Association for Computing Machinery
Welcome to the January 24, 2014 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


Expert Panel to Investigate Internet Governance
Financial Times (01/22/14) Sam Jones; John Grapper

The Global Commission on Internet Governance will open a two-year investigation into the way governments use Internet data following revelations of mass online spying by the United States and its allies. The 25-member panel, an international think tank launched this week at the World Economic Forum, will scrutinize the way governments use Internet data and propose rules to protect citizens' rights online. The commission will investigate a range of topics, focusing on Web governance and the growing state control of online activities. The Internet is under threat from a "loss of trust" by its users and rising state authoritarianism, according to the commission. "We need the ability to help our users understand how many requests we are getting because we need to rebuild trust with [them]," says Yahoo CEO Marissa Mayer. The independent commission will report on the possibility of creating international laws to protect human rights online and ensure Internet stability. "We need some rules of the road that everyone can live with," says Cisco CEO John Chambers. "It has been the wild, wild west around the world and we need some guidelines that we can all live by."
View Full Article - May Require Free Registration | Return to Headlines | Share Facebook  LinkedIn  Twitter 


Big Web Crash in China: Experts Suspect Great Firewall
The New York Times (01/22/14) Nicole Perlroth

A massive Internet failure in China on Tuesday prevented most of the country's 500 million Internet users from accessing websites for up to eight hours, in an incident that technology experts believe stemmed from Web censors known as the Great Firewall of China. Experts speculate the Chinese government mistakenly redirected traffic to sites that are typically blocked in China. The state-run China Internet Network Information Center said the incident was tied to the country's domain name system (DNS). About 75 percent of China's DNS servers were impacted, according to Qihoo 360 Technology. The servers routed traffic from some of China's most popular sites to an Internet address reportedly registered to a company called Sophidea. Although Sophidea's physical server location is unclear, the firm appears to operate a service that reroutes Internet traffic from one website to another to conceal a person's whereabouts. Such a service could be used to circumvent firewalls, which prompted experts to believe the Chinese government might have caused the outage by trying to block traffic to Sophidea's website. Further supporting this theory was the concurrent redirecting of Chinese Internet traffic to Internet addresses owned by Dynamic Internet Technology, which helps users get around China's Great Firewall and ordinarily is blocked.


Ford, Stanford, and MIT Research Giving Self-Driving Cars 'Intuition'
Los Angeles Times (01/22/14) Jerry Hirsch

Ford Motor is working with researchers at the Massachusetts Institute of Technology (MIT) and Stanford University to give self-driving cars intuition. For example, an automated vehicle that is behind a big truck would be able to determine what is in front of the truck. Moreover, if the truck slammed on its brakes, the vehicle would know if the area around it is clear to safely change lanes. "Our goal is to provide the vehicle with common sense," says Ford's Greg Stevens. "Drivers are good at using the cues around them to predict what will happen next, and they know that what you can't see is often as important as what you can see." Stanford University researchers will examine how a car's LiDAR infrared sensors could see around obstacles so it can make evasive maneuvers when needed. Ford will work with MIT to generate a three-dimensional map of a vehicle's surrounding environment through the use of LiDAR, and algorithms will be used to help cars predict the location of moving vehicles and pedestrians. The mapping technology potentially could enable cars to plan a safe path for avoiding pedestrians, bicycles, and other vehicles.


Tor Exit Nodes Attempt to Spy on Encrypted Traffic, Researchers Find
IDG News Service (01/22/14) Lucian Constantin

Karlstad University computer scientists Philipp Winter and Stefan Lindskog say they have identified several exit relays on the Tor privacy network that engaged in man-in-the-middle attacks and other malicious behavior targeting the traffic that passed through them. Winter and Lindskog designed a scanning tool called exitmap and ran it on the Tor network over the course of four months. They identified 19 exit relays that executed man-in-the-middle attacks using HTTPS traffic sniffing, SSH traffic sniffing, or both. Other relays forced HTTPS connects to convert to plain HTTP, injected HTML code into traffic, or blocked access to certain websites. Winter and Lindskog believe the majority of the nodes performing man-in-the-middle attacks were run by the same person or group, as they used similar self-signed certificates and most of them were located on the same Russian virtual private server hosting service. However, the researchers say the measures taken to conceal the attacks made them largely ineffective. "I'm not even sure if they captured passwords," Winter notes, pointing out the attacks may have just been an experiment. However, Winter and Lindskog say the attacks show that traffic over the Tor network can be vulnerable, and they introduced a browser extension that can inform Tor users if a man-in-the-middle attack is potentially in progress.


