Association for Computing Machinery
Welcome to the October 24, 2012 edition of ACM TechNews, providing timely information for IT professionals three times a week.

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


Federal Budget Limits Affect Scientific Conferences
New York Times (10/23/12) Laura Dattaro

The Obama administration recently imposed new guidelines that limit the amount of money federal agencies can spend on regional conferences. However, several science and technology organizations say the federal budget limits are negatively affecting the scientific community's ability to share research and collaborate. "This is a problem not just for the computing research community, but for almost anyone who's involved in scientific work," says ACM president Vinton G. Cerf. Many organizations have written to Congress and federal officials asking for an exemption from the spending policy for recognized scientific, technical, and educational meetings, as well as meetings of national and international standards bodies. "The inability of the government researchers and program managers to participate in these conferences is actually very damaging," Cerf says. The new policy also has created confusion among federal agencies, because individuals do not know how many other agency employees also are attending, and do not know if the agency is close to its federally enforced $100,000 spending limit. Cerf notes that in an era of high unemployment and shifting markets, scientific collaboration is crucial, and limiting those opportunities is not good for the United States.

Contextual Intelligence: Smart Phones to Become Big Brother?
BYTE (10/22/12) Boonsri Dickinson

PARC researchers are developing contextual intelligence technologies that could enable enterprises and governmental bodies to use data in the same way citizens use mobile phones. The researchers say data mined from email messages, Facebook conversations, and phone sensors could be used for intelligence gathering, marketing and application design, and employee relations. PARC's Oliver Brdiczka says researchers are working on a project that predicts a person's personality based on their online behavior, with the goal of marketing this data to enterprises that want to know people's intent for targeted advertising or developing content customization. The researchers found that activity levels can help predict levels of happiness. The government could use this information to identify depressed soldiers, and companies could use it to determine if employees are unhappy. Unhappy people "get much less expressive, so the amount of sentiment goes down--both positive and negative," Brdiczka notes. Companies also could use the data to customize ads so they are more likely to be perceived as recommendations than advertisements. Brdiczka predicts that in the future contextual intelligence systems will be built into operating systems and will be able to assist users.

Training Your Robot the PaR-PaR Way
Lawrence Berkeley National Laboratory (10/23/12) Lynn Yarris

Researchers at the U.S. Department of Energy's Joint BioEnergy Institute (JBEI) have developed Programming a Robot (PaR-PaR), a high-level, biology-friendly, robot programming language that enables scientists to make better use of liquid-handling robots. "The syntax and compiler for PaR-PaR are based on computer science principles and a deep understanding of biological workflows," says JBEI researcher Nathan Hillson. "Our vision was for a single protocol to be executable across different robotic platforms in different laboratories, just as a single computer software program is executable across multiple brands of computer hardware." He says PaR-PaR enables biologists to manually instruct robots in a time-effective manner, and improves the utility of biological design automation software tools. The open source PaR-PaR language uses an object-oriented approach that represents physical laboratory objects as virtual objects. "Flexible and biology-friendly operation of robotic equipment is key to its successful integration in biological laboratories, and the efforts required to operate a robot must be much smaller than the alternative manual lab work," Hillson says. "PaR-PaR accomplishes all of these objectives and is intended to benefit a broad segment of the biological research community, including nonprofits, government agencies, and commercial companies."

In Cyberattack on Saudi Firm, U.S. Sees Iran Firing Back
New York Times (10/24/12) Nicole Perlroth

The cyberattack that took down nearly three-quarters of the computers at the Saudi Arabian oil company Saudi Aramco in August was among the most damaging acts of computer sabotage on a corporate entity to date, and many security experts and intelligence officials believe it originated in Iran. Investigators believe one or more company insiders infected Aramco's network with a virus dubbed Shamoon, possibly using a USB drive, which was programmed to activate itself at a specific time. Shamoon blitzed the Aramco corporate network, deleting three-quarters of its computers’ files, documents, and emails and rendering computers useless by overwriting their system files with images of a burning American flag. U.S. Defense Secretary Leon E. Panetta recently called the Aramco attack "a significant escalation of the cyber threat." File names and other clues in the code of Shamoon have led officials to conclude that the attack was meant as retaliation for the Flame and Wiper malware that have been spying on and undermining Iranian oil companies earlier this year. It still remains unclear who created Flame and its related malware, but many believe it is the product of the same joint U.S.-Israeli cyberwarfare program that generated the Stuxnet computer worm.

