Although MRI has made impressive advances over the past 50 years, along the way we have witnessed several examples of “bogus” science – biased reports, outright fraud and human mistakes. These episodes serve as examples of behaviors to avoid, and they also highlight the hubris of journal impact factors. Selected anecdotes may serve to enlighten newcomers to the field. Of more scientific interest, biological tissues are intrinsically heterogeneous and have a characteristic microstructure, but MR images portray the values of tissue signals averaged over voxels whose dimensions are limited by spatial resolution. A number of approaches have been developed to probe tissue microstructure and heterogeneity beyond the resolution limit. For example, diffusion MRI techniques provide information on the compartmental microstructure within cellular and porous media. A variety of methods have been developed which may be used to quantify microstructural properties based on models of diffusion within tissues. Time dependent diffusion is one promising approach which makes use of diffusion encoding gradients with different temporal waveforms e.g. cosinusoidal, that probe diffusion over different scales. This allows the derivation of cell sizes and intrinsic diffusion rates within tissue compartments based on a relatively simple set of acquisitions and data analyses. This method is of particular interest as it can be translated to clinical applications using current MRI scanners.