Archived tweet #3559 from the @leastfixedpoint Twitter Archive

@jonharrop @psnively Algorithm W is not well presented in that paper - go for Damas+Milner 1982 "Principal type-schemes for fn'l progs"