| Abstract: | Although it has been established observationally beyond doubt that broadline stripped envelope supernovae (SNe) of type Ic produce long duration gamma ray bursts (GRBs), that neutron star mergers produce short hard GRBs (SHBs), and that phase transition of neutron stars in high mass X-ray binaries (HMXBs) may produce SN-Less GRBs, their production mechanism is still debated. The two leading theoretical models of GRBs and their afterglows, the fireball model and the cannonball model, have been widely confronted with the mounting observational data on GRBs and SHBs. Both have claimed success in reproducing the observational data, despite their complexity and diversity. This claimed success, however, may reflect multiple choices and the use of many free adjustable parameters, rather than the true validity of the models. Only confrontation of the key falsifiable predictions of the models with solid observational data can test their validity. Such critical tests will be reviewed. |