How's the Thermometer of Abstraction Coming?

It's still mainly bottlenecks on getting the math to a point where computations are actually tractable. The results so far have basically been like take three different ways of trying to formulate what sorts of things would be natural abstractions. And then we find that all three of them end up roughly equivalent. But it's still, they're still extremely hard to compute. A lot of what I've been working on over the past few weeks, past month or so, is hammering it out and getting to the point where we can efficiently compute it. Do you think sometime this year? Yeah, I would be surprised if by the end of March we're not at that point.