Association for Computing Machinery
Welcome to the August 7, 2013 edition of ACM TechNews, providing timely information for IT professionals three times a week.

As a benefit of ACM SIGCHI Membership and SIGCHI Chapter Membership, ACM is pleased to introduce a new ACM TechNews - SIGCHI Edition, which will be provided to SIGCHI Members and SIGCHI Chapter Members on a monthly basis.

Please click on the link below to access your new ACM TechNews - SIGCHI Edition. CLICK HERE.


Spy Agencies Want Low-Energy System to Solve 'Interesting Problems'
Computerworld (08/06/13) Patrick Thibodeau

U.S. intelligence agencies want to promote the development of superconducting supercomputers because they believe the technology could offer a low-power alternative to current systems in the quest to develop exascale systems. The Director of National Intelligence (DNI) recently published a solicitation to demonstrate a small-scale computer based on superconducting logic and cryogenic memory that is energy-efficient, scalable, and able to solve interesting problems. Superconducting technologies can reduce power demand for 100 petaflops to about 200 kilowatts. Superconducting supercomputing uses extremely cold temperatures to get metal to a state in which there is little to no resistance to electrical current. "Keeping superconductive electronics cold is like keeping a block of ice frozen in the freezer," says Retief Gerber, founder of NioCAD, which is developing superconducting technologies. "Once it's frozen, one needs a lot less energy to keep it frozen." Gerber says superconducting electronics have improved over the last 20 years to the point where complex circuits can be operated at clock frequencies surpassing 100 GHz. The DNI solicitation notes that conventional computing systems "appear to have no path to be able to increase energy efficiency fast enough to keep up with increasing demands for computation."

Robots Are Getting Closer to Having Humanlike Abilities and Senses
Washington Post (08/05/13) Eric Niiler

As robots gain increasingly human-like abilities to sense their environment, they are functioning more independently and taking over tasks typically performed by people. Sensor technology and software advances are enabling robots to make their own immediate decisions. "They are gaining human capabilities, whether it's smell, or touch, or recognizing our voices," says robotics doctorate Daniel H. Wilson. "If they are going to solve human problems, they will have to have human abilities." Robots are gaining visual depth perception, the ability to understand voices, and dexterity that almost equals that of humans. However, robots with advanced sensing abilities are generally in the experimental stage and not yet able to navigate unpredictable real-world situations. For example, Industrial Perceptions has created a robot that uses two- and three-dimensional cameras and software to sort boxes into the correct bins, which is effective as long as the boxes are rectangular and do not move. Meanwhile, University of Tokyo researchers are working on enabling robots to smell, and recently released a robot that is driven by a male silkworm moth responding to a female moth's pheromone scent. The researchers say the "odor-tracking behaviors of animals [could be] applied to other autonomous robots so they can track down smells, and subsequent sources, of environmental spills and leaks when fitted with highly sensitive sensors."

Reliable Communication, Unreliable Networks
MIT News (08/05/13) Larry Hardesty

Massachusetts Institute of Technology (MIT) researchers recently unveiled a framework for analyzing ad hoc networks in which the quality of the communications links fluctuate, with new algorithms to reach maximal efficiency. Past theoretical analyses of ad hoc networks have been based on the assumption that network communications links are stable, which often is not borne out with real-world wireless devices. "There's been a discrepancy between the theory, with its idealized models, and the reality of wireless networks," says MIT professor Nancy Lynch. "When people start designing theoretical algorithms, they tend to rely too heavily on the specific assumptions of the models. So the algorithms tend to be unrealistic and fragile." The team modeled the fluctuation in link quality as the intentional interference of an adversary, who cannot control all of the network links. Some links continue to function during the execution of the communication algorithm, but the adversary can change the bandwidth of other links and the network designer does not have prior knowledge of which links are reliable. The algorithms defeated the adversary by randomly assigning each node a probability of transmitting during a round of communication, instead of using a preassigned sequence.

Japan Sends Talking Robot Into Space as Part of Program to Help Lonely People
IDG News Service (08/05/13) John Ribeiro

A talking robot is expected to reach the International Space Station (ISS) in the next six days, launched on a cargo transfer vehicle from the Japan Aerospace Exploration Agency's Tanegashima Space Center. Kirobo is a black and white robot with red boots that stands a little more than 13 inches tall. The robot combines speech, voice, and face recognition, and other communications functions, and its first task will be to communicate with Japanese astronaut Koichi Wakata. "The Kibo robot has a special mission: to help solve the problems brought about by a society that has become more individualized and less communicative," says the Kibo Robot Project. The program seeks to use devices to provide people living alone, including the elderly, with companionship. Kirobo, which speaks Japanese, will spend 18 months on the ISS.

