We use two screens: one which the user provides a slice of bytes for the graphemes, and the user owns the bytes. We copy those bytes to our internal model so that we can compare between frames Signed-off-by: Tim Culverhouse <tim@timculverhouse.com>
