When Paul Rosenbloom became the director of new directions at USC’s Information Sciences Institute 15 years ago, he helped apply timely computing to a diverse range of enterprises, including earthquake modeling, building construction, military training, art heritage and biomedical research.
However, he never imagined the job title would come to refer to a literal new direction for him. The multidisciplinary work propelled him to explore and explain the very nature, structure and stature of computing and its role in the world.
The results of that journey are presented in Rosenbloom’s new book, On Computing: The Fourth Great Scientific Domain (MIT Press).
In the book, the USC Viterbi School of Engineering computer science professor argues that computing deserves to be considered a “Great Scientific Domain” on par with the physical, life and social sciences.
“Thinking of computing as only hardware, software, mathematics, tools or engineering could not be further off the mark,” said Rosenbloom, who is also a project leader at the USC Institute for Creative Technologies. “There is an overarching coherence to the field of computing that makes it greater than the sum of its parts.”
To prove his point, Rosenbloom developed a model that he dubbed the relational approach — a new way of understanding computing and ultimately the structure of science as a whole.
“The three traditional scientific domains each study the interactions among a characteristic set of structures and processes,” Rosenbloom said.
“The story is the same for the computing sciences, where the focus is on information plus its transformation, and the span includes all the fields that contribute to the understanding and shaping of that information transformation,” he explained.
Rosenbloom conceded that he is not a philosopher and his forays into defining science may be considered beyond the boundaries of an artificial intelligence researcher. But he feels that narrow attitudes about computing need countering. Overstating his case seems less of a risk than allowing the future of computing to remain constrained by institutional or ideological walls.
“My goal is to appeal to that broad notion of what computing is all about,” said Rosenbloom, who calls his book a rallying cry to embrace the intertwining of all its parts.
The fields of computer science and engineering, information science and technology informatics, and, perhaps controversially, even mathematics fit into his unified model of computing. He goes further, too, noting all the ways that computing is integral to other domains as well.
Rosenbloom illustrates this by reviewing recent lists of Time magazine’s “Best Inventions of the Year.”
He found that more than one in every three inventions — including 2007’s iPhone, 2008’s Hulu and 2009’s robotic unicycle — were the result of multiple forms of computing and integration with physical, life and social sciences.
Looking forward, Rosenbloom is excited by what is next.
His book expresses optimism that the grand adventure he calls computing will continue to stretch across multiple directions, including looking inward to redefine its place in schools, science and society at large.