Clarence “Skip” Ellis - first African American to receive a PhD in computer science

Clarence “Skip” Ellis was the first African American to receive a PhD in computer science. Ellis also helped create the idea of clicking on a graphic to start a computer program or to issue a command, rather than typing in lines of computer code which became a defining feature of the Apple computer and was later incorporated into the Microsoft Windows operating system. 

A native of Chicago, Ellis first became interested in and acquainted with computers as a teenager. He landed a job as a night watchman at a local insurance company that had recently purchased an expensive computer. His employer was Dover Corporation, a manufacturing company, where it was his job to guard their large mainframe computers. The computers, which took up several rooms and had over 2400 vacuum tubes, needed to be guarded because they were displayed in picture windows for visitors to see. Computers were a novelty then, because the first commercially successful computer, the UNIVAC, or UNIVersal Automatic Computer, was completed just a few years early in 1951. During his free time, Ellis read the computer’s manuals and became a self-taught expert. One day, an important project had to be run, but there were no more punch cards (which were the only means of input at the time). Using what he had learned from the manuals, Ellis showed the company’s employees how to reuse old punch cards.

Ellis attended Beloit College in Wisconsin, where he received a bachelor’s degree in mathematics and physics. Ellis was one of very few African-American students at Beloit then. At Beloit, he helped set up the school’s first computer laboratory, a place where he spent many hours developing his interest in computers. At Illinois, Ellis continued to work on computer systems, in particular the hardware, software, and applications of ILLIAC IV supercomputer.  Ellis wrote his Ph.D. research dissertation in 1969 on the subject of “Probabilistic Languages and Automata.” He was interested in the probabilities of successful results for computer languages. His intention was “to take the first step forward toward developing a quantitative tool for analyzing programming languages and their translators.”

Over the course of his career, he worked at Bell Telephone Laboratories, IBM, Xerox, Microelectronics and Computer Technology Corporation, Los Alamos Scientific Labs, and Argonne National Lab. Ellis was at the Palo Alto Research Center (PARC) from 1976 to 1984. There he headed a group that invented and developed Officetalk, the first office system to use icons and Ethernet to allow people to collaborate from a distance. Ellis continued to work in this area, and is considered one of the pioneers of the field of operational transformation, which examines functionality in collaborative systems. Operational transformation is now found in a host of computer applications, including Apache Wave and Google Docs.

He held teaching positions at Stanford University, the University of Texas, MIT, and Stevens Institute of Technology. He also taught in Taiwan under an AFIPS teaching fellowship. Ellis joined the University of Colorado at Boulder in 1992 and retired in 2010. He was an early leader in that university’s research on human-centered computing. 

1. Ellis, Clarence A. Participatory Design in Groupware Systems. Proceedings of the Groupware'96 Conference, August 1996. 44-52.

2. Ellis, Clarence A. A Workflow Architecture to Support Dynamic Change. Workshop on Distributed Systems, Multimedia, and Infrastrtucture, March 1995. 23-30.

3. Ellis, Clarence A. Goal Based Workflow Systems. International Journal of Collaborative Computing, 1,no.1, 1994. pp.61-86.

4. Ellis, Clarence A. Technological Challenges in Groupware. Proceedings of the Third International Workshop on Intelligent and Cooperative Information Systems, April 1992. pp.121-132.

5. Ellis, Clarence A.; S. Gibbs; G. Rein Groupware: The Research and Development Issues. Communications of the ACM, 34,1, 1991.

6. Ellis, Clarence A.; G. Rein rIBIS: A Real Time Group Hypertext System. International Journal of Man Machine Systems, 34 (1991), 21-51.

7. Yeh, Show-Way; Ellis, Clarence A.; Ege, Aral; Korth, Henry F. Performance analysis of two concurrency control schemes for design environments. (English) [J] Inf. Sci. 49, No.1-3, 3-33 (1989).

8. Ellis, Clarence A. Probabilistic models of computer deadlock. Inform. Sci. 12 (1977), no. 1, 43--60. 68B20 (68D25 68J10)

9. Ellis, Clarence A. Analysis of some abstract measures of protection in computer systems. Internat. J. Comput. Inform. Sci. 7 (1978), no. 3, 219--251. 68B20

10. Ellis, Clarence A. Optimal scheduling of homogeneous job systems. Information Sci. 9 (1975), no. 4, 323--358.

11. Ellis, Clarence A. The halting problem for probabilistic context-free generators. J. Assoc. Comput. Mach. 19 (1972), 396--399.

12. Ellis, Clarence A. Probabilistic tree automata. Information and Control 19 (1971), 401--416.