Lunch served at 12:45; talk starts at 1pm.
Please subscribe to our mailing list to receive seminar announcements.
Members of the MobiSocial Lab, MobiSocial Partners, MobiSocial Friends, and POMI project will automatically receive notices of the seminar and need not subscribe.
Seminar Schedule, Spring Quarter 2013
Abstract: We will talk about how Cooliris is redefining media experiences on mobile and web platforms. We will also speak about how we bridge the gap between media centric apps and messaging plattforms. Today the limitation with Messaging apps is that they are linear and will allow you only to share one thing at a time. Cooliris provides a solution to fill this divide.
Bio: Venkat Krishnaraj is a Senior Software Engineer at Cooliris and leads the mobile development team for Cooliris 2. Before Cooliris he was Sr Software Engineer at Glu Mobile. Where he was working in Game Development.
Abstract: Stress silently but steadily damages physical and psychological wellbeing, relationships, productivity, and our ability to learn and remember. Chronic stress is often thought to be exasperated by interactive technology but Stanford's Calming Technology Lab challenges this assumption by designing and evaluating techniques for designing experiences and interactions that either mitigate known stressors or induce calm. The outputs of the lab are both novel interactive systems and heuristics and frameworks for designers and researchers to reduce stressors in their designs. In this talk, I will discuss the emerging field of calming technology and dive deep into Breathware, the lab's flagship research project and the one we are now commercializing at www.breathware.com.
Bio: Dr. Neema Moraveji is Director of the Calming Technology Lab at Stanford University. He teaches “d.compress: Designing Calm” at the dschool and his work has been covered in the NY Times, WSJ, NPR, and others. Prior to Stanford he led the Asia Center for Interaction Design at Microsoft Research Asia in Beijing, China. Dr. Moraveji received his B.S. in Computer Science from the University of Maryland at College Park's HCILab and his M.S. from Carnegie Mellon University's HCI Institute.
Apr. 30, 2013
Designing Large-scale Nudge Engines
Professor, Department of Electrical Engineering and Computer Science
Abstract: In many of the challenges faced by the modern world, from overcrowded transportation systems to overstretched healthcare systems, large benefits for society come about from small changes by very many individuals. We survey the problems and the cost they impose on society, and describe a framework for designing ``nudge engines"---algorithms, incentives and technology for influencing human behavior. Some salient themes are: the use of lotteries and games for paying monetary rewards, and the use of mobile, cloud and big data technology for sensing and influencing behavior. We review results from transportation pilots conducted in Bangalore, at Stanford and in Singapore, and a wellness program for the employees of Accenture-USA.
Bio:Balaji Prabhakar is Professor of Electrical Engineering and Computer Science at Stanford University. His research interests are in computer networks; notably, in designing algorithms for the Internet and for Data Centers. Recently, he has been interested in Societal Networks: networks vital for society's functioning, such as transportation, wellness, energy grids and recycling systems. He has been involved in developing and deploying incentive mechanisms to move commuters to off-peak times so that congestion, fuel and pollution costs are reduced.
He is an IEEE Fellow, a Terman Fellow at Stanford University, and a Fellow of the Alfred P. Sloan Foundation. He has received the NSF CAREER award, an IBM Faculty Development Award, the Erlang Prize from the INFORMS Applied Probability Society, the Rollo Davidson Prize from the University of Cambridge, and delivered the Lunteren Lectures. He is a co-recipient of several best paper awards. He serves on the Advisory Board of the Future Urban Mobility Initiative of the World Economic Forum.
Apr. 23, 2013
Wellness Monitoring in Home and Office: Personal Recommendations via Behavior Modeling
Director, Ambient Intelligence Research (AIR) Lab, Department of Electrical Engineering
Abstract: In this talk we present recent results on the use of cameras and sensor networks for developing a model for the behavior of users in their home or office environments. The objective is to produce personalized and adaptive recommendations that would assist the user in maintaining wellness practices. The first step is inference of activities, poses, locations, and postures that relate to the well-being of the user in the target environment. In a home, the wellness of an elderly occupant is indicated by attributes of the user’s lifestyle such as eating and sleeping habits, mobility at home, social contacts, and leisure time activities. In an office, the wellness of an office worker is indicated by the computer usage habits and social interactions with others. Then, a set of generic recommendations provided by experts in each case can be employed and adapted to the specific behavior of the user in order to offer the user recommendations for maintaining their wellness. The use of a network of cameras and networks consisting of other sensors is examined to address the needs and preferences of the different user types. Examples for elderly monitoring and office worker behavior modeling will be provided.
Bio: Hamid Aghajan is director of Stanford's Ambient Intelligence Research (AIR) Lab, and Wireless Sensor Networks (WSN) Lab, where he has supervised research on multi-camera and sensor networks for smart environments since 2003. Focus of research in Hamid’s group is on methods and applications of Ambient Intelligence with an emphasis on activity recognition with networks of cameras, smart phones, and other sensors. Specific research topics include behavior modeling using long-term sensory data, occupancy modeling of smart buildings for resource efficiency, detection of anomaly or shift in behavior in elderly care, improving office worker's well-being through personalized ergonomic recommendations, analysis of meetings, and avatar-based social interactions.
Hamid is Co-Editor-in-Chief of "Journal of Ambient Intelligence and Smart Environments", and has served as guest editor for IJCV, IEEE Trans. on Multimedia, CVIU, IEEE JETCAS, and IEEE J-STSP. Hamid was a general chair of ACM/IEEE ICDSC 2008, AMI 2011, ICMI 2012, and program chair of ICDSC 2007. He has organized workshops, special sessions, or tutorials at ECCV, ACM MM, CVPR, ICCV, ICMI, FG, ECAI, EI, and ICASSP. Hamid received his MS and PhD degrees from Stanford University.
ABSTRACT: __everyware__ is an application platform that allows developers to build applications across devices. The platform is developed with the vision that society is converging toward a configuration in which people will be surrounded by pervasive displays as well as multiple personal devices, such as tablets, mobile phones, and other small computational devices. In this talk, I will describe __everyware__, the infrastructure and primitives it provides, and some of the applications that have been developed. I will also discuss current locations where the system is installed and the data and its analysis that have been collected to date. Lastly, I will discuss future plans for __everyware__.
Bio: Michael Fischer is a PhD student in Computer Science at Stanford University. His adviser is Professor Monica Lam. Michael is interested in the theory of collaboration and how it can be applied to the Internet to promote innovation and creativity with large groups of people. Michael is also interested in applying computer science to problems in physics and mathematics.
