Association for Computing Machinery
Welcome to the February 15, 2013 edition of ACM TechNews, providing timely information for IT professionals three times a week.

Please note: In observance of the Presidents' Day holiday, TechNews will not be published on Monday, Feb. 18. Publication will resume Wednesday, Feb. 20.

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


Sound Waves Used to Increase Disk Drive Capacity
Computerworld (02/14/13) Lucas Mearian

Oregon State University (OSU) researchers have found a way to use high-frequency sound waves to improve magnetic data storage techniques. The researchers say their work could allow greater amounts of data to be stored on hard disks and NAND flash-based solid-state drives. "There's always a need for approaches that could store even more information in a smaller space, cost less, and use less power," says OSU professor Pallavi Dhagat. The researchers note it has been problematic to use the technology, which is known as acoustic-assisted magnetic recording, to increase disk capacity because it involves the complex integration of optics, electronics, and magnetics. In acoustic-assisted magnetic recording, ultrasound is directed at a specific location on the platter while data is being stored, creating elasticity that enables a small portion of the material to bend. After the ultrasound is turned off, the material immediately returns to its original shape, but the data stored during the process remains in a dense form. "This technology should allow us to marry the benefits of solid-state electronics with magnetic recording, and create non-volatile memory systems that store more data in less space, using less power," says OSU professor Albrecht Jander.

Hidden System Makes It Easier for Elderly People to Live at Home
Technical University Munich (Germany) (02/14/13)

Technical University of Munich (TUM) researchers, in collaboration with business partners, have designed an assistive system for helping senior citizens live at home by embedding a tablet computer in the wall. The tablet provides a central location where users can access all of the information they need, such as weather forecasts, bus schedules, and family phone numbers. The wall panel is designed for an entrance hall area and can alert occupants if they forget to take the front door key when they open the front door. It also can keep track of other items that are frequently misplaced. When residents feel poorly, biosensors can measure vital signs so the system can recommend exercise or medication, or alert a physician or mobile nursing service if the health problem is critical. Building automation functions such as air conditioning also can be managed by the system, which would maintain the circulation of fresh air if the occupant forgets to air the living space. The researchers plan to develop similar wall panels for every room in a house, and TUM professor Thomas Bock stresses that "we want people to retain as much of their independence as possible."

Can New Software Testing Frameworks Bring Us to Provably Correct Software?
CIO (02/13/13) Matthew Heusser

Provably correct software has advanced with new testing frameworks, but no one has yet been able to perfect functionally correct software that delivers the correct output for any input. Software testing will likely remain necessary in the future, but recent advances offer to greatly improve software quality before the exploratory testing phase. Proof of correctness involves proofs that show a requirements document is correctly written in code. One approach is transformational computing, which is based on the idea that a precise software definition can be transformed into the actual program. Similarly, behavior-driven development separates the specification from the code and starts with examples of the software's actions in high-level "Given/When/Then" statements that are similar to English. Model-based testing creates a code map of the known states of a Web application and uses a test program to take "random walks" through test software while providing random input to ensure correct transitions, thus providing intelligent automation to catch obvious errors. ACM A.M. Turing Award winner Fred Brooks suggests using multiple "bronze bullets" to incrementally improve each activity, instead of trying to find a "silver bullet" to automate programming. Brooks says eliminating testing through provably correct software can only marginally cut the expense of delivering a system.

Supercomputing Crucial to Clean Energy Production
HPC Wire (02/13/13) Tiffany Trader

The U.S. Office of Fossil Energy's National Energy Technology Laboratory (NETL) has acquired a 500-teraflop SGI supercomputer to advance energy and environmental research. Slated to go live in early spring, the High-Performance Computer for Energy and the Environment (HPCEE) has 24,192 2.6-gHz Intel Xeon E5-2670 cores with 48,384 GB of memory in 1,512 computational nodes. NETL's Chris Guenther says the system ranked 55th on the latest Top500 list, and 403rd on the Green500 list. However HPCEE offers cooling and power with a Power Usage Effectiveness (PUE) in the 1.03 to 1.06 range without any special modifications, meaning that just 1 percent of total electrical consumption is used to cool equipment. This efficiency boost is expected to save NETL an average of $450,000 per year. Guenther says NETL previously has relied on many small computer clusters, but users have sometimes been unable to find available cores to perform their work. The new supercomputer will provide researchers with more cores, enabling them to model complex problems. NETL is developing commercial sorbent-based carbon dioxide capture systems, experimenting with chemical-looping technology to reduce the pollution and cost associated with electricity generation from coal, and working with coal gasification.

