On Friday, October 26, 2012 11:06:59 AM UTC-5, Brian Craft wrote: > I've read about four tutorials on monads so far, but it still escapes me. > > In fact, I'm still not sure what problem it solves. I'm familiar with the > problem of having, say, three functions like f(a) -> b, g(c) -> d, h(e) -> > f, which you'd like to chain like f(g(h(x))), but you can't because b is a > different type from c and d is a different type from e. The monad tutorials > all start with a problem like this, but I still can't tell if they're > actually providing a solution, because it appears every monad is specific > to a particular type. E.g. a sequence monad. So, great, I have something > that takes a scalar and returns a sequence. That might solve g(h(x)) if f > is a scalar and c is a sequence by letting me write g(s(h(x))), but it > doesn't solve the whole problem, since I still have f() to worry about. > > <snip> >
Brian, you may have looked at it already, but I found Konrad Hinsen's 4-part tutorial linked from the README in Clojure's algo.monads repository to be quite helpful: https://github.com/clojure/algo.monads By the way, the links in that README to Jim Duey's tutorials are broken; here is the correct link to Part 1 of Jim's tutorials: http://www.intensivesystems.net/tutorials/monads_101.html I read Jim's tutorials first, and then tried Konrad's. I found the latter a bit more helpful. Also, Konrad's explanation of the state monad in Part 3 really "clicked" with me, so I felt it was worth working through Parts 1 and 2, even though I was still scratching my head a bit when I finished Part 2. Jim also has a more recent set of blog posts dedicated to exploring monads: http://www.clojure.net/archive.html I worked from the bottom of that list to the top, after reading the other tutorials, and came away with a basic but sound understanding of the core concepts: m-result, m-bind, the Monad Laws, etc. Also note that Jim has developed a new monads library for Clojure, implemented with protocols: https://github.com/jduey/protocol-monads Working through that library's test suite seems like a good way improve one's understanding of monads; also, I'm going to try to "port" the examples from algo.monads over to protocol-monads. I haven't done that yet, but when I do you will be able to find them in my fork of protocol-monads (if things go well, I'll submit a pull request): https://github.com/michaelsbradleyjr/protocol-monads/blob/examples/src/examples/monads.clj The original examples for algo.monads: https://github.com/clojure/algo.monads/blob/master/src/examples/clojure/examples/monads.clj -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en
