Tomlinson Sends the First E-Mail

Date November-December, 1971

Ray Tomlinson sent the first e-mail between two side-by-side computers located in a Bolt Beranek and Newman laboratory in Cambridge, Massachusetts, signaling the beginning of a new era in computer network communications.

Locale Cambridge, Massachusetts

Key Figures

• Ray Tomlinson (b. 1941), engineer for Bolt Beranek and Newman
• Larry Roberts (b. 1937), director of the Defense Advanced Research Projects Agency

Summary of Event

By the end of the twentieth century, e-mail had become almost as ubiquitous as the telephone. In the early years of the twenty-first century, more than half of all Americans were using e-mail at least half an hour a day, and it was apparent that e-mail had become a part of everyday life in the United States. However, when Ray Tomlinson invented network e-mail in 1971, he had no idea that his invention would reach such heights.

In 1965, Tomlinson—a graduate of Rensselaer Polytechnic Institute (B.S. in electrical engineering, 1963) and the Massachusetts Institute of Technology (M.A. in electrical engineering, 1965)—had not even considered the possibility of e-mail when he was asked by his employer, Bolt Beranek and Newman (now known as BBN Technologies), to work on the time-sharing computer operating system called TENEX. While working with SNDMSG, a program that allowed users to append a file called “mailbox” without overwriting what had previously been written, Tomlinson began to consider the possibility of sending electronic messages between networked computers. Since the SNDMSG program allowed a user to send only messages to other common users of a given computer system, a new program had to be created.

Tomlinson began developing his new program after examining the components of the CYPNET program, which allowed users to send and receive files from computers connected to the Advanced Research Projects Agency Network (ARPANET), a government-sponsored computer network that linked fifteen computers in laboratories across the country. Tomlinson believed that one could alter the SNDMSG program to function like CYPNET, thereby allowing users to send and receive network messages. With this insight, he modified the SNDMSG program and created a companion program called READMAIL, allowing ARPANET users to send and receive messages to and from other ARPANET computers. In order to test these two programs, Tomlinson, sometime in November or December of 1971, sent a test message between two computers that sat only a few feet apart. The message is considered to be the first e-mail ever sent. At the time, Tomlinson was unaware of the impact it would have.

The first e-mail was composed on a BBN-TENEXA (BBNA), a Digital Equipment Corporation (DEC) KA-10 computer with sixty-four thousand bytes (64K) of core memory, which is approximately 288K by early twenty-first century standards. This computer’s limited capabilities meant that sending and receiving e-mail represented a very arduous task. Consequently, the first test e-mail consisted of a meaningless line that was so forgetable that Tomlinson later had only a faint recollection of what it may have been. Some believe that the first message was “QWERTYUIOP,” which Tomlinson has stated may or may not be the case. The first substantive e-mail announced the e-mail’s existence to the world and was sent to fellow ARPANET users to explain how to use the new program and the purpose of the “at” sign (@), which separated the log-in and location names.

Even though the at sign had existed for centuries, Tomlinson is credited with utilizing it for e-mail. Tomlinson used the typographic symbol because it was traditionally used to indicate a unit price, such as 10 at $1.50. Additionally, the symbol did not have any significance for TENEX editors, so it was easy to annex for Tomlinson’s use. Unfortunately, some editors using MULTICS (Multiplexed Information and Computing Service), a rival time-sharing operating system, complained that the symbol was used as the line-erase character in many of their programs, a point of contention that was never resolved.

Because Tomlinson was using a Model 33 Teletype keyboard, the at sign did not appear on the same key as the 2, where it came to be placed commonly on computer keyboards. Rather, it appeared to the right of the P, before the bracket keys. Tomlinson has been quick to point out that the at sign existed long before he ascribed it to e-mail addresses, but he does believe that he could have contributed to the symbol’s revitalization and eventual emergence in networked computers. By the early twenty-first century, the sign was being used in more than 125 million electronic addresses and had become an integral part of the Internet.

Significance

E-mail was popularized by Larry Roberts, the director of the Defense Advanced Research Projects Agency (DARPA)—the government agency that oversaw the ARPANET. Roberts, a fabled administrator and scientist at the time, decided to deliver all DARPA communications by e-mail. With this decision, all scientists and academics associated with the ARPANET had to become familiar with Tomlinson’s invention. Eventually more than 75 percent of all ARPANET traffic was e-mail, a result developers attribute more to an unanticipated natural phenomenon than to the deliberate development of a new technology.

Even though he has been credited as the inventor of e-mail, Tomlinson has noted the difference between interuser and network e-mail. Interuser e-mail programs, such as SNDMSG, had been around since the 1960’s. These programs were the first e-mail programs in existence. However, the kind of e-mail that became widely used by the end of the twentieth century was in fact network e-mail—that is, it allowed users to send messages between two computers. Tomlinson has noted that he believes Richard Watson (later at the Lawrence Livermore National Laboratory) was the first person to develop the idea for network e-mail. Watson’s RFC 196 program sent messages to numeric mailboxes, which then printed out the message. Tomlinson’s program, developed in November and December of 1971, differed in two main respects. First, “numeric mailboxes” became a definite log-in and host name separated by an at sign, allowing users to have more descriptive addresses. Second, instead of a printer output, Tomlinson envisioned simply exchanging e-mails between computers, with no hard copy produced.

Despite Tomlinson’s humbleness, his accomplishments have not gone unnoticed. In 2000, he received the George R. Stibitz Computer Pioneer Award from the American Computer Museum. This award, which is given to “living pioneers of the computer, communications, and the information age,” was named in honor of Dr. George R. Stibitz, who in 1937 pioneered the use of relays for digital computation at Bell Laboratories. Additionally, in 2001 Tomlinson received a Webby Award from the International Academy of Digital Arts and Sciences, honoring his lifetime achievement in Internet development.

Bibliography

Abbate, Jane. Inventing the Internet. Cambridge, Mass.: MIT Press, 2000. This narrative account focuses on the social and cultural forces surrounding the development of the Internet. Includes detailed accounts of key players and events.

Gillies, James, and Robert Cailliau. How the Web Was Born: The Story of the World Wide Web. New York: Oxford University Press, 2000. A mix of theory and facts that provide a detailed outline of the development of the World Wide Web. Includes time lines and biographical entries.

Hafner, Katie. Where Wizards Stay Up Late: The Origins of the Internet. New York: Simon & Schuster, 1996. Written in a journalistic fashion, this book outlines the origins of the Internet, beginning in February of 1966. Detailed biographies of key figures are included.