Association for Computing Machinery
Welcome to the July 31, 2013 edition of ACM TechNews, providing timely information for IT professionals three times a week.

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

HEADLINES AT A GLANCE


Computer Scientists Develop 'Mathematical Jigsaw Puzzles' to Encrypt Software
UCLA Newsroom (CA) (07/29/13) Matthew Chin

Researchers at the University of California, Los Angeles (UCLA), IBM Research, and the University of Texas at Austin have developed a system to encrypt software so that it only allows someone to use a program as intended while preventing any deciphering of the code behind it. "The real challenge and the great mystery in the field was: can you actually take a piece of software and encrypt it but still have it be runnable, executable, and fully functional," says UCLA professor Amit Sahai. He says the system makes it impossible for a cybercriminal to reverse-engineer the software without solving mathematical problems that take hundreds of years to work out on today's computers. The researchers say their mathematical obfuscation method can be used to protect intellectual property by preventing the theft of new algorithms and by hiding the vulnerability that a software patch is designed to repair when the patch is distributed. Through this mechanism, attempts to determine why and how the software works will be stopped by a nonsensical jumble of numbers. Sahai notes that their method for software obfuscation has lead to the development of functional encryption, which offers a much more secure way to protect information. "Through functional encryption, you only get the specific answer, you don't learn anything else," he says.


Researchers Build Jumping, Flipping, Hand-Standing Robot
Computerworld (07/29/13) Sharon Gaudin

University of Pennsylvania researchers have created the X-Rhex Lite (XRL), a robot with six curved legs that can do a handstand, jump over gaps, and climb over rocks, all movements that could be used in hazardous situations. "Overall, we're really interested in making a robot that can go anywhere," says Pennsylvania graduate student Aaron Johnson. The researchers designed the robot with legs instead of wheels because legs can more readily maneuver on rough and uneven terrain. Furthermore, the researchers designed XRL with curved, one-joint legs, which enable it to manipulate its body into handstands, flips, and jumps. "The way the legs work naturally help get it over the rocks but they're also useful in forward jumps," Johnson says. "The [legs'] rolling contact allows Rhex to throw itself forward in a more effective manner. The challenge has really been trying to push as much performance out of the machine as possible, trying to get over the biggest gap or over the largest rock." The researchers currently are working on making the robot more proficient in traveling over rough terrain.


NIH Big Data to Knowledge (BD2K) Initiative
CCC Blog (07/26/13) Shar Steed

The U.S. National Institutes of Health (NIH) recently announced the Centers of Excellence for Big Data Computing in the Biomedical Sciences funding opportunities, the first in its Big Data to Knowledge (BD2K) initiative. BD2K aims to help biomedical scientists fully utilize the enormous amounts of data being generated by research communities. "The BD2K Centers will work in areas spanning big data science, producing tools and resources from early-stage to mature development for the biomedical research community," according to the BD2K website. Each BD2K Center award will support a multi-investor, interdisciplinary team working collaboratively to develop innovative approaches to address a major challenge to the effective use of big data for a wide range of biomedical research problems. NIH wants the BD2K Center program to cover collaborative environments and technologies, data integration, analysis and modeling methodologies, and computer science and statistical approaches.


Microsoft Launching TV White-Space Pilots in Two Nations
FierceBroadbandWireless (07/28/13) Tammy Parker

Microsoft recently announced new TV white-space (TVWS) pilots in South Africa and the Philippines. The company views TVWS spectrum as a way to bring broadband to underserved areas. In South Africa, a TVWS-based broadband network will use solar-powered base stations to deliver inexpensive Internet access to five secondary schools in remote parts of the Limpopo province. Each school will receive Windows tablets, projectors, teacher laptops and training, education-related content, solar panels for device charging where there is no access to electricity, and additional support. Microsoft hopes to show the government that TVWS is a viable option for providing low-cost broadband to a majority of South Africans by 2020. In the Philippines, the network will be used to improve registration for fishing in the Danajon Reef Marine Key Biodiversity Area in Bohol, and to access a central database for monitoring compliance in gear and vessels. Microsoft TVWS pilots also have been completed or are underway in other countries, including Kenya, Tanzania, the United Kingdom, and Singapore. Microsoft is lobbying governments to enact laws and regulations that support the technology and spectrum sharing.


