Real scientists make their own data (2013)

>Use old data to practice techniques, but if you want to get serious about being a scientist, make your own data.

I generally agree, though Einstein is the obvious counterpoint.