Type 1 diabetes can now be ameliorated by islet transplantation, although this treatment is restricted by the insufficient supply of islet tissue. The need for an essentially limitless supply of a substitute for primary human islets of Langerhans has led to research into the suitability of stem/progenitor cells to generate insulin-producing cells to use in replacement therapies for diabetes. Although there has been much research in this area, an efficient and reproducible protocol for the differentiation of stem cells into functional insulin-secreting β-cells that are suitable for transplantation has yet to be reported. In this commentary we examine the minimum requirements for replacement β-cells and outline some of the potential sources of these cells. We also argue that the generation of the ‘perfect’ beta-cell may not necessarily lead to the most suitable tissue for transplantation.