Studies at the UJI Advance in Recognizing People by the Way They Walk
Jaume I University (Spain) (07/30/13) Irene Rodrigo

Jaume I University researchers are developing a biometric security technique that is based on the way people walk and their silhouette. The researchers note that their technique offers significant advantages since recognition can be done remotely and does not require the cooperation of the subject. "Although it is easy to manipulate and consciously change, each person walks in a different way," says Jaume I University's Ramon Mollineda. The researchers used a video of a subject walking to create a system that distinguishes a background silhouette. The background image becomes a sequence of silhouettes, placed one upon the other, resulting in a summary image. The final representation stores all of the subject's physical appearance and walking movements, which creates a unique mark for each subject. Mollineda notes their technique would be much more effective if it is combined with facial recognition. "They are complementary methods: the way you walk can be detected from a distance and does not require a high-resolution image, while face recognition is performed close-up and with a high-resolution image," he says.


Quantum Machine Learning for Big-Data
EE Times (07/26/13) R. Colin Johnson

Massachusetts Institute of Technology professor Seth Lloyd has developed Quantum Machine Learning, which he says is the first quantum application. "What can we use quantum computers for?" Lloyd asks. "I would like to propose the first quantum app, or q-app, which I call 'Quantum Machine Learning for Big Quantum Data.'" Q-app encodes Google-like queries with q-bits that enable quantum computers to perform real-time searches through massive databases while ensuring their absolute privacy, because attempts to eavesdrop on the query by the search engine provider would disturb the q-bit's superposition of states. The technique involves using the same type of machine-learning algorithms that have already been developed for handling big data, but encodes their queries with q-bits that can only be processed by quantum computers. Lloyd says q-app uses quantum mechanics to map large arrays of vectors onto a tensor product space that results in an exponential compression of the data set. "We get a kind of quantum compression of the classical bits into much smaller q-bit space, then use conventional machine-learning algorithms, but on the much smaller q-bit space," he says. "Many of the most popular machine-learning techniques would work in this much smaller quantum space."


Kindergarten Coders Can Program Before They Can Read
New Scientist (07/26/13) Michael Reilly

Tufts University researchers are studying the ability of kindergarten-aged children to program computers using a graphics-based coding language called ScratchJr, which is based on the Scratch programming language. Developers created the original Scratch program to teach eight-year-olds the basics of programming using graphical blocks instead of text. Targeting even younger children aged four to five years, ScratchJr works like Scratch but with easier commands. ScratchJr teaches children the essential concept of sequencing code, and complexity grows as the student advances. By the third day, ScratchJr students learn to program tasks in parallel by making a snake move through a meadow with one string of commands, while a bird flies using a separate string that works simultaneously. The benefit of early exposure to coding was highlighted earlier this year when Google engineer Neil Fraser polled more than 100 co-workers about when they first learned to code. Fraser then gave respondents a skills test and found that those who wrote their first code between the ages of eight and 11 were most likely to learn advanced coding skills. However, the U.S. government does not consider computer science a core subject, and "most computer science classes won't start until grade 10 [ages 15 to 16], if you're lucky," Fraser notes.


Make It Yourself and Save--a Lot--With 3D Printers
Michigan Tech News (07/29/13) Marcia Goodrich

Personal manufacturing from three-dimensional (3D) printers is about to enter the mainstream, says Michigan Technological University professor Joshua Pearce. He notes that 3D printers can deposit multiple layers of plastic or other materials to make almost anything, and free designs that direct the printers are now available by the tens of thousands on several websites. Users also can download designs to make their own products using open source 3D printers. After conducting a lifecycle economic analysis on 3D printing in an average U.S. household, Pearce concluded that the typical family can save a lot of money by making things with a 3D printer instead of buying them off the shelf. The researchers chose 20 common household items with downloadable designs for 3D printers, and used Google Shopping to determine the maximum and minimum cost of buying those items online, excluding shipping charges. The researchers compared the costs with the cost of making the items with 3D printers. The researchers found that it would cost the typical consumer between $312 and $1,944 to buy those items, compared to $18 to make them at home using a 3D printer. Open source 3D printers cost between $350 to $2,000, so the printers could pay for themselves in as little as a few months, according to the researchers.


Gadget Genius--Nanotechnology Breakthrough is Big Deal for Electronics
University of Akron (07/25/13) Denise Henry

University of Akron researchers say they have developed new materials that function on a nanoscale and could lead to the creation of smaller electronics. The researchers, led by Stephen Z.D. Cheng, used nanopatterning to combine functioning molecular nanoparticles with polymers to create the new materials, called giant surfactants. The giant surfactants are similar to macromolecules, but they function like molecular surfactants on a nanoscale, according to Cheng. He says that when the new materials are integrated into electronics, they will lead to the development of ultra-lightweight, compact, and efficient devices because of their unique structures. The self-assembling materials differ from common block copolymers because they organize themselves in a controllable manner at the molecular level. "The IT industry wants microchips that are as small as possible so that they can manufacture smaller and faster devices," Cheng says. The current technique can produce the spacing of 22 nanometers, but cannot go down to the 10 nanometers necessary to created tiny yet powerful devices. "This is exactly what we are pursuing--self-assembling materials that organize at smaller sizes, say, less than 20 or even 10 nanometers," Cheng says.


