I just discovered Scalable Vector Graphics. One code idea that I have had previously was derived from working with html. Html in several ways seems so arcane and baroque. particularly when using an Apple II+ seems to provide a superior experience and design philosophy. I mean, whatever happened to the simple methodology of GR & HGR. where you write one command to start session, some numbers to specify your graphics area size, and then sets of coordinates to draw, single pixels, lines and other shapes. you can make whole projects that look cool and serve a real purpose in 50-80 lines of code, Plus, you can teach a little kid to do it because it makes so damn much sense. well anyway, I started speculating about perhaps writing a firefox plugin that embeds SDL and create an interpreter, (which I may still do if svg sucks). so that a client can dl a very small amount of graphics instruction from a server, that create client side graphics, but the real fun is that instead of letting a web-browser push you around and place your text and table borders in the wrong places, that you get to specify everything explicitly by just telling the interpreter the exact X,Y coordinates. ahh what a dream.