The properties and functionalities of inorganic glasses can be tuned by adjusting their chemical composition and, in turn, their atomic-scale structure. However, accurate prediction of glass properties from composition has traditionally been impossible. Recent progress in temperature-dependent constraint theory paves the way for the design of new multicomponent glasses with tailored properties. Atoms in network glasses are constrained by their chemical bonds and bond angles, and the strength of these constraints depends on the local topology and the chemical nature of the elements. By counting the number of constraints around both network-forming and network-modifying atoms as a function of both composition and temperature, it is possible to make quantitative connections among composition, structure, and certain macroscopic properties. Here, we review recent developments in glass-structure determination and modeling. We then demonstrate how the structural information is used as input for topological predictions of glass properties such as viscosity and hardness. These predictions enable the design of novel industrial glasses with desired properties and manufacturing attributes.