As corporate vice president of the External Research Division within Microsoft Research, Tony Hey is responsible for the worldwide collaborative-research strategy across Microsoft. He leads the company’s efforts to build long-term, public-private partnerships with global academic, industrial, and governmental communities to advance computer science, education, and key areas of scientific research that increasingly rely upon advanced computing. The division focuses not only on supporting specific, collaborative research projects, but also on the process of research and its role in the innovation ecosystem, including developing and supporting efforts in open access, open tools, open technology, and interoperability. 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. Previously, he helped transform the School of Electronics and Computer Science at the University of Southampton into one of the top five computer-science research institutions in England. A fellow of the U.K.’s Royal Academy of Engineering and a previous member of the European Union’s Information Society Technology Advisory Group, Hey received the award of Commander of the Order of the British Empire in the 2005 U.K. New Year’s Honours List. In preparation for Microsoft Research’s annual Faculty Summit, to be held at the Microsoft Conference Center on July 28-29, Hey found time to discuss his first year in his current role―and his plans for the organization.

Q: What opportunities have you been able to identify in your first year as the head of External Research?

Hey: There is a sea change happening in science: It’s increasingly being driven by data and computation. The practice of science is now enhanced by collecting and analyzing massive quantities of data rather than small, focused experiments. The data are coming from instruments such as satellites, high-throughput biometric screening systems, networks of sensors, and telescopes, as well as from massive computer simulations. In the next five years, we undoubtedly will collect more scientific data than we’ve collected so far in the whole of human history. Scientists won’t be able to look at every piece individually but instead will need to use computing tools to aid in the interpretation of vast amounts of information. There’s a unique role right now for computer science and the IT industry to help scientists explore their data. Soon it will be impossible to do any kind of science without computational tools―and the more advanced and powerful, the better for the scientist and the science.

My personal excitement since joining Microsoft derives from seeing where we have been able to play our part in the transformation of the scientific discovery process through offering unique combinations of breakthrough research, software assets, algorithms, and collaboration to accelerate the process of reaching insight. I see the future as holding more opportunities for Microsoft and computer science to innovate, such as through the emergence and adoption of cloud computing, improved openness and access to technology, and deeper collaboration in our focal areas. All of these should help foster broader community sharing of scientific data, algorithms, and results. Such action is needed to allow science to make any significant progress in addressing some of the world’s broader social and environmental challenges.

Q:Tell me about your background and how that has prepared you to step into a role like this.

Hey: My Ph.D. was in theoretical physics. I was in physics for about 15 years, and then I got interested in parallel computing, putting computers together to solve problems faster and to solve bigger problems.

At the University of Southampton, I became chair of the School of Electronics and Computer Science. Because of my physics background, I understood the electronics people, and I think there is synergy in having electronics and computer science in the same place. I then became dean of Engineering. I think that “IT in all of engineering” is the future, and that was the mission I took on.

Then I worked for the [U.K.] Office of Science and Technology, leading the national ‘eScience’ program; in the U.S., the equivalent would be the Cyberinfrastructure Initiative. By eScience, I mean research that involves some combination of the adjectives “data-centric,” “multidisciplinary,” “collaborative,” and “distributed.” I interacted with Jim Gray on eScience for five or six years before his sad departure, and I first began to realize that Microsoft was a serious company for looking into the scientific domain. I therefore welcome the fact that Microsoft is now addressing not only consumers and business, but also scientists, researchers, educators, and students.

I’ve worked with industry for the last 15 years, in industrial projects as well as pure research projects. I came to Microsoft because I know that here I can make a huge difference in reaching out to the academic and research communities.

Q: What are the goals you have for Microsoft External Research? Have they changed since you joined this group?

Tony Hey

Hey: Broadly, my goal of wishing to see Microsoft’s computer-science expertise and technologies be instrumental in facilitating research breakthroughs is largely unchanged, but my organization and my views on how we can best achieve this have evolved over the last 12 months.

Following the thread of the challenge we see in the increasing data quantities, I see a corresponding need and trend toward open science data, open access to text and publications, open standards, and open collaboration around computational tools that best serve the science community. I know Microsoft can make a major contribution because, although we have not previously been seen as being a technology provider or collaborator for disciplines such as climatology, biology, or oceanography, this has changed over the last few years. We have now convincingly demonstrated that our research capabilities, technologies, and software assets can have an enormous impact on both common issues faced by researchers and scientists and on some of the more extreme challenges of data aggregation, collection, and processing.

