Everything is 0s and 1s to a computer. What a pattern of 0s and 1s encodes is decided by people--often arbitrarily. Over the years there have been attempts to standardize encodings but, for legacy reasons, older encodings are still valid.

The 0s and 1s that encode ' in UTF-8 (a standardized encoding) are the same 0s and 1s that encode ’ in CP-1252 (a legacy encoding).

The � symbol is shown when the 0s and 1s don't encode anything of meaning.