In this paper, aspects of observable and non-observable based models are discussed. A survey of recent literature was done to show how using non-observable-based language carelessly may cause disagreement, even in professional research programs and incorrect assertions, even in prestigious journals. The relation between physical measurements and observables is discussed and it is shown that, in contrast to general belief, this relation may be complicated and not always straightforward. The decomposition of the system into basic subsystems (physical or conceptual) is (...) traced as the origin of non-observable-based languages. The possibility of defining new quantum mechanical observables for open quantum subsystems and of replacing them with non-observable-based concepts has been mentioned and the AIM theory is explained as an example. An account of some current non-observable-based models for molecular geometry is discussed and it is shown that not all non-observable-based languages possess the same effectiveness. In the end, the need to develop a clear chemical language is stressed. (shrink)