On Sun, Jun 28, 2009 at 9:49 AM, Roger Mason<[email protected]> wrote: > Mark, > > Mark Knecht <[email protected]> writes: > >> So the main question is what sort of language (and possibly >> programming environment) should a complete novice look at to get his >> feet wet with GUI programming. I'd like something fairly light - >> performance probably won't be a huge problem - that I could run under >> Cygwin or maybe compile to run native in Windows should that ever >> become useful. For now it's probably a relatively simple Linux app >> that I'd likely run once a week on Saturday morning on 15 to 20 >> databases I collect on Friday night. > > One possibility is R (http://www.r-project.org/). It has very good > graphing facilities, can access various database engines, is > multi-platform and unless you process immense quantities of data, should > be fast enough. There may be people on the R mailing list doing the > kind of thing that you want and there may be an add-on package that > matches your needs (there are hundreds of add-ons). Lightweight? No, > but you don't need to learn all of it, just the bits relevant to your > usage. R is in portage. > > Another possibility is Root (http://root.cern.ch/drupal/) but it > requires you to program in C++ (but there are Python and Ruby bindings) > and is probably a steeper curve to ascend than R. However, Root is > capable of processing huge amounts of data quickly -- that is what it > was designed for. Anything you can do in R you can do in Root, but you > will write more of the application yourself rather than using canned > routines. I have seen messages on the Root mailing list from people > working with fiscal data. Root is not lightweight, but is _is_ very > powerful. Root is in portage. > > I'm sure there are other very capable systems out there, these are two > that I use (or have used). > > Cheers, > Roger > > Actually, I have R on one machine now. I haven't done much with it. There's a good Google University YouTube statistics course you can take online for free - was taught at Google, recorded an put on you Tube, on data mining that uses R. I went through about 4 hours of that but got distracted by life and didn't finish it.
R might be a good solution in that I could play on Linux but always be able to go to windows if the need arose. Thanks, Mark
