Description
Originally radio and optical observations were at the core of blazar selection and classification and the term blazar was used to group flat-spectrum compact radio sources with smooth nonthermal continua which showed strongly polarized, violently variable emission. The term was widely used under the framework of unification of radio-loud sources where relativistic beaming turns the observed pro[perties strongly orientation dependent. In this scheme blazars are the aligned, Doppler-boosted jet population of underlying radio galaxies and they came in 'two falvours': BL Lacs and Flat spectrum radio quasars.
However, with observation opening across the elemagnetic spectrum, initialy X-ray and later Gamma-ray surveys have almost largely forgotten about some of the original defining blazar criteria such as strong polarization and variability, or even radio charateristics. The term seems to have evolved to be synonym of a source that shows a two humped spectral energy distribution. Since the spectral energy distribution is not a snapshot of the source at any given time, the shift in classification begs the question of what exactly do we mean when we talk about a blazar. This communication is an attempt to bring attention back to the question of classification and the selection biases that we encounter in defining samples of these still rather remarkable sources .