Ok, so I was just looking more into octal and it seems exactly the same as hexadecimal except using groups of 3 bits rather than 4 (which I guess makes sense, since 3 bits can hold half the amount of numbers as 4 bits, just as 8 is half of 16). Then it's not as mystifying as I previously thought... though it still seems a bit messy (8 bits in a byte, yet 8 isn't a multiple of 3...)