Nature: In a classical computer, data compression is standard practice. At its simplest, any number of identical bits can be compressed to just a single bit. For quantum bits, or qubits, Aephraim Steinberg of the University of Toronto and his colleagues have shown that it requires measuring the quantum states of multiple identical qubits and averaging the results. For example, a set of three qubits can have a total of eight possible combinations of binary states (000, 001, 010, 011, 100, 101, 110, 111). Averaging the value of the bits in those combinations results in just four states (0, 1/3, 2/3, 1). That average value can then be represented by three qubits: two of the qubits use the polarization and path information of one photon and the third uses the polarization of a second photon. For actual quantum computing, that method may not be effective because many qubits are entangled, which results in varying states when measured. And scaling up the method is also likely to be difficult because managing a large number of photons is not easy.