This paper explores the limitations of objective-based search in evolutionary computation (EC) and proposes an alternative approach based on novelty search. The fitness function in EC typically measures progress towards an objective, but it can sometimes prevent the objective from being reached due to deception. The paper argues that objective-based search is limited by the potential for local optima and that searching for behavioral novelty can lead to more effective solutions. Novelty search focuses on finding new behaviors rather than directly pursuing an objective, which can lead to increasing complexity and better performance in certain tasks. The paper presents experiments in maze navigation and biped walking tasks, where novelty search outperformed objective-based search. In the maze navigation task, novelty search found novel behaviors that led to the goal, even though the objective function did not reward those behaviors. In the biped walking task, novelty search evolved controllers that walked significantly further than those evolved by objective-based methods. These results suggest that ignoring the objective can sometimes lead to better solutions. The paper also discusses the concept of open-ended evolution, where the goal is to create systems that can continuously generate new and complex behaviors. Novelty search is proposed as a method to achieve this by searching for novel behaviors rather than modeling natural evolution. This approach decouples open-ended search from artificial life worlds and can be applied to real-world problems. The paper concludes that objective-based search is not always the best approach and that novelty search offers a promising alternative. By focusing on behavioral novelty, the search process can avoid deception and local optima, leading to more effective solutions in certain tasks. The results suggest that some problems may be best solved by methods that ignore the objective.This paper explores the limitations of objective-based search in evolutionary computation (EC) and proposes an alternative approach based on novelty search. The fitness function in EC typically measures progress towards an objective, but it can sometimes prevent the objective from being reached due to deception. The paper argues that objective-based search is limited by the potential for local optima and that searching for behavioral novelty can lead to more effective solutions. Novelty search focuses on finding new behaviors rather than directly pursuing an objective, which can lead to increasing complexity and better performance in certain tasks. The paper presents experiments in maze navigation and biped walking tasks, where novelty search outperformed objective-based search. In the maze navigation task, novelty search found novel behaviors that led to the goal, even though the objective function did not reward those behaviors. In the biped walking task, novelty search evolved controllers that walked significantly further than those evolved by objective-based methods. These results suggest that ignoring the objective can sometimes lead to better solutions. The paper also discusses the concept of open-ended evolution, where the goal is to create systems that can continuously generate new and complex behaviors. Novelty search is proposed as a method to achieve this by searching for novel behaviors rather than modeling natural evolution. This approach decouples open-ended search from artificial life worlds and can be applied to real-world problems. The paper concludes that objective-based search is not always the best approach and that novelty search offers a promising alternative. By focusing on behavioral novelty, the search process can avoid deception and local optima, leading to more effective solutions in certain tasks. The results suggest that some problems may be best solved by methods that ignore the objective.