Classical computers are made from semiconductors, but semiconductor-based architectures for quantum computers are still less developed than those based on superconducting circuits, trapped ions, or neutral atoms. I will try to convince you that there are good reasons to still pursue the "semiconductor path" to quantum computers, and discuss challenges that need to be overcome on this path. I will also discuss what kinds of interesting physical problems from solid state physics and open quantum system physics one can grapple with, while contributing to a somewhat "engineering-like" task of building a scalable quantum computer.