The extreme difficulty of scaling production of new technology is not well understood.

The answer was we thought it could be done.

Falcon One is going to be the lowest cost per flight to orbit of any production rocket.

I think the reason it’s cheaper is, first of all, we are a private entity and we have a very lean system here. What we have been able to do here at SpaceX is to cherry-pick, you know, the top one or two percent and give them, you know, capital to execute well and a clear mission, which is low cost, reliable access to space, and no other constraints.

Well, I have tried to learn as much as possible from prior attempts.

I think we’ve got the risks pretty well characterized. I think we are at least avoiding the mistakes that have been made in the past.

I want to be able to make sure that we have enough capital to survive at least three consecutive failures.

I like to be involved in things that change the world. The Internet did, and space will probably be more responsible for changing the world than anything else. If humanity can expand beyond the Earth, obviously that's where the future is.

Starting and growing a business is as much about the innovation, drive and determination of the people who do it as it is about the product they sell.

My approach is simply to seek out very talented people, ensure that the environment at SpaceX is as motivating & enjoyable as possible and establish clear & measurable objectives.

Rocket engineering is not like ditch digging. With ditch digging you can get 100 people and dig a ditch, and you will dig it a hundred times as faster if you get 100 people versus one. With rockets, you have to solve the problem of a particular level of difficulty; one person who can solve the problem is worth an infinite number of people who can’t.

I think it is a mistake to hire huge numbers of people to get a complicated job done. Numbers will never compensate for talent in getting the right answer (two people who don't know something are no better than one), will tend to slow down progress, and will make the task incredibly expensive.

  If the design takes a long time to build, it's the wrong design. This is the fundamental thing. 

Over and over, the tendency is to complicate things. And I have another thing which is, the best part is no part. The best process is no process. It weighs nothing, costs nothing, can't go wrong. So, as obvious as that sounds, the best part is no part.

The thing I'm most impressed with, when I have the design meetings at SpaceX, is:

"What did you undesign?"  

Undesigning is the best thing. Just delete it. That's the best thing.

Possibly, the most common error of a smart engineer is to optimize something that should not exist.

Everything works in PowerPoint; but if you have the physical item or some demonstration software, that's much more convincing to people than a PowerPoint presentation or a business plan.