This is a talk on work in (slow) progress. I will begin with a brief overview of the technical and conceptual issues and problems that deontic logic has faced since its inception in the early 1950s.

Eventually, I will argue that any adequate system of deontic logic ought to encompass two essential features: 
— it should be many-sorted, involving separate sorts for norms, actions, and propositions, in their interrelations and interaction, and
— it should be multi-agent, involving normative modalities not only for individual agents, but also for group obligations, permissions and prohibitions — again in their interaction. 

I will then outline some technical work in progress towards developing a more adequate, many-sorted and multi-agent framework for normative reasoning and deontic logic.