Seminar Schedule, Winter Quarter 2012-2013
Abstract: Every computer revolution changes our lives dramatically; so will mobile devices. Mobile devices enable billions of people to capture, share, interact, and consume real-time personal media in new and creative ways. In addition, being devices owned by individuals, they can form an autonomous computing fabric that frees us from the domination of existing centralized proprietary social networking services. This talk presents a system architecture called Musubi (Mobile, Social, and UBIquitous) that combines a novel and natural mobile social experience with a clean architecture that lets users choose different cloud backup services. In addition, Musubi is an app platform that makes it easy to create privacy-honoring social apps. This can open up new markets for social and collaborative apps in fields like education, health and businesses, where centralized proprietary services are inappropriate. A fully working prototype of Musubi is available on both the Android and iPhone app store (http://mobisocial.stanford.edu/musubi/public/).
Bio: Monica S. Lam has been a Professor in the Computer Science Department at Stanford University since 1988, and the Faculty Director of the Stanford MobiSocial Computing Laboratory. She received her PhD in Computer Science from Carnegie Mellon University. Her current research interest is in creating open social computing platforms. She has worked in the areas of high-performance computing, computer architecture, compiler optimizations, security analysis, virtualization-based computer management, and mobile/social software architectures. She is a co-author of the "Dragon Book". Together with her students, she founded MokaFive Inc. in 2005 and MobiSocial Inc. in 2012. Monica is an ACM Fellow.
Abstract: The design studio epitomizes a culture of openness and visibility. It is a place to talk, brainstorm, reflect and ideate. How can we bring the design studio online, so students around the world engage in meaningful conversations around creative work?
By tapping into the scale and diversity of online class, we can develop techniques that provide high-quality feedback for open-ended assignments such as design and writing, as well as help students learn and reflect. I will first introduce peer assessment techniques that enable students in massive online classes provide calibrated, open-ended feedback on each others' assignments. I will share the surprising robustness of using peer assessment at grading thousands of students on open-ended work such as design projects. Second, I'll report on our experiments helping students complete work, and providing them personalized feedback. Finally, I invite your suggestions for future work where we are developing ways to provide students feedback that will help them reflect and learn from their peers' work.
Bio: Chinmay Kulkarni is a PhD student in Stanford’s Computer Science department. Chinmay’s research focuses on peer processes in massive online learning. Before starting at Stanford, Chinmay worked at Microsoft Research India on projects that enabled tools for end-user interactive storytelling, and search technology for Bing.
Abstract: Modern smartphones can detect the user's physical activities by continuously capturing and processing data from the phone's accelerometer sensor. However, continuous sensing reduces the phone's battery life, which might discourage users from using activity recognition applications. In this talk, we present a power-efficient activity recognition system that classifies accelerometer-only data into user activities. We propose an adaptive duty cycling mechanism that dynamically varies both the sampling duration and sleep duration to reduce power consumption. Using naturalistic accelerometer data totaling over 36 days from 8 subjects, we show that our adaptive duty-cycling approach achieves 90% of the accuracy of an always-on activity recognition system, while reducing the wake time and power consumption of an always-on activity recognition system by 96.7% and 81.9% respectively. As a proof of concept, we present two applications that display a snapshot of the user's activities to raise awareness on physical activity levels.
Bio: Vijay Srinivasan is a Mobile Sensing Researcher at Samsung Research America, where he researches core technologies focusing on mobile user activity and context sensing, to enable highly personalized mobile devices. Before joining Samsung, Vijay finished his PhD in computer science in 2011 from the University of Virginia, where his research focused on the unsupervised inference of user activities, identities, and resource usage patterns in smart homes. Vijay has published his research in top-tier conferences on context sensing, including Ubicomp, Pervasive, Sensys, and IPSN.
Abstract: Like letters in the past, email archives capture personal correspondence of great historical value. The Department of Special Collections & University Archives at Stanford leads the archival community in the use of forensic and natural language processing software to process and deliver born digital archives. We will talk about new ways that Stanford University Libraries is collecting, processing, and delivering email archives to benefit its patrons. The new tools are based upon a collaboration with Stanford Computer Science, and build upon the Muse project. We will describe the path we went through to bring Muse to Special Collections, and demonstrate the current prototype of the discovery module of the ePADD (Email--Process, Appraise, Discover, Deliver) system. We will also talk about University Archives' efforts to capture, preserve and provide access to email to enable research into archival email collections.
Bios: Daniel Hartwig is the Stanford University Archivist. He actively works with Stanford faculty, departments, and alumni to take in Stanford records, including increasing amounts of electronic records/email. He holds an MLIS from Wisconsin, an MA from Indiana, and a BA from Iowa.
Glynn Edwards is the head of the Technical Services, focusing on manuscript collections and born-digital materials. The core functions of this position encompass acquisition, cataloging, arrangement, description & strategies for metadata reuse as well as future developments of tools and processes for working with born-digital materials in collections. She received her MLIS and MA in History from Simmons College and AB in Classics from Brown University.
Peter Chan is responsible for developing workflows to process born-digital archives. He received an MLIS from San Jose State University and an MBA from University of Illinois at Urbana-Champaign.
Abstract: The wide availability of personal communication devices in even developing countries, combined with the global publishing power of social networking platforms like Twitter, Facebook, and Tumblr, is changing the journalistic landscape, making it possible to document and report on events in even the most inaccessible locations. From the Arab Spring to Occupy Wall Street, mobile phones and social networks are becoming core reporting devices for both citizen and professional journalists. Yet using these technologies today presents important *risks*: the communication networks on which they rely are often proprietary, thus exposing users’ information to the scrutiny of collaborating companies or governments. They are also inherently *fragile*: in conflict situations, devices may be destroyed or confiscated to prevent publication, or authorities may simply disconnect the links connecting an entire country to the internet. Those who do succeed in publishing may risk retribution either during or after reporting, if they are identified with their device or their publishing records are traced. Dispatch is a collaboration between computer scientists and journalists at Stanford and Columbia seeking to develop a mobile communication and publishing solution for journalists that reduces risks to reporters, increases tolerance to network fragility, and matches the usability of popular social networks.
Bio: Kanak is a graduate student in Computer Science at Stanford. He is working under the direction of Monica Lam as part the Mobile and Social Computing Research Group, in which he has explored applications of mobile devices in new domains including finance, student experience, and journalism. He received his BS in Computer Science from UCLA in 2011. Along the way, he has interned at several different companies covering a wide range of the Computer Science spectrum.
Abstract: The number of connected devices in one’s personal collection of consumer electronic (CE) devices is already large, and growing rapidly. Today, most “connected” CE devices exist in their own, isolated, silos, and they are typically accessed through a dedicated application running on a mobile device (i.e., a smartphone or tablet), rendezvousing through the cloud (and frequently making use of cloud-based services as well). We assert that there is great value to be unlocked by enabling CE devices to interact among themselves (and cloud-based resources) in a peer-to-peer manner. Today, this kind of interaction is limited by the lack of open access to (suitably abstracted) programmatic interfaces to the devices. We propose the creation of a secure, distributed communications framework which we think will enable the development of many new and useful applications which involve the interoperation among multiple CE devices and cloud-based systems. We have implemented a prototype of such a communications infrastructure based on publish/subscribe messaging technology, which offers solutions to many of the associated technical challenges – e.g., security, scalability, ease-of-use, extensibility, efficiency, robustness, location transparency. We have recently demonstrated some example applications which were built upon this prototype communications framework, and we are in the process of creating a collection of partners to help turn this into a cross-platform, industry-wide, open standard.