We understand that the research and teaching environment consists of a complex web of software, tools, programming models, languages, technologies, and devices. I am looking to help research communities exploit Microsoft tools when appropriate and to make sure that these tools work alongside existing software technologies from other providers or that are developed by the community itself. One simple example of this is that I seek to develop open-source plug-ins to Microsoft platforms that make it easier for scientists to do their work. For example, we are examining the entire chemistry research lifestyle, especially around semantics and enabling tools in Microsoft Word. One research project is designed to add chemistry-related drawing in Microsoft Office via an add-in, to make it easier for those working in the chemistry realm to document their research.

A more difficult challenge for me and my team is to understand how we can best work with the computer-science community. Much of this community currently makes little use of Microsoft tools and technologies, and the challenge is to see how we can best add value to their research and teaching. I want to understand better how Microsoft can work synergistically with the global science community, and I see opportunities with our new generation of cloud services that are beginning to emerge.

Computing plays an incredibly important role in tackling some of the world’s toughest challenges in domains such as health, education, and environment. Microsoft External Research projects are addressing these challenges by creatively exploring the power of computing to advance potential solutions to these tough challenges. We could help deliver great social impact on some of these problems.

Q: You have decided on a model that includes four research areas you’re interested in pursuing. Could you talk about those?

Hey: Broadly, there are four themes I am looking to explore:

One is Computer Science, which is the core business of Microsoft Research. Here we want to explore groundbreaking research that has the potential to be both high-impact and disruptive, focusing on the pursuit of the hottest new computing technologies and advancing the state of the art in computing and computational sciences. We believe that cloud services will play an increasingly important role in the coming years. 

The second area is in Education and Scholarly Communication, all the things that support education, both research and teaching. In that area, we are working closely with Microsoft’s Education and Public Sector teams and with the Developer and Platform Evangelism folks to make sure we arrive at a strategy that can win mind share in education. Microsoft External Research is working both at the research level, with the world’s leading research universities, and with the company’s digital-inclusion programs, where we have projects in India and in Latin America with underserved communities. Those are two important areas in which we can assist the company.

On the scientific side, Health and Wellbeing is a particularly exciting area. Here we’re helping researchers explore technologies that can empower people to make better choices about their health, thereby saving lives. We have data-mining technologies that can help with things like identifying subtle correlations between genetic variations. We believe we have powerful machine-learning technologies that can help data management in that space. In addition, we are exploring ways to use mobile phones as a platform for health care, as an example of how creative use of advanced information and communication technology can help address pressing issues in healthcare accessibility and availability.

In this theme we are working closely with the Health Solutions Group on how we can help provide the pharmaceutical community with powerful new tools and technologies. It’s clear there is the possibility of some major scientific wins, and it’s extremely exciting. It certainly would be wonderful if Microsoft could play a role in helping scientists ‘solve’ HIV or diabetes. I think there are real chances that we will help make significant scientific advances.

The fourth area is typified by Jim Gray and his SkyServer with Alex Szalay and astronomy data. Accessibility to astronomy data has now been taken to a new level with Curtis Wong’s WorldWide Telescope. We call this area collectively Earth, Energy, and Environment. Our goal is to improve our understanding of the world by helping researchers more effectively address challenges facing fields such as energy efficiency, climate prediction, air pollution, and hydrology.

Standing back and looking at our themes, it is not by chance that we have chosen to focus on issues related to environment, education, and health. These are of major social importance and areas where Microsoft External Research can contribute significantly. Indeed, I am extremely proud of the work we are able to undertake and the potential breakthroughs we seek to facilitate in research communities worldwide.

Q: What benefit does Microsoft External Research bring to the research ecosystem?

Hey: The benefits are relatively diverse and affect a broad spectrum of people. Let me start by saying that a key component of Microsoft External Research’s engagements is to empower researchers through the development of advanced technologies and services to support every stage of their research process. This sounds relatively ambitious, but it is also a realistic aspiration for a company such as Microsoft, and it manifests itself in a range of ways, best evidenced by some examples.

