a swiftly TILTing planet
I have been reading a lot of papers. I mean a LOT of papers. So that's going pretty well. I have made somewhat less progress on the HOT compiler (I really wish I would have been able to attend lectures and take notes for myself, I don't do well with other people's notes). So my questions for you all are as follows:
1) Is it a bad idea to do the 2006 HOT assignments instead of this year's? I still am not sure how I'm supposed to have them graded, but I guess if Karl would be able to autograde/check for accuracy any of them, it would be the more recent one. Still, having actual handouts that summarize all the relevant rules is pretty nice.
2) Is there a definitive paper on type inference/decidability thereof? I know 2nd order and higher stuff is not inferrable, but since I'm relying on type inference, it would be nice to be able to have some sort of theoretical backing beyond "here is an algorithm, go have fun".
1) Is it a bad idea to do the 2006 HOT assignments instead of this year's? I still am not sure how I'm supposed to have them graded, but I guess if Karl would be able to autograde/check for accuracy any of them, it would be the more recent one. Still, having actual handouts that summarize all the relevant rules is pretty nice.
2) Is there a definitive paper on type inference/decidability thereof? I know 2nd order and higher stuff is not inferrable, but since I'm relying on type inference, it would be nice to be able to have some sort of theoretical backing beyond "here is an algorithm, go have fun".