The space shuttle was often used as an example of why you shouldn't even attempt to make something reusable. But one failed experiment does not invalidate the greater goal. If that was the case, we'd never have had the light bulb.

So we originally expected to make about 35 gigawatt hours at the cell level [solar] and about 50 gigawatt hours at the module or pack level. Now we are expecting to do about 150 gigawatt hours in the same volumetric space as the original design.

If we drive down the cost of transportation in space, we can do great things.

Obviously Tesla is about helping solve the consumption of energy in a sustainable manner, but you need the production of energy in a sustainable manner.

There are really two things that have to occur in order for a new technology to be affordable to the mass market. One is you need economies of scale. The other is you need to iterate on the design. You need to go through a few versions.

When Henry Ford made cheap, reliable cars people said, 'Nah, what's wrong with a horse?' That was a huge bet he made, and it worked.