You seem to think the word was in use prior to the development of taxonomy, but the word "dinosaur" began as a specific taxonomic term, coined by Richard Owen.

You seem to think the word was in use prior to the development of taxonomy, but the word "dinosaur" began as a specific taxonomic term, coined by Richard Owen. From the outset, it was recognized that not all Mesozoic reptiles were dinosaurs. If conformance to common usage is an important taxonomic principle, then maybe taxonomists should put the label "fish" in their taxonomies so that whales and cuttlefish are included, as these were commonly regarded as a type of fish prior to the development of scientific taxonomy? This example shows why this principle does not work in general: "fish" would then include a great many species - including ourselves - that are not regarded as fish by anyone.