Simulating 1 Second of Real Brain Activity Takes 40 Minutes and 83K Processors (08/02/13) Derrick Harris

The world's fourth-fastest supercomputer needed 40 minutes to simulate one second of actual brain activity on a network equivalent to 1 percent of a brain's neural network. A team of Japanese and German researchers were behind the effort to simulate the activity of 1.73 billion nerve cells connected to 10.4 trillion synapses, the largest-ever simulation of neural activity in the human brain. The simulation involved 82,944 processors on the K supercomputer and 1 petabyte of memory, amounting to 24 bytes per synapse. If computing time scales linearly with the size of the network, it would take nearly two and a half days to simulate 1 second of activity for an entire brain. "If petascale computers like the K computer are capable of representing 1 percent of the network of a human brain today, then we know that simulating the whole brain at the level of the individual nerve cell and its synapses will be possible with exascale computers hopefully available within the next decade," says project leader Markus Diesmann. The simulation was a test of the open source NEST simulation software.

How Computer Clouds Could Help Cure Cancer (08/05/13) Joseph Marks

The National Cancer Institute plans to sponsor three pilot computer clouds that will enable researchers to access genomic cancer information. George Komatsoulis, who is leading the cloud initiative, says the largest source of data on cancer genetics is NCI’s Cancer Genome Atlas, which currently contains half a petabyte of information and could increase to 2.5 petabytes by 2014. Although only a few research centers can afford to store that volume of data on their servers, if that data could be stored inside a cloud, Komatsoulis says researchers worldwide would be able to access and study it, "without requiring millions of dollars of investment in commodity information technology." He notes that if one or more of the pilots proves successful, a private-sector cloud vendor may eventually store the information and make it available to researchers on a fee-for-service basis, similar to the Thousand Genomes Project at Amazon. "As one reviewer from our board of scientific advisers put it, this means a smart graduate student someplace will be able to develop some new, interesting analytic software to mine this information and they'll be able to do it in a reasonable time frame," Komatsoulis says.

The unPC PC: Computer Writes Sexist Jokes (08/06/13) Richard Gray

University of Edinburgh researchers have developed software that can make up its own jokes following a simple set of rules. The system was programmed to exploit a popular comedy technique, in which a statement is followed up with a surprising comment. However, the computerized comedian has developed a rather sexist sense of humor. The system most commonly came up with jokes comparing men or women with another object. When the researchers tested the jokes on volunteers they found that they laughed, although not as much as they laughed at man-made humor, according to Edinburgh's David Matthews. He notes that for the computer-generated jokes to improve, the software would need to develop cultural awareness. "Computers have an advantage over people in that they can process masses of information, so we fed computers a wealth of material from which they extracted creative and unusual word combinations to fit our joke template," Matthews says. "The holy grail for machine-generated comedy would be to include cultural references, but these are very hard to capture."

Shadows and Light: Dartmouth Researchers Develop New Software to Detect Forged Photos
Dartmouth College (08/05/13)

Newly developed software for detecting doctored photos uses a geometric algorithm to find inconsistent shadows that are not obvious to the naked eye. The method analyzes a variety of shadows in an image to determine if they are physically consistent with a single illuminating light source. This significant step in the field of digital forensics is the work of researchers at Dartmouth College and University of California, Berkeley. The method enables people and computers to do what they excel at in understanding scene content and assessing the validity of geometric constraints, respectively, says Dartmouth professor Hany Farid. "Our method shifts the dialogue from 'does the lighting/shadow look correct?,' which is well known to be highly unreliable, to a discussion of whether an analyst has correctly selected the location of cast and attached shadows in an image, a far more objective task," Farid notes. Researchers presented their study at the recent ACM SIGGRAPH 2013 conference.

