NJIT Research in 1970s Became Vital Parts of Today's Social Media Recipe

Long before social networks, instant messengers, web forums, Internet Relay Chat, AOL, Compuserve, and dial-up bulletin board systems, there was EIES – Electronic Information Exchange System, pronounced like the word eyes – developed here at NJIT in the 1970s.

The main purpose of EIES was to be a rudimentary version of online classes and information sharing. Faculty and students typed back and forth in scheduled group conferences. Users made public and sometimes anonymous posts. They also looked up contact information in a directory feature, read news and sent each other private messages. None of that may impress 21st-century students, but it was groundbreaking at the time.

EIES was not the first online service. There were predecessors such as EMISARI (Emergency Management Information System and Reference Index), developed for the U.S. government; PLATO (Programmed Logic for Automatic Teaching Operations), built as an online education system at the University of Illinois; and Community Memory, built by counterculture-inspired hackers in San Francisco as a tool for the general public. EIES went online in 1976 with influences from such systems. Pundits at the time referred to these services as computer-mediated conferencing.

EIES was the brainchild of professor Murray Turoff. He worked at IBM on the 1620 computer, which also happened to be the first model of computer purchased by NJIT in the early 1960s. He developed experimental decision-making software for a U.S. defense organization, and then made EMISARI – for a White House department before coming to NJIT in 1974. His research partner was Starr Roxanne Hiltz, a sociologist at the former Upsala College in nearby East Orange, who was passionate about collaborative virtual education. Her role was to study acceptance and impacts of EIES. They are both now NJIT distinguished professors emeritus.

Turoff was the lead designer and Hiltz developed the evaluation proposals funded by the National Science Foundation. One of her articles on that subject led to an infamous rejection letter from the American Sociological Association, when the editors told her that doing such things by computer was outside the domain of sociology. "In a fit of pique, I burned the letter," she said, "and decided to publish in computer science journals from then on." She and Turoff co-authored an influential book in 1978 called The Network Nation: Human Communication via Computer. Hiltz joined NJIT in 1985, and their professional partnership led to marriage the next year.

The first generation of the EIES server ran on an Interdata 7/32 minicomputer, as "mini" was a relative term used in the 1960s-1970s to indicate systems less powerful than mainframes and merely the size of a desk or refrigerator. Interdata was based in Oceanport, N.J. and often worked with government customers. Most users interacted with the server through terminals and teletype machines, although it's known that ambitious hackers built custom interfaces for microcomputers. National Science Foundation grants for EIES during this time totaled $1 million, or about $3.7 million adjusted for inflation in 2019.

Turoff said that when he taught an online course and wanted students to have a discussion, he would post something controversial using a pen name. "This always broke the bottleneck so other students would start to disagree with me," he recalled. "This made things very interesting and actually encouraged the other students to disagree with the teacher."

This made things very interesting and actually encouraged the other students to disagree with the teacher.

There were two versions of the second-generation EIES server. One primary version, EIES-2, ran on Unix platforms. The other was tailorable EIES – TEIES – built with IBM grant money. It also had a personal computer interface called PTEIES. IBM's intention was to have a service that could be customized for corporate applications. Quinnipiac University historian Ramesh Subramanian said the projects in total employed 20-30 programmers and administrators at NJIT.

Hiltz said she received more than $2 million in grants during the 1980s-1990s to build and evaluate an online learning application inside EIES called Virtual Classroom. Grants came from the Corporation for Public Broadcasting and the Alfred P. Sloan Foundation.

Grant money started drying up in the late 1990s as the web caught on. NJIT decided to shut down EIES in 2000. It didn't go smoothly. "Despite the obsolescence of VC/EIES we are having difficulty finding a replacement product with its key features (gradebook, quiz, search, and roles) and scalability (1,000 simultaneous people using 6GB data store and <15 second average response)," officials stated at the time.

The announcement continued, "Currently three alternatives to VC/EIES are in production, and which one will be used for a particular course is at the discretion of the instructor of the course. Other products, were/are being evaluated ... We will continue to consider and evaluate other replacement systems and technologies as they become available. There is no current effort at NJIT to develop a replacement in-house."

Almost 20 years later, there remains nostalgia for EIES. Turoff has his technical papers on an archived NJIT website. The NJIT library has its own archive including a user guide from 1977. Fans compile information and stories through a wiki called EIES Legacy.

Some EIES alumni would like to see a modern emulator on the web. One source of assistance may be Kevin Walsh, who was instrumental to the shutdown in 2000 and still works at NJIT. He has EIES-2 disks in his office which may or may not still contain working source code written partially in mainstream C and partially in Smalltalk, which was an educational language designed by the influential Xerox Palo Alto Research Center. He wants to give those to the university library, where officials could then decide whether to open-source the code. Outside of the university, various hobbyists and historians in recent years built modern emulators for PLATO, so there is precedent. But it's unlikely the first-generation EIES code, programmed in Fortran, will ever turn up. "A lot of people have looked for that over the years and not found it," Walsh said.

Another person who would like to know about the original EIES code is former NJIT student employee Ken Schreihofer, who now lives and works in Baltimore. Schreihofer literally grew up with the system, as his mother worked at NJIT and announced his birth on an EIES channel. A computer professional himself, "We did do a lot of searching, everyone who's part of that community did some checking. There are some design documents but as far as I know there is no downloaded source copy," he said. "A lot of people involved wish at some point in time they would have printed or downloaded or somehow saved that history."

Editor's note: this is part two of a three-part series about the history of computing at NCE/NJIT. Part one is here. Part three is here.