Microsoft Researcher Honored by Leading Technology Institute
As a young child, Charles P. (“Chuck”) Thacker didn’t wish for new toys at Christmas. Instead, he’d ask for a better set of screwdrivers, coveted by a boy who at age 6 knew he wanted to an engineer.
By the time Thacker entered college, he’d shifted his sights to physics, and after graduating in 1967, he chanced upon a job opening in a fledgling computer project. It was a detour that proved fortuitous for Thacker, who recalls being immediately captivated. In the four decades since, his change of course has proved equally fortuitous for the computing industry.
This weekend, Thacker’s contributions are being recognized at an honors ceremony in Philadelphia, where he will receive the prestigious John von Neumann medal for outstanding achievements in computer-related science and technology. Awarded annually by the Institute of Electrical and Electronics Engineers (IEEE), the world’s leading professional association for the advancement of technology, the von Neumann medal honors Thacker for playing “a central role in the creation of the personal computer and the development of networked computer systems.”
“Mr. Thacker’s influential work at Microsoft and in the computer industry -- as well as within the greater technology industry -- has resulted in fundamental breakthroughs and technology upgrades in personal and network computing, and has paved the road for so many of the innovations that affect the lives of people everyday,” said IEEE President Leah H. Jamieson. “We are pleased to recognize Mr. Thacker, a true pioneer and visionary, for his important contributions.”
“It’s a great honor,” says Thacker, who joined Microsoft Research in 1997 and is one of only 18 Microsoft Technical Fellows, a designation that acknowledges the highest order of technical leadership within the company. “But in reality, the work that’s being honored, which is essentially the development of the first PC, was a large team effort.”
Thacker points out that the core team included Alan Kay, Robert Taylor and Butler Lampson, the last also a Technical Fellow and Microsoft Research colleague. In 2004, the National Academy of Engineering jointly awarded the four men the Charles Stark Draper Prize -- considered the Nobel Prize of engineering -- for the same set of work recognized by the IEEE’s von Neumann medal.
A Career Built on Technology Innovation
For Thacker, that work began when he joined the Genie research project at the University of California at Berkeley, where he’d earned his undergraduate degree.
The Genie project, which focused on building a time-sharing computing system, soon morphed into a startup called Berkeley Computer Corp. (BCC). Two years later, when the company failed during the 1970 recession, Thacker and a core group of his BCC colleagues joined forces with Taylor, who was then starting up a new computer science lab at the Xerox Palo Alto Research Center (PARC).
Thacker would eventually spend 13 years at PARC. Early on, he led the effort to develop a time-sharing system called the Multiple Access Xerox Computer (MAXC). It was one of the first systems to use DRAM, a semiconductor memory technology introduced at the time by Intel. Thacker recalls understanding exactly what Intel founder Gordon Moore said about silicon density. Clearly, as transistors became cheaper, hardware designers could use more of them at less cost, a trend that lessened the importance of developing large-scale systems intended to be shared.
“Instead of building a system where the human is servant to the computer -- the way it had worked up until then -- we decided to build a system where the computer is in the service of the human,” Thacker explains.
That new way of thinking led the PARC team to search out a better user interface and ultimately choose a novel device called a mouse, which had been invented shortly before by Douglas Englebart of SRI. Another innovation of the PARC lab was a bit-mapped display in which every pixel on screen was represented by one bit of memory.
“For the first time, you could do something that closely approximated the properties of paper,” Thacker explains. “So anything you could print in black and white, you could display on screen.”
Both innovations -- mouse and bit-mapped display -- became part of the first personal computer, the Alto, a project for which Thacker served as chief hardware designer. Around the same time that he and his colleagues were building that first PC, they built another system that he remembers Xerox liking a lot — the laser printer. The idea of servers also grew out of the PARC lab, which in turn led to a desire to hook all those pieces together using a network. The PARC researchers considered several alternatives, then settled on one that eventually became Ethernet.
Thacker, a co-inventor of Ethernet local area network technology, notes that it has evolved from 3 megabits per second to today’s 1,000 megabits (1 gigabit) per second, but the fundamental ideas of transferring information as packets and the routing function that guides those packets through the network remain essentially the same 30 years later.
When Hardware Met Software
By 1983, Thacker had built several generations of systems and was looking for new things to do. But when Xerox management decided to spend more money on physics and less on computing going forward, he and others in the laboratory left PARC. It was about then that Thacker first visited the campus of a small company called Microsoft and met its co-founder, Bill Gates, over lunch. A former PARC colleague, Charles Simonyi, who’d joined Microsoft in 1981, had invited Thacker to visit, suggesting that he might want to work there.
“[Not taking that job] was the biggest financial mistake I ever made,” Thacker says, laughing ruefully. “But Microsoft was completely a software company then, and there was no place for a hardware guy.”
