Abstract: The electron spin is a natural choice for encoding a quantum bit (qubit), the fundamental building block of a quantum information processor. In this talk I will review semiconductor quantum dot approaches to realizing spin qubits, and some of the key challenges to scaling up. I will describe recent efforts in my research group to develop CMOS-like silicon quantum dots for scalable, long-lived spin qubits, and ideas for implementing active error correction in such devices. Results will also be presented from experiments on quantum devices based on III-V nanowires, including superconductor-semiconductor junctions that show unusual effects such as Andreev bound states and proximity supercurrent. These nanowire junction devices are useful in the exploration of Majorana bound state physics as a possible basis for topological quantum computing.