Self-Driving Car Given UK Test Run at Oxford University
BBC News (02/14/13) David Lee

Oxford University researchers have developed a self-driving car that uses lasers and small cameras to memorize regular routes. "It's not depending on [global positioning systems], digging up the roads, or anything like that--it's just the vehicles knowing where they are because they recognize their surroundings," says Oxford professor Paul Newman. The technology allows the car to gain control when driving on roads it already has traveled. At first, a human drives the car, and it builds a three-dimensional model of its environment. When the car goes on the same route again, an iPad built into the dashboard gives a prompt to the driver, offering to let the computer take control of the car. The driver can regain control of the car at any time by tapping on the brake pedal. "Our approach is made possible because of advances in 3D laser-mapping that enable an affordable car-based robotic system to rapidly build up a detailed picture of its surroundings," Newman says. Google also is developing driverless-car technology, but Newman says the U.K. effort uses all of its own technology. "I would be astounded if we don't see this kind of technology in cars within 15 years," he says.

Robots With Lift
Harvard Gazette (02/13/13) Peter Reuell

Harvard University researchers have developed a soft robot that can leap up to one foot in the air using small explosions produced by a mix of methane and oxygen. The researchers say the ability to jump could help future robots avoid obstacles during search-and-rescue missions. "Using combustion allows us to actuate the robots very fast," says Harvard's Robert Shepherd. "We were able to measure the speed of the robot’s jump at four meters per second." One of they key innovations was incorporating a valve into each leg of the robot. "We flow fuel and oxygen into the channels, and ignite it. The heat expands the gas, causing the flap to close, pressurizing the channel, and causing it to actuate," Sheperd says. He notes that in the future, internal-combustion systems could be used to develop walking or running robots. "When we do develop those systems, it would be useful to have a power source that can deliver a high volumetric energy density for a long time, and burning hydrocarbon fuels is a proven way to do that," Shepherd says.

Google App Engine Research Awards Go to 7 Innovative Projects
eWeek (02/13/13) Todd R. Weiss

Google recently awarded its first-ever Google App Engine Research Awards to seven projects that will use the App Engine platform to work with large data sets for academic and scientific research. "We selected seven projects and have awarded each $60,000 in Google App Engine credits recognizing their innovation and vision," says Google's Andrea Held. "We are excited about the proposals' creativity and innovation and look forward to learning about their discoveries." The winning projects include the California Institute of Technology's Cloud-based Event Detection for Sense and Response project; the University of Texas at Dallas' Software Benchmark and Simulation Forecaster; the University of Bristol's Personalized DNA Analysis system; the Massachusetts Institute of Technology's Vision Blocks project; Carnegie Mellon University's Mapping the Dynamics of a City & Nudging Twitter Users project; the University of Washington's Sage: Creating a Viable Free Open-Source Alternative to Magma, Maple, Matlab, and Mathematica; and Arizona State University's Cloud Computing-Based Visualization and Access of Global Climate Data Sets project.

When Google Got Flu Wrong
Nature (02/13/13) Declan Butler

The latest U.S. flu season appears to have flummoxed the Google Flu Trends data-mining algorithms, as evidenced by wide disparities between its estimates and those reported by the U.S. Centers for Disease Control and Prevention (CDC). Several researchers think widespread media coverage of the flu outbreak may lie at the heart of the algorithms' difficulties by triggering many flu-related Web searches by healthy people. Despite these problems, many feel Google Flu will recover its accuracy following the refinement of its models. "You need to be constantly adapting these models, they don’t work in a vacuum,” says Harvard Medical School's John Brownstein. “You need to recalibrate them every year.” Meanwhile, several projects are underway to calculate flu outbreaks by crowdsourcing via citizen volunteers. Lyn Finelli with CDC’s Influenza Surveillance and Outbreak Response Team sees great potential in such efforts, particularly because the questionnaires are based on clinical definitions of influenza-like illness (ILI) and so generate very clean data. Some research groups also have published work suggesting that a close match can be made between official ILI data and models derived from analysis of flu-related Twitter messages.

