by Brian B
The article I chose this week is named “Toward a Next Generation Data Modeling Facility: Neither the Entity-Relationship Model nor UML Meet the Need” by David Kroenke and Cary Donald from the Journal of Information Systems Education. It starts off by talking about the importance and the purpose of data modeling. The article states that “the main use for data models is to describe and document the user’s view of the world. (Kroenke & Gray, 2006)” The authors then put forth a list of criteria that they believe are required to create a new and better suited data model for today’s world. The first criterion that they go into depth about is that a model needs to be “sufficiently robust.” This means that the model needs to be to represent the user’s idea of how objects are defined. They then talk about their next requirement, that a model be as simple as possible. The requirement states that the models need to be easy to understand to someone with even the most basic training in data modeling. The next require is that a data model must be independent from anything else. They say that “A data model should be a representation of the users’ semantics, and nothing more. (Kroenke & Gray, 2006)” The next requirement that they list is that a data model must “utilize domains with inheritable properties.” This would allow for a data model to be easily changed later down the road without having to go back and manually change every attribute that is incorrect. The last requirement is that a data model support data migration. This would allow for data models to be more easily changed as new requirements are added to the model. The article then goes into a brief history of the E-R diagram and talks about the Erwin software and its ability to convert between IDEFIX & IE notations. After that the author starts to test the IE notation of E-R diagrams. He finds that while it is robust, it does not meet any of the other criteria that they have laid out in the article. They conclude that “Based on years of data modeling experience, we are convinced that there is a fundamental conceptual problem with the E-R model: namely, it represents entity relationships! Most users do not think of relationships as things.” Because of this many users have problems learning E-R modeling. They also have a small section on UML which finishes with them saying that it has roughly the same problems that E-R diagrams do. The article ends with a conclusion that there is a need for a new data model and that it should follow the criteria that had been laid out earlier in the article.
The article is related to our class because it talked about the E-R diagrams and that is what we covered in class this week. It also deals a lot with the deficiencies that are in E-R diagrams and how creating a new modeling system would improve learning how to model data. While the article is 6 years old I think that it still has relevance in this class because we are learning about E-R diagrams and because it presents an interesting argument for creating a new system that would replace the models that are widely used.
I liked this article because it talked in depth about the subject matter that we are talking about in class. I also liked it because it points out a lot of the problems with the E-R diagram and a desire to create better forms of modeling that would make it easier for more people to learn how to do.
Kroenke, D. M., & Gray, C. D. (2006). Toward a Next Generation Data Modeling Facility: Neither the Entity-Relationship Model nor UML Meet the Need. Journal of Information Systems Education, 29-37.