A PLATO “dataset” is a file of records kept in the permanent (magnetic disk) storage. You can write some data out to the 5th record of the dataset, then get it back months later simply by reading back the 5th record of that dataset. Each record is made up of many words, and the record word size is specified at the time the dataset is created. (Currently the minimum record size is 64 words.) One record might, for example, hold exam scores for a particular student.

In order to perform operations on a dataset, you first must execute a -dataset- command to tell PLATO which of your datasets you are going to be working on at the moment. You can then execute any number of -dataout- commands to send data out to the dataset, and any number of -datain- commands to read such information back. You can use a -reservecommand to reserve specific records, similar to using a “reserve common”. You must use a -release- command to permit others again to manipulate those records. (For details, see the PLATO on-line “aids”.)

When manipulating a data base it is often necessary to sort a list of items into alphabetic or numeric order. The -sort- (numeric) and -sorta- (alphabetic) commands will transform a disordered list into a sorted list. These commands will also sort an associated list of items at the same time. For example, you might have student names in one part of a common, and corresponding grades in another part of the common. You could use a -sorta- command to place the names in alphabetical order, and at the same time you could have the -sorta- command similarly re-order the grades to correspond with the altered order of the students. (See the PLATO on-line “aids” for details.)