Governments around the world have been releasing raw data to their citizens at an increased pace. The mixing and linking of these government datasets enhances their value and makes new insights possible. The use of mashups, i.e., digital works in which data from one or more sources is combined and presented in innovative ways, is a great way to expose this value. Mashups enable end users to explore data that has a real tangible meaning in their lives. Although there are many approaches to publishing and using data to create mashups, we believe linked data and semantic web technologies solve many of the true challenges in open government data and can lower the cost and complexity of developing these applications. In this chapter we discuss why linked data is a better model and how it can be used to build useful mashups.