Hi Quinn -- I can more-or-less reliably reproduce the same assert (with not exactly the same backtrace) by doing something particularly convoluted with zillions of lines of code. I'm going to try to reproduce it later this afternoon with as small of a case as possible. It'll still be weird but I think it'll trip the same assert -- and if I'm successful, I can tar it up and send it off. Would this be useful to you?