Discussions of conceptual modelling assume that notation details are of secondary importance, a matter of taste and past experience rather than of science. For example, it does not really matter if cardinality is shown wi th a 'crow's foot' or with the symbol '1..*'. This paper argues that such an assumption is wrong and that the notation is extremely important in the process of modelling relationships because of the normative language that the notation specifies. Normative language is shown to be a useful way of understanding and comparing relationship notation. Barker's practical relationship definition, using a formal notation, is shown to be sufficiently well formed to allow the modeller to make sense of the domain i n their own linguistic context. Less formal notations are shown to disadvantage the less experienced modeller
This paper discusses the use of interpretive research to learn about the practicality of entityrelationship modelling in ternary relationship situations. The findings can be compared with previous studies of novice modellers who used the same invented scenario. Previous research excluded the interaction between practitioners that would occur during a modelling session because novices were used to complete tasks in isolation. A team of experienced practitioners are shown to use entity-relationship modelling in a business context of social interaction about design. The interaction proves to be a key part of the modelling process. Practitioners 'talk with the notation' as well as using the notation to draw a diagram. The entity-relationship model constrains the social interaction because the model provides a way of talking about design. The practitioners use the model to talk about a normalized relational data structure in a way that undermines the idea of the entity-relationship model as an independent conceptual model. The findings show that theories from the field of linguistics explain why the model is used in this design-dependent way and suggest that this dependency may be inevitable. When the design conversation is about a normalized relational data structure there is no benefit to the practitioners from using a special notation for ternary relationships. On the contrary, the practitioner's design dependence seems to enable them to expose aspects of a domain that do not make business sense. The wider implication is that interpretive research's role is important in generating insights about the extent to which conceptual modelling is usable by practitioners. Interpretive research highlights the importance of being able to distinguish between ideas about conceptual modelling and ideas about how to apply modelling to practice.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.