Plain text database

A flat file database is a database that is a plain text file, and that does not conform to the record structure database, the hierarchical database or the relational model. The term refers to the file's data, and to the files's structure of that data, known as a description. The software used to create a flat file is initially called ALGOL (short for ALGOrithmic Language), COBOL, its name is an acronym for COmmon Business-Oriented Language. Sometimes software is mistakenly called the database, for instance the Query language SQL for a SQL database. Flat Files are nowadays archaic and for the most part not existing in their original form anymore. When they were introduced however they were modern. Flat files or string files were at that time the only possible way to store information on a magnetig medium because Random Access of data was impossible and the main storage medium were reel tapes. Files were a consecutive order strings containing information. Different files were placed one after another on these reel tapes. Scientific research into computer science, databases and the technology were by and large the products of spin-offs from other scientific areas. Big Blue wasn't that Big at that time but became the moloch we now know in these days.

File Structure

A Flat File Database is known from its basic structure, it only is a repetitive string of characters of a predeterminate size. The computer language used to create and handle these strings determined the possibilities to use the content of these strings, other then for character use, but for instance for usage as dates, times or numericals. With the introduction by IBM of the programming language PL/1 (short for Programming Language 1), IBM and the compiler that could use this sub-compiler (COBOL) introduced advanced numerical possibilities to the standard languages used until that time. A flat file basically is a row of ended strings in a file. The strings are of one length, the position and the amount of bytes within that string from a start position to an end position determine the content of a subfield in the string. COBOL is capable to match the name associated to a certain element within the string. The sub-string obtained can be used for further calculations.

Data handling

Flat Files even though they are only strings can contain semi-records. Where a part of the string defines the total length for the information within a record, and sub-parts of that field can refer to other fields in that record.