Let’s start with the Research Information Centre, a virtual research environment jointly developed with the British Library in which we support collaborative discovery and information sharing. Such efforts, one of a number of free software tools to help scholars and researchers share knowledge, are showing how existing professional software solutions can be applied to support teams as they share and orchestrate research findings, data, scientific search results, and discussions. Clearly, this directly impacts researchers and research groups by improving their productivity.

Another, quite different example comes from our engagement in digital inclusion, where we are seeking to make the benefits of computing affordable, accessible, and relevant to some of the world’s most challenged populations.

Other investments include genetic analysis to explain the behaviors of viruses and to improve the design of treatments and cures. In addition, we are working on water flow in various hydrology studies to do a better job of predicting flooding or drought scenarios for river and land management. We are even looking at how social networks form and evolve, as a way to encourage collaboration and community building. In all these cases, we are enabling Microsoft researchers to collaborate with leading researchers outside Microsoft and endeavoring to ensure that the results can enrich the entire research ecosystem.

External Research works closely with various national and international funding bodies to help make sure that research agendas reflect both academic and industrial needs, to ensure that, for example, there are good long-term job opportunities for researchers who wish to move into a commercial business or industry. Although we do this broadly, we have specific interests in supporting computer-science faculty members because there is a great need for skilled computer-science graduates across the IT industry.

There are great challenges in core computer science, such as in multicore and parallel computing. There is challenging and important research to pursue. We also have supported creative higher-education projects in areas such as robotics, the use of Tablet PCs, gaming, and collaborative technologies. These are all areas where we feel that Microsoft External Research is uniquely positioned to make a positive difference.

Q: How about the contribution from your Advanced Research Tools and Services group?

Hey: As I said, we aim to empower researchers through every stage of the research process. As well as having a thematic focus in four specific domains, we need to be cognizant of the opportunity for reuse between domains. A problem solved in one area can often be of value in another. The Advanced Research Tools and Services group provides the cross-theme architecture and development skills essential to ensure we are not building stand-alone solutions.

Roger Barga, who heads this group, is working with oceanographers, in a project called Trident, to help with their workflow and data-management challenges. Our intent is that, over time, this will be extended to include equivalent groups and requirements from the atmospheric scientists, bioinformaticians, and astronomers. This is possible because, although the context changes, the underlying needs and workflow behaviors are very similar.

My prior example of our work with the British Library also was led by Roger’s team. It builds on SharePoint as a good way for collaboration and sharing and brings in Web 2.0 technologies like wikis, RSS feeds, and blogs into your normal work. Originally, this was designed for the biological community, but it can be adapted readily to pretty much any research field. We might achieve the same capability even more easily when we have, for example, a SharePoint service “in the cloud,” so that scientists can focus on their research and not on having to install and manage a local software infrastructure.

Q: Any other areas you’re interested in exploring?

Hey: Multicore continues to be important to the company, and it also will be important in academia. We’re doing exciting things in that area.

We’re producing a fair amount of parallel tools within Microsoft, and Microsoft External Research is supporting and collaborating with parallel-computing research centers in Barcelona, the University of California, Berkeley, and the University of Illinois at Urbana-Champaign. And we just awarded seven academics with a total of $1.5 million to stimulate and enable bold, substantial, and impactful research in multicore software. We hope that all this will lead to breakthrough research that rethinks the relationships between computer architecture, operating systems, run times, compilers, and applications. Part of my role has been to emphasize the scale and importance of the multicore challenge and to catalyze the academic computer-science community to work on the problems facing the entire IT industry. This is a major challenge, and, in truth, the research agenda is still in the early stages.

The IT industry needs consumer and business applications that require the power of multicore chips. To leverage the performance advantages from this on-chip parallelism, we have to rethink how we develop software. How do we make parallel code development easy enough for the vast majority of developers and not be restricted to a small cadre of expert programmers? This is the real challenge of the multicore revolution.

This revolution will also impact the scientific community—they’re going to have clusters of multicore nodes. At the moment, it’s dual-core, but it’ll soon be quad-core, and then it’ll be 16 and so on, until you get to hundreds of cores. How do you use all these cores efficiently? How do we convert the existing software base? Can we develop a new class of more intelligent, aware, reliable, secure, and useful applications and tools? Those are exciting challenges.

Q: What’s your engagement model with the greater research community?

