ARCHIVE / What is a data structure?
Data is literally raw information and, according to information theory, raw information can be viewed as the equivalent of an extremely large number. Just as we have no way to determine the meaning of any given number without contextual information, raw information is effectively useless without its being related to something in the real world. Relating raw information to the real world is done in two ways. The first is through decoding, or converting, a number into a representation we can understand like text in a human language or a picture, bit of audio and so on. The second is through relating a piece of data to other pieces of data. Such pieces of data are referred to as nodes or data types. These relationships among nodes are often referred to as references and collections of interrelated nodes are referred to as data structures.
To provide an example of a data structure, consider a family tree. A given node in a family tree might contain a person's name and references to that person's mother and father, each consisting of another node. By having enough nodes and knowing the context of the references, we never need to write down the words "aunt", "cousin" or "paternal grandfather" in order to determine the name of each for any given individual recorded as a node in the data structure.
Since a collection of numbers can itself be treated as a number, data structures can refer to other data structures and anything else which can be represented as a number. This includes computer code itself, allowing for programmatic functions to be referenced by data structures, thus giving rise to the methodology of object-oriented programming.
last updated 2008.02.19
What is data normalization?
What is a data tree?