I heard that text rendering on GPU is hard, can the proposed approach pave the road to a reliable and widespread solution?
They are based on this work, Vello is the vector graphic rendering library and Xilem a high level ui framework.
And he is the first author of the paper (and he is also commenting in this same thread)
I'd give my interest in Hell for a graphical interface for METAFONT which would take a series of pen strokes and provide the outline --- which reminds me that I never heard back from the METAFOG author on a license....