Online Gamers Harnessed to Help Disaster Response
New Scientist (08/02/13) Hal Hodson

Internet Response League co-founders Patrick Meier and Peter Mosur want to use the World of Warcraft's more than 8 million active players to analyze the digital information that accumulates during a real-world disaster. Gamers could choose to help sort through crowd-generated images from ongoing disasters, labeling the damage levels they see to give emergency responders an idea of where the most help is needed. Players would receive quests to explore and tag damage in a virtual recreation of the affected city. "If we want to engage hundreds of millions of gamers in real-world problem-solving, we need to learn how to do it inside existing games, not outside of them," says gaming advocate Jane McGonigal. The Digital Humanitarian Network, a crowdsourcing network consisting of volunteers, already has carried out work such as translation between English and Arabic for such organizations as the Red Cross and the United Nations. "Problem-solving within games like Minecraft, World of Warcraft, or even Candy Crush Saga would be genius," McGonigal says.

Making a Mini Mona Lisa
Georgia Tech News (08/05/13) Jason Maderer

Georgia Tech researchers have painted a miniature copy of the Mona Lisa on a substrate surface approximately 30 microns in width, demonstrating a technique that could be used to develop nanomanufacturing technologies because the team was able to vary the surface concentration of molecules on short-length scales. The image was created with an atomic force microscope using a process called ThermoChemical NanoLithography. The researchers positioned a heated cantilever at the substrate surface to create a series of confined nanoscale chemical reactions, and by varying the heat at each location, they were able to control the number of new molecules that were created. "By tuning the temperature, our team manipulated chemical reactions to yield variations in the molecular concentrations on the nanoscale," says Georgia Tech professor Jennifer Curtis. The researchers produced chemical gradients of amine groups, but they say the technique could be extended for use with other materials. "This technique should enable a wide range of previously inaccessible experiments and applications in fields as diverse as nanoelectronics, optoelectronics, and bioengineering," Curtis says.

Toward Harmonized Aircraft Communication
CORDIS News (08/01/13)

The European Union is funding a project that seeks to merge a full range of aircraft communications applications and services into a coherent digital architecture. Modern aircraft networks often combine decades-old analog communications and a separate satellite-based system, which makes cockpit communications complex and inefficient, says Markus Werner of project partner TriaGnoSys. The Seamless Aeronautical Networking through integration of Data links Radios and Antennas (SANDRA) project is considering ways to coordinate cabin crew operations, in-flight and on-ground passenger services, air traffic management, and security services. The four main areas that would require integrated solutions include service, network, radio, and antenna-based systems. The researchers say in the next 10 years there will be a shift from voice communications between air traffic controllers and pilots to air traffic control management systems based on data communication between computers. "SANDRA is bringing cockpit communications into the 21st century by simplifying the process for pilots and providing the platform for many more advanced services," Werner says. The project partners report that a trial of a new avionic communication system in June was a success.

Smart Phone App to Help Treat Psychosis Benefits From Government Funding
University of Manchester (07/31/13)

University of Manchester researchers recently received a grant to develop and test a smartphone app to deliver cognitive behavioral therapy (CBT) to people who have experienced a first episode of psychosis. The Active Assistance for Psychological Therapy (Actissist) app will enable patients to manage their own care at home through a mobile device. Actissist will deliver personalized CBT strategies that will enable patients to identify and manage their symptoms as part of their everyday lives. "Our ultimate goal is to make helpful treatments more widely accessible and to provide more choice about how people receive treatment, with a view to reducing the number of psychotic episodes people experience, keeping people well and out of hospital," says Manchester's Sandra Bucci. The proof-of-concept study will provide 24 first-episode psychosis patients with a CBT smartphone, and 12 people with an app designed to just monitor their psychosis symptoms. If successful, the researchers hope the app could be further developed for psychosis and applied to other mental health conditions.

An App to Lead the Blind
EurekAlert (08/01/13)

People with the debilitating disease night blindness, or nyctalopia, stand to benefit from a smartphone app designed to track the location and distance walked from home or a hotel. Developed by researchers at the Federal Urdu University of Arts, Science, and Technology, the app will warn users when they are likely to be caught out after dark. Kamran Ahsan and colleagues have designed the app to calculate both the remaining daylight hours available and estimate how long it will take the person to reach their base before nightfall. The app is geo-aware, knows the time of sunset around the world, and has access to online mapping software than can offer users shortcuts back to their base. Moreover, the app can locate nearby hotels, which would make it useful to travelers who find themselves too far from base to get home safely by nightfall.

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