Faster Chips 'Cut Cloud-Computing Bills'
BBC News (10/23/12)

Researchers at Deutsch Telekom Laboratories and Aalto University have found that customers of Amazon's EC2 cloud service do not receive the same level of performance. Amazon says it uses generic hardware, but the team used tools to examine the software that controls the groups of servers customers rent, and was able to identify the chip at the heart of each server in a group or instance of computers. Measurements taken over the course of a year revealed instances running newer, faster chips, and they were much faster than clusters that used older hardware. "In general, the variation between the fast instances and slow instances can reach 40 percent," the researchers wrote in a paper, noting that for some applications the newer clusters worked about 60 percent faster. The faster instances would enable users to reduce their server bills by up to 30 percent because the newer machines are able to crunch data faster. The team is now working on tools that can determine the performance characteristics of particular clusters and push work to more powerful groups.

A Bandwidth Breakthrough
Technology Review (10/23/12) David Talbot

Academic researchers have developed coded Transmission Control Protocol, a method for improving wireless bandwidth by one order of magnitude that involves using algebra to overcome the network-clogging task of resending dropped packets. The technology provides new ways for mobile devices to solve for missing data, and can combine data streams from Wi-Fi and LTE. Massachusetts Institute of Technology researchers tested the system on standard Wi-Fi networks, where 2 percent of packets are normally lost, and found that a normal bandwidth of 1 Mbps was boosted to 16 Mbps. The researchers also tested the technology in the Amazon cloud. Internet Protocol traffic was sent to Amazon, encoded, and then decoded as an application on phones. The technology sends algebraic equations that describe a series of packets. That way, if a packet is lost, the receiving device can solve for the missing packet instead of asking the network to resend it. If the technology works in large-scale deployments as the researchers expect, it could help delay a spectrum bottleneck.

Popular Android Apps Vulnerable
InformationWeek (10/22/12) Thomas Claburn

Researchers from Leibniz University and Philipps University of Marburg have found that 8 percent of Android apps featured weak implementation of secure sockets layer (SSL) encryption standards. The researchers analyzed 13,500 free Android apps available in the Google Play app store and found potentially vulnerable SSL/Transcript Security Layerst code in 1,074 of them. The researchers say these apps were potentially vulnerable to man-in-the-middle attacks, and found such vulnerabilities in 41 out of 100 apps subjected to a manual audit. The researchers were able to capture credentials for credit cards, bank accounts, social media, and email accounts, and numerous other services from those 41 apps, three of which had subscriber bases of 10 million to 50 million users. The researchers note that the default Android browser was "exemplary in its SSL use" and that most of the problems came from third-party apps, and they offer several suggestions for improving security on the Android platform. The suggestions include disallowing the use of custom SSL code in Android apps, implementing HTTPS-Everywhere, and better presenting Internet permissions and security information to users.

Using Big Data to Save Lives
UCR Today (10/22/12) Sean Nealon

University of California, Riverside researchers have developed a method for mining data derived from pediatric intensive care units to help doctors treat children and cut health care costs. "This data has the potential to be a gold mine of useful--literally life saving--information," says Riverside professor Eamonn Keogh. He notes that modern pediatric care units are equipped with a variety of sensors that record up to 30 measurements. The researchers developed a technique that makes it possible to search the sensor datasets, which can contain more than one trillion objects. The researchers also are exploring ways to capture and store data from five or more sensors, and capture multiple data points per second. In the next few years, the researchers plan to study archived pediatric intensive care unit data to find common patterns that can help doctors in diagnosing and predicting medical episodes. The researchers also want to incorporate those patterns into intensive care unit sensors. However, the difficulty is in finding medically useful patterns because there are an infinite number of trivial patterns. "We have to find those that aren’t known but are useful and that can benefit from intervention," Keogh says.

Intel Strives to Develop Tiny Chips to Run Wearable Computers
Computerworld (10/22/12) Sharon Gaudin

Intel researchers are developing tiny microprocessors that would power wearable computers. "These will be wearable computers that are very small and unobtrusive," says Intel's Manny Vara. "Imagine wearing something that would tell you, before you shake someone's hand, that she's Mary and it will tell you where you met her last." In order to make wearable computers feasible, Intel needs to develop computer chips that run on milliwatts of power, and are less than half the size of the company's low-voltage Atom processors. Other challenges include creating devices with sufficient memory and battery life. As a method for minimizing the size and power requirements for these next-generation chips, Intel will likely forgo the instruction set normally found in today's microprocessors, Vara notes. "Right now, it's early in the game so we're looking at what makes sense to put in there," he says. "You'd want to have some memory built in and maybe some graphics, because you'd want to have one chip ... maybe two chips, but size-wise you want to keep it small." Vera says wearable computers that record users' activities and conversations would likely lead to the development of privacy and security technologies.

