As with the collection of data, this is one of the most important and time-consuming tasks, however, a well designed
database not only will save time and effort but a carefully constructed plan will provide a blueprint for the database structure,
thus offering the maximum display for analysis.
One of the most important features of the design is to aim for a productive
database, one that will give the maximum amount of information allowing for a representative breakdown of the data. A second
criteria is an understanding of the difference between the way the data is used and the way a Relational Database Management
System (RDBMS) stores data. After all data is information, facts presented in a useful manner, i.e. using the same facts for
a number of purposes and combining this fact with additional facts.
Therefore, Relational Databases need to have facts
about different subjects stored in separate tables. . Ultimately, to combine facts in a meaningful manner the database application
needs to know how the subjects are associated with each other, or connections.
Much thought went into the planning
of the Akhmim Database, and to a certain extent it is a flexible design, whilst still retaining its basic structure. Certain
concepts and questions were considered while in the design process, such as purpose, what nature of facts are needed to store?
The data has to be collated in an efficient manner as possible
in order to provide a speedy and accurate retrieval system,
essential for analysing the distribution of material and in-depth breakdown on the genealogy of families. In addition, a forward-looking
approach was needed as to what application the database would be associated with in the future. This may involve not only
the development of a CD-ROM and application for Internet use, but the design process should allow for the subsequent conversion
of more advanced computer techniques, such as 3D and VR work.
By utilising these design criteria in conjunction with an
acceptable software application, the ultimate goal may be easier to realise, that of full textual and graphical analysis.
With these concepts in mind, the initial task was to determine what type of tables were required, that is the division
of information into separate subjects and the type of information required in each table. Each category of information is
known as a field and displayed as a column describing the subject, while relating directly to the subject of the table. With
this in mind a design sheet was developed, otherwise known as a Caption sheet (fig.13) consisting of a number of relevant
subject headings. This Caption sheet was to provide the basic card index, whereby all information relating to each individual
object would be recorded with its own individual unique reference number prior to entry into the database.
See Author's Preface