Bio: Duane Northcutt is currently the CTO of the Connected Home Division of Technicolor (formerly Thomson). From 2010 to 2012 he was the CTO and SVP at Trident Microsystems, where he was responsible for corporate technical strategy. Prior to this (from 2002 to 2010) Duane was the CTO and VP Technology at Silicon Image, where he was responsible for the company’s technical strategy, led the Systems Architecture group, and created the company’s Steelvine storage product family. Duane was a Distinguished Engineer at Sun Microsystems from 1990 to 2001, where he conceived, and led the development of, the Sun Ray desktop product, and was twice awarded the Sun President’s Award for technical contributions to the company. Previously, Duane was a Principal Engineer at Olivetti Research Lab where he worked on network and operating system support for digital audio/video. From 1986 to 1989 he was a member of the Research Faculty in Carnegie Mellon’s School of Computer Science, where he developed the Alpha real-time operating system and the Archons multiprocessor test bed. Duane holds over two dozen issued US patents (with multiple foreign filings and other applications pending issue), is the author of a research monograph on real-time distributed operating systems, and has authored over a dozen articles and technical reports on time-critical computing, distributed systems, multimedia, and thin client technology. Duane has presented a number of invited talks and has served on a variety of different conference program committees. He received a BSEE from Auburn University, and both an MSEE and a Ph.D. in Computer and Electrical Engineering from Carnegie Mellon University.
Abstract In this talk, I will present REACH, which is an intelligent, people-finding system that helps users find someone in their social directory, especially those whom they do not fully remember or barely know. It analyzes a user’s communication (e.g., emails and online chats) and social networking behavior (e.g., LinkedIn and online forums) to automatically extract all the contacts and derive a “social genome” of each contact that characterizes the unique relationship between the contact and the user from multiple facets. It then employs a personalized, faceted search to retrieve and present a ranked list of matched contacts based on their properties. In addition to using the derived social genomes to find existing contacts/relationships, I will also briefly talk about how to leverage the social genomes to help understand people, especially strangers whom one meets in the real world.
Bio: Dr. Michelle Zhou is a Senior Research Manager at IBM Research – Almaden, where she manages the User Systems and Experience Research (USER) group. Prior to her current post, she was a Senior Research Manager at IBM Research, China, managing the Intelligent User Interaction and Social Collaboration group, and a Research Manager of the Intelligent Multimedia Interaction group at IBM T. J. Watson Research Center. Zhou received a Ph.D. in Computer Science from Columbia University. Her expertise is in the interdisciplinary areas of intelligent user interaction, information visualization and smart visual analytics, and analytics-driven social computing and crowdsourcing. She has published over 70 peer-reviewed, refereed articles and filed over twenty patents in above areas. Zhou is active in several research communities, including Intelligent User Interfaces, Information Visualization and Visual Analytics, and Multimedia. She has co-organized/co-chaired conferences and workshops, and served on the technical program committees for key conferences in these areas. She was the general conference co-chair for ACM IUI 2007, and the technical program co- chair for ACM Multimedia 2009 and Intelligent User Interfaces 2010. Michelle currently serves on the editorial board of three journals, ACM Transactions on Multimedia Computing, Communications, Applications (TOMCCAP), ACM Transactions on Intelligent Systems and Technology (TIST), and ACM Transactions on Interactive Intelligent Systems. Zhou was named an ACM Distinguished Scientist in 2009.
Abstract: Imagine an imaging chamber placed around an entire community. What if we could, with permission, record and display nearly every facet of behavior, communication, and social interaction among its members as they live their everyday life? This potential would afford rich insights into humanity - how societies operate, how real world relationships form and change over time, and how behavior and choices spread from one person to another. We could diagnose the health of a community, and of its individuals. We could even measure the effects of feeding this information back to them. At the MIT Media Lab, we have built the beginnings of what we call "The Social fMRI." You don’t need a huge chamber – just a bunch of modern smartphones. Using our mobile sensing software, we transformed a residential community into a living laboratory for over 15 months. Many signals were collected from each participant, altogether comprising what is, to date, the richest real-world dataset of its kind. As part of our continuing research, we are developing new tools to realize "the quantified self", and architectures to do all of this from a user centric perspective – where individuals own their data, and privacy is embedded into the framework. The talk will describe the Social fMRI approach, highlight results from the study, and introduce the system supporting the methodology, implemented, field-tested, and released as an open-source framework with a growing community of users.
Bio: Dr. Nadav Aharony is co-founder and CEO at Behavio. He completed his Ph.D. at the MIT Media Lab's Human Dynamics group, where he investigated the use of mobile phones as social and behavioral sensors, conducted one of the largest mobile data experiments done in academia, and initiated the open source mobile sensing platform that became Funf.org. He was a fellow for three years at the MIT Center for Civic Media, working on mobile and social activism topics. Most recently, Nadav worked at Google as a product manager in the Android team. He has over 10 years of industry experience in engineering, product management, and business development roles, in organizations ranging from startups to corporate environments. He holds a PhD and MS degrees from MIT, and a bachelor's degree in electrical engineering cum-laude from the Technion - Israel Institute of Technology. Nadav holds patents in social mobile networking, machine learning, network algorithms and sensor technologies. His work has been featured in both academic and popular press (Technology Review, Businessweek, Wall Street Journal, Wired UK, and TechCrunch, among others), and received awards of recognition (including Best and Distinguished Paper awards, Knight News Challenge, SXSW Accelerator, and three Google Research Awards).
Seminar Schedule, Fall Quarter 2012-2013
Note unusual location: Gates 463 (4B open area)
Abstract: In the past learning how to use complex visual design software, such as Adobe Photoshop, involved taking classes and reading books. In contrast, today's photographers and visual designers are more likely to learn how to use software opportunistically, as they need to adjust a photo, create a family album, or design a website. Together with my collaborators I have been developing and studying a variety of techniques for making it easier to learn new visual design and computational photography skills. In this talk I will describe our research on instructional design, video tutorials, and games and discuss the challenges and opportunities for learning in the context of visual design tools.
Bio: Mira Dontcheva is a Senior Research Scientist at Adobe Systems. Mira's research focuses on learning in the context of software, search interfaces, and creativity. Since joining Adobe, Mira has been learning about designers and developers and building new tools to support their information needs. In 2010 Mira published her first edited book - No Code Required: Giving Users Tools to Transform the Web. Mira completed her PhD at the University of Washington in 2008. She was an undergraduate at the University of Michigan and completed her BSE in computer engineering in 2000.
