[ The Types Forum, http://lists.seas.upenn.edu/mailman/listinfo/types-list ]
Hello, On Thu, Sep 10, 2015 at 10:48:33AM +0200, Gabriel Scherer wrote: > In particular, the work of Christian Mossin on flow analysis seems > relevant. It was done in a typed setting but, in "Higher-Order Value Flow > Graphs" (1997) a dynamic type is considered; it could be interesting to > translate your language to his system to compare your analyses. It may be worth noting that the paper "Constraint Abstractions" by J. Gustavsson and J. Svenningsson offers a pleasant, constraint-based view of flow analysis and a significant improvement in complexity compared with Mossin's algorithm. http://www.cse.chalmers.se/~josefs/publications/ca.pdf -- François Pottier [email protected] http://gallium.inria.fr/~fpottier/
