This came to mind for me as well. This, from Burdensome Details, popped out at me: “Moreover, they would need to add absurdities—where the absurdity is the log probability, so you can add it—rather than averaging them.” All this does for me is pattern-match to a Wikipedia article I once read about the concept of entropy in information theory; I don’t really know what it means in any precise sense or why it might be true. And the essay even seems to stand on its own without that part. I’ve come to ignore my fear of not understanding things unless I don’t understand pretty much everything I’m reading, but I think a lot of people would get scared that they didn’t know enough to read the book and just stop reading.
Come to think of it, we could collect proposed rewrites / deletions to some wiki page: this seems suitable for a communal effort. The “deletions” wouldn’t actually need to be literal deletions, they could just be moved into a footnote. E.g. in the Burdensome Details article, a footnote saying something like “technically, you can measure probabilities by logarithms and...”
I like the idea of turning a lot of these jargony asides, especially early in the book, into footnotes. We’ll be needing to make heavier use of footnotes anyway in order to explicitly direct people to other parts of the series in places where there will no longer be a clickable link. (Though we won’t do this for most clickable links, just for the especially interesting / important ones.)
You’re welcome to use a wiki page to list suggested changes, or a Google Doc; or just send a bunch of e-mails to errata@intelligence.org with ideas.
This came to mind for me as well. This, from Burdensome Details, popped out at me: “Moreover, they would need to add absurdities—where the absurdity is the log probability, so you can add it—rather than averaging them.” All this does for me is pattern-match to a Wikipedia article I once read about the concept of entropy in information theory; I don’t really know what it means in any precise sense or why it might be true. And the essay even seems to stand on its own without that part. I’ve come to ignore my fear of not understanding things unless I don’t understand pretty much everything I’m reading, but I think a lot of people would get scared that they didn’t know enough to read the book and just stop reading.
Come to think of it, we could collect proposed rewrites / deletions to some wiki page: this seems suitable for a communal effort. The “deletions” wouldn’t actually need to be literal deletions, they could just be moved into a footnote. E.g. in the Burdensome Details article, a footnote saying something like “technically, you can measure probabilities by logarithms and...”
I like the idea of turning a lot of these jargony asides, especially early in the book, into footnotes. We’ll be needing to make heavier use of footnotes anyway in order to explicitly direct people to other parts of the series in places where there will no longer be a clickable link. (Though we won’t do this for most clickable links, just for the especially interesting / important ones.)
You’re welcome to use a wiki page to list suggested changes, or a Google Doc; or just send a bunch of e-mails to errata@intelligence.org with ideas.