Introduction - Desktop GIS: Mapping the Planet with Open Source Tools - page 6

GIS Data Types in a Nutshell
You are about to be exposed to a bunch of new terms as
we launch into our discussion of what you can do with OSGIS.
Among these are GIS data types. Essentially you can divide GIS
data into two types: vector and raster.
Think of vector data as things you would draw with a pencil
and paper. We could draw points, lines, and polygons. In GIS,
the features have a location in the real world, allowing us to
examine their relationship to other features.
Taking it one step further, we can attach attributes—
information about the feature. Our vector data can have one
or more attributes. For example, we might create a polygon
that represents the outline of a lake. The attributes for the lake
might be name, area, perimeter, and mean depth. Attributes
are stored in fields in our dataset, whether they be in a file or a
database table.
These two characteristics, location and attributes, are what
make GIS different from a simple drawing or paint program.
The other type of GIS data is raster data. In a raster, the infor-
mation is represented by cells (in some cases, pixels) where the
value of each cell represents a quantity or color. Examples are
a photograph where the cells represent a color and an image
where each cell value represents an elevation.
In GIS, we use both types of data, depending on what we are
trying to accomplish. In the simplest case, we might use a raster
image—an aerial photograph in this case—showing our neigh-
borhood. We then would overlay our vector data in the form
of streets. It not only makes a nice picture to look at, but with
attributes attached to the streets, we can also learn the name
of each.