for another example of Rationalist crank shit trying to pass itself off as computer science, check out the article for Minimum Description Length:
Selecting the minimum length description of the available data as the best model observes the principle identified as Occam's razor. Prior to the advent of computer programming, generating such descriptions was the intellectual labor of scientific theorists. It was far less formal than it has become in the computer age. If two scientists had a theoretic disagreement, they rarely could formally apply Occam's razor to choose between their theories. They would have different data sets and possibly different descriptive languages. Nevertheless, science advanced as Occam's razor was an informal guide in deciding which model was best.
With the advent of formal languages and computer programming Occam's razor was mathematically defined. Models of a given set of observations, encoded as bits of data, could be created in the form of computer programs that output that data. Occam's razor could then formally select the shortest program, measured in bits of this algorithmic information, as the best model.
note that this is uncited nonsense, but it sounds exactly like a LessWrong post. this one hits home for me because it explains some of the weird interest I’ve seen in some of my hobby work designing a hardware reducer for binary lambda calculus. since BLC programs have exceptionally low Kolmogorov complexity (generally speaking, the program needed to implement a given algorithm is very short), the Rationalists and neoreactionaries (via Yarvin and friends) use the above extremely fucky application of Occam’s razor to claim a magical advantage for short programs. while I really like playing with BLC and I feel it has interesting potential for exploring alternative caching and optimization strategies, its actual performance is kind of hilarious:
- BLC programs take up a shitload of memory (around 500mb to a few gigs for a basic Lisp REPL) because their simple program strings expand to extremely complex garbage collected in-memory representations (which this fucky version of Occam’s razor elides, of course)
- mathematical performance is awful because Church numerals are a unary system and operations are very expensive. this can be somewhat fixed by implementing binary operators (lambda calculus doesn’t really have a native concept of numbers at all, so you effectively get to choose your numerical base), but more efficient numbers are one reason why practical lambda calculus derivatives usually choose more complex encodings
but hey, speaking of Algebraic Information Theory, look whose weirdo fingerprints are on that article! that’s right, it’s the Burgin fucker from the super-recursive algorithms article and formerly of the Solomonoff induction article! fuck me I hate what the Rationalists are doing to my current hobby obsession.