Obituaries: David M. Young, Jr.March 21, 2009
David M. Young, Jr. (1923–2008), at the Elliptic Problem Solvers Conference, Santa Fe, New Mexico, summer 1980.
David M. Young, Jr., an emeritus professor of mathematics and computer sciences at the University of Texas at Austin, died on December 21, at the age of 85. He was an international pioneer in numerical analysis whose career corresponds exactly with the first fifty years of the modern scientific computer era.
David grew up in Quincy, Massachusetts, the son of David Monaghan Young and Madge Tooker Young. He and Mildred Acker met in the eighth grade; they were married on October 9, 1949, the beginning of their fifty-nine years together.
From an early age, David was a skillful and avid tennis player. In the fall of 1941, David enrolled as an undergraduate, on a full scholarship, at the Webb Institute of Naval Architecture in New York City. On discovering that the president of the institute, an admiral in the Navy, shared his love of tennis, David would frequently volunteer to play tennis with him on Sunday mornings.
Soon after the attack on Pearl Harbor, David and most of his classmates at Webb enlisted in the Navy. Because the Navy needed officers, the students were allowed to finish their degrees, but on an accelerated basis. David graduated from Webb in 1944, and was sent to midshipman's school at Cornell University for naval training. He was then assigned to the David Taylor Model Basin, in Washington, DC, as an assistant naval architect. (He would later become a lieutenant commander in the U.S. Naval Reserves.)
David received a master's degree in mathematics from Harvard University in 1947 and went on to study for a PhD, under the direction of Garrett Birkhoff. Sir Richard Southwell, visiting Birkhoff early in David's graduate career, had discouraging words about David's research on mathematical relaxation methods: "Any attempt to mechanize relaxation methods would be a waste of time!" Nevertheless, David continued with the work. His propensity for making numerical errors was so great, he reasoned, that he would never be able to solve a significant problem without discovering a mathematical procedure that could be used on the new computing machines.
David finished his PhD in June 1950. Given the technology of the time, almost all of the mathematical symbols and equations in the final three copies of the thesis were beautifully handwritten, in ink, by David's wife Mildred (who was a nurse). Birkhoff considered David's thesis a classical piece of mathematical research. It established the mathematical framework for the successive overrelaxation (SOR) method, among other things. In the summer of 1951, Richard Varga, a graduate student at Harvard, shared an office with David, who was by then an instructor in mathematics. Varga found the thesis to be a wonderful mixture of function theory and matrix theory.
A page from David Young's dissertation.
Getting a journal to publish a paper based on the dissertation posed considerable difficulty. Journals that would even consider numerical analysis papers were scarce! Moreover, David had a hard time condensing his dissertation; it was painful for him to throw out results that he believed to be interesting. One referee, Hilda Geiringer, wrote, "This paper is far from ready for publication." It would take four years and many revisions before the paper was published. Acknowledging the help of his adviser, David said: "Without Garrett Birkhoff's continued interest and encouragement, this paper would never have seen the light of day!" (Additional information on the rich history of iterative methods can be found at http://history.siam.org/.)
In 1951, then working at the computing laboratory at Aberdeen Proving Ground, David took the train to Philadelphia one evening to attend the first organizational meeting of SIAM. He would be an active member of SIAM throughout his career. At this time, David and Mario Juncosa, of the RAND Corporation, often worked together. An early contribution to SIAM arose from an encounter on the New York subway, where David and Mario happened to see Ed Block, SIAM founder and, later, first managing director. Ed told them about his difficulties in finding suitable papers for the new Journal of the Society for Industrial and Applied Mathematics. Mario immediately pulled out a copy of a manuscript that he and David had been working on and handed it to Ed. The paper, "On the Order of Convergence of Solutions of a Difference Equation to a Solution of the Diffusion Equation," appeared in Volume 1, Issue 2 (1953) of the journal.
As a member of the faculty at the University of Maryland (1952–55), David taught the first course in numerical analysis offered by the Mathematics Department. In response to his request for a raise, he was promoted to associate professor and given tenure instead! But because David and Mildred had a growing family and the aerospace industry paid more than academia, he took a job in Los Angeles as manager of the mathematical analysis department at Ramo–Wooldridge Corp. (later Space Technology Labs). During that time (1955–58), George Forsythe of Stanford University visited the lab, accompanied by a young Gene Golub, who had recently moved to California after finishing his PhD at the University of Illinois. David wondered why Forsythe would bring what he mistakenly believed to be a "mere" graduate student to a consulting job. David and Gene would be friends and colleagues for the rest of their lives!
Beginning in 1958, David spent the rest of his career at the University of Texas at Austin. He was a professor of mathematics and the first director of the university's Computation Center. In 1966, he was among a small group of professors from various disciplines who formed the Computer Sciences Department. In the mid-1960s, with the help of a million-dollar grant from the National Science Foundation, the University of Texas was able to purchase a CDC 6600 "supercomputer," the fastest computer at any university at the time. During these years, David worked closely with Robert T. Gregory and Charles H. Warlick, among others. He was the founding director of the Center for Numerical Analysis, which was established in 1970. Some years later, he was appointed to a distinguished Ashbel Smith Professorship and became a member of the university's Institute for Computational Engineering and Sciences. He supervised 14 PhD students and 25 master's students, and was an author of three books--a two-volume textbook, with Gregory, and a graduate monograph, with Louis A. Hageman. These books have been reprinted by Dover Press.
David was named a fellow of the American Association for the Advancement of Science, and in 1990 he was honored by the Association for Computing Machinery for "Outstanding Contributions to Computer Science." Several conferences, with published proceedings, were held in his honor, and special issues of scientific journals were dedicated to him.
As to his other major talent, David often teamed up over the years with one of his sons, Arthur or William, to play in doubles tennis tournaments. In 1969, he and William were ranked the number one father–son doubles team in the state of Texas. Many times, David would arrange to speak at conferences in England around the time of the Wimbledon Tennis Tournament, which he would then attend. Before Wimbledon became an open tennis tournament, David once applied to play in the amateur ranks. On these trips, he would frequently stop over in Boston to work with Garrett Birkhoff at Harvard and to visit colleagues at Yale.
To graduate students, David was friendly and generous with his time. At conferences, he always spent time talking to students, and giving them advice and counsel. Many of his friends in scientific computing can trace their first meetings with him to such occasions.
A big man, both physically and in his profession, David had a huge impact on the lives of his students, colleagues, and friends. We have lost an inspiring teacher and a fine gentleman who was an international leader in his era. We will miss him, but his spirit will remain alive in our memories.
In addition to Mildred and their two sons, William David and Arthur Earle, David is survived by their daughter, Carolyn Ellen, and two grandchildren, John Robert and Sara Noelle.
In David's memory, the University of Texas at Austin has established the David M. Young, Jr., Instructorship in Computational and Applied Mathematics to support young postdoctoral researchers. Donations to this endowment can be directed to the University of Texas at Austin, College of Natural Sciences, Office of the Dean, Development Office: DMY Instructorship in Math, 1 University Station G2500, Austin, TX 78712–0548, or made online at http://cns.utexas.edu/development/outright.as#endowment. Additional information can be obtained from the College of Natural Sciences Development Office (firstname.lastname@example.org or (512) 471–3299).---David Kincaid, University of Texas at Austin, Richard Varga, Kent State University, and Paul Saylor, University of Illinois, Urbana–Champaign, and Louisiana State University.