solar panels. It's a shame we never deployed a deep space probe with a large solar sail and that there is no Aldrin Cycler station in place yet.
That implies only the first attempts are bad and NASA would be able to do stuff at the same level of efficiency SpaceX does currently given both have access to the same previous attempts.
The fact NASA is not as efficient may indicate either NASA produces more research data or it is indeed less efficient. I would bet the answer lies somewhere in the middle.
I think we could accomplish many more firsts with less money if we used the X-Prize model. (If we did this, we wouldn't have to abandon the NASA model either!)
I'm not sure, but I suspect that some things are expensive enough to be beyond private investment. Mega-projects like the Apollo missions, the Hubble telescope, the LHC and such seem likely to be too risky to embark on with a prize as the incentive.
Edit: I know the LHC has nothing to do with NASA; it is just in a similar vein.
I think a $20 billion prize for a manned Mars landing could pay off handsomely for both the winning contestant and the government, particularly if the government received a license to the technology used to accomplish the prize.
$2 billion for a fully reusable TSTO lofting 5000 pounds to LEO would also be money well spent.
That's most definitely not what I intended. NASA have done things that make me feel proud to be a human being. I was more aiming to highlight the value in having government organizations spearhead research into areas that may not be immediately, clearly profitable. Private industry wouldn't have had much to gain from Apollo...
Also, compare the PR impact of "The pilots safely ejected from the enormous explosion" vs. "the pilots died screaming in the enormous explosion". One instance of the latter cost us about five years. Even just on a PR basis the money may be well spent.
And no, I don't think just accounting for the PR is a good idea, the human factor is very important. I'm just saying that the PR factor alone is adequate to justify it before you consider the human factors.
The greatest threat to private spaceflight is the government stomping on it, and the most likely way that will happen (/most likely excuse the government will use) is fatalities in the program, because our culture not only doesn't want to take risks like that, it doesn't want anyone to be able to choose to take risks like that, for better or worse. (I'm serious about the ambivalence in "for better or worse"; I can muster good arguments either way. It is not sarcasm or snark.)
Not to say that NASA has been doing design-by-politics for a very long time. When practicality gives way to what politicians want in order to increase their own constituency, the technological endeavor is pretty much doomed.
Of course, Feynman said it much better than me.