Abstract: Through the Migo project, we aim to give high-school students a taste of the excitement of computer science research and to dissolve the false stereotypes that dissuade them from pursuing computer science as a possible career path.
This summer, the MobiSocial Computing Laboratory at Stanford University sponsored a successful summer pilot program that introduced high-school students to computer science research. Leveraging Musubi, the students were able to create multi-party communication games of their own design within a month. More importantly, the students’ views of computer science as a field changed drastically by the end of the program in that they realized computer science was not just programming but also involved creativity, design, and collaboration. We plan on iterating on this summer program to create a targeted program for high school students designed to emphasize these aspects of technology.
Joy Kim is a 2nd-year PhD student in the HCI group at Stanford currently working on various computer science education and online storytelling research projects. She graduated from the University of Washington in 2011 from the Computer Science department. While there, she worked on various accessibility projects, including MobileASL, a video phone application designed for people who speak sign language, and TapBeats, a music game for people with visual impairments.
Henry Qin is a first-year Phd student who is currently rotating with Monica Lam. He spent a year working at Microsoft, leveraging machine learning and crowdsourcing to improve ranking and relevance for the Xbox Bing team. He graduated from Duke University in 2010 with a bachelor's degree in computer science. While at Duke, he worked on mobile security and privacy on the Android platform as part of the YouProve project. He also worked with middle and high school teachers and students to bring computer science into the classroom by leveraging the Alice programming environment.
Abstract: When information is abundant, it becomes increasingly difficult to fit nuggets of knowledge into a single coherent picture. Complex stories spaghetti into branches, side stories, and intertwining narratives; search engines, our most popular navigational tools, are limited in their capacity to explore such complex stories.
We propose a methodology for creating structured summaries of information, which we call metro maps. Our proposed algorithm generates a concise structured set of documents that maximizes coverage of salient pieces of information. Most importantly, metro maps explicitly show the relations among retrieved pieces in a way that captures story development. Pilot user studies with real-world datasets demonstrate that the method is able to produce maps that help users acquire knowledge efficiently.
Bio: Dafna Shahaf is a postdoc at the Computer Science Department at Stanford University, working with Prof. Jure Leskovec. She received her Ph.D. from Carnegie Mellon University, her M.S. from the University of Illinois at Urbana-Champaign, and her B.Sc. from Tel-Aviv university. Her research focuses on helping people make sense of large amounts of data by creating structured summaries of information. She has won a best research paper at KDD 2010, received the Microsoft Research Fellowship, and is also a Siebel Scholar.
Abstract: We all rely on our families, friends, and broader social circles for guidance and personal news in our daily lives, more and more of which happens online. Web search engines recognize the value of online social information and have started incorporating personal content and context to improve the coverage and relevance of search results. The shift toward a customized, more social and personal search experience introduces a set of interesting new challenges for the underlying systems that were previously fine-tuned to identify and serve globally relevant public web content. The talk will present the social features currently available in leading search engines; provide an overview of the issues around and solutions for collecting, processing, and retrieving social and personal data in search systems; as well as giving an overview of what we have learned about the kind of social information seeking people do, and how social annotations affect user search behavior.
Zoltan Gyongyi has been a research scientist at Google since 2008. He is a member of the web search team in Mountain View, California, working on personal search features. Previously, he received his PhD in computer science from Stanford University, where his research focused on web link analysis.
Ed H. Chi is a Research Scientist at Google since 2011, where he focuses on research relating to social search, recommendation, annotations, and analytics. He was previously an Area Manager and a Principal Scientist at Palo Alto Research Center's Augmented Social Cognition Group, where he led the group in understanding how Web2.0 and Social Computing systems help groups of people to remember, think and reason. Ed completed his three degrees (B.S., M.S., and Ph.D.) in 6.5 years from University of Minnesota, and has been doing research on user interface software systems since 1993.
Oct 23, 2012
Mobile Search Applications for the Cloud: Future Paradigms
Principal Architect, eBay
Abstract: The Cloud has become a very important technology architecture today and there is huge adoption by the industry because of the ease of use for consumers and the flexibility for businesses. One of the biggest drawbacks of computing in the cloud is that all the heavy weight computation is borne by the server. This tends to add to data center costs – in terms of hardware, cooling, square foot space, maintenance etc. This puts great pressure on cost reduction, and the development of new computing paradigms for mobile cloud applications. This is where mobile comes in.
Today’s mobile phones have so much compute power. This combined with the ability to run native applications in Java & C++ makes mobile a great partner for the cloud. In this talk I’ll provide many examples of mobile search applications where the mobile shares a lot of the computation and reduces server costs.
Bio: Naren Chittar is a Principal Engineer in eBay’s Search Engine group where he has taken several of his ideas from prototypes to production. He has worked on a wide range of problems in Search from core engineering to Machine Learning in Ranking and Recommendation Systems. He is the co-creator of eBay’s Image Search technology. In 2009 he won eBay’s worldwide innovation competition. He currently works in a small incubation team for the VP of Engineering. He has 15 years of experience in IT and previously worked on Mobile Speech Recognition at Promptu and security applications at Citibank. He has several pending patents in the area of Image Search, Speech Recognition and Recommendation Systems. He has published numerous papers in top tier conferences and was on the program committee for posters at WWW 2012. He has a bachelor’s degree in Electrical Engineering from IIT Bombay and a master’s degree in Electrical Engineering from Stanford University. He can be contacted at nchittar at alumni dot stanford dot edu.
Seminar Schedule, Spring Quarter 2012
In this talk, Paul Kim will talk about why numerous so called "innovative technologies" fail in the education space while sharing his own learning experiences from technology implementation projects around the world. He will also talk about SMILE (Stanford Mobile Inquiry-based Learning Environment) which is based on the early POMI work of Monica Lam's team: Junction Framework - Ad Hoc mobile social networking. SMILE introduces a new "flipped" learning environment to ultimately cause a pedagogical paradigm shift in classrooms around the world. Recently, SMILE caused industry leaders to adopt the technology. For example, Marvell Technologies, the OLPC manufacturer, developed a commercial-grade SMILE Plug( http://www.marvell.com/solutions/cloud-computers/assets/Marvell-SMILE-Plug-Platform-Brief.pdf) and also caused various telecommunication companies and educational organizations to scale up SMILE for their own regions: Argentina Telecom, Argentina, ANTEL, Uruguay, Magma, Brazil, CIDE, Mexico, etc.
Bio: Dr. Paul Kim is the Chief Technology Officer and Assistant Dean for Stanford University School of Education. He is currently a senior researcher for Programmable Open Mobile Internet, a National Science Foundation project aimed at developing and evaluating wireless mobile computing and interactive systems for formal and informal education systems. He works with numerous international organizations to develop mobile empowerment solutions for underserved communities. In his recent experiments in Latin America, Africa, and India, he investigated the effects of SMILE (Stanford Mobile Inquiry-based Learning Environment), ROSE (Remotely Operated Science Experiment), and PocketSchool Network. His learning technology innovations focus on improving educational systems in developing countries and causing a pedagogical paradigm shifts in classrooms. He serves as a committee member of the National Academies on Grand Challenges in International Development. Dr. Kim also established seedsofempowerment.org, a global education non-profit organization for children in need.
