> Well, the Unicode people certainly didn't `screw up' the encoding. > They simply had to make a decision which of the overloaded meanings of > the character ` should be retained, and they finally decided to follow > the available *character* standards like ISO-8859 for round-trip > compatibility.
Well, they could have left positions 0x27 and 0x60 untouched, to be used according to prevalent practice, and defined separate code points for the specific characters. That would not have broken anything while still allowing precise specification of characters where necessary. But the really serious point where they screwed up was by specifying quotes by shape, not intended usage. They missed defining code points for languages which use conventions other than those in English (e.g., German left and right quotes, etc.). So now we have the dilemma that you can't simply switch a font using one style of quote (e.g., Times) for one that uses another style (like Verdana or Tahoma, which use Courier-style quotes) without manually fixing all the quotes in the text. (Which is impossible on the Web, where the author doesn't know what fonts the reader uses...) We've been over this before...
