The main priority for Flint 2 is to meet the current language specification, as the Swift implementation of Flint currently (mostly) does. Features that were not prioritised as part of the Quartz research project should now be focused on to make sure Flint 2 is a good replacement, and covers at least the same capabilities as the original.