Earthquake location is a well-defined inverse problem to which the mathematical fundamentals of existing methodologies were established nearly a century ago. However, in quantitative seismology, achieving accurate, bias-free earthquake locations still remains to be the one of most important and challenging tasks. In this article, we give an overview on various earthquake location methods, that vary from linearized to nonlinear, from grid search to probabilistic algorithms. We review single and multiple-event location techniques, along with computational complexities of each algorithm. An example from a real-world earthquake location problem is given to highlight the importance of data availability in achieving bias-free earthquake locations. We discuss earthquake location accuracy, and uncertainty estimation that originate from measurement and modelling errors. We end with a list that summarizes publicly available earthquake location software packages. We conclude with an outlook for future directions towards data driven machine learning techniques in earthquake location research.