Follow-up on effect masking and backtracking

A follow-up on yesterday's post about backtracking with state: an anonymous commenter reminded me that persistent data structures are a useful and lightweight (compared to e.g. first-class stores) approach. In fact, probably my favorite paper from ESOP 2008 was Conchon and Filliâtre's Semi-Persistent Data Structures, which was addressing exactly this problem. Rather than introducing an entirely transactional semantics for the language, it's just a data structure with a transactional interface.