Corporations, organizations and other entities use email as a communication device. Every day, billions of emails are sent by people and computers to recipients in domains all over the world (some valid and some not). With the ability to transmit documents, spreadsheets, slide decks and other attachments, email is our most used, if not our most powerful tool for interaction and collaboration. When the systems and networks are operating properly, email conveniently blends the advantages of written correspondence with delivery times that border on instantaneous.

Information workers need to communicate and collaborate, and organizations are continuously looking for more efficient and economical ways to facilitate this. The de facto method is email. In just a short time, email has become a necessary component of business. For many businesses, email is mission-critical. At least one study has found that the overwhelming majority of business workers consider email more necessary than the telephone, and that the loss of email would present a greater hardship than the loss of the phone service.1

The success of many companies is tied to their information technology infrastructure and how it is used. In business today it is critical for information to get to the right people at the right time so that informed decisions can be made quickly. Corporate information can generally be put into one of two categories: that which is used to support the purpose of the business, such as sales and marketing data, and that which is used to support the business’ infrastructure, such as payroll and accounting data.

Today, a significant percentage of corporate information flows through corporate email systems, and over the Internet. It is generally accepted that in order to do this, messaging and collaboration systems must be:

  • Reliable—the system must continuously work as expected, without loss of service or failure of core functionality.
  • Secure—the system must support secure messaging features, such as encryption of transmitted data, verification and authentication of messages and message origins, and privacy features that protect confidential content. In addition, the system must be secured against tampering, unauthorized access, unauthorized use of services and other maliciousness.
  • Interoperable—the system must interoperate with all other email systems on the Internet and inside an organization.
  • Accessible—end-users should be able to access messaging and collaboration services regardless of their location, platform, or available bandwidth.
  • Standards-based—in order to interoperate and be accessible from any location and any platform, the system must use public protocols for communication and transmission of data.
  • Economical—the system must be inexpensive and cost-effective, with the lowest total cost of ownership (TCO) possible.

But how did email get its start, and where did it come from?

A Brief History of BBN

You may not have heard of a company called Bolt Beranek and Newman (BBN) or a BBN engineer named Ray Tomlinson. Nor might you be aware of their impact on business messaging and collaboration. In 1948, two MIT professors, Richard Bolt and Leo Beranek started a small consulting firm in the field of acoustics. A year later, they hired one of Bolt’s former students—Robert Newmann—and founded BBN. Between 1948 and 1997, BBN was a small, independent R&D company. They worked on various ways to improve sound and sound quality, reduce noise pollution from aircraft, and establish standards for noise metrics.

GTE purchased BBN in 1997, and in 2000, GTE merged with Bell Atlantic to create a new company—Verizon. Certainly Verizon has had an impact on business messaging with its cell phone, wireless and broadband services. But the real impact on business messaging; that which brought business messaging to the masses, happened more than 30 years ago.

Although they began their work in the field of acoustics, they later expanded their research into the most sophisticated areas of computer science. In 1960, BBN developed a priority-interrupt system for a Digital Equipment Corporation (DEC) computer called the PDP-1. For the first time, a computer could multi-task: processes could consume computing resources based on priority, as scheduled by the system’s kernel. After demonstrating the first time-sharing computer system in 1962, BBN designed the world’s first computer modem (the Data-Dial) in 1963.

Throughout the next several years, BBN worked to create new computer languages, refine calculations and formulas for predicting structural and noise vibrations, and even created the Grafacon—a tablet-based device that digitized graphical data for input into a computer. The Grafacon is literally the first ancestor of the Tablet PC, and other pen-based computing devices.

Building ARPANET

Taken individually, these accomplishments have had major impacts on the human race, but they pale in comparison to work BBN began in 1968, when the government awarded BBN a contract to create the Advanced Research Projects Agency Network (ARPANET), a network that would connect military and educational institutions using packet-switching protocols. ARPANET would later undergo several name changes, but ultimately settled on the one we all use today—the Internet.

In less than a year, they had established a network with four sites, with one node at each site. Not that it was a reliable network, however. The first transmission of data was successfully from a node in Stanford to a node in UCLA on August 29, 1969. However, when that same data was sent to a third node, it crashed the third node. Once the network was stabilized, additional nodes were added, and the network was primarily used for transferring files and for logging on to remote systems using technology called Telnet that is still in use today.

One of the scientists at BBN who worked on the ARPANET project was Ray Tomlinson. Tomlinson and his team worked on a project called CYPNET. CYPNET research and development concentrated on a primitive file transfer protocol that was the forbearer to today’s File Transfer Protocol (FTP). One of Ray’s other projects was an equally primitive method of sending messages called SNDMSG and its companion RDMAIL. These components worked together to allow users on the same machine to send messages to each other. A user creates a text file which was then delivered to one or more mailboxes on the same machine. On this system, a mailbox was simply a text file to which text was appended. The name of the mailbox was nothing more than the name of the text file. In addition, data could be added, but not deleted or changed. Of course, the bigger limitation was that all users had to use the same computer; SNDMSG could not send messages to a user on a different computer.

