Some of the very first programs written for the first timesharing computers
in the early 1960's were designed to support real-time chat -- it was almost
immediately recognized as a natural use of this new technology to support multi-user
human communication. Many different chat programs
were written by various people in university and research organizations.
These early chat programs were
limited to use by people connected to the same computer. The first computer
terminals were electronic typewriters without a computer screen, and usually situated in the same room as the computer. As technology improved the terminals could be distributed throughout a building, and even among
different buildings that were close together, but they couldn't be
used by geographically distributed elements until the development of networks.
The very first chat programs could only transmit messages between two
users at a time. Later enhancements supported broadcast of messages
to more than one user, but only if the sender specified each address. One of the best known of these simple chat programs was called "talk", a utility supported by many Unix
computers on the ARPANET in the 1970's. Participants in a talk conversation had to be logged in directly or over the
network to the same computer. A user could ask the talk system for a list of people currently
on the system, and then send a message to any of those users. An incoming message would
pop up on the user's terminal in the middle of whatever they were doing, which
could be handy, or annoying, depending on the circumstances.
Although rudimentary
and limited compared to today's sophisticated chat systems, these early programs
were a brand new use of computer technology and a huge step forward for the
time, and seemed almost as magical to the people who created and used them
as the Internet does
for us today.