This is very sketchy at this point, but it should give a sense of our intended scope. ## Side-effects and mutation ## 1. What difference imperativity makes 2. Side-effects in a purely functional setting, via monads 3. [Phil/ling application]Semantics for DPL, using state monad 4. Passing by reference 5. [Phil/ling application] Fine and Pryor on "coordinated contents" (see, e.g., [Hyper-Evaluativity](http://www.jimpryor.net/research/papers/Hyper-Evaluativity.txt)) 14. Curry-Howard isomorphism between simply-typed lambda and intuitionistic propositional logic

15. The types of continuations; continuations as first-class values ## Continuations ## 0. [Phil/ling application] Partee on whether NPs should be uniformly interpreted as generalized quantifiers, or instead "lifted" when necessary. Lifting = a CPS transform. 1. Using CPS to handle abortive computations (think: presupposition failure, expressives) 2. Using CPS to do other handy things, e.g., coroutines 3. Making evaluation order explicit with continuations (could also be done earlier, but I think will be helpful to do after we've encountered mutation) 4. Delimited (quantifier scope) vs undelimited (expressives, presupposition) continuations 5. [Phil/ling application] [Barker/Shan on donkey anaphora](http://dx.doi.org/10.3765/sp.1.1) 6. The continuation monad ## Preemptively parallel computing and linear logic ## 1. Basics of parallel programming: semaphores/mutexes 2. Contrasting "preemptive" parallelism to "cooperative" parallelism (coroutines, above) 3. Linear logic 4. [Phil/ling application] Barker on free choice, imperatives ##Other## 17. [Phil/ling application] Expletives

18. Some references: * [de Groote on the lambda-mu calculus in linguistics](http://www.loria.fr/%7Edegroote/papers/amsterdam01.pdf) * [on donkey anaphora and continuations](http://dx.doi.org/10.3765/sp.1.1) * [Wadler on symmetric sequent calculi](http://homepages.inf.ed.ac.uk/wadler/papers/dual-reloaded/dual-reloaded.pdf) 19. Dependent types