Tomlinson changed all of this with a series of ideas. First, he realized that the file sharing technology in CYPNET very closely met the functionality needed to transfer messages between computers. It dawned on him that just as computers could exchange files, they could also exchange messages. Combining a protocol that can transfer files with one that can transfer messages was described by Tomlinson as a “no-brainer” and as a “neat idea.”2 Once he realized this, he modified the file sharing code from CYPNET so that it could append to a file. This was a simple and quick change. After that, Tomlinson made another quick decision that would impact every Exchange administrator and user, and every Internet email user that was to come.

The problem was location; or put slightly different, although email from one user to another user on the same computer had been used for some time, network email—email between users on different computers—did not exist. If the user did not have a local mailbox, and instead had their mailbox located on another computer, you could not send them a message. A mechanism would be needed so that a message could be sent to a remote user at a remote computer. As it turned out, the key word here was at or in computer symbols, @. According to Tomlinson, after less than a minute of thought on the problem, he chose the @ symbol to separate the username (also called the mailbox name) from the computer name (also called host name) because it enabled him to indicate that the user was at some other host and not local. Effectively, Tomlinson created the format of a user’s email address by using a mailbox name, an @ symbol, and the name of the host containing the mailbox.

After making some minor code modifications in late 1971, Tomlinson did some tests and sent the world’s first email between two different computers that sat side-by-side.3 Legend has that the first email ever sent had something like “QWERTYUIOP” or possibly “TESTING 1 2 3 4” in its body, however, nothing was ever saved or recorded for history or posterity, and Tomlinson himself no longer remembers exactly what it said. Nonetheless, history was made that would forever change human communication. Tomlinson had created networked email—messaging for the masses.

OK, maybe not yet for the masses. Network email had no features other than the ability to send a text message to another user on another computer. There were no attachments, no rich-text, no spell checking, no file or folders, and no ability to reply to or forward messages. There was also no electronic address book or directory. Instead, email addresses were listed on printed paper directories and distributed to email users.

Email at the Speed of Light

By 1972, the first email management program had been written and distributed. RD (no fancy marketing names here), by programmer Larry Roberts at ARPANET, enabled users to list incoming messages, store them in a file-system like environment, and forward and reply to messages. RD would later inspire others to create email management programs (now generically referred to as email clients), including some that are still in use today, such as Eudora.

Within two years of Tomlinson’s first email to himself, 75% of all traffic on ARPANET was email. By 1978, ARPANET was providing email services to a large number of government institutions and universities, as well as a few private businesses, such as BBN and DEC. Although it wasn’t called it at the time, the first unsolicited commercial email (UCE)—also known as spam—was sent by a DEC employee on May 3, 1978, in violation of the acceptable use policy for ARPANET in effect at the time. It is not truly known if this spam was accidental or intentional, but one thing is clear: the baby Internet was already a miniature model of things to come—email is hugely popular among those with access to it, and the system is wide-open for abuse.

Between 1971 and 1976, Request for Comments proposals (RFCs) for standards-based mailbox and messaging protocols had been published. The very first RFC for a mailbox protocol—published on July 20, 1971—was RFC 196, "Mail Box Protocol," by Richard W. Watson. Later came RFC 561, "Standardizing Network Mail Headers", and RFC 680, "Message Transmission Protocol.” These were later rendered obsolete by RFC 733, “Standard for the Format of ARPA Network Text Messages.”

In November 1975, Jon Postel wrote RFC 706, "On the Junk Mail Problem.” In it, Postel noted that ARPANET had no mechanism for that allowed a host to selectively refuse messages. Each host that accepted email had to accept every message it was sent. There was no way to filter out email; it was simply all or nothing (a spammer’s paradise if spammers had existed back then). On September 1, 1980, Postel and fellow researcher S. Sluizer published RFC 772, "Mail Transfer Protocol." This in turn led to a series of RFCs which eventually led to RFCs 821 and 822, which initially defined one of the Internet email protocols in use today—the Simple Message Transfer Protocol (SMTP). Those RFCs were superseded by RFCs 2821 and 2822, respectively, which were then obsoleted by RFC 5321 in October 2008.  By the way, if you’re going to administer Exchange, you should read these RFCs, which you can do at http://www.ietf.org.

Conclusion

So that’s where email came from.  Email has come a long way since Ray Tomlinson sent the world’s first message between two different computers in late 1971. Now more than forty years later, billions of email messages are sent around the globe every day.  And some of them are even worth reading!

        

1 April 22, 2013: META Group, Inc. (Senior Executive Panel)

2 http://www.rit.edu

3 http://openmap.bbn.com/~tomlinso/ray/ka10.html