Association for Computing Machinery
Welcome to the May 2, 2012 edition of ACM TechNews, providing timely information for IT professionals three times a week.

ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).


California Chosen as Home for Computing Institute
New York Times (04/30/12) John Markoff

The Simons Foundation has chosen the University of California, Berkeley to host the Simons Institute for the Theory of Computing at U.C. Berkeley. The new institute will explore the mathematical foundations of computer science and aims to help solve problems in fields such as health care, astrophysics, genetics, and economics. The $60 million institute will have about 70 visiting researchers, including faculty, postdoctoral researchers, and graduate students. “We’ve been talking to astronomers, climate scientists, fluid mechanics people, quantum physicists, and cognitive scientists,” says Berkeley professor Richard M. Karp, who will be the institute’s director. The announcement is part of a broader trend toward expanding support for research in computational theory. For example, Boston University recently launched the Rafik B. Hariri Institute for Computing and Computational Science and Engineering. "It’s analytics with big data, it’s the ability to compute and analyze in massive parallel architectures," says Carnegie Mellon University professor Jeannette M. Wing. “All the science and engineering disciplines realize this is part of the future.”

Tempering the Rise of the Machines
Inside Higher Ed (05/01/12) Steve Kolowich

A report written by former Tufts University president Lawrence S. Bacow and former Princeton University president William G. Bowen analyzes the state of online education and machine learning in the university system. The report assessed the potential roadblocks that might prevent traditional research institutions from integrating sophisticated, machine-guided learning tools into their curriculums. Proponents of interactive learning online (ILO) say that it can help teach students new concepts and help human instructors collect data on how students interact with those ideas. "A wide variety of such systems, of varying quality and sophistication, will proliferate in the next three to five years," say Bacow and Bowen. They say the data harnessed by ILO systems could result in a greater understanding of learning outcomes and greater institutional accountability. However, there are potential hurdles to the widespread adoption of these systems, mostly involving faculty skepticism. Many faculty members fear that instructors could lose their jobs to robotic systems. "To date, no sustainable platform exists that allows interested faculty either to create a fully interactive, machine-guided learning environment or to customize a course that has been created by someone else (and thus claim it as their own)," Bacow and Bowen say.

Robot Sensing and Smartphones to Help Blind Navigate
New Scientist (05/01/12) Helen Knight

Edwige Pissaloux and colleagues at Pierre and Marie Curie University's Institute of Intelligent Systems and Robotics have developed technology that could eventually let blind users navigate their surroundings without assistance. The system features glasses outfitted with cameras and sensors like those employed in robot exploration, and it generates a three-dimensional map of the user's environment and their position in it, which is continuously updated and displayed on a handheld electronic Braille device. The system produces nearly 10 maps each second, which are transmitted to the Braille device and displayed as a dynamic tactile map. Pissaloux says the Braille map's update speed is sufficient for a visually impaired wearer to navigate an area at walking speed. Other robotics technology being applied to help the visually impaired includes software that predicts how far a robot has traveled according to on-board sensor data, which is being tweaked to track a person's movements based on the length of their stride. The system, under development by researchers at the University of Nevada in Reno, would help blind users walk around buildings with the help of a smartphone, using freely available two-dimensional digital indoor maps and a built-in accelerometer and compass. Synthetic speech would provide directions.

A Stock Exchange for Your Personal Data
Technology Review (05/01/12) Jessica Leber

The creation of a marketplace for personal information would be a way for people to regain control of their data in the information age, says Hewlett-Packard (HP) Labs senior fellow Bernardo Huberman. In a recent paper, Huberman, director of HP Labs' Social Computing Research Group, and co-author Christina Aperjis describe a New York Stock Exchange for personal data. A trusted market operator could take a small cut when people decide to sell their data, and help arrive at a realistic price for the transaction. On the proposed market, a person who values her privacy might choose to sell her shopping patterns for $10 at a risk of not finding a buyer, for a guaranteed payment of 50 cents, or might opt out and keep her privacy entirely. Huberman says such pricing options gauge how a person values privacy and risk, and thus helps address how to put a realistic dollar value on data to make transactions worthwhile to sellers and buyers, and how to sell unbiased data so buyers can use small samples of people to infer information about larger populations. Huberman and Aperjis also say a trustworthy market could encourage people to share more and new kinds of data.

