File systems

Data should be stored in the computer in a way that will allow quick access and cross-referencing.

There are three ways to store, retrieve and organise data:

Simple lists

This is the simplest form of a database. As a new item is added to the database, it is added to the bottom of the list. Data input is easy using this method but data retrieval is inefficient. Some form of structure to the list is necessary to help speed up the retrieval process.

Ordered sequential files

This type of database is organised like names are in a telephone book. That is, they are ordered on name and structured alphabetically. Using this approach stored items can be retrieved much quicker than the previous approach.

Indexed files

The previous two approaches require that data be retrieved according to a key attribute, eg the name of individuals. However, in the GIS, items (be they pixels, points, lines or polygons) will have a key attribute as well as information about their associated attributes, and it is often this information that is required from the database.

Access to data in indexed files can be speeded up by using direct files and inverted files.

• Direct indexed files use the records to provide access to other information ie the record for each items contains enough information so that any search will skip over unnecessary items.
  
  
• Inverted files require an initial sequential search of the data so that the data is ordered on the attributes of interest. The results of this search are stored in the inverted file which provides the key for further data access (Burrough and McDonnell, 1998).