Bio: Ken Wood is Deputy Managing Director at Microsoft's Cambridge Research Lab with responsibility for the lab's business-facing activities, including technology transfer, incubation, licensing, spin-outs, and other models for utilizing the intellectual property generated by the lab's research groups. As Deputy Director, Ken also oversees the lab's marketing and communications activities.
Ken also heads the Computer-Mediated Living research group (CML) which he founded in 2003. CML's vision is fundamentally interdisciplinary, bringing together hardware engineering, computer science, psychology, and sociology to address the problem of designing innovative technology to support everyday life in its widest sense. Ken's personal research interests include human-computer interaction, information retrieval, digital media management, and ubiquitous computing. He has numerous publications in these areas and holds several patents.
Ken joined Microsoft in September 2002 from RealVNC, a start-up he co-founded with colleagues from AT&T Labs Cambridge. Previously, in his seven years at AT&T Labs, Ken led research in the areas of multimedia information retrieval and communications, and was involved in the incubation and business planning of a number of projects that were spun out from the lab as successful independent companies. Earlier posts include several years as an academic at Oxford University, three years at Nortel Networks, and a year's sabbatical at the London School of Economics.
Ken is a Fellow of the British Computer Society and sits on the Board of the Cambridge Network. He holds a doctorate in Computation from Oxford University and an AB in Applied Mathematics and Economics from Harvard University.
Abstract: In today’s world, information technology is at the core of our everyday existence, not just while we’re sitting in front of screen typing on a keyboard, but while we use a myriad of devices to communicate, travel, play, relax, socialize, and work. This world presents new challenges for software: it must interoperate more than ever while adapting itself to radically different hardware platforms and interaction mechanisms; it must make use of the cloud to provide the power and flexibility that people expect while still allowing for disconnected operation; and it must provide a unified experience that addresses what people really want and need, not just basic functionality driven by technological capabilities. Microsoft Research is addressing these challenges in many ways, and the Computer-Mediated Living Group in Cambridge is particularly interested in doing so by exploring the boundary between software and hardware. In this talk, I will present several projects that illustrate how this boundary can be blurred and I will discuss specific novel hardware that we are making available to our academic partners so that they can join us in this exploration.
Bio: Wolfram Schulte is a principal researcher and the founding manager of Microsoft’s Research in Software Engineering (RiSE) team in Redmond, Washington. His research concentrates on improving software development. At Microsoft he co-designed runtimes like Linq and the Task Parallel Library, test tools like Spec Explorer and Pex, verifiers like Spec# and VCC, and modeling tools like AsmL and Formula. Schulte has coauthored more than 80 refereed papers, and holds numerous patents. He has a PhD from TU Berlin, a state doctorate from the University of Ulm, and he worked for a couple of years as a software engineer.
Abstract: Formal Methods (FM), which are based on precise mathematical principles, have always held the promise to change the way that software systems are developed. Now the time has come: with modern language and tool support FM will impact every stage of the development process including requirements, specification, design, implementation, testing, and even documentation. But what has enabled this sudden breakthrough of FMs and what comes next?
The last 10 years have seen amazing progress in scalability and widespread availability of basic formal methods infrastructure like model checkers, automatic theorem provers, standardized intermediate languages, compiler infrastructure, etc. This basic infrastructure has also enabled the amazing success story for the wide spread adoption of static and dynamic program analysis tools. 10 years ago, static program analyses were imprecise and slow. In addition they often required adding complex annotations. We can remove this impediment by using other artifacts like existing assertions, tests, and even earlier code versions as specifications against which the implementations are checked. In fact, users of modern program analysis tools often don’t even recognize that formal methods are used under the hood.
The next 10 years will see the wide adoption of lightweight FMs, which emphasize partial specification and focused application, but for other phases than just implementation and testing. For instance, going to the cloud requires new ways to protect data and privacy. Logic based policy languages can fill this hole. The increasing complexity of modern software/hardware systems requires that we need to analyze these systems at design time, since it will simply be too expensive to build the wrong thing. New constraint-based languages are already emerging that support these analysis. But of course computer scientists’ ultimate dream is to teach the computer to program itself. Can FMs help realize this dream, too?
Abstract: Open source development as a distinct set of engineering methods used at web scale is relatively new, compared to source code sharing dating back to the 1950s. My talk will recount the history of open source software engineering, with particular emphasis on Mozilla's experience. I will detail some specific lessons learned, with measurements of contributions over time. A volunteer-friendly open source development ethos is at least as important as the particular engineering methods built upon it. This ethos has evolved through "cathedral and bazaar" models, from volunteer-only projects to initially-or-mostly-commercial ones and back again, with critical feedback effects on and from evolving (at first local, now distributed) version control systems. The culture of open source development and its several "systems" aspects are exemplified currently by sites such as github.com. I will close by looking to the future and making a few predictions.
Bio: As corporate vice president of the Microsoft Research Connections division of Microsoft Research, Tony Hey is responsible for worldwide external research collaboration in Microsoft Research. He leads the company's efforts to build long-term public-private partnerships with global scientific and engineering communities, spanning broad reach and in-depth engagements with academic and research institutions, related government agencies, and industry partners. His responsibilities also include working with internal Microsoft groups to build future technologies and products that will transform computing for scientific and engineering research. Hey manages the U.S.-based external research group for North and South America, and the multidisciplinary eScience Research Group. He also has dotted-line management responsibility for Microsoft Research's Connections teams in Asia, Europe, and India.
Before joining Microsoft, Hey served as director of the U.K.'s e-Science Initiative, managing the government's efforts to provide scientists and researchers with access to key computing technologies. Before leading this initiative, Hey worked as head of the School of Electronics and Computer Science, and dean of Engineering and Applied Science at the University of Southampton, where he helped build the department into one of the most respected computer science research institutions in England.
His research interests focus on parallel programming for parallel systems built from mainstream commodity components. With Jack Dongarra, Rolf Hempel and David Walker, he wrote the first draft of a specification for a new message-passing standard called MPI. This initiated the process that led to the successful MPI standard of today.
Hey is a fellow of the U.K.'s Royal Academy of Engineering. He also has served on several national committees in the U.K., including committees of the U.K. Department of Trade and Industry, and the Office of Science and Technology. He is a fellow of the British Computer Society, the Institute of Engineering and Technology, the Institute of Physics, and the U.S. American Association for the Advancement of Science (AAAS).
Tony Hey also has a passionate interest in communicating the excitement of science to young people. He has written "popular" books on quantum mechanics and on relativity.
Abstract: Commercial technology companies face a continual need to adjust their business models and approaches to determine how best to benefit from, work with, and contribute to technology standards, interoperability and open source software. Over the past decade, Microsoft has made important adjustments to its software and services strategy, and to its participation in the technology community with this evolving open software ecosystem in mind. Microsoft Research is building software and services that are both open and interoperable through our collaboration and development collaborations with the computer science and scientific research communities. In this talk, I will address some of these activities related to standards and interop, and discuss our Research Accelerators Gallery in the Outercurve Foundation.