2.1.4 Architectural Principles and what they mean to you
Any data model should be developed with a consistent look and feel.  This will enable developers and users to comprehend the technical design, once and then spend more time on the actual implementation.  It also creates a framework that can be used for any enhancements that a company wishes to do with the PPDM.  These practices can be summed up in the following table.

Practice  #
Practice Rule
Practice Description & Example
Every table has a synonym
This allows for the use of the synonym in SQL statements, rather than the full name.  An example out be using BA rather than BUSINESS_ASSOCIATE.
Every table/column has a comment.
Each object’s comment starts with the name of the table/column followed by an ‘:’ and then the actual description.  An example is: SOURCE: The individual, company, state, or government agency designated as the source of information for this row.
Synonyms are used in the names of the relationships
The following relationship name: BA_R_S_FK tells me that it is a foreign key (FK) relationship between BUSINESS_ASSOCIATE (BA) and R_SOURCE (R_S) with the parent table being R_SOURCE (second in the name)
Synonyms are used in the naming of indexes.
The following index name: BA_PK tells me that this index is based on the Primary Key for the BUSINESS_ASSOCIATE table.
Domains are used for numbers
Examples include depths set to a data format of 10,5.  Other examples include: Lat & Long set to 12,7; volume set to 14,4; currency set to 12,2.
Reference tables are prefixed with a ‘R_’
This allows the grouping of all reference tables together.  This makes the review/use of the model in a tool like TOAD a lot easier.
All tables have a foreign key relationship with R_SOURCE & R_PPDM_ROW_QUALITY
This was explained earlier in the chapter.
All reference data types are   set to VARCHAR2(20)
This consistency is recognized throughout the model and therefore easier to program to and plan for data content.
Consistent reference table design
All reference tables have the following columns (and data types):
in addition to the eleven standard  columns. 
Use of _SEQ as a  column name
This indicates that there is a chronological ordering of the entities.
Use of _OBS_NO as a column name
This indicates that there is not an order to the values listed.