The discussion is initiated over and over again, whether a standard software or a project is the better solution for lotteries. Is it not true that lottery projects are driven by fixing one bug after the other just before GoLive? And this repeats itself over and over again with each new software delivery?