Significance of Compression Rate Method — LessWrong