The vast majority of entities use the simpler ASCII equivalent to the Unicode character. I would prefer if we always used these simpler chars as follows:

Left Quote - Apostrophe
Right Quote - Apostrophe
Left Double Quote - Double Quote
Right Double Quote - Double Quote
Hyphen - Hyphen Minus
En Dash - Hyphen Minus

because it:

Makes the data more consistent
Makes editing easier
Makes it easier for applications using the data

and existing data should automatically be updated.

I understand the view that the Unicode chars are in some sense more correct, but it causes alot of problems. If the consensus is that we should stick to using unicode chars then we should at least update the data the other way where possible so editors dont waste their time converting from ascii to unicode, of course when going from Apostrophe to Left quote, Right Quote is harder coz have to make a choice but should still be easy enough in most cases by checking for text, whitespace to each side of the apostrophe.