I feel bad for the poor unicode implementers these days, what a metric ton of work. How would folks go about partial implementations? Or using fallback fonts for missing points?

Unicode implementation was always a ton of work. Far as I can tell there have been few novel features beyond ZWJ sequences maybe? I don’t know if these are treated separately or as ligatures by engines.

The big thing is that historically those features were ignored due to being low demand, the desirability of emoji made emoji great vessels for getting those features implemented and tested by engines.

I think creating fonts is what’s gotten a lot more intensive, fallback fonts are a thing but the integration between different fonts is not always enjoyable.