A database is described as a permanent organised structure of data. This means that’s most organised data can be described in some form as a database.
Consider a wall calendar appointments or meetings are stored by writing into the space is identified for each day of the month: as the days of the month are organised into a structure and the details of the appointments are written into these spaces, this can be considered as both organised and permanent and therefore is a form of paper database.
Flat File Databases
When a digital database has data which is stored on a single table it is identified have a flat-file. Although a flat-file may initially appear much easier to access using program code, finding data and keeping it correct over a number of records becomes very difficult as the size of the database grows.
For example, if a teacher kept on record of the feedback given to each student after every class in a flat file they would need to write out the student’s name each time a new piece of feedback was added to the file. this would mean but if the teacher spell the student’s name incorrectly for one entry it would not be shown when the data was filtered by the student’s name.
Relational databases try to negate this problem by saving data in multiple tables with records as are linked together. This attempts to remove duplicated data and has the added benefit that’s the change in one record (for example the student’s name) would be reflected when the data for that record Hey is shown across the whole database.
In the example above, we saw that one student can have many pieces of feedback. We can represent this in a diagram called an entity-relationship diagram which shows each table as an entity in the database and uses a format called crow’s foot notation to show the degree of the relationship between the two entities.