It's on the list as a research project. It's not obvious to me at this stage that the bottlenecks for more advanced parsers are necessarily going to be in the same place as they are for JSON. It might make more sense to look at a state-of-the-art parser and see if we can contribute a few tricks instead.