Breakthrough Offers New Route to Large-Scale Quantum Computing
Princeton University (10/19/12) John Sullivan

Princeton University researchers have developed a method that allows for the quick and reliable transfer of quantum information throughout a computing device, and could help engineers build a quantum computer that consists of millions of qubits. "The whole game at this point in quantum computing is trying to build a larger system," says Princeton professor Andrew Houck. The researchers used a stream of microwave photons to analyze a pair of electrons trapped in a quantum dot. "The microwaves are affected by the spin states of the electrons in the cavity, and we can read that change," says Princeton professor Jason Petta. The method combines techniques from materials science and optics. To make the quantum dots, the researchers isolated a pair of electrons on a small section of material known as a semiconductor nanowire, and then created small electron cages along the wire. The researchers found that electrons of similar spin will repel, while those of different spins will attract. "The methods we are using here are scalable, and we would like to use them in a larger system," Petta says.

Supercomputer Crunches Big Data
Collegiate Times (10/18/12) Andrew Kurlak

Computer science and engineering researchers at Virginia Tech have created a supercomputer to address the challenge of handling big data. HokieSpeed, listed on the Green500 ranking of the world's most energy efficient supercomputers, is at least 10,000 times more powerful than a standard personal computer, says Virginia Tech professor Wu Feng. The researchers tapped into graphics processing units (GPUs), in addition to traditional central processing units (CPUs), to enable the supercomputer to capture and process enormous amounts of information more efficiently than previous systems. The team had to make a translator because they operate in different languages. A processor called AMD Fusion fuses the CPUs and the GPUs together. Feng likens the task of determining how to best divide the labor between processors to the left and right side of the brain. "How do you make use of these brains that people haven't used in the past?" Feng asks. "And once you've programmed them, how do you extract performance?" Big data information processing holds promise for fields such as DNA sequencing, drug design at the molecular level, and the study of disease.

HarvardX Marks the Spot
Harvard Gazette (10/17/12) Tania DeLuzuriaga

Harvard University recently launched its first two courses on the edX digital education platform, and more than 100,000 students worldwide began taking online versions of an introductory computer science course and a course in epidemiology and biostatistics. HarvardX is a university-based organization that supports Harvard faculty as they develop content for the edX platform. Over the past six months, HarvardX has developed a leadership team, research committees, and interactive online courses. The HarvardX courses include short video-lesson segments, along with embedded quizzes, immediate feedback, student-ranked questions and answers, online laboratories, and student-paced learning. Students can "replay if they don’t understand something, and they can speed up when they grasp something quickly," notes Harvard professor Marcello Pagano. However, HarvardX officials emphasize that online learning does not mean the end of traditional classes. "Students can choose the learning process that works best for them, and have the option of exploring other topics even if we don’t have the time to cover them in class," says Harvard professor David Malan. The HarvardX platform also will provide data for researchers to study patterns of student achievement with the goal of making course material and methods more effective for students on campus and online.

Fighting Cybercrime, Making the Future More Secure
CORDIS News (10/16/12)

The European Commission plans to publish a European cybersecurity strategy focusing on preparedness, prevention, and response, as well as a permanent Computer Emergency Response Team. European Union funding also is being used for a variety of pan-European projects aimed at improving cybersecurity. For example, Nessos focuses on fostering the design and development of secure software and systems for the future Internet. The goal is to ensure that engineers and developers address security issues at the beginning of system analysis and design. SecureChange involves researchers from nine European countries collaborating to develop the methods, techniques, and tools needed to make the software life cycle more efficient, flexible, and more secure. The Aniketos project concentrates on bringing security and trust to heterogeneous environments by developing new technology, tools, and security services to support the design-time creation and run-time dynamic behavior of secure composite services. The TECOM project has helped bring trusted computing to embedded systems by developing technology that can run everything from smartphones to electricity meters.

Abstract News © Copyright 2012 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