We show how water with its small proton bridge between tetrahedrally coordinated oxygen centers, forms a provocative series with (i) the archetypal glassformer and strong liquid, SiO2, (ii) the analog ionic compound BeF2 in which the Be ions are bridged by fluoride ions, and (iii) elemental Si with no bridge at all. As the size of the bridging unit decreases from the polarizable oxide of silica through the unpolarizable fluoride of BeF2, the tiny proton of water, and finally vanishes at silicon, the constraints on the network tighten. As the constraints become more severe, the departures from simple activated diffusion behavior of SiO2 during heating of the amorphous phase at ambient pressure, become more pronounced. In the case of Si, the extreme of a first-order transition from strong to extremely fragile liquid state is manifested in the supercooled regime. With water the behavior is intermediate. In both cases the anomalous regime is located in the supercooled state which greatly complicates its investigation and leads to much controversy. In the case of BeF2, for which the exceptional viscosity behavior can now be understood, new MD results show that a weakened form of the anomalies exists. The weakened anomaly should be observable in detail in experimentally accessible ambient pressure conditions, because now it occurs in the thermodynamically stable domain, at temperatures which lie above those studied to date. The observable anomalies should include not only a density maximum but also a density minimum, and a smeared lambda type heat capacity anomaly. In view of the phase diagrams deduced for various water models, our observations suggest that the second critical point scenario proposed for water, if correct, might be observable under thermodynamically stable conditions at high pressures in the case of BeF2. In the cases of water and Si, in which the interaction potentials cause the critical point to lie at relatively low or negative pressures, respectively, the glassy state proves to have unusual properties intermediate between those of normal glasses and crystals. It is suggested that such glasses, which will include many Si analogs such as Ge, InSb, etc, might constitute a distinct class of amorphous material with the attributes (low residual entropy, etc.) of the hypothetical perfect glass state.