By Rob Knies
November 11, 2009 3:00 PM PT
For the Japanese people, Mt. Fuji long has been an object of national pride and a subject of artistic inspiration.
On Nov. 4, it became an object of technological inspiration, as well, with the announcement of the Mt. Fuji Plan, a comprehensive platform for collaboration between Microsoft Research and Japan’s academic community. This effort―followed immediately by the Tokyo installment of Microsoft Research Asia’s two-day Computing in the 21st Century conference―will provide a multimillion-dollar investment to foster long-term cooperation for the development of advanced computing technologies that will change the way we work and live, addressing such issues as health care, energy, and education.
“Together, Microsoft and Japan’s academic community will rethink computing and its power to turn social challenges into economic opportunities,” said Rick Rashid, senior vice president of Microsoft Research, during a press conference in the Fujiwara Memorial Hall on the Hiyoshi campus of Tokyo’s Keio University. “The Mt. Fuji Plan represents our commitment to technological exploration, as well as our confidence in Japan’s future.”
The announcement, which opened with comments from Yasuyuki Higuchi, Microsoft corporate vice president and president of Microsoft Japan, furthers Microsoft Research’s robust record of technological collaboration with leading academics in Japan. The effort, described in detail by Hsiao-Wuen Hon, managing director of Microsoft Research Asia, includes four key objectives: research collaboration, talent fostering, academic exchange, and curriculum innovation. Its significance was underscored by comments from influential members of the Japanese academic community.
“The Mt. Fuji Plan is about researching and exploring the computing technologies that will define tomorrow,” said Sadaoki Furui, professor at the Tokyo Institute of Technology and current director of the Microsoft Institute for Japanese Academic Research Collaboration’s Academic Advisory Committee. “I believe that this collaboration between Microsoft Research and the Japanese research community will create unique technological assets and exciting economic opportunities.”
As evidence of the focus on encouraging young academic talent, an initiative led by Lolan Song, senior director of Microsoft Research Asia’s University Relations program, two young Japanese professors, Atsushi Igarashi of Kyoto University and Toshihiro Kamiya of the National Institute of Advanced Industrial Science and Technology, have been named as recipients of the 2009 Japan New Faculty Award. That, too, attracted notice.
“The energy that Microsoft Research devotes to encouraging the scientific pursuits of young Japanese academics is remarkable,” said Makoto Nagao, director of the National Diet Library of Japan and chairman of the Microsoft Research Japan New Faculty Award screening committee. “By working together, we believe we can cultivate a new generation of talented researchers that will extend Japan’s technological legacy well into this new century.”
The importance Microsoft Research places on its collaboration with the Japanese academic community was already apparent by the staging of the Computing in the 21st Century conference in Tokyo and, on Nov. 6, at Kyoto University. It marked the first time the event was held entirely outside China.
“We’re honored to partner with Microsoft in this prestigious conference,” said Toshiaki Makabe, vice president of Keio University. “We are excited about future collaboration with Microsoft Research as we bring together the world’s brightest minds to create the brightest possible future.”
Wei-Ying Ma, assistant managing director of Microsoft Research Asia, served as chairman for both sessions of the conference, which had the theme “3 Screens and 1 Cloud: Rethinking Computing” and featured keynote addresses from Rashid; Takeo Kanade, U.A. and Helen Whitaker University Professor at Carnegie Mellon University; Dan Reed, Microsoft corporate vice president of the eXtreme Computing Group; and Alan Kay, president of the Viewpoints Research Institute and winner of both the 2003 Turing Award and the 2004 Kyoto Prize.
In Tokyo, after a welcome from Makabe, the keynotes were followed by a question-and-answer session featuring the keynote presenters and by a panel discussion moderated by Hon and featuring computer-science luminaries Butler Lampson, Microsoft technical fellow and recipient of the 1992 Turing Award; Hideyuki Tokuda, professor with the faculty of Environment and Information Studies at Keio University; Katsushi Ikeuchi, professor in the Graduate School of Interdisciplinary Information Studies at the University of Tokyo; and Satoshi Matsuoka, professor with the Global Scientific Information and Computing Center at the Tokyo Institute of Technology.
At Kyoto University’s Centennial Hall, Nobutaka Fujii, the university’s executive vice president, welcomed attendees, and the panel discussion included Lampson; Katsumi Tanaka, professor in the Graduate School of Informatics at Kyoto University; and Katsuro Inoue, professor in the Graduate School of Information Science and Technology at Osaka University.
The keynote talks at both venues examined recent computing trends stemming from the rapid advance of computing hardware and software both in clients―PCs, mobile phones, and televisions―and the Internet-based cloud. Each speech was received with rapt attention by hundreds of academics, industry leaders, researchers, students, and journalists.
