This is a special post for quick takes by Maxwell Peterson. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page.
Whenever I have an idea for a program it would be fun to write, I google to see whether such programs already exist. Usually they do, and when they do, I'm disappointed - I feel like it's no longer valuable for me to write the program.
Recently my girlfriend decided we had too many mugs to store in our existing shelving, so she bought boards and other materials and constructed a mug shelf. It was fun and now we have one that is all her own. If someone walked in and learned she built it and told her - "you know other mug shelves exist, right? You can get them online", I'd view them as somewhat deranged. Obviously a big part of the value is that she built it, that building is cool and fun, and that there aren't others like it.
But that's the kind of thing I've been telling myself when I find other programs exist! So I should look at myself as somewhat deranged, and change my attitude.
One thing I often think is "Yes, 5 people have already written this program, but they all missed important point X." Like, we have thousands of programming languages, but I still love a really opinionated new language with an interesting take.
I was reading a comment (linked below) by gwern and it hit me:
Jaynes’s Probability: The Logic of Science is so special because it presents a unified theory of probability. After reading it, I no longer think of “probability” and “statistics” as being different things. As many understand evolution - feeling there is a set of core principles, like selection and evolutionary pressure and mutation, even if the person isn’t familiar with many of the technical findings or machinery they’d need to actually do an analysis good enough to make good predictions from - this is how I feel about probability after reading Jaynes.
The direct practical value of the book is quite low! But it can give you a mind that feels probability is an intuitive field, and nothing like a collection of tricks. I might have gotten a lot of help on this front by reading the sequences, but it’s Jaynes who really brought it together for me. I even skipped a lot of the algebraic math in his book and still got so much out of it.