Bringing the World Reboot-less Updates
MIT News (01/24/14) Rob Matheson

In 2008, a group of Massachusetts Institute of Technology (MIT) researchers began developing Ksplice, software that automatically applies patches to an operating system without having to reboot. Ksplice compares changes between the old and updated code and implements the changes into a running Linux kernel. The software also removes the need for programmer intervention with the code, which decreases the cost and risk of error, says Jeff Arnold, who led the research, which eventually became a commercial product. "The aim is to allow administrators the benefit of the update while eliminating both the cost and downtime for the users," Arnold says. Ksplice constructs hot patches using the object code with pre-post differencing, which creates object code before a patch and object code are modified by the patch on the fly. Ksplice also uses a technique called run-pre matching, which computes the address in computer memory of ambiguous code by using custom computation to compare the "pre" code with the finalized, running kernel. "Something that was key to our success was we had a good network of mentors and advisors, and many were part of the MIT community," Arnold says. Ksplice has only ever run on Linux systems, but its developers say it could be ported to other operating systems.


Microsoft Unveils Primary School Suite for New Computing Curriculum Teachers
ComputerWeekly.com (01/21/14)

U.K. primary school teachers who have never taught computer science before can turn to a new technology suite from Microsoft to help prepare for the task. The United Kingdom will usher in a new national computing curriculum in nine months. The Switched on Computing material, developed in partnership with educational publisher Rising Stars and available for free, is part of Microsoft's First Class Computing program. The material is divided into separate units for each year-group from one to six, and is aimed at developing logical thinking and problem-solving skills. Primary school teachers will be introduced to new concepts such as algorithms. Teachers compiled the materials and tested them to develop computer science skills in children as young as five. "We welcomed the news of the new computing curriculum alongside others in the industry because it is absolutely critical for the future success of our young people," says Microsoft U.K.'s Steve Beswick. "The challenge now is to ensure that primary teachers are equipped to deliver it by September." Rising Stars managing director Andrea Carr notes the new national curriculum will be challenging for primary schools. "We think these free resources are exactly what teachers need to get started with computing," Carr says.


Neural Nets: Now Available in the Cloud
Technology Review (01/21/14)

University of Vienna computer scientists Erich Schikuta and Erwin Mann have developed a system called N2Sky that sets up neural networks in the cloud so their services can be shared like other computing resources. Creating neural networks is challenging, and researchers in different labs have established disparate, mostly incompatible methods for accomplishing neural network-based tasks. Schikuta and Mann wanted to enable any user to access neural networks as a shared resource to solve problems or work with geographically remote people on neural networks. "We present the N2Sky system, which provides a framework for the exchange of neural network-specific knowledge, as neural network paradigms and objects, by a virtual organization environment," the researchers say. N2Sky will enable anyone to sign in and use neural networks in a validated method via the cloud. In addition, Schikuta and Mann plan to create a search engine for neural networks that will let users type in a problem to find nets that have successfully solved the specific problem or a similar one.


CU-Built Software Uses Big Data to Battle Forgetting With Personalized Content Review
CU-Boulder News Center (01/21/14) Laura Snider

Researchers at the University of Colorado (CU) Boulder’s Institute for Cognitive Science have developed software that helps students remember the content they have already studied. The software works by tapping a database of past student performance to suggest what material a specific student most needs to study. "If you have two students with similar study histories for specific material, and one student couldn't recall the answer, it's a reasonable predictor that the other student won't be able to either, especially when you take into consideration the different abilities of the two students," says CU Boulder professor Michael Mozer. He says the process of combing big data for performance clues is similar to strategies used by e-commerce sites. As part of the study, the researchers divided the material students were learning into groups based on when the material was originally covered. In a cumulative test taken a month after the semester's end, "personalized-spaced" review boosted remembering by 16.5 percent over massed study and by 10 percent over "generic-spaced" review.


Cryptography: Unsafe and Sound
The Economist (01/18/14)

Weizmann Institute professor Adi Shamir (who received the 2002 ACM A.M. Turing Award, along with Ronald Rivest and Leonard Adleman) and colleagues have exploited acoustic leakage to eavesdrop on a computer. Their method, known as acoustic cryptanalysis, involves listening to the noise capacitors and coils make as they vibrate in response to the amount of power being drawn by the processor. The team applied their method to GnuPG, a popular version of RSA, extracting full 4,096-bit keys from a range of laptops in less than an hour, which would enable anyone to read encrypted messages sent to those computers. The researchers sent to laptops encrypted emails carefully crafted so the acoustic leakage produced by encrypting them was specifically related to the value of particular bits in the key. A series of such texts, each building on knowledge gleaned from the previous attack, gradually builds up the whole number. The decrypted text will appear as nonsense, but emails can be made to look like spam or backdated so the attack goes unnoticed. An old-fashioned bugging microphone or a parabolic mike can be used, but an attacker also could hijack the target computer's microphone or place a programmed smartphone close enough to both send emails and do the analysis. The researchers suggest a spy could even touch a target computer when it is decrypting an email.