IT Skills Shortage Hampers UK Response to Cyber Threats
ITPro (02/12/13) Caroline Donnelly

The information technology (IT) security skills gap could take as long as 20 years to close, hindering Britain's ability to protect itself from cyberthreats, according to a recent National Audit Office report. The number of people training to become cybersecurity professionals has not kept pace with the growth of the Internet, the report notes. "This shortage of [information and communications technology (ICT)] skills hampers the U.K.'s ability to protect itself in cyberspace and promote the use of the Internet both now and in the future," the report says. The report also notes that "interviews with government, academia, and business representatives confirmed the U.K. lacks technical skills and the current pipeline of graduates and practitioners would not meet demand." To overcome this shortage, the British government has pledged to overhaul the ICT curriculum in schools. The report also says other professionals will need to bolster their cybersecurity skills to meet the challenge of securing cyberspace. “Other professionals, such as lawyers and accountants, also need to understand cybersecurity in order to assess, manage, and mitigate the business risk of cyberthreats," the report says.

Sandia Researcher Looks for Bad Guys in Cyberspace
Sandia National Laboratories (02/11/13) Sue Holmes

Sandia National Laboratories researcher Jeremy Wendt wants to reduce the number of visitors that cyber-analysts have to check as possible attackers among the tens of thousands who search Sandia's Web sites every day. Wendt has developed algorithms that separate robotic Web crawlers from people using browsers, which he says could improve security because it allows analysts to look at groups separately. Even if an outsider gets into a Sandia machine that does not have much information, that access makes it easier to get into another machine that may have something, Wendt notes. The algorithm looks for computers that do not identify themselves by saying they are one thing but acting like another, trolling Web sites in which the average visitor shows little interest. First, the algorithm measures how many HTML files are pulled down by a user, because bots tend to pull down HTML files far more often. The algorithm also measures politeness, because bots tend to ask for other users to send them pages instead of just taking them. Wendt says the next step is bridging the gap between splitting groups and identifying targets of ill-intentioned emails.

Mapping the Underworld
University of Southampton (United Kingdom) (02/05/13)

Researchers from the universities of Southampton, Birmingham, Bath, Leeds and Sheffield are participating in the Mapping the Underworld project, which is developing a multi-sensor platform that can locate, map in three dimensions, and record the position of the United Kingdom's buried utility assets. The platform would provide utilities with new ways to accurately track the exact location of buried gas, electric, water, and telecommunications pipes and cables, without having to excavate them. Researchers on the project, which is nearing the end of its second four-year phase, are using ground-penetrating radar, low-frequency electro magnetics, vibro-acoustics, and magnetic-field technologies. About four million holes are dug each year to lay, repair, or remove buried pipes and cables. However, the researchers want to give utilities the ability to assess the condition of their pipes and cables and determine whether they may need replacing without having to dig them up.

Moshers, Heavy Metal and Emergent Behavior
Technology Review (02/02/13)

Although the collective movement of concert-goers in a mosh pit appears random, it is mathematically similar to a disordered 2D gas and has the properties of self-organized emergent behavior, say Cornell University researchers. They studied YouTube videos of mosh pits with crowds from 100 to 100,000 people and corrected for perspective distortions and camera motion. Applying particle image velicometry techniques to study collective movement, the researchers found that mosher speed distribution strongly resembles molecules of a 2D gas at equilibrium. Wondering why an inherently non-equilibrium system showed equilibrium properties, the researchers used flocking-simulation software to simulate moshers while altering parameters. When noise dominates flocking behavior and self-propulsion parameters, mosher behavior is easily replicated, the researchers found. Furthermore, when flocking terms dominate behavior, highly-ordered vortex-like patterns emerge, as sometimes occur at heavy metal concerts.

Why We're Building a 1-Billion Euro Model of a Human Brain
New Scientist (02/11/13) Jessica Griggs

The Swiss Federal Institute of Technology's Henry Markram has won a 1-billion euro research prize for his Human Brain Project, which will recreate the human brain in a supercomputer to significantly advance neuroscience. Markram says the supercomputer will serve as a CERN for research, bringing together neuroscientists in genetic, behavioral, and cognitive areas, as well as informaticians, chemists, and mathematicians. He says the model can explain the interaction of all molecules, and notes that drugs are molecules that hit other molecules, not a cell or the entire brain. One goal of the project is eventually to couple the brain model to a robot and study the robot learning, following the sequence from molecules to cognition. In addition, the model will be used to discover biological signatures of disease based on global hospital data, which could yield a new classification of brain disease with objective diagnoses not based on symptoms alone. Markram also intends to create neuromorphic computers with processors that can learn, imitating the human brain. Once a model is coupled to a robot, the robot's neurons will fire and it will behave, but whether the robot will ever be conscious is an unresolved philosophical issue, Markram notes.

Abstract News © Copyright 2013 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