Big Data's Big Problem: Little Talent
Wall Street Journal (04/29/12) Ben Rooney

A shortage of data scientists capable of providing analysis is an impediment to realizing the advantages of mining big data, according to a recent McKinsey report. chief scientist Hilary Mason says key skills for data scientists include modeling and understanding the data set mathematically, and extracting insights and narratives from the data. Transforming data into information and action is the most challenging aspect of a data analyst's job, and University of Southampton professor Nigel Shadbolt points to the current lack of courses for training data analysts. "As an integrated discipline it is only just starting to emerge," he notes. EMC Corp. president Pat Gelsinger says data science will become a new discipline much like computer science did 30 years ago. "Now nobody has a data-science department; in 30 years every school on the planet will have one," he says. DataSift CEO Nick Halstead says companies will largely have to teach themselves big data analysis in the meantime. However, Autonomy's Fernando Lucini believes most of the big data challenges can be addressed via algorithms. "They can do the heavy lifting for you so that anyone in a business can use them and ask the questions they need to answer," he says.

21st Century Cities--and Microsoft's Energy-Smart Buildings
CCC Blog (04/26/12) Erwin Gianchandani

Microsoft's Innovation & Policy Center recently hosted a panel discussion on 21st Century Cities as part of the @Microsoft Conversation series. The panel discussed the technology and policy opportunities and challenges involved in making cities smarter and more energy efficient. Microsoft's Rob Bernard says Microsoft researchers are working with partners to develop analytics and visualization tools that allow for the effective integration and analysis of data from new and existing sensors and systems. He says the project, which is in its first phase of deployment, is already producing benefits and savings. One source is the integration of information that enables Microsoft to operate its buildings more efficiently, according to Bernard. Future opportunities include improving algorithms, rules, and tools that could lead to more effective presentations of the information and better decision-making. In addition, local governments might want to expand from one building to whole cities, and international organizations might want to have a global view of facilities in different countries. As the systems become more sophisticated, deployment strategies will need to be sensitive to security and privacy goals. All of these projects aim to more effectively use constrained infrastructures and environments.

New Graphene-Based Material Could Revolutionise Electronics Industry
University of Exeter (04/27/12)

Researchers at the University of Exeter say they have developed the most transparent, lightweight, and flexible material ever for conducting electricity. The material, called GraphExeter, is based on graphene, which at just one atom thick is the thinnest substance capable of conducting electricity, is very flexible, and is one of the strongest known materials. The researchers from Exeter's Center for Graphene Science created GraphExeter by sandwiching molecules of ferric chloride between two layers of graphene. Ferric chloride enhances the electrical conductivity of graphene without affecting its transparency. The researchers say GraphExeter is much more flexible than indium tin oxide, the main conductive material currently used in electronics. The new graphene-based material could be used to create wearable electronic devices, such as clothing containing computers, phones, and MP3 players, or smart mirrors or windows that have computerized interactive features. "GraphExeter could revolutionize the electronics industry," says lead researcher Monica Craciun. "It outperforms any other carbon-based transparent conductor used in electronics and could be used for a range of applications, from solar panels to 'smart' teeshirts." The researchers are developing a spray-on version that could be applied directly onto fabrics, mirrors, and windows.

What Online Social Networks May Know About Non-Members
Heidelberg University (04/30/12)

Heidelberg University researchers have found that through network analytical and machine-learning tools the relationships between members and the connection patterns to non-members can be evaluated with regards to non-member relationships. The researchers used contact data to correctly predict that two non-members know each other with about 40 percent reliability. "Once confirmed friendships are known, predicting certain unknown properties is no longer that much of a challenge for machine learning," says Heidelberg professor Fred Hamprecht. In an online social network it is possible to infer information about non-members using friend-finder applications. The Heidelberg researchers made their calculations using a standard procedure of machine learning based on network analytical structural properties. The separation of members and non-members was modeled using a wide possible range of simulations, and these partitions were used to validate the study results. Using standard computers, the researchers were able to calculate which non-members were most likely friends of each other in just a few days. "Based on realistic assumptions about the percentage of a population that are members of a social network and the probability with which they will upload their email address books, the calculations enabled us to accurately predict 40 percent of the relationships between non-members," says Heidelberg's Michael Hanselmann.

Research Breakthrough Takes Supercomputing Out of the Lab
University of Toronto (04/30/12)

University of Toronto researchers have developed a system that they say will make the production of a special class of photons used in supercomputing faster and easier. In order for advanced technologies, such as ultra-secure communication systems and optical quantum computers, to work a photon must be tightly coupled with another photon. The Toronto researchers have designed a new integrated system that could produce the entangled photon pairs using an integrated circuit, which could lead to the production of photons using a single chip. "The research offers the prospect of unleashing the potential of the powerful and underutilized quantum technologies into the main stream commercial world, out of the lab," says Toronto professor Amr Helmy. The researchers say that utilizing quantum optical computing will be important in solving difficult computational problems, such as complex data sorting.

