Since error correction is crucial to any large-scale quantum computer, there is significant effort in designing more efficient and robust protocols and verifying their efficacy. On the theoretical side, we have witnessed code families of much better parameters with or without Euclidean locality constraints and new decoding strategies are proposed to reduce the time overhead of logical operations. Experimentally, small instances of error correction protocols have been demonstrated in labs, showing that various tasks can be done better at error corrected settings. I will give my reading of the status and discuss opportunities.