1Data management is ultimately concerned with representing some aspect of reality that must be recorded, analyzed, or communicated. The complete representation of reality itself – our world, the universe – is an intractable problem as we will see shortly, but even representing some small aspect of it can be a daunting a task.
The information management professional must remember that many of these issues are often not obvious to people outside of the discipline.
The aspect of reality that is chosen to be represented depends upon two main determinants. The first determinant is the nature of the domain for which the data are to be managed. Common generic data management domains are:
- objects: These can be physical or conceptual entities such as automobiles, train or airplane reservations, or health records. Data management in this domain involves recording and tracking objects.
- events: These can be any type of occurrence for which a record is desired, such as a business transaction or a bank deposit. Data management in this domain involves recording in a highly reliable fashion any facts about events necessary for reconstructing them at a later time.
- organizations: These can be individual businesses, communities, governmental agencies, or departments within larger entities. Organizational data management in this domain involves recording and tracking: objects within the organization, including people; events and processes within the organization; and relationships between processes and entities.
- physical phenomena: These can be any observable occurrences such as: geological conditions, weather conditions, or astronomical processes. Data management in this domain involves recording any measurements necessary for deducing, at a later time, an understanding of a phenomenon.
- multiple domains: This can be the management of data across several domains given above or multiple instances of the same domain. A common example is the need to manage data about multiple organizations, each having their own data management scheme, in a coordinated fashion. Data management in this context may involve the development of a global data model that encompassed the data models of the individual domains or it may involve the creation of mechanisms to map between the data models in the constituent domains.
One requirement that is common to most data management domains is the need to support queries. A query in this context is a question that is posed to and answered by a data management system using the data it is managing.
The second determinant of what aspects of reality are to be represented is the set of tasks to be performed with the data that are to be managed. Suppose we are designing a data management scheme for a grocery store. For example, suppose that one of the tasks the store needs to perform is tracking its inventory so that it can know when certain items must be replenished. The aspects of reality that must be represented then include not only product numbers and their quantities for the inventory, but information about which of those products have been purchased by customers.
The concepts discussed in this section apply to many forms of data management. Most can be applied using pencil and paper. That is, they do not necessarily require computers and database management software. Technology will be discussed later. Concepts surrounding the nature of data will be discussed in the remainder of this chapter.
- 1325 reads
