“The Principle of Computational Equivalence: when one sees behavior that isn’t obviously simple, it’ll always correspond to a maximally sophisticated computation.

“Nobody in academia seems to like Hofstadter or Minsky.

The neuroscientists deplore the absence of rigorous experiments and the refusal to canvass the relevant experimental literature thoroughly. Where are the data? The philosophers of mind find few formal arguments and a frustratingly cavalier refusal to define their terms at the outset. Where are the proofs? The cognitive psychologists are offended that they don’t see the need to adjudicate between all the competing models and theories that have been painstakingly developed and defended, and instead offer their own impressionistic and oversimplified sketches. Where are the models that make testable predictions? The artificial intelligence crowd wants to see a running demo program. Where is the code? It’s all just speculation! And then there is the style, too clever and playful, apparently written for bright high school students, not professors and graduate students.

“The business model of today’s internet giants might best be called vampiric. Their overriding goal is to know us, to transfer into their databases the informational life-blood of our selves. Their thirst is insatiable. To survive, they must uncover ever more intimate details of our lives and desires. And we are not averse to the seduction. We embrace these companies, welcome them into our homes, because we desire the gifts they bear and the conveniences they provide. We tilt our necks to them freely.

“Imperative programming is more susceptible to premature optimization [than functional programming].

“If you and I want to stir up a resentment tomorrow that may rankle across the decades and endure until death, just let us indulge in a little stinging criticism—no matter how certain we are that it is justified.

“The real reason people drink: it takes effort to have fun if you are sober.

“Ruby extension modules are slightly easier to write and read than Python ones, but slightly less portable. Main reasons: no reference counting, less obtrusive exception propagation using setjmp/longjmp, and a more uniform notion of types and classes. There are also some minor conveniences, such as C methods which take an ordinary argument list instead of a vector.

“Plastic bags that float into storm drains in the west coast make their way to the so-called Great Pacific Garbage Patch—a heap of debris twice the size of Texas floating in a no-man’s land between San Francisco and Hawaii. This patch has been growing, along with other ocean debris worldwide, tenfold every decade since the 1950s.

At this point, cleaning it up isn’t an option. It’s just going to get bigger as our reliance on plastics continues.