[plt-scheme] immutability and parallelism (was ... immutable pairs)

From: Paul Schlie (schlie at comcast.net)
Date: Sat Mar 11 10:35:46 EST 2006

Matthias Felleisen wrote:
> Please take a look at Mike Ernst's paper in the 2005 OOPSLA.
> You might find it interesting. -- Matthias

Thank you; I read through both papers, and like their combination.

More specifically, would love to see object predicates being formally
supported by scheme, whereby then objects (including operand declarations)
may be annotated as having particular formally defined (such as primitive
type and/or mutability) and/or user defined (such as some user defined type
and/or requiring the invocation of some correspondingly defined resolution
function upon an objects reference or assignment) properties; such that
they may then be utilized upon a programs static analysis to aid in more
efficient and/or provably correct code generation, and/or extend an object's
runtime reference semantics.

Posted on the users mailing list.