> When the workshop returned the wooden balls, he tested building some molecules. This was when he had a moment of insight: there was a vast amount of information baked into the holes’ positioning. The model molecules automatically had the correct form and structure, because of where the holes were situated. This insight led to his next idea: what would happen if he utilised the atoms’ inherent properties to link together different types of molecules, rather than individual atoms? Could he design new types of molecular constructions?