Laurenz

Some disorganized thoughts and observations: - I think this doesn't work as intended because before select_fallback is invoked, a well-known list of fallback fonts is traversed, which don't trigger the...

I also want to note that just adding an emoji will already trigger this warning as few people will explicitly list an emoji font in the fallback chain. One might...

I've posted some thoughts about warning supression and spam in the [discord forge](https://discord.com/channels/1054443721975922748/1176473207910436935). In particular, I am now of the opinion that we - should not have `fallback: "warn"` but...

Someone else also [started looking into this](https://github.com/typst/typst/issues/1967#issuecomment-2030789774) and asked a few questions, [the answers](https://github.com/typst/typst/issues/1967#issuecomment-2031543856) to which might also be relevant here.

cc: @LaurenzV

I'm not sure. I somewhat agree that we shouldn't add too many special cases, but the current behaviour is clearly not what anyone would expect. The fact that shorthands can...

The subsetter indeed doesn't produce minimal sizes, yet.

CFF font subsetting is unfortunately rather complicated.

KB is very widely used, I don't see the problem here since the difference between 1000 and 1024 barely matters in the grand scheme of the differences here.

This issue is not the right place to discuss this.