Visualization has remained an important means for understanding the structural aspects of networks, social and otherwise. The design of techniques for supporting the visual analysis of large, complex networks, however, remains an open challenge. In this talk, I will present GraphPrism, a technique for visually summarizing arbitrarily large networks for the purposes of characterizing, comparing, and classifying such data. By abstracting away some details about individual nodes and edges, GraphPrism diagrams allow viewers to quickly ascertain important aspects of network structure, such as the effective diameter, small-world properties, and the presence of structural holes. This poster will describe how these diagrams are constructed and present results from an evaluation conducted with network analysis experts; these results show that even static paper prototypes of GraphPrism diagrams can aid in basic network analysis tasks after only minimal training.
Bio: Sanjay Kairam is a second-year Ph.D. student advised by Jeffrey Heer in the Computer Science Department at Stanford University. His research focuses on modeling and visualizing human behavior in large social and information networks and communities. Prior to starting graduate school, Sanjay worked with Peter Pirolli and the Augmented Social Cognition group at the Palo Alto Research Center on several projects pertaining to social search and information-seeking. Sanjay received his B.S. in Mathematics and M.A. in Philosophy in 2006, also from Stanford, along with a minor in the interdisciplinary Symbolic Systems program.
The payment rails that handle trillions of financial transactions are based on 40 year old technologies and methodologies. To authorize a single transaction, data can sometimes flow through four or more companies. Today, many companies are trying to build their own rails, but are discovering the challenges of entrenched technologies and business practices. This session will review the major payment schemes and their architecture, and then contrast them to some of the recent startups such as Bling (unsuccessful) and Dwolla (growing). Smartphones, NFC and other technologies are creating opportunities for new disruptive approaches to payments.
Bio: David Pinski is with Capital One (Formerly ING Direct Bank) where he has held a number of roles, most recently as head of Architecture and Discovery Services (Research and Development). A current focus is guiding the future of payments, and building APIs to support 3rd party development on bank platforms. Previously, he has worked for MBNA (now part of Bank of America), startups and commercial software companies.
Kineograph is a distributed system that takes a stream of incoming data to construct a continuously changing graph, which captures the relationships that exist in the data feed. As a computing platform, Kineograph further supports graph-mining algorithms to extract timely insights from the fast-changing graph structure. To accommodate graph-mining algorithms that assume a static underlying graph, Kineograph creates a series of consistent snapshots, using a novel and efficient epoch commit protocol. To keep up with continuous updates on the graph, Kineograph includes an incremental graph-computation engine. We have developed three applications on top of Kineograph to analyze Twitter data: user ranking, approximate shortest paths, and controversial topic detection. For these applications, Kineograph takes a live Twitter data feed and maintains a graph of edges between all users and hashtags. Our evaluation shows that with 40 machines processing 100K tweets per second, Kineograph is able to continuously compute global properties, such as user ranks, with less than 2.5-minute timeliness guarantees. This rate of traffic is more than 10 times the reported peak rate of Twitter as of October 2011.
Bio: Raymond Cheng is a first-year PhD student at the University of Washington. He works with Tom Anderson and Arvind Krishnamurthy in the UW CSE Networks Lab. Kineograph was joint work performed with the Microsoft Research Asia Systems Research Group. Before attending UW, Raymond attended MIT for his undergraduate and master's degrees in EECS and physics.
Abstract: mon.ki aims to turn the basic concept of social media on its head – instead of making you sift through your social networks looking for information, it helps you understand how the people you know relate to the places, companies and topics that interest you – a “social compass” for your browser , smart phone or tablet. There is a ton of useful information on the web, and in all likelihood several of your thousands of contacts on Twitter or Facebook or LinkedIn have something to say about it. But who has the time to pick over thousands of tweets and status updates to find out whether that person whose name you can’t remember who you think works at Apple tweeted that crucial insight about the article you’re reading about the iPad 3? mon.ki works to solve that problem by unearthing the people buried in your social networks and bringing them into context. Instead of being a distraction from the task at hand, mon.ki turns your social networks into a “social compass” that helps you take action. The early beta of mon.ki brings your people from Twitter into the context of your browser. It automatically extracts and collects information about the page you are looking at and provides you with relevant profiles and their tweets. The value proposition is simple convenience – no jumping from page to page, no searching one by one for relevant profiles and keywords – just the people you need, at your fingertips. In the coming weeks, the mon.ki team will add key features to the product that will take it way beyond Twitter, progressively incorporating more socially relevant tweets, additional social networks and different types of information, fine-tuning the product’s functionality. Advisors to mon.ki include industry veterans Jonathan Smarr (Google+, Plaxo), Jay Bhatti (Spock) and Shane Mac (Gist, Zaarly). The company recently started raising its seed round and closed investors from the US, Spain, France, Chile and Brazil.
Bio: Tim Delhaes is currently the CEO of MonkeyContact, a Silicon Valley based startup currently in stealth mode. In addition to MonkeyContact, Tim is also actively involved in a number of other projects including First Tuesday and Founder Institute Chile. In the late 90's Tim co-founded First Tuesday Chile eventually becoming Director for the Americas in 1998. First Tuesday is now a prominent entrepreneurial organization with representation all over the world. In Chile, Tim inaugurated the first semester of the Founder Institute, a four month pre-seed incubator program for entrepreneurs looking to launch meaningful and enduring technology companies. In the past, Tim co-founded MSM Interactive which was acquired by Nurun - NUR:CN and Tigabytes.com, Google’s largest Enterprise partner in Latin America. Tim regularly acts as adviser to entrepreneurs, venture capitalists and ONGs and has consulted for multinational companies, government agencies and ONGs including Microsoft, GTZ and the former president of Chile. Tim is a self-proclaimed resident of the world having lived in Germany, Portugal, Indonesia, Chile and the United States. He spends his free time surfing, kite boarding or snowboarding.
Seminar Schedule, Winter Quarter 2012
Abstract: Social networks today are social intranets: the entire social graph and all the information communicated is accessible to and monetizable by the social network provider. The trend is for a monopoly to emerge that owns the world’s social graph, which not only puts user privacy at risk, but threatens competition and innovation as well. This talk proposes an open system called the Egocentric Social Platform (ESP), leveraging on Identity-Based Cryptography (IBC), that enables users to easily communicate securely on their mobile phones with their existing identities using their local address book as a sliver of the distributed social graph. This system is integrated into the Musubi social app platform that enables users to share text and photos, play games, all without a central third-party intermediating all communication.
