Almost 2 decades ago, it was suggested that the peculiar emission of the Soft Gamma Repeaters and Anomalous X-ray Pulsars was caused by the decay of superstrong magnetic fields, in a (very) small population of neutron stars (NSs) with extreme properties. Originally considered an oddity, the model has progressively received strong support from a huge body of observations. Nowadays, the role of such objects in several astrophysical setttings is much debated, and their implications for forthcoming observational windows may be far reaching. I discuss new observational evidence for decay of the dipole magnetic field in magnetar candidates, that has accumulated in recent years thanks to the improved burst detection capabilities of the new generation of high-energy observatories. Making contact with previous studies of field decay in the population of "normal" radio \ pulsars, I stress the great advantages that magnetars offer in this respect. I further show that, although certainly decaying, the dipole field of these objects cannot power their persistent X- ray emission. The latter indeed requires the dissipation of an additional, stronger field component, presumably hidden in the interior of these neutron stars. Using existing models it is even possible to determine the salient properties of this internal field component, a work that can be much developed in parallel with a better understanding of NSs interior structure. Finally, I will sketch preliminary considerations on the possible evolutionary links between magnetars and other classes of neutron stars, that have nearly as strong dipole fields but do not show magnetar-like activity.