On 26/01/06, John Hughes
I had promised myself not to propose anything that was not already implemented, tried, and tested, but I just can't resist bringing up a proposal I've made in the past to fix the monomorphism restriction. Maybe now is the time to do so.
I know many will proclaim "just get rid of it", but consider this: without the M-R, some programs can run exponentially slower than you expect. This actually happened to me, which is how we discovered something of the sort was needed in the first place. But the current design is surely a wart.
Do you have an example of such a program handy? Perhaps I'm just doing something stupid, but I can't seem to replicate the lack of sharing that's supposed to happen with the MR turned off in GHCi, so it's hard to play around with doing various translations by hand and seeing the results. Perhaps the optimiser is getting to the code and commoning things up? If this is the case, can't we just show that it always does so, or write that commoning into the translation, as suggested by Philippa, and forget about it? - Cale