[galaxyLogic]

I think it's worth studying and developing further, for instance study how to make AI which will not "fail in unexpected ways"