In order for their content to be exploited to the fullest, true electronic dictionaries must store data using underlying structures and categories. Specialised terminology dictionaries, especially multilingual ones, can strongly benefit from a design where the data is organised into highly granular categories and structured following a monosemic ‘spoke-hub’ model. A representation consistent with the ISO 16 642 standard (TerminologicalMarkup Framework, 2003) would be a very appropriate way to implement this model. However, using XML markup that meets this standard set by the International Organization for Standardization is not a guarantee of the data’s quality or interoperability.