All functional knowledge is compression and contextual decompression. Sometimes this compression can be lossless or nearly so. Compression doesn't automatically mean loss of information.