Ten-Year U.S. Exascale Roadmap Crystalizes
HPC Wire (01/17/14) Tiffany Trader

In late December, President Barack Obama signed a bill into law that included the directive for the U.S. Department of Energy (DOE) to develop exascale computing capability over the next decade for a nuclear stockpile stewardship program. However, a month before, DOE officials laid out a 10-year roadmap as part of the Advanced Scientific Computing Advisory Committee (ASCAC) meeting in Denver. The DOE Office of Science envisions an exascale computing system that is productive and based on marketable technology. The proposed execution strategy covers the research, development, applications, facilities, and integration needed to deploy a system in the early 2020s. The Office of Science is pursuing partnerships with government, the computer industry, DOE labs, academia, and the international research community to reach this goal. The ASCAC exascale subcommittee study on the top technical challenges includes energy efficient circuit, power, and cooling technologies; high performance interconnect technologies; advanced memory technologies to dramatically improve capacity and bandwidth; scalable system software that is power and resilience aware; and data management software that can handle the volume, velocity, and diversity of data storage.


Facebook Memes Can Evolve Like Genes
University of Michigan News Service (01/16/14) Nicole Casal Moore

Researchers at the University of Michigan and Facebook have found that changes in a Facebook meme in support of health care reform as it spread over the network were similar to the evolution of biological genes. During an 18-month period between 2009 and 2011, the researchers studied 460 million shares of more than 1,000 text status-update memes, which are intended to be shared by users agreeing with the message. The researchers found that 89 percent of copies of the health care reform message were exact replicas, and 11 percent mutated the message in some way. The team modeled the meme evolution using the Yule process, which states the frequency of various gene mutations within a population. "It surprised me that the model fit really well," says Michigan professor and Facebook data science team member Lada Adamic. "We can't claim that memes evolve exactly like genes, but there are a bunch of very convincing parallels." One such parallel is lateral gene transfer, in which useful genetic information is passed laterally among different individual organisms, instead of down between generations. Mutations in both genes and memes also occur more frequently at the beginning or end of a sequence, with shorter sequences of similar information spreading more widely.


Human Arm Sensors Make Robot Smarter
Georgia Tech News Center (01/16/14) Jason Maderer

Georgia Institute of Technology researchers have developed a control system they say makes robots more intelligent. The system is designed to improve the safety and efficiency of manufacturing plants by eliminating unnecessary vibrations through the use of sensors worn on a controller's forearm. The device sends muscle movements to a computer, which provides the robot with the operator's level of muscle contraction. The technology assesses the operator's physical status and adjusts how it should interact with the human. "Instead of having the robot react to a human, we give it more information," notes Georgia Tech researcher Billy Gallagher. "Modeling the operator in this way allows the robot to actively adjust to changes in the way the operator moves." The research is designed to benefit groups interested in the adaptive shared control approach for advanced manufacturing and process design, including the automotive, aerospace, and military industries. "Future robots must be able to understand people better," says Georgia Tech professor Jun Ueda. "By making robots smarter, we can make them safer and more efficient."


E-Whiskers: Berkeley Researchers Develop Highly Sensitive Tactile Sensors for Robotics and Other Applications
Berkeley Lab News Center (01/20/14) Lynn Yarris

Researchers at Lawrence Berkeley National Laboratory and the University of California, Berkeley have created tactile sensors from composite films of carbon nanotubes and silver nanoparticles similar to the highly sensitive whiskers of cats and rats. The researchers say their e-whiskers are very sensitive and could give robots new abilities to "see" and "feel" their surrounding environments. "Our electronic whiskers consist of high-aspect-ratio elastic fibers coated with conductive composite films of nanotubes and nanoparticles," says Berkeley professor Ali Javey. "In tests, these whiskers were 10 times more sensitive to pressure than all previously reported capacitive or resistive pressure sensors." The researchers used a carbon nanotube paste to form an electrically conductive network matrix with excellent bendability. They then loaded a thin film of silver nanoparticles that endowed the matrix with high sensitivity to mechanical strain. "The composite can then be painted or printed onto high-aspect-ratio elastic fibers to form e-whiskers that can be integrated with different user-interactive systems," Javey says. The researchers say they successfully used their e-whiskers to demonstrate highly accurate two- and three-dimensional mapping of wind flow. "The ease of fabrication, light weight, and excellent performance of our e-whiskers should have a wide range of applications for advanced robotics, human-machine user interfaces, and biological applications," Javey predicts.


Abstract News © Copyright 2014 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