Some of the improvements in the release are:
- Canonical codes
- The subdivision codes have been changed to all have the bcp47 format.
- The locales in the language-territory population data are in canonical format.
- The timezone ID for GMT has been split from UTC.
- There is a mechanism for identifying hybrid locales, such as Hinglish.
- Emoji 5.0
- Short names and keywords have been updated for English. (Data for other languages to be gathered in the next cycle).
- Collation (sorting) adds the new 5.0 Emoji characters and sequences, and some fixes for Emoji 4.0 characters and sequences.
- For Emoji usage, subdivision names for Scotland, Wales, and England have been added for 65 languages.