Rashid’s presentation, “Creating Connections at Microsoft Research,” traced the vision of Microsoft Research and its role in keeping Microsoft abreast of developing academic and industrial trends. He explained how his organization works to connect people, sensors, and data in today’s three-screens-and-the-cloud reality, using a set of projects from Microsoft Research’s labs across the world to illustrate the innovation and collaboration that makes such work successful.
But first, he shared a personal recollection of an epiphany he had while a student at Stanford University, working on a programming assignment for an early, experimental mini-computer.
“Even though it was 37 years ago,” Rashid said, “I can remember, with crystal clarity, the moment I fell in love with computing. … It was an enormously exciting feeling. In that moment, I felt that my brain, my intellect, had animated this piece of equipment. I began to realize then the power of computing, the ability to take your ideas and your dreams and your thoughts and turn them into something real that could change the world and impact people in an important way.”
As exhilarating and life-defining a moment as that was, Rashid said, it can’t compare to the prospects encountered by today’s computer-science students.
“Today, the opportunity in cloud computing, in tying the cloud into your desktop, your phone, your television, and the other devices you have―that’s really an exciting opportunity,” he said. “I envy the students in the audience. You have the chance to do with computing things I would never have imagined when I was young.”
Kanade delved deeply into one such possibility: aiding the rapid advancement of biological methods to provide new ways of healing. His talk, “Large-Scale and Complete Analysis of Cells in Time-Lapse Microscope Images to Aid Biological Sciences,” was a dazzling, occasionally droll examination of tissue engineering―culturing cells with hormones to induce growth of tissue for restoration.
One approach, Kanade explained, uses a technique similar to ink-jet printing, in which hormones are injected into damaged tissue and the movements of the affected cells are tracked over time.
“If we are able to do this during time and space,” he said, “how will these cells move, and what is the behavior?”
The answers to such questions could transform the way medicine is performed―and could lead to significant advancement in fields such as robotics.
“Computer visualization can be useful for medicine, biologists say,” he concluded. “If perfected, that will be a new way to use computing in the future.”
Reed took a more expansive view in his talk, “Ubiquitous Experiences: The Future of Computing.” He described the affect the Internet, social networks, digital media, broadband networking, smartphones, and electronic commerce will have in a connected world in which the physical and virtual worlds coalesce into an entirely new experience.
“With every technology change, challenge brings opportunity,” Reed stated. “The question we might ask as researchers is: Could multicore alone, can chip parallelism supplant what we’ve been doing over the last few decades of driving clock rates and commensurate performance increases? This is where the opportunity arises. It’s an opportunity to reinvent computing in fundamental ways. Innovation about parallel algorithms, about parallel architectures, about new programming models, about abstractions and applications that take advantage of those.”
The answer, he suggested, is to take an expansive view of the future.
“The challenge for those of us who work in computer architecture,” Reed continued, “is to think long-term, not to think about what we do with two or four or eight cores per processor, but what do we go with tens or hundreds or perhaps even thousands? What kinds of new architectures can enable the new kinds of applications that we care about?”
And, he made it clear, he relishes the work ahead.
“My group, the eXtreme Computing Group, has what I think is the coolest job in the world,” Reed said. “It’s to take a blank sheet of paper and think about the future of computing at radical scale, from quantum computing at one extreme to massive cloud data centers at the other, and think about how we enable a new set of applications with those technologies.”
Kay, speaking via video due to illness, concluded the keynote session by emphasizing coding concision. In his address, titled “‘T-Shirt Computing’? Steps Toward the Reinvention of Programming,” he discussed the aesthetic need to express mathematical exploration as clearly and simply as possible. Using the example of John McCarthy’s concise expression of LISP that fit on a single T-shirt, Kay said such clarity provides powerful ways to think about programming languages.
“How many T-shirts would personal computing require,” he asked, “if its parts were redesigned into ‘active mathematics’?”
He proceeded to describe a project called Steps that attempts to find out, using metaprogramming.
“Could we produce runnable code that is many orders of magnitude smaller?” he asked, “small enough to be expressed on a few hundreds of mathematical-equation T-shirts, versus tens of millions of pages of code?”
The exercise, Kay said, has been instructive.
“We feel,” he said, “all of personal computing, from the end user down to the CPU, is probably at least a thousand times smaller than any [current operating system] and maybe as much as 10,000 times smaller. We might be able to illuminate something about what higher-level programming might be in the future.
“Since this conference is thinking about what’s to come in the 21st century, I think there’s no higher goal in computing than to try to come up with a much better approach to programming the size and scale problems that we’re working on today.”
As Hon noted, such issues are precisely what the Computing in the 21st Century conference and the Mt. Fuji Plan are designed to address.
“Japan has long been a leader and pioneer in consumer electronics,” he said, “powering the three-screen scenario. Cloud computing presents a unique opportunity for Japan. Three screens and the cloud will have a pervasive influence and impact in every part of our life and our society. The need to rethink computing is an important subject, not only for computer scientists and engineers, but also for other domain professionals and users.”