Re: [Haskell-cafe] Re: Shared thunk optimization. Looking to solidify my understanding

On 9/25/10 3:43 PM, Jan-Willem Maessen wrote:

No one seems to have mentioned that this is a non-optimization in call-by-need lambda-calculus (Ariola et al.), where it follows from the standard reduction rules.

Exactly. Then again, call-by-need gives a form of partial evaluation, which was what I was pointing out in in the post described as:

On Wed, Sep 22, 2010 at 11:10 AM, David Sankel wrote:

...

wren ng thornton provided an evaluation using another operational semantics (reference?). Under this semantics, this optimization would be called partial evaluation. Unfortunately I couldn't follow the steps or the reasoning behind them, perhaps a reference to the semantics would help.

-- Live well, ~wren