Facebook Invents a PHP Virtual Machine
IDG News Service (07/26/13) Joab Jackson

Facebook says it has developed a virtual machine that can run the PHP Web programming language nine times as fast as PHP natively on large systems. Facebook says it has been using the HipHop Virtual Machine (HHVM) across all of its servers since earlier this year. The HipHop compiler, devised to translate PHP code into C++, provided Facebook with considerable performance gains for several years, but the company continued to seek other ways to speed the delivery of dynamically created Web pages to its 1 billion users. Under development for about three years, HHVM works on the same principle as the Java Virtual Machine, and uses a just-in-time (JIT) compiler that converts the human-readable source code into machine-readable byte code when it is needed. The JIT approach enables the virtual machine to "make smarter decisions at runtime," says Facebook's Joel Pobar. He estimates that HHVM is about twice as fast as the previous HipHop, which has been retired within the company. Facebook has posted the code for HHVM on GitHub to give others an opportunity to use it to accelerate their PHP websites.


Computer Algorithms Reveal How Brain Processes, Perceives Sound in Noisy Environments
Scientific Computing (07/29/13)

Johns Hopkins University professor Mounya Elhilali is researching how listeners perceive sounds in various acoustic environments, with the goal of ultimately improving products that will enhance communication. "We want to understand how the brain processes and perceives sounds in noisy environments, when there are lots of background sounds," Elhilali says. "The ultimate goal is to learn how the brain adapts to different acoustic environments." A brain exposed to many competing noises engages in bottom-up processes that occur when a person's hearing is dominated by many surrounding noises, and top-down processes that take place when the focus is on a particular sound, such as a conversation, that is controlled by a person's mindset. Elhilali and her colleagues plan to write algorithms that mimic the brain's perception of sound in a complex environment to create computers "that can hear like a human brain can hear," she says. "We are trying to build a computer brain that can process sounds in noisy environments like the human brain, so you can talk to the computer and it can understand what you are saying." The technology is likely to have many engineering and communication applications, for example, in cellphones, hearing aids, and telephone-automated systems.


The 'Maker Movement' Inspires Shift in STEM Curriculum
Center for Digital Education (07/22/13) Tanya Roscorla

New curriculum from the Digital Harbor Foundation, BatelleEd, and Arizona State University reflects a shift that is occurring in education, from passive to active learning, inspired by the maker movement. The mindset is one of wanting producers and someone who is not just making a website to learn, says Digital Harbor executive director Andrew Coy. Students will get to apply what they learn immediately. The science, technology, engineering, and math (STEM) Core curriculum emphasizes inquiry, projects, and production around app development, and at the end of the process students will have an app to market in app stores. The schools that plan to use STEM Core this fall will not be giving students information to regurgitate, but will have students find solutions to problems that are rarely addressed. Students will use Corona SDK as a mobile development framework and Lua as the programming language, and their apps can be exported to the native iOS and Android platforms once they are finished. "All innovations and innovation economies rely on this ability to solve a currently unsolved problem, but so much in education revolves around solving questions that already have known answers," Coy says.


Q&A: Tiffani Williams, Computer Scientist, on Creating an Open Source Tree of Life
SmartPlanet (07/29/13) Christina Hernandez Sherwood

In an interview, Texas A&M University computer scientist Tiffani Williams discussed the Open Tree of Life project, which aims to develop an open source compendium of existing knowledge on thousands of plant and animal species that would serve as a research tool for both scientists and the public. "We as human beings have this notion of appreciating our family history," Williams says. "All the tree of life does is take that to another level. Instead of thinking of a family in terms of your human ancestors, the tree of life is the world’s ancestry, which includes all of the world’s organisms." Scientists are currently aware of 1.7 million species, but estimates run as high as 100 million organisms in existence. Williams notes that one of the challenges of the project will be to merge the various trees of life into one unified tree. "It could be that two findings just outright conflict each other," she says, noting that researchers are uncertain as to how best to represent conflicting opinions in the unified tree. The project depends on contributions from the community, and Williams hopes the tree will ultimately achieve a level of usefulness in phylogenic research that equals that of Wikipedia for general research.


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