Thanks!

We can maybe start with just Bytes, I'm not sure Byte would be used (typing UInt8 is almost as easy) and Chars has no use case right now. I just mentioned those other in a theoretical way.

Byte feels nice for working with "binary buffers" rather than UInt8 which could be seen as more ambiguous. Obviously would go along well with Bytes.
Chars would be better off as CString or so for use cases I'd think of, no?

LibC::UnicodeString then? Though hard to argue whether it should be Slice(Char) or Char*.

The use of Slice(Char) is for IO, when you'd want to read a few chars instead of bytes with some encoding. For example:

io = ...

bytes = UInt8[1024]
io.read(bytes.to_slice)

chars = Char[1024]
io.read(chars.to_slice)

This would be similar to C# (read chars from a text reader, or read bytes from a binary stream). However, we haven't found a real use case for that yet, so for now I wouldn't introduce a Chars alias.

Ah, right, Char = 32bit, my mind was in C-world X-|