Contemporary software systems are beset by problems that create challenges and opportunities for broad new OS research. To illustrate, we describe five areas where broad OS research could significantly improve the current user experience. These areas are dependability, security, system configuration, system extension, and multi-processor programming. In each area we explore how contemporary systems fall short. Where we have thought of possible solutions, we offer directions for future research. To prove our point that opportunities for new OS research exist, we describe Singularity, a research project at Microsoft Research. Singularity is a new operating system designed to explore solutions to four of the challenges we have identified. Singularity incorporates three specific design decisions in order to increase system dependability and improve system security, configuration, and extension. These design decisions include the adoption of an abstract instruction set as part of the system binary interface, a unified extension architecture for both the OS and applications, and a first-class application abstraction.