Lots of documentation examples still use encoding APIs with encoding names
rather than encoding constants. I think it would be preferable to direct
users toward constants as it can help with auto-completion, static analysis
and such.
The example code in the documentation for the `String#encode` method has
a typo in the `fallback` option. The example code uses `h` as the
fallback option, but it should use `hash` instead to match the variable
name in the example code.
