From “How to Do Great Work” by Paul Graham:

Schools also give you a misleading impression of what work is like. In school they tell you what the problems are, and they’re almost always soluble using no more than you’ve been taught so far. In real life you have to figure out what the problems are, and you often don’t know if they’re soluble at all.

Schools sometimes also give students the misleading impression that learning is not fun for its own sake and that writing should be boring.

// 01.04.24

From “The Bitter Lesson” by Rich Sutton:

In speech recognition, there was an early competition, sponsored by DARPA, in the 1970s. Entrants included a host of special methods that took advantage of human knowledge—knowledge of words, of phonemes, of the human vocal tract, etc. On the other side were newer methods that were more statistical in nature and did much more computation, based on hidden Markov models (HMMs). Again, the statistical methods won out over the human-knowledge-based methods. This led to a major change in all of natural language processing, gradually over decades, where statistics and computation came to dominate the field. The recent rise of deep learning in speech recognition is the most recent step in this consistent direction. Deep learning methods rely even less on human knowledge, and use even more computation, together with learning on huge training sets, to produce dramatically better speech recognition systems. As in the games, researchers always tried to make systems that worked the way the researchers thought their own minds worked—they tried to put that knowledge in their systems—but it proved ultimately counterproductive, and a colossal waste of researcher’s time, when, through Moore’s law, massive computation became available and a means was found to put it to good use.

[…]

We want AI agents that can discover like we can, not which contain what we have discovered. Building in our discoveries only makes it harder to see how the discovering process can be done.

// 01.03.24

In addition to being New Year’s, this site turned 15 years old (!) today. It contains hundreds of posts, over a half million words, and oodles of my time.

Thanks for reading!

// 01.01.24

My crystal ball is as cloudy as ever.

Earlier this month I wrote about Radiology Partners loaning a group money to help shore up radiologist compensation. That happened, but I was also wrong in my estimation of the likelihood of repayment: it turns out RP may be getting some of that money back after all.

The ultimate outcome is still up in the air, but I’ve addended my previous post with an update.

// 12.29.23