---

Motivation for the Relationl Data Model

---

• Level of "abstraction"
  • What is abstraction:

    Abstraction = removing details so that you can focus in what is essential (= important)

    ---

  • Facts:

    To different audience, different things are important Consequently: We have different levels of abstract

    ---

    ---

  • High level of abstraction:

    Many details are "removed" to focus on a small set of essential properties

  • Low(er) level of abstraction:

    Few(er) details are "removed" and must work with a larger set of essential properties

  ---

  ---

  ---

  ---

• The ER-model
  • The ER-model is:

    a high level of abstraction

    Why:

    The ER-model is only concern about the accuracy of the information that we want to store The ER-model does not concern itself with many details that a database system must consider, such as: Storage efficiency, e.g.: The ER-model can contain Multi-valued attributes whcih cannot be stored efficiently Unique identication of items, e.g.: The ER-model can contain weak entities whcih does not have a key attribute used for identification

  ---

  ---

  ---

  ---

• A data model with a lower level of abstraction
  • The Relational (data) model:

    We will now study a data model that has a lower level of abstraction (than the ER-model) The Relational Model is suitable to describe the data that we will store in the database