The Future
Here are some ideas that I have cooking away slowly. Who knows what stage they'll get to... Realistically, I expect about 20% of them to reach some level of usefulness. The list is in approximately descending order of interest at time of writing.
- I'm working on an extension for MediaWiki that takes an OS coordinate and outputs a nicely formatted OS coordinate with a link to the OS web site map (the get-a-map service), and does approximate conversion to Longitude and Latitude. The tricky bit is the conversion to Long/Lat. There is an algorithm for it available in the perl Geography::NationalGrid module by P. Kent which I am trying to convert to PHP, but not being a bit of a PHP newbie, progress is slow. In particular I'm missing the convenience of using regular expressions that perl offers. Ho hum.
- Ogre 3D, a library for 3d graphics that looks like it will mesh with ODE well. Ogre's got a lot going for it - cross platform, well featured, nice clean API, good documentation, and a nice Wiki with some good tutorials. It also turned v1.0.0 recently. I submitted my first ebuild to Gentoo for it. I hope to extend the tutorial a little and write some of my own. There are two simple goals I have right now: 1) howto, "From blender to ogre - how to put you own model into ogre on Linux". 2) howto, use ODE with Ogre on Linux".
- Blender is a powerful 3d modelling and animation package. It seems to be the primary Free Software 3d tool for Linux so I'm trying to learn it. The learning curve is pretty steep - blender has a unique interface with a lot of keyboard shortcuts and such. According to advocates fans this makes it a really productive tool once it has been learned, but it's not so easy to get starterd. I'm not sure my little brain can hold emacs and blender keys at the same time!
- I'm currently playing about the Open Dynamics Engine, and should I produce anything of general interst I will post it. At the moment my two main ideas are to produce some sort of electronic domino rally program, and something like a 3d version of the 90s computer game "The incredible machine". These are quite ambitious projects though, so it might not be for a while!
- ASCII mode graph drawer. This would be useful for quick and dirty analysis of stats coming out of files and the like. See the TODO file in the Stuff Tools archive file for more details.
- Binary File Pretty Printer. I once wrote a program that "knew" about various binary file formats, and pretty printed the contects of the files in a nice tabulat format, or dumps them in a verbose format. This was a great tool for diagnosing problems with the system I was working on at the time. I'd like to extnd the idea and write a generic binary file reader. It would still have to "know" about the data in the files. This would be written as a simple script deifing the file structure. Some sort of generic output interfaces would also be required. Nothng too complex, but something a little more exciting than using od to get at the content...
- Thesaurus. It'd be nice to have a simple off-line thesaurus running under Linux. Right now the options are using some on-line resource (wordnet or the like) or the one embedded in open-office. The biggest problem in writing such a project is finding a good source of data - the program itself would be very simple. With Wiktionary, maturing at the pace it is, it might be a good candidate for a truly open-source way of getting that data. It might also be possible to get the data that wordnet uses?
- Gimp batch tools. After some fun making an SVG -> raster converter with Script-Fu in The Gimp, I think maybe a wider batch converter tool that uses Gimp as the back end would be nice. Perhaps a set of Script-Fu programs and a small GUI that calls them in various conbinations. Something maybe for making animations - select a bunch of filters to apply to a series of images, batch process them and then wrap it all up into an MPEG... Who knows. The sky's the limit.