In quantum field theory, a model's predictions at sufficiently low resolution often depend only on relatively few details of the model's construction. This is called universality. Thanks to universality, finding models that accurately reproduce the results of low-resolution measurements doesn't require getting all of the high-resolution details exactly right. The key is knowing which combinations of high-resolution details do matter at much lower resolutions. This article introduces some of the basic ideas that lead to that insight, emphasizing how those ideas relate to the act of taking (near-)continuum limits of models that were initially defined by treating spacetime as a lattice. Models that involve only scalar fields are used as examples.
