Time-sharing system evolution

As of the middle 1960's, all computer systems were single program systems known as "batch processing": One job in and one job out. Some very large and powerful mainframes had been built, and most all of them were powered by vacuum tubes. The invention of the solid-state transistor by Texas Instruments in the early 1960s made possible the development of mainframes with many times the computing power of the vacuum tube computers. It was also the first step toward miniaturization of computers, a trend that has continued ever since. Solid-state computers were much smaller, generated far less heat, and were many times more reliable than their vacuum tube predecessors. Maintenance also became several orders of magnitude less expensive and immensely more simple to identify the problem and repair it.

As solid-state computers developed, even the largest and most powerful ones had to be operated in batch processing mode. Thus developed the need for multitasking, multiprogramming operating systems that allowed multiple users to run their individual programs at seemingly the same time as several dozen other users using teletypewriter terminals that operated at 10-characters per second over ordinary voice-grade telephone lines.

The following tables provide links to major early time-sharing operating systems, showing their subsequent evolution.

• To avoid listing every multi-user system ever built (including virtually every system in use today), the goal here is to list:
  • Influential systems
  • Built between 1960-1990
• For clarity, the direct successors of these systems are also included, as well as several important interactive systems that, although not multi-user, had an impact on user interface design.
• Family relationships have been shown where practical, to help structure the tables. However, the intent is to provide a simple, compact set of links to all these systems – not to illustrate every relationship. See each system's main article for further details; all had many direct and indirect influences.

About the term time-sharing:

In the 1960s, time-sharing was a new concept, a departure from the batch processing approach previously used with computers. The idea of an individual user being able to initiate a computer job at a particular time, and to see results immediately, was regarded as strange – and probably wasteful. Computers were very expensive, and individual users had to conform to the computer's schedule, not vice versa. Time-sharing systems were thus a major change and, for a generation, represented a distinct category of operating system. Famous political battles were waged at IBM and elsewhere over this issue. Today, of course, virtually all operating systems are time-sharing systems.

Time-sharing system families

See details and additional systems in the table below. Relationships shown here are for the purpose of grouping entries and do not reflect all influences (e.g., OS/2 was more influenced by VAX/VMS than by MS-DOS, but its legacy is as an x86 platform).

System descriptions and relationships

See also

• History of CP/CMS has many period details and sources.
• Timeline of operating systems

References