Cooperative Oracles: Introduction — LessWrong