Back in 1974, Richard Feynman, one of the most famous physicists of the 20th century, gave a commencement address at the California Institute of Technology. He called it "Cargo Cult Science" (PDF). It contains what I think is the best succinct summary of the scientific process:
The first principle is that you must not fool yourself—and you are the easiest person to fool.
Then, he said, "it's easy not to fool other scientists. You just have to be honest in a conventional way after that." The entire structure of science—in universities and journals, online and at symposia, and in the principles of being able to repeat experiments, falsify theories, and predict results—is an attempt to keep from fooling ourselves about reality.
That shows how hard it is to avoid being foolish: scientists and the rest of us still fool ourselves all the time. But the process also does a good job correcting mistakes, so our foolishness doesn't last. As a human system, science isn't perfect, but it does well at refining our knowledge, letting us weed out imperfections in what we understand about our world and universe.
A bunch of stuff I've been accumulating over the past few months: