Association for Computing Machinery
Welcome to the March 21, 2016 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).


Johns Hopkins Researchers Poke a Hole in Apple's Encryption
The Washington Post (03/21/16) Ellen Nakashima

A research team from Johns Hopkins University has successfully cracked Apple's iMessage encryption by exploiting a bug that would enable hackers to decode photos and videos sent as secure instant messages. The researchers wrote software to emulate an Apple server, and the encrypted transmission they targeted had a link to a photo stored in Apple's iCloud server and a 64-digit decryption key. The team guessed the digits by changing a digit or a letter in the key and sending it back to the target phone to see what would be accepted, and they repeated this process thousands of times until the key was revealed. Johns Hopkins professor Matthew D. Green says this breakthrough disproves the notion that strong commercial encryption is hack-proof for either hackers or law enforcement. He also says such methods make court orders compelling companies to create software to open security unnecessary. "It scares me that we're having this conversation about adding backdoors to encryption when we can't even get basic encryption right," Green notes. He urges users to update their phones and laptops to iOS 9.3 as a preventive measure. The American Civil Liberties Union's Christopher Soghoian says the exploit illustrates the danger of companies building their own encryption without independent vetting.
View Full Article - May Require Free Registration | Return to Headlines | Share Facebook  LinkedIn  Twitter 

Utah Republicans Open Caucuses to Online Voters
The Wall Street Journal (03/18/16) Byron Tau

Utah residents this week will be able to cast ballots in the Republican presidential contest using computers, tablets, and smartphones, representing the largest online presidential voting experiment since 2004. Republican voters must complete a separate registration to vote online, and then they are given a personal identification number that will permit them to vote. The online polls will be open all day Tuesday and will produce a receipt that verifies the correct recording of each participant's vote. "We expect all the jurisdictions across the U.S. to take notice and to look at this experience as something to study and, hopefully, follow," says Smartmatic Group CEO Antonio Mugica, whose company is running the Utah election. He notes Smartmatic has security protocols and backups of the election data in case a recount is needed. E-voting's adoption has been slow over the past two decades, partly due to security issues concerning election integrity. "The Internet...was not built for security when it was built," says Verified Voting president Pamela Smith. "It was built for open communications." President Barack Obama recently said the Internet should be used to expand voter participation and called on the technology community to determine "how can we create safe, secure, smart systems for people to be able to vote much easier online."
View Full Article - May Require Paid Subscription | Return to Headlines | Share Facebook  LinkedIn  Twitter 

When Slower Is Faster
MIT News (03/17/16) Peter Dizikes

The amount of vehicular traffic using roads could be doubled with a traffic-signal-free transportation design involving communication between sensor-equipped vehicles, according to a new study co-authored by Massachusetts Institute of Technology (MIT) researchers. The study mathematically modeled the theoretical effect of slot-based intersections (SIs), which eliminate waits for traffic lights, on the speed of traffic flow. Carlo Ratti with MIT's SENSEable City Lab says this setup "moves control from the [traffic] flow level to the vehicle level. Doing that, you can create a system that is much more efficient, because then you can make sure the vehicles get to the intersection exactly when they have a slot." The SENSEable City Lab's Paolo Santi notes the SI system's increased capacity stems from establishing a more consistent flow of traffic at an optimal middle speed, and its basis is on the "slower is faster" effect. "If you need to slow down the vehicles because there is a lot of traffic, you slow them down early in the road, so they approach the intersection at slow speed, but then when they cross, you use the best speed," Santi says. The modeling in the study determined the most efficient traffic flow batches vehicles together and then slides them through the intersection accordingly.

Why Students Are Throwing Tons of Money at a Program That Won't Give Them a College Degree
The Washington Post (03/21/16) Karen Turner

Computer-coding "boot camps" are surging in popularity, leading to an explosion of technology job applicants who graduate from such programs. The catch is these non-accredited and little-regulated programs do not offer a college degree in computer science, so graduates who invest a lot of money in such courses are often denied the high-paying positions such boot camps promise. Colleges and universities are now partnering with these for-profit outfits or launching their own accelerated coding workshops for students, but tech recruiters are concerned about this trend. "A lot of the people that may get coerced into signing up for these boot camps may end up with a lot of debt and not a lot of job offers," warns recruiter Dave Fecak. Driving the boot-camp trend is an abundance of market demand for skilled tech workers, but while typical computer science degrees emphasize theoretical programming principles as well as coding, boot camps exclusively concentrate on programming. "You emerge from a boot camp fit to do an oil change, but not design a car," contends Monetate's Brian O'Neill. Some experts see advantages to the boot camps, such as their adaptability to industry trends and their tendency to focus on skills rather than theory. However, critics are worried boot-camp graduates may lack the basics to acquire new tool and skill sets that are necessary to have an evolving career.
View Full Article - May Require Free Registration | Return to Headlines | Share Facebook  LinkedIn  Twitter 

App to Tackle Impact of Injuries on NFL Players
The Wall Street Journal (03/17/16) Ron Winslow

Researchers last week rolled out a new smartphone app to enable former National Football League (NFL) players and the general public to collectively participate in a clinical study to evaluate how on-the-field injuries affect the long-term health of players after they retire. The TeamStudy app, which was developed with Apple's ResearchKit, uses questionnaires and smartphone functions such as recording steps, distance walked, and heart rate to glean data feeding into players' physical and cognitive health. Researchers want members of the public to download TeamStudy and act as a control group. Having information from a diverse cohort of people will help researchers differentiate between "what is an artifact from your playing career and what is happening because you're 50 years old and live an active lifestyle," says project adviser and former NFL player Chad Brown. The app's features initially will look for data on pain and mobility, heart health, and cognitive difficulties, says Harvard Medical School professor Alvaro Pascual-Leone. He notes the study will give researchers the means to collect real-time data on ex-players' daily experiences, while rapid feedback will enable subjects to compare their health status against the average of colleagues who played similar positions or for a similar number of years. Researchers intend to link player clinical records to data compiled by TeamStudy to gain knowledge about the interplay of such factors with the impact on the players' careers.
View Full Article - May Require Paid Subscription | Return to Headlines | Share Facebook  LinkedIn  Twitter 

Scientists Develop a 100 Times Faster Type of Memory Cell
Moscow Institute of Physics and Technology (03/16/16)

A new type of memory cell based on superconductors would be hundreds of times faster than the types of memory devices commonly used today, according to scientists in Russia. Researchers from the Moscow Institute of Physics and Technology (MIPT) and the Moscow State University propose creating basic memory cells based on quantum effects in "sandwiches" of a superconductor. Called Josephson junctions, the electrons in these sandwiches are able to tunnel from one layer of a superconductor to another, passing through the dielectric like balls passing through a perforated wall. The researchers note the data in Josephson junctions could be encoded in the value of the superconducting current. They say the system could be switched from "0" to "1" and back again by using injection currents flowing through one of the layers of the superconductor. The method requires only one ferromagnetic layer, so there is no need to create a new architecture for a processor. "A computer based on single flux quantum logic can have a clock speed of hundreds of gigahertz, and its power consumption will be dozens of times lower," says Alexander Golubov, head of MIPT's Laboratory of Quantum Topological Phenomena in Superconducting Systems.

Rumor Has It an Algorithm Could Scope Out Gossip
Inside Science (03/15/16) Marcus Woo

Mathematical studies focused on tracing the point of origin of a rumor that has spread across a network could help track down the sources of memes on social media, trends, computer viruses, and epidemics, but recent research says network complexity influences whether such points of origination are findable. "The network structure basically dictates when you can and can't find someone," says the Massachusetts Institute of Technology's (MIT) Tauhid Zaman. "People hadn't made that connection yet, that the network complexity tells you when you can hide or when you can't hide." Zaman says overly simple networks thwart the tracing of rumor sources, but there is a one-in-three chance the source can be found with just a little network complexity. Finding a rumor's source requires counting how many ways it could have spread from each network member, and the one who could spread the rumor through the most number of paths is likely the originator. To calculate that probability, Zaman and fellow MIT researcher Devavrat Shah first applied an algorithm to a tree network, and later used computer models to show their results were applicable to most other networks. "I won't even pretend it's a really game-changing algorithm," Zaman says. "It's a cool idea, but it is theoretical in nature."

Social Networks Used in the Assessment of Damage Caused by Natural Disasters
Charles III University of Madrid (Spain) (03/14/16)

Data from social networks can be use to determine the damage caused by natural disasters in just a few hours, according to an international scientific study. Researchers from the Charles III University of Madrid, the National Information Communications Technology Australia, and the University of California, San Diego analyzed hundreds of millions of geo-located tweets before, during, and after Hurricane Sandy in 2012 and found a strong correlation between the route of the storm and activity on the social network. They examined the data alongside data on the levels of aid granted by the U.S. Federal Emergency Management Agency and insurance claims, and also found a correlation between the mean per capita of social network activity and economic damage per capita caused by disasters in the areas where such activity occurs. The team was able to demonstrate the same dynamic in the case of floods, storms, and tornadoes. The researchers believe social media can provide additional information for assessing the damage from disasters. Moreover, the distribution space of messages could help authorities monitor and evaluate emergencies as well as improve response, according to the researchers.

Kindergarten for Robots Is a Lot Like Kindergarten for Kids
Motherboard (03/14/16) Daniel Faggella

The Max-Planck Institute for Brain Research's Danko Nikolic believes robots can be taught using a similar approach to how children are educated in the primary grades. His vision of an artificial intelligence (AI) kindergarten is of human trainers and educators teaching robots in a playroom as they exercise calibrated interactions while sensing the real world, beginning with simple actions and eventually generating their own rules for learning new information. Nikolic sees limits with current models of deep learning in terms of creating a truly intelligent entity that can learn from its mistakes and adjust its knowledge according to its environmental engagement. His idea is to apply the theory of practopoiesis to deep learning; the theory holds that organisms function and behave based on how their systems are structured. Different adaptive systems are assigned to different adaptive categories defined by the number of levels of organization at which the system receives environmental feedback, also known as traverses. Nikolic says AI systems must have three traverses--or T3-systems--to match a human: they can learn from mistakes, store past experience abstractly, and apply that knowledge in a more adaptive and efficient manner than two-traversal systems. Nikolic wants to equip robots with basic learning rules via a machine genome.

Record Setting Leap Towards Ultra-Secure Communication
Swinburne University of Technology (Australia) (03/11/16) Lea Kivivali

An international team of researchers has set a new record for the complexity possible on a quantum computing chip, which they say brings the field one step closer to ultra-secure telecommunications. The researchers created entangled photon states with unprecedented complexity and over many parallel channels simultaneously on an integrated chip. The researchers note the chip also was created with processes compatible with the existing semiconductor industry, a breakthrough they say could lead to the incorporation of quantum devices directly into laptops and cell phones. The researchers, led by Swinburne University of Technology professor David Moss, used optical frequency combs that help to "tangle" photons on a computer chip. The research set a new record in both the number and complexity of entangled photons that can be generated on a chip to help lead to ultra-secure communications. In addition, the research has direct applications for quantum information processing, imaging, and microscopy. "Not only can we generate entangled photon pairs over hundreds of channels simultaneously, but for the first time we've succeeded in generating four-photon entangled states on a chip," Moss says. The team also included researchers from City University of Hong Kong, University of Sussex, Herriot University, Yale University, and the Xi'an Institute.

Toward the Internet of Things--a Framework for Data Analytics on Digital Device Networks
Rensselaer Polytechnic Institute (03/11/16) Mary L. Martialay

Rensselaer Polytechnic Institute (RPI) researchers are developing tools and frameworks that can be used to perform data analytics over a range of devices connected to the Internet of Things. "Rather than developers designing custom algorithms for each network of devices, we're going to build a framework of software that sits on all these devices and the cloud that will automatically manage communication between the devices and deal with device and network failures," says RPI professor Stacy Patterson. The research is supported by a five-year, $618,661 U.S. National Science Foundation Faculty Early Career Development Award. The project is an extension of Patterson's current research to enhance the utility of sensors embedded in automobiles by creating real-time networks that enable cars to combine their individual data into a larger shared picture of the driving environment. To achieve her goals, Patterson must develop a system that can quickly analyze and learn from the massive amount of generated data. She aims to develop a computational framework that reduces the problem to an abstraction, anticipating considerations such as the type and quality of data, the number of devices, and how the data is related across devices. The researchers also want to provide a stable platform that masks the differences between devices, and compensates for failed devices or computers and list data.

Brain scan: Bruno Michel
The Economist (03/12/16)

Bruno Michel, head of the advanced micro-integration team at IBM's Zurich research lab, says the inherent inefficiency of today's computers in terms of space and electricity consumption has a bearing on their information-processing abilities. "Nowadays, the cost of buying a computer or a data center is less than the cost of running it for a few years," Michel notes. His current focus on more energy-efficient computer chips is targeting a solution inspired by biology. Michel's specific inspirations are the fractally branching blood vessels feeding energy into the brain, enabling the bulk of its volume to be dedicated to data processing. He estimates this architecture makes the human brain's efficiency about 10,000 greater than that of the most advanced silicon-based computers. Michel hopes to mimic the density of the brain--and its efficiency--as part of a project to build an electronic version of the blood that diverts energy to the brain. "It was something like 200 years after the invention of the steam engine before mechanical engineering began to catch up with biology in terms of efficiency," Michel notes. "It would be good if computing could accomplish the same thing in half the time."

Abstract News © Copyright 2016 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