https://github.com/candlerb/string19/blob/master/soapbox.rb

It's buggy as hell. I found loads of bugs just in the process of
  documenting this. To me this two could imply two things:

  - even Ruby's creators, who are extremely bright people, don't understand
    their own rules sufficiently to implement them properly. In that case,
    what chance do the rest of us have?

  - very few people are actually using this functionality, in which case,
    what's it doing as a core part of the language?

* Of course, it's very hard to categorise something as a "bug" if you don't
  know what the intended behaviour is.  Almost all the behaviour given in
  string19.rb is undocumented.  By that I mean: when I look at the
  documentation for String#+, I expect at minimum to be told what it
  requires for valid input (i.e.  under what circumstances it will raise an
  exception), and what the properties of the result are.

  If Ruby has any ideas of becoming a standardised language, the ISO and
  ANSI committees will laugh down the corridor until all this is formally
  specified (and what I have written here doesn't come close)

* Even when I explicitly tag an object as "BINARY", Ruby tells me it's
  "ASCII-8BIT".  This may seem like a minor issue, but it annoys me
  intensely to be contradicted by the language like this, when it is
  so blatently wrong. All text is data; the converse is *not* true.

* It solves a non-problem: how to write a program which can juggle multiple
  string segments all in different encodings simultaneously.  How many
  programs do you write like that? And if you do, can't you just have
  a wrapper object which holds the string and its encoding?