A 100-Gigbit Highway for Science
Lawrence Berkeley National Laboratory (04/30/12) Linda Vu

The U.S. Department of Energy's Energy Sciences Network (ESnet) is laying the foundation for a high-speed network that can transport an increasing amount of scientific data. "Over the last decade, the amount of scientific data transferred over our network has increased at a rate of about 72 percent per year, and we see that trend potentially accelerating," says ESnet director Greg Bell. ESnet researchers worked with the Internet2 consortium to develop the Advanced Networking Initiative (ANI), which is part of a 100 Gbps national prototype network and a wide-area network testbed. More than 25 groups have taken advantage of ESnet's wide-area testbed, which connects the National Energy Research Scientific Computing Center, the Argonne Leadership Computing Facility, and the Oak Ridge Leadership Computing Facility. "Our 100G testbed has been about 80 percent booked since it became available in January, which just goes to show that there are a lot of researchers hungry for a resource like this," says ESnet's Brian Tierney. For example, Brookhaven National Laboratory researchers have used the ANI testbed to design an ultra-high-speed, end-to-end file transfer protocol tool to move science data at 100 gigabits per second across a national network.

New Software Matches More Kidney Donations, Faster
National Science Foundation (04/30/12) Miles O'Brien; Ann Kellan

The Organ Procurement and Transplantation Network (OPTN) Kidney Paired Donation pilot program uses a suite of computer programs co-developed by Harvard University's Alvin Roth to match living kidney donors with recipients. Roth and colleagues from other universities developed software that analyzes all donors and recipients in the program, inspired by optimization theory, theory of graphs, and probabilistic techniques. The software matches program participants with compatible blood types and antibodies. "It can put together an amazing string of different potential transplants that you just could not do manually," notes OPTN director Ruthanne Hanto. Pairing optimization depends greatly on an understanding of game theory and market dynamics. "If you're trying to organize an exchange, you need a marketplace and a clearinghouse, and that's what we tried to help our surgical colleagues put together," Roth says. "Game theory turns out to be a giant thing for thinking about big systems in which there are lots of different incentives." Boston College's Utku Unver says the design of centralized kidney exchange clearinghouses was helped enormously through the study of other allocation and exchange programs, such as dormitory room allocation in colleges.

A Ride on MIT Media Lab's Digital Bandwagon
CNet (04/27/12) Martin LaMonica

Massachusetts Institute of Technology's Media Lab recently hosted its annual Inside Out conference where speakers from several projects discussed the future of technology and design. A recurring theme is that more everyday objects are becoming digital and that shift opens up new ways of engaging with the physical world. For example, the WristQue project is developing a wristband with a few basic sensors designed to control building temperature and the lighting of smart buildings, while the Augmented Product Counter project enables computer control through tapping on the surface of a table. The Sparsh project enhances the touch interface by letting a person transfer information from a smartphone to a PC. The Recompose project allows users to manipulate an actuated interface of small squares by moving their hands above them. The Media Lab's Changing Places group is studying several ways technologies can improve transportation and buildings. The City Car project is a two-seat electric car with robotic wheels that can be folded to take up less space in crowded cities. Meanwhile, the Mobility-on-Demand project is building a system for managing the distribution of shared vehicles, such as CityCars and bicycles.

Intel Researchers Plot a Smarter, Personalized Cloud
IDG News Service (04/26/12) Agam Shah

Intel researchers recently launched a project to populate neighborhoods with sensors that provide a more accurate picture of elements such as pollution and weather. Intel's Terrance O'Shea says the plan involves gathering weather and air quality information from the sensors, finding the user's exact position, and delivering accurate information for that location using a personalized cloud service. Intel has designed a pollution sensor chip that can be installed in stores and other locations in the neighborhood. The stores carrying sensors can make money by delivering advertisements through cloud services. Intel already is planning a future redesign of its chips that will be equipped with near-threshold voltage technology, which enables central processing units to operate at extremely low voltage levels. That technology could help make it practical to include the sensors in mobile devices. Intel also aims to make cities smarter, and the company has several research projects that use sensor kits for energy, traffic light, and gas station management.

Abstract News © Copyright 2012 INFORMATION, INC.
Powered by Information, Inc.

To submit feedback about ACM TechNews, contact: [email protected]
Current ACM Members: Unsubscribe/Change your email subscription by logging in at myACM.
Non-Members: Unsubscribe