26.6.09
QML: Explicit first-class polymorphism for ML
Claudio Russo and Dimitris Vytiniotis, QML: Explicit first-class polymorphism for ML, ML Workshop, Edinburgh, August 2009.
This simple solution is along different lines than ML^F, FPH, and HML. I like the idea of allowing both explicit and implicit quantification, which was explored years ago by O'Toole and Gifford. I'm surprised that to make it work they needed to avoid permitting bound variables to be in scope, so their explicit constructs look (alas) quite different than the type abstraction and application of System F. I'd always thought that having 'rigid' type variables in explicit types was a good idea, but this paper makes an intriguing case otherwise.
This simple solution is along different lines than ML^F, FPH, and HML. I like the idea of allowing both explicit and implicit quantification, which was explored years ago by O'Toole and Gifford. I'm surprised that to make it work they needed to avoid permitting bound variables to be in scope, so their explicit constructs look (alas) quite different than the type abstraction and application of System F. I'd always thought that having 'rigid' type variables in explicit types was a good idea, but this paper makes an intriguing case otherwise.
