Entity Relationship Modeling (ER modeling) is by far the most common way to express the analytical result of an early stage in the construction of a new database. In this ebook, Alf Pedersen describes the principles for ER modeling, as well as the most important terms used in modeling a new database.
Entity Relationship Modeling - Attributes in entities (Page 4 of 21 )
Most common metadata on attributes- We take for granted, that doing a major analysis study in a given field, you use tools that are appropriate. As for the choice of a modeling tool, I give no concrete recommendations: I have used Oracle Designer (formerly Oracle CASE*Method) for the last 15 years, and I have found it to be a powerful and rich system, which delivers in many more areas than I have needed to use it for. I am probably a little biased here.
However, a toolset should include reusable objects: The results from the analysis phase should be the basis for generating tables and all other database objects for use in the design phase, as well as functions should be used for generating candidate modules.
Furthermore, the database objects and candidate modules should be used to generate the DDL (Data Definition Language) scripts for physically building all the elements of the database, as well as the candidate modules should be used to generate running program modules. Not that I expect a system to be 100% generated – far from it.
However, with such functionality you could show a prototype, which illustrated the resulting, needed functionality, but without the last finishing touch, or the most advanced business constraints built into it.
As you enter attributes, it is important that you can define their descriptions, optionality, data format and –length restrictions for later use in design. Here is what I would recommend you to: If at all possible, place each and every attribute in a domain.
A domain is a container that holds general descriptions, as well as data types and length, and in certain circumstances also can keep legal values or legal ranges for a domain. By relating every attribute to a domain definition, you ensure that if you change the domain, the change is reflected for all attributes that belong to that domain. A short example, taken from Oracle Designer:
We see that the domain MONEY has a legal range between 0.01 and 999,999.99. It sounds fair: Money must have some minimum value. How about maximum? This domain was created for use in transactions, and we have a rule in the business saying that no transaction larger than a million is allowed, for security reasons.