Author doesn't seem to care that there is a difference between Unicode the standard and utf-8 the encoding. While the changes on the fringes to the system are d

Author doesn't seem to care that there is a difference between Unicode the standard and utf-8 the encoding. While the changes on the fringes to the system are debatable, they are also in a way sensible. Internal to your application everything should be encoding independent (unicode objects in Py2, strings in Py3) while when talking to stuff outside your program (be it network, local file content or filesystem names) it has to be encoded somehow. The distinction between encoding independent storage and raw byte-streams forces you to do just that! Stop worrying and go with the flow. Just do it as it is supposed to be done and you'll be happy.