Who: Sean Murphy When: 9th April, 7pm Title: Wolfram Mathematica and functional programming Abstract: I've been using mathematica now for what amounts to 13 years. As a maths student it was a obvious good program to use for doing symbolic manipulation of equations: Integration, differential equations, solving systems of equations and that. Most people think of it in those terms I think. As I learned more about it, the range of problems I used it for grew. It had a built in programming language that seemed to encourage a bizarre style of programming. Years later I learned that this was functional programming: Functions as variables, code as data and an ability to represent common patterns at a very high level. It was good then but its even better now. Its suite of functionality covers algebra, calculus, plotting, graphing, 2D and 3D vector graphics, linear algebra, number theory, graph theory, statistics, probability theory, import and export of data, text processing, image processing, parallel programming, interfacing with external languages, CUDA support, distributed programming, web services, typesetting, user interfaces .... probably other stuff. All (most) done in a functional style. Like all good Narcissists, Stephen Wolfram is convinced that his creation is the answer to all of life's problems. I'm not far behind him but I will also tell you what its *not* good at. I'm not sure what I'll show but it will probably be a random walk through some of this functionality, with an emphasis on the functional aspects. |

Coming Up >