[Haskell-cafe] Runtime strictness analysis for polymorphic HOFs?

[Jason Dagit]

On Sun, Jun 14, 2009 at 8:18 PM, Eugene Kirpichov <ekirpichov at gmail.com>wrote:

> The idea looks cool, but perfect strictness analysis is not possible,
> t.i. the problem of determining whether f _|_ = _|_ is undecidable,
> since it is a non-trivial property of f (there exist f's for which it
> is true, and ones for which it is false) and non-trivial properties
> are undecidable, thanks to Rice theorem.


Unless you remove _|_ from the language.  If you have a "total" functional
programming language[1] it becomes a trivial property, right?  I guess this
would also extend to some dependently typed languages too.

[1] http://lambda-the-ultimate.org/node/2003

Jason
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.haskell.org/pipermail/haskell-cafe/attachments/20090615/0c4edaf9/attachment.html