Hey: Jim Gray used to say: “Find a scientist who’s drowning in data and is desperate.” You can then say, “Maybe using these technologies or a database or some of our data-mining tools might help.” We’ve put together a small number of projects in the Americas, Europe, Asia, and India, either where there is technology in Microsoft Research, or building upon Microsoft’s products, that we believe can help the scientists make more rapid progress in solving their problems.

This is one type of engagement. The strategy with the tools and the platform is to take what we learn from these engagements so we can get something that is more broadly usable. In this way, we can have an in-depth strategy, but also develop a breadth strategy for the needs of a much wider set of research communities.

Another engagement model involves Microsoft issuing requests for proposal. In fact, we supported hundreds last year alone. This is an extremely effective way of casting a broad net and receiving innovative and thought-provoking proposals. It’s a lot of work for my team to process all those proposals, but it is absolutely worthwhile.

Even without direct collaboration with Microsoft Research, we aim to provide a set of free software tools that will help researchers and scholars share data and knowledge more easily and preserve it more reliably. Similarly, improving the workflow for researchers and scholars can open new frontiers in academic research and help accelerate scientific discovery. The design process for our new scholarly communication tools illustrates our basic engagement model, which centers on our partnerships between academia and industry.

Q: Tell me about your leadership team.

Hey: I have Daron Green, who’s my senior leader for the activities of those four research themes. Daron has deep experience in the IT industry and research from his time with SGI, IBM, and British Telecom. I have Roger Barga, my architect who leads the tools-and-services activity. I have a business manager, Michael Aday, and David Heckerman is a senior Microsoft researcher who has taken over the leadership of Jim Gray’s multidisciplinary eScience research team.

Q: How do you integrate the activities of your team in Redmond with those occurring in other Microsoft Research labs?

Hey: This is an essential part of my role and is much helped by my interactions with [Microsoft Chief Research and Strategy Officer] Craig Mundie and [Microsoft Research Senior Vice President] Rick Rashid. In the coming year, I hope to develop close ties with the heads of Microsoft Research’s labs in Cambridge, India, Asia, Silicon Valley, and New England so we can best leverage the great research that is going on in these labs.

Having talked about integration, it is very important that we strike a balance between global coordination and local needs and sensitivities. In China, we are very focused on computer science and education and our internship and fellowship programs, and partnerships with the China Ministry of Education and universities are having a very positive effect on education there. In India, while originally focused on just computer science and education, we’re now adding a science theme through new research projects like India Digital Heritage and e-Hydrology.

Microsoft Research Cambridge contributes many additional dimensions to this picture. As well as running internships, postdoctoral research, and awards programs, it has roughly 100 researchers and deep expertise and engagement in areas as diverse as environmental science, computational biology, medicine, machine learning, computational systems, and networking. We work closely with the team in the U.K. and thrive on being able to support the connection of Microsoft’s world-class researchers with leading research teams.

It’s always tricky to find the right balance between global programs—after all, the computer-science and scientific communities are global—and local customizations to fit local and regional interests and challenges. The good news is that we have such strong presences in our labs around the world, with researchers who are actively collaborating with academics. That gives us a wealth of information and context to inform the decisions we make. Part of my job is to make sure we have a constant conversation between my team and the people in our labs around the world so that we’re all aware of what’s going on, and we can align our efforts and make smart decisions.

Q: What is your vision of the value you want External Research to bring to Microsoft Research and to Microsoft?

Hey: I want External Research to be recognized as bringing great value both to Microsoft Research and to the company as a whole. I want us to help create an ecosystem of innovators who are inspired to add value on Microsoft’s platforms across the complete research lifecycle because they see the intrinsic value of what Microsoft brings to the table with regard to the work that they are doing. The people we are working with in universities worldwide are going to be the next generation’s leaders. It is vitally important that they see the value that Microsoft can bring to their research, their business, or their recreation.

I also want people outside of Microsoft to know that we actively participate in advancing computing and scientific research, that we can partner with them, and that we can contribute in significant ways to the academic research community around the world.

In order to achieve this, we need to make sure we do a very good job of telling our story, so that people know the ways in which we’re helping on scientifically challenging and socially relevant problems.

I think that we have the potential to be one of the most important groups in Microsoft for the company’s medium to long-term future. It’s my job to see that we live up to that potential.