The upcoming European design study 'Einstein gravitational-wave Telescope' represents the first step towards a substantial, international effort for the design of a third-generation interferometric gravitational wave detector. It is generally believed that third-generation instruments might not be installed into existing infrastructures but will provoke a new search for optimal detector sites. Consequently, the detector design could be subject to fewer constraints than the on-going design of the second-generation instruments. In particular, it will be prudent to investigate alternatives to the traditional L-shaped Michelson interferometer. In this paper, we review an old proposal to use three Michelson interferometers in a triangular configuration. We use this example of a triple Michelson interferometer to clarify the terminology and will put this idea into the context of more recent research on interferometer technologies. Furthermore, the benefits of a triangular detector will be used to motivate this design as a good starting point for a more detailed research effort towards a third-generation gravitational-wave detector. © 2009 IOP Publishing Ltd.