采访: Deborah Estrin#
Interview: Deborah Estrin
Deborah Estrin is a Professor of Computer Science at Cornell Tech in New York City and a Professor of Public Health at Weill Cornell Medical College. She is founder of the Health Tech Hub at Cornell Tech and co-founder of the non-profit startup Open mHealth. She received her Ph.D. (1985) in Computer Science from M.I.T. and her B.S. (1980) from UC Berkeley. Estrin’s early research focused on the design of network protocols, including multicast and inter-domain routing. In 2002 Estrin founded the NSF-funded Science and Technology Center at UCLA, Center for Embedded Networked Sensing (CENS http://cens.ucla.edu.). CENS launched new areas of multi-disciplinary computer systems research from sensor networks for environmental monitoring, to participatory sensing for citizen science. Her current focus is on mobile health and small data, leveraging the pervasiveness of mobile devices and digital interactions for health and life management, as described in her 2013 TEDMED talk. Professor Estrin is an elected member of the American Academy of Arts and Sciences (2007) and the National Academy of Engineering (2009). She is a fellow of the IEEE, ACM, and AAAS. She was selected as the first ACM-W Athena Lecturer (2006), awarded the Anita Borg Institute’s Women of Vision Award for Innovation (2007), inducted into the WITI hall of fame (2008) and awarded Doctor Honoris Causa from EPFL (2008) and Uppsala University (2011).
Please describe a few of the most exciting projects you have worked on during your career. What were the biggest challenges?#
In the mid-90s at USC and ISI, I had the great fortune to work with the likes of Steve Deering, Mark Handley, and Van Jacobson on the design of multicast routing protocols (in particular, PIM). I tried to carry many of the architectural design lessons from multicast into the design of ecological monitoring arrays, where for the first time I really began to take applications and multidisciplinary research seriously. That interest in jointly innovating in the social and technological space is what interests me so much about my latest area of research, mobile health. The challenges in these projects were as diverse as the problem domains, but what they all had in common was the need to keep our eyes open to whether we had the problem definition right as we iterated between design and deployment, prototype and pilot. None of them were problems that could be solved analytically, with simulation or even in constructed laboratory experiments. They all challenged our ability to retain clean architectures in the presence of messy problems and contexts, and they all called for extensive collaboration.
What changes and innovations do you see happening in wireless networks and mobility in the future?#
In a prior edition of this interview I said that I have never put much faith into predicting the future, but I did go on to speculate that we might see the end of feature phones (i.e., those that are not programmable and are used only for voice and text messaging) as smart phones become more and more powerful and the primary point of Internet access for many—and now not so many years later that is clearly the case. I also predicted that we would see the continued proliferation of embedded SIMs by which all sorts of devices have the ability to communicate via the cellular network at low data rates. While that has occurred, we see many devices and “Internet of Things” that use embedded WiFi and other lower power, shorter range, forms of connectivity to local hubs. I did not anticipate at that time the emergence of a large consumer wearables market. By the time the next edition is published I expect broad proliferation of personal applications that leverage data from IoT and other digital traces.
Where do you see the future of networking and the Internet?#
Again I think its useful to look both back and forward. Previously I observed that the efforts in named data and software-defined networking would emerge to create a more manageable, evolvable, and richer infrastructure and more generally represent moving the role of architecture higher up in the stack. In the beginnings of the Internet, architecture was layer 4 and below, with applications being more siloed/monolithic, sitting on top. Now data and analytics dominate transport. The adoption of SDN (which I’m really happy to see is featured in this 7th edition of this book) has been well beyond what I ever anticipated. However, looking up the stack, our dominant applications increasingly live in walled gardens, whether mobile apps or large consumer platforms such as Facebook. As Data Science and Big Data techniques develop, they might help to lure these applications out of their silos because of the value in connecting with other apps and platforms.
What people inspired you professionally?#
There are three people who come to mind. First, Dave Clark, the secret sauce and under-sung hero of the Internet community. I was lucky to be around in the early days to see him act as the “organizing principle” of the IAB and Internet governance; the priest of rough consensus and running code. Second, Scott Shenker, for his intellectual brilliance, integrity, and persistence. I strive for, but rarely attain, his clarity in defining problems and solutions. He is always the first person I e-mail for advice on matters large and small. Third, my sister Judy Estrin, who had the creativity and courage to spend her career bringing ideas and concepts to market. Without the Judys of the world the Internet technologies would never have transformed our lives.
What are your recommendations for students who want careers in computer science and networking?#
First, build a strong foundation in your academic work, balanced with any and every real-world work experience you can get. As you look for a working environment, seek opportunities in problem areas you really care about and with smart teams that you can learn from.