`decodeUtf8` / `decodeUtf8'` situation

There are a few issues with the current situation:

+ Decoding/encoding from UTF-8 is a really common use case, but requires a separate import (`Data.Text.Encoding`)
+ The best name (`decodeUtf8`) is taken by the partial version of the function
+ `decodeUtf8'` doesn't mean anything, you just have to know the `'` means it's the total version

Is there anything we could do to improve this situation? It doesn't bother me, but it's tough on people that are new to haskell. I know we're constrained by backwards compatibility here because so many packages use `text`, but I thought it would be worth thinking about.