Watts Humphrey
November 2, 2010
There was a series of
television commercials in the 1960s that began like this
Young Adult No. 1 "Certs is a candy mint!"
Young Adult No. 2 "Certs is a breath mint!"
Announcer "Stop! You're both right! New Certs is two mints in one!"
Computing faces a similar identity problem. Is computing a science, or is it a field of engineering? Mixed signals abound, since a Google search will reveal many university departments of "Electrical Engineering and Computer Science" and "Computer Science and Engineering." There's no question that modern computing had its origins in neither science nor engineering, but in mathematics. After a time, the physicists embraced computing, and it inherited some aspects of science. It's obvious that without some very capable electrical engineers, computing would not have developed as quickly as it has; but that's all about hardware. Software is something different, isn't it? Watts S. Humphrey, a proponent of software engineering, died on October 28, 2010, at age 83.[1-3]
As it was for many of his generation, Humphrey's career was delayed by World War II, in which he enlisted at age 17 in the Navy. Fortunately, the war ended before he completed his training, and when his stint in the Navy was over, he obtained a bachelor's degree in physics from the University of Chicago and a master's degree in physics from the Illinois Institute of Technology. After that, and what set him apart from other physicists in that era, was his completing an MBA from the University of Chicago and not pursuing a Ph.D.
With academic credentials in hand, Humphrey started his career in 1953 with Sylvania, then a prominent electronics company, in Boston. In those days, Sylvania was interested in computing, as its participation in the development of COBOL will attest. Sylvania sent him to attend a computer course at MIT taught by Maurice Wilkes, himself. Returning to Sylvania, he became manager of a team to design and build a cryptographic system. This device was quite complex in that day's technology, having 5,000 subminiature vacuum tubes, and it needed to withstand military field conditions.
Having become interested in computers, Humphrey asked about computer courses he might take at nearby Northeastern University. Not surprisingly, none were offered, but the university convinced him to teach one. To prepare to teach a course on computer design, he learned as much about computing as he could from the MIT and Harvard libraries. Eventually, he wrote a book on the topic.
In 1959, possibly at the urging of his wife,[4] he went to work for IBM, initially as a computer designer and architect, finally becoming vice president of technical development to whom 4,000 software engineers reported. In 1986, at age 60, Humphrey accepted a position at Carnegie Mellon University as director of the Software Process Program in the Software Engineering Institute. His objective was to change software development from the traditional coding/testing model, to a planned development approach. The fruit of this effort was the Capability Maturity Model (CMM), which was based on Humphrey's 1989 book, "Managing the Software Process." The CMM is formally specified in the 1993 SEI Technical Report CMU/SEI-93-TR-024 ESC-TR-93-177, "Capability Maturity Model SM for Software, Version 1.1." As explained by Larry Druffel, SEI director and CEO from 1986 to 1996, "We all understood the importance of things like version control, configuration management and methodology, but I don't think anyone knew how to put those into a transferable form."[3]
Business Week called Humphrey the "father of software quality," and he received the National Medal of Technology in 2005. Humphrey was a named inventor on five US patents (see figure), and he was the author of twelve books and numerous magazine and journal articles. That's quite a comeback for a boy who failed first grade because of dyslexia.
Fig. 3 from US Patent No. 3,094,610 by Watts Humphrey, et al. [6]
I'll conclude this article with Humphrey's comment on Open Source Software that appears in the transcript of his 2009 interview with the Computer History Museum.[4]
"... it's become a very interesting niche, which I think is likely to really cause Microsoft enormous trouble. I think the Microsofts, maybe the Oracles -- people that are building the widely used, general-purpose programs I think are quite exposed. And the reason I think they're exposed is that the companies that are building them are motivated to get continuing revenue... So they keep trying to come up with these added little gimmicks and features that will make people want to buy it. And they're not terribly successful with that..."
Oh, one more thing. The earlier reference to Certs is actually more relevant to this blog than you might believe. Internet security experts use the acronym CERT to mean "Computer Emergency Readiness Team." There are two such organizations in the US; namely, the CERT Coordination Center, which is at Humphrey's Software Engineering Institute (SEI) at Carnegie Mellon University, and the United States Computer Emergency Readiness Team which is part of Homeland Security.
References:
- Ann Belser, "Obituary: Watts S. Humphrey / Software engineer with five patents, national medal, July 4, 1923 - Oct. 28, 2010," Pittsburgh Post-Gazette, October 30, 2010.
- Dr. Dobb's Journal, "Software Process Pioneer Watts Humphrey Dies," October 28, 2010.
- National Medal of Technology Winner Watts Humphrey, 1927-2010, Software Engineering Institute, October 28, 2010.
- Grady Booch, Interviewer, "Oral History of Watts Humphrey (June 17-22, 2009)," Computer History Museum Reference number: X5584.2010.
- Susan Kushner, "Writing the Book on Process Improvement: An Interview with Watts Humphrey," News at SEI, March 1, 2005.
- Watts S. Humphrey, Jr., John Terzian and Franz Bosch, "Electronic Computers," U.S. Patent No. 3,094,610 (June, 1963).