On Thu, May 13, 2010 at 3:01 PM, Edgar Z. Alvarenga
On Thu, 13/May/2010 at 18:57 +0100, Maciej Piechotka wrote:
Hmm. What GDAT/existential do you use (for lazy people who do not want to read paper)?
The GADT that I refered was from my faileds attempts.
How is it programmed in Lisp?
The paper don't give much details, but by what I undertood, if a random generated program don't compile it's discarted.
The impression I had from your first email is that your GADT approach worked but you didn't want to use it here. Have you seen the research on higher-order abstract syntax (HOAS)? What you're doing looks fairly similar to my naive eyes. I didn't speak up because I thought you had already been down that road. I was just trying to find a paper on the subject (I think there was an ICFP paper that explained them in GADTs), and I found this: http://www.haskell.org/pipermail/haskell/2007-January/019012.html I think this is the paper I had in mind: http://adam.chlipala.net/papers/PhoasICFP08/ I hope you find those resources helpful! Jason