ICU4X-WG discussed approving the MessageFormat 2.0 spec today (with @eemeli). We had a bunch of questions answered which helped our planning for the implementation

We also had some feedback / requests, specifically around the ability for data-dependent options for functions (u:locale, notation, etc) to depend on external input (or even be expected to change between locales). This does not work well with ICU4X's data loading model.

I plan to write a longer doc explaining the situation, but @eemeli asked me to file an issue for this so it can be tracked by MF2.0. Once I have a doc I shall post it, as a personal contribution, with pending approval from ICU4X-TC as its official position.

Raw discussion here: unicode-org/icu4x#3028 (comment)