Bio: T.J. is a Stanford Computer Science PhD student. He is working under the direction of Monica Lam in the Mobile and Social Computing Research Group. He received his undergraduate degree in CS from MIT in 2004. After graduating, he worked for a startup developing real-time rendering systems for novel 3D displays and subsequently a distributed desktop virtualization company where he built several OS virtualization technologies.
Abstract: What follows is an exploration of innovative new ways to deliver privacy notice. Unlike traditional notice that relies upon text or symbols to convey information, emerging strategies of “visceral” notice leverage a consumer’s very experience of a product or service to warn or inform. A regulation might require that a cell phone camera make a shutter sound so people know their photo is being taken. Or a law could incentivize websites to be more formal (as opposed to casual) wherever they collect personal information, as formality tends to place people on greater guard about what they disclose. The thesis of this Article is that, for a variety of reasons, experience as a form of privacy disclosure is worthy of further study before we give in to calls to abandon notice as a regulatory strategy in privacy and elsewhere.
Bio: M. Ryan Calo runs the research around privacy and robotics, including the Disclosure by Design and Legal Aspects of Autonomous Driving Projects. Prior to joining the law school in 2008, Calo was an associate at Covington & Burling, LLP, where he advised companies on issues of data security, privacy, and telecommunications.
Calo researches and presents on the intersection of law and technology. His work has appeared in the New York Times, the Associated Press, the Wall Street Journal, and other news outlets. Calo serves on several advisory and program committees, including the Future of Privacy Forum, the Mozilla Legal Advisory Board, and National Robotics Week. He also co-chairs the American Bar Association Committee on Robotics and Artificial Intelligence.
Calo received his JD cum laude from the University of Michigan Law School, where he was a contributing editor to the Michigan Law Review and symposium editor of the Journal of Law Reform, and his BA in Philosophy from Dartmouth College. In 2005-2006, he served as a law clerk to the Honorable R. Guy Cole Jr. of the United States Court of Appeals for the Sixth Circuit. Prior to law school, Calo was an investigator of allegations of police misconduct in New York City.
Graph analysis is becoming increasingly important with the rise of social networks. Because SQL lacks the expressiveness and performance needed for graph algorithms, general-purpose languages are often used instead.
For greater ease of use and efficiency, we propose SociaLite, a high-level graph query language based on Datalog. With SociaLite, users can provide high-level hints on the data layout and evaluation order; they can also define recursive aggregate functions, which can be evaluated incrementally and efficiently.
We evaluated SociaLite by running eight graph algorithms on two real-life social graphs. Our experiment demonstrates that SociaLite queries are very succinct, and the queries run as fast as highly optimized Java programs.
Bio: Jiwon Seo is a PhD student at Stanford University, advised by Monica Lam. His current research interests include query language for graph analysis, parallel graph processing, and social network analysis.
Abstract: The world of digital marketing has evolved beyond the banner ad. As advertising, media and technology continue down this path of convergence, digital worlds are colliding and marketers now find themselves confronting unchartered territories – privacy and government regulation, new platforms, the “netizen” consumer, big data and analytics, mobile, social media… Advertising and communications has become a whole new world. If you have ever wondered about the role of digital and technology in advertising and media, please join us for and learn more about how our world works, what data is valuable to us, and different ways digital and technology is shaping the future of marketing and communications.
About: WPP is the world leader in marketing communications services, employing 153,000 people working in 2,400 offices in 107 countries. It is made up of leading companies in: Advertising; Media Investment Management; Consumer Insight; Public Relations & Public Affairs; Branding & Identity; Healthcare Communications; Direct, Digital, Promotion & Relationship Marketing; Specialist Communications.
Bios: Joanna Jen is Director of Digital at Kantar, with a focus on social media and mobile. She is responsible for finding new technologies and integrating those platforms and data into existing operating companies and product, service and solutions. Joanna is an ex- WPP MBA Fellow, spending previous rotations with Grey Group/ G2 (digital creative) and GroupM/ Mindshare (digital media). Prior to her agency career, Joanna was in management consulting. She has a graduate degree from the Tuck School of Business at Dartmouth and completed her university studies at UC Berkeley, where she double majored in Business Administration and Mass Communications.
George Pappachen is Business Development Director and Chief Privacy Officer at Kantar/WPP. George pursues collaboration and partnership opportunities for the Group where data (consumer, trading or trends) is the primary consideration. On privacy matters, George routinely leads cross-functional teams to advance the Group's privacy principles and to establish a culture of transparency for the Group’s data operations. A recognized contributor to the rapidly moving industry dialogue on digital, data, privacy and innovation, George has appeared before various governmental, business and academic forums in multiple regions of the world and as Guest Lecturer in some of world’s top universities and business schools. George draws on his extensive experience in media, advertising, and research and his training and experience as a lawyer in the field of technology. George is a graduate of the University of Florida and Fordham University School of Law.
ABSTRACT: As people increasingly engage in online health-seeking behavior, the volume of medical text authored by patients and medical novices is growing rapidly. Patient Authored Text (PAT) includes media such as blog entries and online health community posts, and has proved a promising resource for medical data mining challenges like discovering disease trends and comorbidities. To date, however, we lack both an effective method for extracting medical concepts from PAT, and a gold standard against which to evaluate one. In this work, we show that crowdsourcing the task of identifying medical words in PAT yields results comparable in quality to medical experts’ for a fraction of the cost. Based on this fact, we build a gold standard dataset comprising 10,000 PAT sentences from MedHelp, the world’s largest online health community. Finally, we conduct a comparative performance analysis between existing medical concept annotation tools MetaMap and NCBO Annotator, and two statistical supervised learning models that we train on the crowdsourced data. Our best model significantly outperforms both annotators.
Bio: Diana MacLean is a third-year PhD student at Stanford University, advised by Jeffrey Heer. Her current research investigates Medical Informatics from a Social Media perspective, tackling questions such as 'How do online health community dynamics vary by medical condition?', or 'How quickly (if at all) is medical misinformation corrected in online health forums?'.
In the digital age, users can have perfect recall of their (online) experiences via their personal digital archives. We explore how this recall can be leveraged during web browsing. We will present a system called the Experience-Infused Browser that indexes a user’s archive, primarily from long-term email and chat records. As the user surfs the web, it observes each page and highlights named entities that the user has encountered in the past. This browser provides two benefits. First, it highlights terms on the page that occur frequently in the user’s archive, effectively personalizing the page for the user. Second, it can remind the user of names that she has encountered in the past but may not remember. Come to the talk to hear more!
