Information Technology Reference
In-Depth Information
Downloading the database
For the analysis of the data, the user will probably want to transfer them from
the server to a local system. It is more convenient for sure, and in some cases
it is not possible for an ordinary user to access the entire data store from the
server. The data storage for the systems cannot be transferred directly from
one computer to another; the data must first be dumped or exported. The
export operation saves the data to a text file, which may be in one of several
formats. Most typically the output file is a sequence of commands that can be
applied to the database program to recreate the database. This is most useful
if the database will be recreated on another computer. Alternatively, the data
may sometimes be exported to an XML format, or perhaps to a comma-
separated-variable (CSV) format; which may be convenient to import into a
different kind of application, like a spreadsheet, or a textual analysis applica-
tion. In any case, the resulting text files can be transferred between comput-
ers using standard network file transfer method such as FTP, or perhaps SCP
(secure shell copy).
One of the benefits of the SQL language standard is that a dump into SQL
from one database system can be used to recreate the data in a different data-
base system from a different vendor. This is not 100 per cent accurate as there
are small differences between the different applications' usage of SQL (despite
the fact that it is a standard), but the differences are minor and can usually be
managed with some editing. So a user could dump data from
MySQL
and load
the data into
MS SQLServer
on their local system.
MS Access
and
FileMaker
Pro
also have import functions to receive data from external sources; but since
neither of them accepts SQL commands, the data will need to be exported
from the database in CSV or perhaps XML form.
Geographic data files
Some data are most usefully displayed using maps. Public health data, census data
and consumer product preferences, for example, often have a geographic compo-
nent. In other words, the data of interest vary by region, and analysis of the data
will tend to include representing the data on a map of some sort. Geographic data
have their own specialized formats, which include the data of interest, say voter
percentages, and a location on a map of where the information comes from, a city
or county for example. Geographic Information Systems (GIS) usually use a par-
adigm of defining a geographic region using a “shape” which may be a point in the
case of a city location, a set of line segments in the case of a road or a river, or a
polygon in the case of a political boundary or physical structure. These shapes are
stored in one or more files. The shapes have attributes which are the properties of
interest to the researcher, and the attributes are stored in a simple database for-
mat and have a pointer to the associated shape, so that the GIS application can
select an appropriate way to display the shape. In general, it is difficult to convert