Instead, Thacker and a number of other PARC veterans joined computer maker Digital Equipment Corp. in 1983 to start up another research lab. His years at DEC saw Thacker leading the development of a multiprocessor workstation and the first Alpha-architecture multiprocessor, as well as working extensively in computer networking.
The idea of working at Microsoft surfaced again in 1997. Thacker helped establish a Microsoft Research lab, led by Roger Needham, in Cambridge, England. By the time that two-year assignment ended, the U.K. Microsoft Research center was up to 40 people and well on its way to its current status as one of the premier systems labs in Europe. Its success paved the way for Microsoft Research to set up additional labs, including facilities in Beijing, China; Bangalore, India; and Mountain View, Calif.
After returning to the United States in 1999, Thacker joined a product group to develop prototypes of Microsoft’s Tablet PC. More recently, he worked on a project focused on designing low-cost computer systems to promote education in developing countries. Like most of the projects he’s worked on throughout his career, that effort needed someone who could “straddle the boundary between hardware and software.”
His current position at Microsoft Research is no different. Thacker is setting up a group at the Silicon Valley lab that focuses on computer architecture research. He describes this work as the discipline within computer science that studies the structure of computers and how they should be built to deliver the highest performance at the lowest cost.
Thacker explains: “Microsoft Research decided to get into architecture research because significant changes are happening in the base technologies out of which computers are built, and those changes are causing a lot of challenges to bubble up to the level of the software.”
Until now, Thacker has been one of a rare few hardware architects at Microsoft, but he’s actively drawing on his 35 years of experience to recruit, interview and hire researchers who can help the new architecture group grow. Microsoft is “in it for the long term,” Thacker explains, and Microsoft Research has strong support from management to develop a greater level of understanding in areas that stretch beyond the company’s core business.
In his current role, Thacker also works with a consortium of academics representing major computer science research universities. Known as the Research Accelerator for Multiple Processors (RAMP) program, the group seeks to enable more controlled experiments in computer architecture by building a machine designed with field-programmable gate arrays (FPGAs) rather than costly, specialized chips. Recognizing the value that such an emulator would have for architecture research, Thacker is deeply involved in the project’s manufacturing and engineering efforts. He envisions those machines eventually becoming available to academics as well as the architecture group at Microsoft Research and other industrial laboratories. He expects they will serve as a good resource for revitalizing architecture research, an area of innovation that he’s especially interested in.
“I think architecture research has been moribund for a decade,” he says. “A lot of computer science has moved to be highly theoretical as opposed to being more practical, and architecture research has become incremental rather than revolutionary. It’s good to have a solid theoretical foundation for what you do, but in the end it’s about people using computers. I’m an engineer. And an engineer wants to make the world better by building better machinery and better technology.”
Still Room for Breakthroughs
Thacker believes that the changes occurring today in the semiconductor industry hold great promise for the future of innovation. He predicts that the fundamental shift toward a multiprocessing architectural model will have as profound an effect on the computing industry as the first two major shifts he’s had the good fortune to witness during his long career -- from time-sharing to personal computing, and from single-machine to wide-area networked computing.
“Computer performance will continue to increase, but it will be delivered to us in new and unfamiliar ways,” he says. “Figuring out how to harness this new and unfamiliar model of more processors rather than faster processors is a tremendous challenge, and it will be with us long after I retire.”
Retirement clearly isn’t a prospect Thacker gives much thought to these days. At 64, he is still inspired by his work in computer architecture, a field he holds in common with the namesake of the IEEE medal he’s being honored with. John von Neumann was a mathematician by profession, Thacker explains, but he was very definitely a computer architect, especially toward the end of his career. In an influential mid-1940s report written with fellow mathematician Herman Goldstine, von Neumann laid out an organizational structure that became the blueprint for building electronic computers. Known as the von Neumann architecture, the approach continues to define the fundamental way a computer works, 60 years later.
“Chuck’s remarkable body of work speaks for itself,” says Roy Levin, director of Microsoft Research Silicon Valley. “We are privileged to have him working at our Silicon Valley lab. It is always an honor when our researchers receive recognition for advancing the state of the art in computer science, as Chuck has done so impressively.”
A longtime resident of Silicon Valley, Thacker has been happily married for 43 years and has two adult daughters. In his leisure time, he enjoys travel and reading, but he confesses that his mind is on work much of the time.
“I’m the ultimate nerd,” admits Thacker, who was also named a Fellow recently by the American Academy of Arts and Sciences. “But I’m also one of the luckiest people around. I’m doing a job that, if my employers were not willing to pay me, I would be perfectly happy to pay them.”
This news content was configured by WebWire editorial staff. Linking is permitted.
News Release Distribution and Press Release Distribution Services Provided by WebWire.