Linear optical quantum computing is a promising approach for practical quantum computing, leveraging single photons, linear optical elements, and projective measurements. The protocol by Knill, Laflamme, and Milburn demonstrated that scalable quantum computing is possible with these resources. Subsequent improvements have bridged the gap between theoretical scalability and practical implementation. This review covers the original theory, improvements, and experimental demonstrations of two-qubit gates. It discusses realistic components, their induced errors, and error correction techniques. The review also explores the use of cluster states, Yoran-Reznik, Nielsen, and Browne-Rudolph protocols, as well as circuit-based optical quantum computing. Realistic optical components, such as photon detectors and sources, are discussed, along with loss tolerance and general error correction methods. The review concludes with an outlook on other promising techniques in optical quantum information processing.Linear optical quantum computing is a promising approach for practical quantum computing, leveraging single photons, linear optical elements, and projective measurements. The protocol by Knill, Laflamme, and Milburn demonstrated that scalable quantum computing is possible with these resources. Subsequent improvements have bridged the gap between theoretical scalability and practical implementation. This review covers the original theory, improvements, and experimental demonstrations of two-qubit gates. It discusses realistic components, their induced errors, and error correction techniques. The review also explores the use of cluster states, Yoran-Reznik, Nielsen, and Browne-Rudolph protocols, as well as circuit-based optical quantum computing. Realistic optical components, such as photon detectors and sources, are discussed, along with loss tolerance and general error correction methods. The review concludes with an outlook on other promising techniques in optical quantum information processing.