Many software systems today control large-scale sociotechnical systems. These systems aren't just entangled with the environment but also with our dwindling resources and mostly unsustainable way of living, while the planet's population continues to grow. Dealing with sustainability requirements and systematically supporting their elicitation, analysis, and realization is a problem that has yet to be solved. Decades ago, the discipline of software engineering dealt with similar shortcomings in its processes by including safety and security as new system qualities. In light of the increasing consequences of inadequately addressing sustainability in developing software systems, software engineers must apply the lessons learned from these prior research efforts and identify the necessary research agenda. Considering sustainability in software engineering means more than energy efficiency and green IT, which are concerned with the first-order impacts of software systems. Software engineers must also take into account the second- and third-order impacts in the system context, even if they're hard to assess. By doing so, engineers have the potential to considerably improve civilization's sustainability. The Web extra at http://youtu.be/VC07j6a1XUw is a video in which author Birgit Penzenstadler talks about how software engineers can considerably improve civilization's sustainability by taking into account not just the first-order impacts of software systems but also their second- and third-order impacts.