I'm about one third of the way through the rewrite of the system so far, this time using doing the job properly (checking that there aren't too many or too few of anything in particular, that the numbers all agree, etc). Even if they had no other worth, the clarity that the monadic approach lends to code (parsing and error handling at the moment) more than justifies the presence of monads in Haskell.
I expect to have the system into a fairly robust state in the next day or two after which I'll knock together a small test suite (De Morgan's Laws, a bunch of tautologies and the like) and then move on. I hope to have the features required to implement K finished well within two weeks.
If anyone knows of a way to define a hypergraph inductively, please drop a note in the comments. If I can't find such a representation then I'm going to have a hard time supporting ternary and higher relations.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment