I had a disturbing experience today: While discussing some intrinsics of a proposed new version of PalmBahn (A tool which retrieves railway departure and arrival tables, stores them in a prc file, and a Palm app to display those (It's specific to Germany, so I didn't think it'd be sufficiently important to be listed here)), someone who appears to understand at least the basics of programming dismissed some of my thought experiments as "too academic". I never really considered myself to be a truly acedemic type; two years ago I'd have laughed at that thought. However, his words got me thinking- while I don't think he's right in this specific case (If using undirected graphs and a weight function, plus Dijkstra (or Moore) is too academic for railway network graphs, what else is it supposed to be good for?) OTOH, he's probably right in pointing out that my thoughts have become more academic than what they used to be. While I don't think that this is a bad thing, I have also noticed that I take much longer to produce code these days. I refactor more often, sometimes discarding perfectly working code for design reasons. I guess that's one of the things that has slowed down the FreeSCI GFX subsystem down (apart from my usual lack of spare time ;-)