But it’s midnight and I’m not tired - even though I was tired all day - now I’m not tired - got a goddamned second wind because it’s midnight. I dream of going to bed early, waking early, having a great productive day tomorrow. Who who rises early has a weak work ethic? Emerson or Franklin or whoever it was was right: early to rise is the way to go. So much of who I am, what I’ve done, what I can’t do, is due, I believe, to the fact I don’t sleep well, early or often.Īnd one thing I despise about the work-ethic moralism of this site - maybe it’s just the conservative side of it - is in not understanding how everything changes when you don’t sleep well. Since there’s a fair bit of useful stuff that can be done without actuators, that’s what they use as their up-front selling point. So the AI research community is going to focus on what they can do in the near term, with some vague handwaving about hiring some MechE types when they’re rolling in money after the IPO. And then it breaks, and no you can’t just restore from backup. By the time you’ve got something useful, you’ve probably spent tens of thousands of dollars. Then it requires parts, many of which will be special-order or custom-made and not in a cheap 3-D printer.

Because a brick of high-grade computronium costs a few thousand dollars off the shelf, can be used by an average coder or comp sci major with no special training, rarely breaks, and if the code you run on it breaks then it’s a few keystrokes to restore from a known-good backup.Īnything with moving parts, requires a completely different skillset that few AI researchers have and the people who do have it won’t work cheap (nor on the speculative hope that they’ll share in the Nobel or IPO for the first AGI, because everybody knows the people with the comp sci degrees will take all the credit).

Dont starve wiki brainy matter