Decision making, whether by people or machines, is often based on a limited set of eventualities. Which candidate will be elected in a primary? How many job offers will I receive in the next month? Will s(he) call tonight? Research characterizing coherent choice behavior in such contexts has typically relied on strong assumptions about available choices and/or the generalization of eventualities beyond those specified in the choice problem at hand. I provide an overview of new research attempting to quantify attitudes towards uncertainty over a limited domain of future events, eschewing strong assumptions. I will attempt to sketch examples where this could be useful in applied work (e.g., experiments and eliciting decision weights from expert systems).