The assumption was that the solutions were chained and that they would never see the things that failed before that. But I guess I only stipulated that the failures were independent, when the math I wrote meant that the successes should be independent, as well.