Where do we need Lazy data structures ? Can anyone suggest me few good examples proving the need of Lazy data structures or provide me a pointer to look for them. By "lazy data structures" do you mean datatypes with non-strict constructors? If so, the literature is full of arguments along the lines of "separating data from control", "allowing infinite data structures", etc. But, there is no "proof" that we need them. Indeed, there is no "proof" that we need data structures at all (:-). Besides I/O where do we need irrefutable pattern matching. Sometime back there was some discussion about irrefutable patterns by Paul Hudak. But, I fear it didn't give egs. where such pattern matching was needed My earlier message tried to argue for the merits of "lazy patterns", a technical term for a kind of pattern matching in Haskell, which is a somewhat separate issue from having lazy data structures to begin with. On the other hand, if I remember right, I included several examples that should motivate lazy data structures, too. In particular, there was an example of a "simulation program" that would be much more cumbersome without lazy data structures. There was also a recent message announcing the availability of a Haskell tutorial that Joe Fasel and I recently wrote. It contains lots of examples motivating pretty much every feature of Haskell (at least, that was our intent!). Another good reference would be Bird and Wadler's textbook "Introduction to Functional Programming" (Prentice Hall, 1988). Cheers, -Paul
