In the spirit of science, there really is no such thing as a 'failed experiment.' Any test that yields valid data is a valid test.

You have failed your own test,

Our success at Amazon is a function of how many experiments we do per year, per month, per week, per day.

I learned at an early age that I could make the things that I wanted. That's a very powerful thing to realize as a kid. LEGOs were a key part of that.

We didn't set out to be educators or even scientists, and we don't purport that what we do is real science but we're demonstrating a methodology by which one can engage and satisfy your curiosity.

We've had our first test, and we've failed miserably,