Yes. From the Model T to the Pentium, we see the winning product is the one that has the best manufacturing process behind it. Often, the product itself isn't anything special compared to the competition.

Nope, that wouldn't work, because the division of labor wouldn't correspond to the basis for the "independent" cost estimate, and the cost "experts" would therefore refuse to endorse it. Then there's the potential cost of lost payloads. In principle, the contractor could insure them, except NASA is forbidden by law to buy insurance.

Except that implicit in that is the idea that every bug is a disaster. SpaceX's approach is to have robust engineering rather than perfection. The idea is that small problems should not cascade into mission failures. That's how "real world" engineering works: for example, we don't use chains to hold up suspension bridges any more, because a single crack can cause a collapse. We use multi-strand cables, where cracks don't propagate from strand to strand. The fragile perfection of old-school aerospace is expensive and hazardous.

Not really. NASA generally goes with what appears most "credible" to them within the cost cap. The most important factor in credibility is matching their detailed estimates of costs, created using "parametric" methods. These methods take historical costs into account and then allow for inflation. Imagine estimating the cost of a computer by scaling from an IBM 709, assuming every performance enhancement costs money, and multiplying by inflation. Then, you refuse to try anything cheaper, because it's "risky".

The result? The bidder must propose not only a high price, but must justify that price based on costs. You must demonstrate the ability to put together and manage very inefficient processes. It usually doesn't even help to have done similar jobs efficiently: the cost "experts" don't find actual experience in conflict with their databases to be "credible". Their databases are full of previous examples of projects approved and planned with the same methodology, so the reasoning is almost perfectly circular.

Historically, nobody has been able to develop an orbital launch vehicle without government subsidy, so this credibility problem has been an impenetrable barrier to exploiting real high tech methods, where deflation, not inflation is normal. But Musk has deep enough pockets, and a talent for PR that has made it impossible to dismiss the success of Falcon as an aberration.

A lot of stuff on github is experimental, "quick and dirty" code. The amount of effort to, say, put GPL boilerplate in every file isn't large, but it isn't zero, either. So, *ask*. You send mail to me, volunteer to do this small job, I'll probably give you commit access to the repo.

Musk doesn't need to diss Ariane 5: his company is taking business away from them without any negativity. The thing that surprises me here is how positive he is about Ariane 6. Given how rapidly SpaceX is improving their product, I don't see how Arianespace, with its slow expensive processes, could ever get ahead with a new vehicle.

The NAS report addresses this. It's a serious possibility.

The thing to understand here is that *in principle* the net required water input is tiny (it provides the hydrogen in the hydrocarbon output stream, but that's not much compared to the water needed as a solvent), and the net required nitrogen and phosphorus inputs are zero (they aren't in the output stream). One issue is therefore the recycling of the waste stream after hydrocarbon extraction. Another is losses (especially water) from open ponds, if that's the technology of choice. The report tells us that projects to date have not adequately addressed these issues.

The two things I take away from this are:

1. The technology isn't ready yet, but it has future potential.

2. Open freshwater ponds are probably not the winning approach.

This report will undoubtedly help steer future research in this area.

There was certainly a lot of discussion of this among us at the time. I recall we wondered whether NASA would go for free return or be more radical and use more delta-V in cislunar space to get the astronauts home sooner.

But call up NASA? Be serious. Which of the 100,000 phone numbers would you call? The critical people were busy: they weren't going to talk to some random student. This was all elementary orbital mechanics, somewhat difficult to calculate and execute accurately, but not conceptually difficult at all. The flight team certainly knew this stuff. The real question was what the damaged systems could still accomplish, and that required information well beyond what we had access to. So it never occurred to anybody I know to try being a back seat driver.

It's generally not the people who create things, make things, or provide useful services who fear change. They can adapt. But if you've spent your life's effort working your way into control of a particular set of cash registers, change is very threatening indeed.

The problem with this model is that it's very hard to control your usage. There's no practical way to know in advance how much a particular click will cost. Of course, the providers love it for exactly that reason.

Some broken clouds, but they weren't much of a problem.

There is nothing more "industry unfriendly" than a genuine free market. Despite propaganda, no businessman ever wants to compete in freedom. What you want is to control the customer and the competition, not let them control you.

The trouble is that most of that cost isn't lifting it to altitude, but getting it moving at the right velocity for the orbit you want. If you put some sort of recycling device in orbit, almost all of the junk that it encounters will be moving at high velocity relative to your device's orbital velocity. Speed will tend to be similar, but direction will be all over the place. Changing the velocity of either the device or the junk is difficult.

Lead is a reasonably valuable metal, but stationing yourself in no man's land between two armies and recycling the bullets that come at you seems a difficult way to obtain it.

Peer reviewers can't check everything, especially when the conclusion results from elaborate analysis of data from complex apparatus. Sometimes you detect bonehead mistakes, but usually your focus is more on clarity than correctness: do the authors explain their methods and reasoning in enough detail that someone else can repeat the research?

Absolutely yes! It is the physician's responsibility to avoid basing treatments on results that haven't been independently confirmed. It is the researcher's responsibility to publish: how else will you get that independent confirmation? Other researchers need to know what they should attempt to confirm or falsify.

We're talking about the science of the journals here. This is raw "source code", checked to some degree, but not debugged. The debugging takes place in the community: if you don't publish, your results will never get properly debugged.

In my experience as a scientist, what has increased is the pressure to publish quickly. So, people publish results that haven't been checked as much as they perhaps should be. But this is not fraud, and perhaps it's even healthy. Better to get crazy results out there than bury them in notebooks: sometimes they turn out to be major discoveries.