I just looked for something to do and stuff had gedit open with a kd-tree class I wanna write and I just though "meh". I reminded that I wrote that while I was at the red cross (alternative service).
When I think back, it seems that I code the most effective while a little stressed, such as school time.
I often thought that I can use the holidays to write a fuckload of code, but that never went true. Much more the opposite: I was watching movies or cartoons or something and playing games the whole day.
Maybe I become lethargic when not really having an immediate obligation to do something.
Coding only seemed shit when I had to force myself to code because of having nothing better to do. It's much more fun when I want to kill time instead when I have to - like right now.
Coding is fun, but you do need to think straight, unlike when watching some movies and different from playing games. Well, I don't really play many demanding games. Demanding as in thinking strategically ahead like a good chess-player would. Not like needing good reactions and making tactically clever moves that must be made quickly.
A little stress is always encouraging if you know that it won't become a burden.
Some people stated that they publicly announced their projects to get feedback from others and to get pushed more into finishing and releasing it.
I am not sure if this is wise with my projects. I did it with ViruZ, but I simply do not have enough experience with real projects as opposed to simple proof-of-concepts.
ViruZ finally froze because I was being too general again. I should have continued with my path to keep it simple, even if that means a reduction in quality of implementation.
I am not sure why Solace died. It was not my fault, I was away while it died and contributed after I was back at home, finding that noone else has uploaded anything in quite some time.
I plan on re-coding ViruZ soonish. I need to properly setup DSSS to get SDL working. Then I will just use sprites (images) as opposed to using raw OpenGL and using programmatic approaches to draw stuff.
Currently I am just doing some experimental stuff.
I can safely tell about my approach to the Google AI Challenge. I downloaded the C++ starter-kit, didn't like it, rewrote it and hoped for the community to like and prefer it.
I released it on the forums but it seemed to slip by rather unnoticed, unlike another kit. I did extend functionality by now, but it seems useless to post.
That post got way larger than I expected. At least I could tie up some loose ends :)