A Brief History of SML#

The origin of the SML# can be traced back to the proposal of Machiavelli published in ACM SIGMOND 1989. This paper showed that ML with record polymorphism can be extended with SQL and reported a implementation of a prototype interpreter.

In 1993, Atsushi Ohori extended the Standard ML of New Jersey compiler at Kansai Laboratory of Oki Electric, and named the experimental prototype “SML# of Kansai”. The Internet still remembers the posting of SML# of Kansai to the types mailing list.

The name ‘‘SML# of Kansai’’ symbolized the field selector #label, which was given a polymorphic type for the first time by this compiler. This prototype implementation of th compiler was reported in the ACM TOPLAS article on record polymorphism as SML#.

To support not only record polymorphism but also inseparability and other practically important features, we decided to develop a new SML-style language from scratch, and in 2003, we started the SML# compiler project at Japan Advanced Institute of Science and Technology as a part of the e-Society project funded by the Japan ministry of science, education and technologies

In 2006, the project moved to Tohoku University.

In 2021, the core activity of development moved to the SML# development team.