Three or four tangents
Do I actually like learning?
I definitely like having learned, because understanding something is wonderful while not understanding hurts and is the worst. But some of my friends seem to actually like learning, to the extent that they would rather do math than play with friends.
I've had feelings of guilt and questions of "am I a bad person" because I don't skip with joy when I learn something. But I think I genuinely like physics, even if I don't crave it. *On the other hand, I can crave programming like drugs.
Maybe the truth is this: Given that I'm learning anything, I would far prefer to be learning physics, and that's enough to be called love. When I'm studying ODEs my thoughts flicker to physics; when I'm reading about biology, I wish I were doing physics. When, last summer, the experience of reading neurotech papers was terrible and unintelligible, I ran away to do physics, which was also confusing but less terrible.
Stories
Aidan recently published his website, "Beatin' Paths", about traveling across America and meeting Americans, reading and writing the Great American Novel. "I dropped out of Harvard to pursue my destiny and live in a bus." It was brave, it was heartwarming. Have you ever heard someone say out loud that they're pursuing their destiny?
I love stories like that, and they're essential for society. If we don't hear about heroes and quests in the news, our imagination withers. We start to believe other tales, like career advancement. In the light of a good story, it's easy to see other things as pallid, naked bribes.
Ohm my god
Consider Ohm's law, V = IR. Naively you would think it expresses linearity in V and R, with a slope of I. But current, not resistance, is the more fundamental quantity, while resistance is a catch-all of physical properties that affect the relation between V and I. The truer, more illustrative expression is V = RI, with R in the role of slope. Yet V = IR is the form we use.
The second problem is that V = IR casts V as the dependent variable. The causal relationships are obscured, if not totally antagonized — the voltage difference drives the current, while the current depends on the voltage in a predictable way. To capture that we should write I = GV and talk about G, the conductance, instead of the resistance.
We can visualize current as a river of electrons, and voltage as a set of level curves, representing the "drop-ishness" of the system. But resistance! I can't see anything when I think about it. Resistance is often the absence of something: the thinness of a wire, the sparsity of material, the absence of electrons. Glass is a resistor, but if I add more glass and make a larger, wider resistor, I've actually decreased the resistance. *Conductance is easier to think about: the metalness of the material plus how much the geometry resembles a pancake.
This is what is called "bad design". It's so terrible! Bad design makes people think it's their fault for not understanding, when in fact the interface is hostile.
How is EEG possible?
The more and more I understand electromagnetism, the less and less I understand EEG. Fields are extremely sensitive to geometry; how could it be that the mess of randomly orriented currents in the brain convey any information at all? For that matter, why doesn't neuron firing emit light?