While smartphones come with state-of-the-art technologies, give access to new applications, and contain the latest contents, all this becomes mostly non-existent when entering a car. This presentation will highlight the available concepts for integrating mobile devices into the car environment and will primarily focus on a new arising technology called MirrorLink(TM). Backed by a consortium of leading car and mobile device manufacturer, representing more than 60% of the respective markets, MirrorLink(TM) offers drivers access to their mobile devices applications and content in a safe and driver friendly way. The talk will give first-hand information, from the original inventors of MirrorLink(TM), about the technology, how its progressed from Research into Products and future opportunities. After the talk you will have the opportunity to see by yourself MirrorLink(TM) integrated to a real car.
Bio: Jörg Brakensiek is an R&D Manager at Nokia Research Center, Palo Alto, California. His research focuses on ubiquitous mobile device interoperability with a specific emphasis on the automotive domain. He is currently heading a research team, working closely with multiple external partners around the globe, and driving seamless smartphone interoperability with in-vehicle infotainment systems. Jörg is the original inventor and main specification editor of MirrorLink(TM) and is currently chairing the Technical Work Group of the Car Connectivity Consortium. Jörg received his PhD in engineering from the University of Dortmund, Germany in 1999. He then joined Nokia Research Center, where he had worked on various projects, including ASIC designs, SoC architecture research, reconfigurable accelerators, multi-processors, software defined radio and mobile virtualization. As part of his research, Jörg has been involved in the definition and execution of multiple national and European funded research projects in this domain. In 2007, Jörg relocated to California, to join Nokia's newly created Palo Alto research lab.
Pulse is a mobile app (on iOS, Android, Amazon, WP7) that started off as a class project at Stanford University. The basic idea of the product was to allow readers to get all their favorite websites in one place. The app was built in 5 weeks, and put on the app store in June 2010. In the last 18 months, it has grown from a small class project to a 20-person company, from a tiny user base to over 11M users today and from an app to a full-fledged service that allows you to keep your content with you. This talk will tell you more about the story thus far, the rapid growth we've seen, the issues we've faced with scaling and the challenges we face ahead. It will cover different aspects of the app, including but not limited to Systems, HCI and AI.
Akshay Kothari - Co-founder and CEO of Pulse. Stanford MSEE '10.
Greg Bayer - Backend Engineering Lead at Pulse. Stanford MSCS '09.
Seminar Schedule, Autumn Quarter 2011
ABSTRACT: Nokia Research Center established the North America Lab in November 2006. This lab started in Palo Alto CA and has now expanded to Berkeley CA and Cambridge MA with a total of over 100 researchers and interns located at these three sites. This talk highlights the innovation culture and the key accomplishments of this relatively young research lab. A number of projects in the domains of Collaboration, Community and Context will be showcased.
Bio: John Paul Shen is the founding Head of Nokia Research Center North America Lab. NRC North America Lab is pursuing a wide range of mobility research projects in the areas of: Mobile Computing Systems and Federated Devices; Mobile Cloud and Web Technologies; and Data Analytics and Data Monetization. Prior to joining Nokia in 2006 John was the Director of the Microarchitecture Research Lab (MRL) at Intel. Prior to joining Intel in 2000, John was a Professor at Carnegie Mellon University, where he supervised a total of 17 PhD students and received multiple teaching awards. He is currently an Adjunct Professor at Carnegie Mellon University Silicon Valley Campus.
Nov. 28, 2011
Using digital archives for personalized web browsing and search
Stanford University (jointly with Sudheendra Hangal and Monica Lam)
Abstract: We present two systems that use personal digital archives to improve personalization for the user's benefit. Our first system, called Slant, extracts links from a user's email and Twitter feed to create personal search engines that slant search results towards the sites that the user and his contacts talk about. Our experiments show that the results from slanted search can perform as well as those from regular web search; and moreover when combined with regular web search can improve satisfaction with search results. In our second approach, we introduce the idea of infusing a web browser with our past experiences, as captured in our digital archives. As a user browses the web, our experience infused browser observes the contents of pages viewed, and highlights named entities on the page that the user has encountered in the past, e.g. in email and chat records. Our browser works entirely on the client side, and thus preserves user privacy while allow him to enjoy the benefits of personalization. We evaluated how users reacted to the browser during organic web browsing. Our ﬁndings were that it was useful to surface content on crowded web pages that users otherwise may have missed, and in recalling serendipitous connections to people that they had forgotten. We have released both these systems publicly and hope you will use them!
Bio: Abhinay Nagpal is a second year Masters student and has been working with the Mobisocial group. His interests include databases, HCI and social networks.
The biggest engineering problems that twitter faces are not due to its large user base, or due to the huge amount of data being pumped through its system -- it is its "real-time constraint". the name of the game at twitter is "now": whether that means getting @ladygaga's, @barackobama's, or @raffi's tweets out to their followers within seconds (or faster!) of them writing their 140 characters; or it means doing a search to find out what's happening right now in the world because an earthquake happened, people are talking about a television show, or you want to know where your friends are. very few other systems have this constraint, and very few engineering teams rise to its challenge.
Bio: At @twittereng, @raffi is the Director of theApplications Services group, the custodians of Twitter's core logic → his teams manage, amongst other things, the business logic, scalable delivery, APIs, and authentication of Twitter's application. His group, from the @twittereng side, helped create the iOS 5 Twitter integration as well as the "The X Factor" + Twitter voting mechanism. Previously, he was the lead of the public APIs as well as being the one of those behind Twitter's Geospatial APIs. Before Twitter he used to create technologies to help people frame their personal energy consumption against global energy production (Wattzon - Business Week's "Best Idea" 2008), and also ran a consulting company building off-the-wall projects. At one point, he used to teach at NYU’s ITP (created the class Every Bit You Make) and spent way too much time as a student at MIT and the MIT Media Lab (Internet 0 - Scientific American September 2004).
Nov. 7, 2011
FamilyNet and Comm Apps: An Alternative to Facebook for Staying in Touch
Wendy Mackay, INRIA and Michel Beaudouin-Lafon, U. Paris-Sud
Today’s social media focus on creating ever larger networks of ‘friends’, many of whom have never met. We are interested in the opposite problem: how to better connect people who already know each other very well. We describe a decade of participatory design research in which we created dedicated ‘always on’ communication channels for exclusive use between pairs or small groups of family and close friends. A key insight was to separate network management from the information communicated. Our goal was to create an interface simple enough for a 70-year old grandmother and her 7-year-old grandson to set up without help from Dad. We created FamilyNet, a tangible interface for creating and maintaining dedicated peer-to-peer communication channels. We also created a wide variety of Comm Apps that support both peripheral awareness and focused communication, ranging from the serious, such as helping the elderly to ‘age in place’, to the more whimsical, such as providing remote awareness for couples who miss each other.
