public personal library

Still, I can’t help but think that people’s priorities are enormously out of touch with what will actually matter most to their future selves. It feels as if people seem to have this mental model where medical progress simply happens. Like there’s some natural law of the universe that says “treatments improve by X% per year” and we’re all just passengers with a dumb grin on this predetermined trajectory. They see headlines about better FDA guidelines or CRISPR or immunotherapy or AI-accelerated protein folding and think, “Great, the authorities got it covered. By the time I need it, they’ll have figured it out.”. But that’s not how any of this works! Nobody has it covered! Medical progress happens because specific people chose to work on specific problems instead of doing something else with their finite time on Earth.

Their eyes are slowly decaying, and that if they manage to hit fifty, there is a one-in-ten chance that there will be a creaking, incurable black hole in the middle of their sight, expanding day after day. Think! Think! Do something about it!

To be fair, most people go through their first few decades of life not completely cognizant how terrible modern medicine can be.

Startups don’t care about marginal features because their success is entirely dominated by finding a single successful feature that anybody wants to pay for.

Why would analogies give genuinely new ideas? Well, intelligence is messy! If everything is so uniform, then there is no way to make a far-flung analogy – everything is pretty much the same already. Besides, just look at all the broken dreams of logical AI – their corpses tell us that no elegant theory of intelligence exists. Programming a genuine AI is a messy job. Messiness is a hideous strength. This is the 4th lesson.

2th lesson: Representation matters a lot. AM worked so well for mathematics, because AM used Lisp code as data. Lisp is the perfect tool if you want to search over the space of interesting mathematical functions.