Well, here for your delight is my first attempt at what some will undoutably think is a broken matrix library. However I would argue that my principal aim is to allow functions to be polymorphic over matrices as well as scalars. Features are: - Scalars and Vectors supported as [1x1] and [Nx1] matrices. - Standard arithmetic operators do standard matrix operations (except "/" is unimplemented - until I write a Gaussian-elimination routine) - 'dot' operators '.*' './' do elementwise operations on matrices, and are composed into a matrix level analogue of the standard Numeric class hierachy. - Any operation involving a scalar applies that operation to each element in the matrix. - operations in Floating that have no obvious matrix equivalents are applied elementwise (sin/cos/tan) - operations in Floating that have matrix equivalents (exp) are applied elementwise for consistancy with the other functions in the Floating class. So functions which generalise over scalars, vectors and matrices can be written... for example element-wise generalisation: f :: FloatingMatrix a i e => Matrix a i e -> Matrix a i e f x = 1./ (1 + exp (-x)) can be used with: f (scalar 2) f (vector [1,2,3]) f (matrix [[1,2,3],[3,4,5],[6,7,8]]) On reflection, am not that fussy over the naming of functions. 'exp' could be matrix exponentiation rather than the element-wise one. It could be that just defining a map for matrices is enough, in which case 'f' could be written: f :: FloatingMatrix a i e => Matrix a i e -> Matrix a i e f z = mmap (\x -> 1 / (1 + exp (-x))) z I will add features as I require them, but if anyone actually wants to use it, I will consider patches and requests for features... Idea's and improvements greatly appreciated... The archive includes the module, and a modified copy of lambdaTeX which uses smaller fonts, more appropriate to modern high resolution printers and displays. Keean.