Bio: Wendy Mackay is a Research Director at INRIA Saclay – Île-de-France where she heads the In|Situ| research group in Human-Computer Interaction. Formerly Vice President of Research for the Computer Science Department at the University of Paris-Sud, she is now on sabbatical as a Visiting Professor at Stanford University, until 2012. She received her Ph.D. from MIT and created a multidisciplinary research group at Digital Equipment that produced the world's first commercial interactive video system (IVIS), a pre- Hypercard multimedia authoring language and over 30 multimedia software products in the 1980s. She then created a research group at Xerox PARC’s EuroPARC lab that was among the first to explore media spaces, tangible computing and mixed reality interfaces. She is a member of the ACM CHI Academy and has published over a hundred research articles in the area of human-computer interaction. She has served as Chair of ACM/SIGCHI, co- editor–in-chief of the journal IJHCS and on the editorial boards of CACM, ACM/TOCHI and RIHM, as well as program or associate chair of ACM CHI, UIST, CSCW, DIS, IUI and Multimedia. She will chair CHI’13 in Paris, France. Her research interests include co-adaptive instruments, tangible computing and multi-disciplinary, participatory design methods.
Michel Beaudouin-Lafon is Professor of Computer Science at Université Paris-Sud (France) since 1992 and was director of LRI, the laboratory for computer science joint between Université Paris-Sud and CNRS, from 2002 to 2009 (280 faculty, staff, and PhD students). Michel has worked in human-computer interaction (HCI) for over 20 years and was elected to the ACM SIGCHI Academy in 2006. His research interests include fundamental aspects of interaction, engineering of interactive systems, computer-supported cooperative work and novel interaction techniques. His current research is conducted in the In Situ group, a joint lab between LRI and INRIA (http://insitu.lri.fr). Michel has advised twenty Ph.D. students, sits on several journal editorial boards and on many program committees. He is the technical co-chair for the ACM CHI 2013 conference in Paris. He founded AFIHM, the Francophone association for human-computer interaction, in 1996 and was a member of the ACM Council and the ACM Publications Board from 2002 to 2008. He has been nominated to the prestigious Institut Universitaire de France in 2011.
Bio: Anders Isberg is working as a researcher within the Technology & Research organization at Sony Ericsson since 2007 and has been working mobility and data communication since 1990. He is a member of W3C Device API and Policy working group and a member of the Webinos EU research project. He was worked on Sony Ericsson WebGL on Android. He is a Sony Ericsson delegate in EASE Theme B, Flexible Execution of Software in Parallel Embedded Systems.
Activists from Tahrir Square to Liberty Plaza confront forces that curtail their ability to communicate effectively. Disruptive technologies, from established circumvention tools like Tor to new ones like Lantern, attempt to address this need by challenging traditional models of network technologies. Josh King, lead technologist for the New America Foundation's Open Technology Initiative, will speak about the field of disruptive networks, review the technical challenges they face, and introduce the Commotion Wireless project. Commotion, which is also known as "Internet-in-a-Suitcase," aims to create a new device-based physical layer that enables activists to construct autonomous networks. King will discuss engineering issues and review both opportunities and challenges that arise when communication devices become communication infrastructure."
We coin the term micro-interactions to refer to the class of small exchanges between devices that occur almost instantaneously. For example, a mobile payment using near-field communication (NFC) is a micro-interaction. The arrival of NFC on smart phones makes possible a wide array of applications using micro-interactions, from sharing photos between a phone and a TV to checking a car into a valet parking service by touching two phones.
We address the challenge of how to create intuitive, frictionless micro-interactions that require no pre-configuration for a large class of applications. We deliver a consistent tap-and-share interface for many forms of micro-interactions through several concepts. We propose interaction manifests as universal descriptors of multi-party, cross-platform applications. Zero-click overheads are made possible by automatically using the foreground application as the context for the micro-interactions. We extend the concept of connection handovers to allow NFC-enabled applications to run unmodified on devices lacking NFC. We also show how these abstractions make it easy to create a variety of applications. All the application and library code is available as open source.
We demonstrate that by focusing on micro-interactions, our mobile phones can provide a single focal point that enables sharing of our digital identity, assets, applications, and personality with friends (with their mobile phones) as well as the larger-screen PCs and TVs all around us.
This is a preview of the presentation of a paper at Mobicase 2011, co-authored with Monica Lam.
October 10, 2011
Fan page borne threats
Founder and CEO, Reppler
Facebook fan pages have millions of followers and have become a key mechanism for brands, celebrities, political movements and many others to communicate with their audience. The large volumes of people flocking to these pages have made them an attractive venue for distribution of spam and various cyber threats. This talk will go over some of the various threats that are propagated through Facebook fan pages as well as techniques that we developed for detecting these threats.
Vlad Gorelik is a founder of Reppler, a social media reputation management company. Prior to developing Reppler, Vlad was the vice president of cloud services at AVG, one of the biggest players in the security software market. He joined AVG after they acquired the previous company he was at, Sana Security. At Sana Security, Vlad was responsible for the overall architecture of their products and technologies focusing on behavioral malware detection and removal. Prior to joining Sana Security, Vlad was a co-founder and CTO of Optivo Corporation. Vlad has also held various technical lead and engineering positions at Sagent Technology, Sybase, and several other Silicon Valley companies.
Our email archives silently record many of our actions and thoughts, forming a passively acquired life-log of our everyday activity. Billions of mainstream consumers are accumulating these life-logs day-by-day. It is easy to imagine that, over time, they will acquire several decades worth of email archives, that have embedded in them the rich stories of their lives. At the very least, these archives are of great interest to the user, their friends and family for reflective, historical and narrative purposes; but they may also be of interest to historians and digital archeologists of the future.
Given the vast amount of data in a long-term archive, users need effective tools to help them discover and explore interesting content in the archive. MUSE (Memories USing Email) is the first system designed specifically for this task. It combines data mining techniques and an interactive interface to help browse a large archive; it analyzes the contents of the messages and generates a set of cues intended to help spark users’ memories. These cues serve as entry points into a browsing interface that supports faceted navigation and rapid skimming of the messages. In our user studies, we found that users generally enjoyed browsing their archives with MUSE, and extracted a range of surprising beneﬁts, from summarizing work progress to renewing friendships and making serendipitous discoveries.
Mobile VAS - Serving 100+M customers on 10 cents a month
President, OnMobile Global (http://onmobile.com)
OnMobile Global is a publicly traded value-added services (VAS) provider to mobile operators worldwide. OnMobile delivers a diverse array of white-labeled VAS applications to operators in the emerging markets of India, Africa, Latin America and more recently, to operators in Europe and North America. OnMobile serves 100m end users daily with entertainment, information and communication services that enable the operator to play at the applications layer and avoid being relegated to becoming a "dumb pipe".
OnMobile is deploying network based address book and community video services that make use of intelligence in the operators network to roll out the next generation of applications that are compelling, sticky and compete effectively with the OTT players.
Bio: Sanjay is a graduate of IIT, Bombay, the University of Wisconsin-Madison and Stanford University. He has worked at large companies and start-ups from HP, Citrix, Webvan and Caymas Systems to OnMobile Global, and now spends time equally between Bangalore, Paris and the Bay Area.