Association for Computing Machinery
Welcome to the June 24, 2015 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


A Disaster Foretold--and Ignored
The Washington Post (06/22/15) Craig Timberg

A group of hackers called L0pht warned a U.S. Senate panel in 1998 that computer security was dangerously lax and companies had neither the incentive nor government the skill or will to remedy the situation. Their warning went unheeded, and 17 years later the Internet suffers from endemic insecurity that criminals regularly exploit, inflicting financial and other forms of damage around the world. Many serious online breaches can be traced to flaws in software developed at the Internet's birth, and growing awareness of the threat in ensuing years has not eliminated or mitigated the root problems, says former L0pht hacker Cris Thomas. The same programs that have enhanced the Internet experience have enabled remote manipulation of systems to which the software is connected, while tech companies fostered a culture that favored profits over security, leaving the burden of system failures on customers. Compounding Web insecurity was the growing competition among tech giants, which emphasized innovation and growth and embedded Internet-related features into their products, creating more opportunities for exploitation. Despite a revitalized push for security from tech firms and the government, cybercriminals remain consistently ahead of both business and federal efforts to protect the Internet and its users. Former L0pht members fear a historic cyber disaster may be the only effective agent for serious change.
View Full Article - May Require Free Registration | Return to Headlines | Share Facebook  LinkedIn  Twitter 


Linux Foundation Launches $500,000 Project to Stop Future Heartbleed Flaws
V3.co.uk (06/22/15) Alastair Stevenson

The Linux Foundation will invest $500,000 in three projects that will help improve the security and services of the open source technology. The foundation's Core Infrastructure Initiative will provide $200,000 to Reproducible Builds, a project that seeks to improve the security of the Debian and Fedora operating systems by letting developers independently verify the authenticity of binary distributions. The feature will help prevent the introduction of flaws during the build process and reduce unneeded variations in distribution code. Meanwhile, $192,000 will go to the False-Positive-Free Testing project, an effort to build an open source TIS Interpreter that will reduce false positive TIS Analyzer threat detections. And $60,000 will go to the Fuzzing Project, an initiative to coordinate and improve the fuzzing software testing technique that identifies security problems in software or computer systems. The funding represents the Linux Foundation's latest initiative targeting future Heartbleed-level flaws. The funding follows the discovery of several critical bugs in widely used open source technologies.


How to Read a Digital Footprint
University of Cambridge (06/23/15) Fred Lewsey

Researchers from more than 100 institutions worldwide avail themselves of a database of psychological scores and social media data furnished by the University of Cambridge's Psychometrics Center. The Cambridge scientists think the database and other tools could be used to usher in a new era of psychological big data that can be channeled toward improving commercial and government services and scientific research--provided such tools are openly accessible. The Psychometrics team's recently introduced Apple Magic Sauce interface can be used for marketing and research to produce psycho-demographic profiles from digital footprints. Psychometrics researcher Vesselin Popov says targeted online advertising may be a burden, but it is less of a burden if consumers actually want the promoted items. "Using opt-in anonymous personality profiling based on digital records such as Facebook Likes or Last.fm scores could vastly improve targeted advertising and allow users to set the level of data-sharing they are comfortable with," he notes. "This data could then, with the permission of users, be used to enrich scientific research databases." Meanwhile, the team is applying psychometric testing toward the development of video games that might benefit job centers by measuring applicants' psychological strength and weaknesses via gameplay. Psychometrics Center director John Rust also thinks future human-computer interaction could be refined with psychometric data, which systems would use to recognize people.


Facebook Can Recognize You in Photos Even If You're Not Looking
New Scientist (06/22/15) Aviva Rutkin

Facebook has developed an experimental algorithm that can recognize people in photographs even when their faces are hidden. The company's artificial intelligence lab has adapted a face-recognition algorithm to look for unique characteristics such as clothing, hairstyle, body shape, and pose, which is what humans do very well, according to Facebook's Yann LeCun. The lab ran nearly 40,000 public photos from Flickr--some of people with their entire faces clearly visible and others where they were turned away--through a neural network. The final algorithm recognized individual people's identities with 83-percent accuracy. The algorithm could power photo apps such as Facebook's Moments, but the tool also could potentially be used to alert privacy-conscious users whenever a photo of them has appeared on the Internet. However, Carnegie Mellon University's Ralph Gross notes the impressive algorithm raises some serious privacy issues. "If, even when you hide your face, you can be successfully linked to your identify, that will certainly concern people," Gross says. "Now is a time when it's important to discuss these questions."
View Full Article - May Require Free Registration | Return to Headlines | Share Facebook  LinkedIn  Twitter 


Faster Computers Coming Soon, as Graphics Chip Answers the Call to Action
CNet (06/22/15) Stephen Shankland

A more efficient graphical interface model that taps hardware access deeper than previous interfaces will give programmers more direct control over the graphics chip and free up the computer's central-processing unit to conduct other tasks. The result will be a significant upgrade in computing speed once the new technology debuts on Macs and Windows machines later this year. Other advantages of the new interfaces is the same interface is functional across mobile devices, and they can manage graphics work as well as computing work that can be performed on the graphics chip. Among the upgrade's challenges for companies will be deciding which of the new interfaces to support, while programmers also must choose between them. An additional consideration is the interfaces may perform better on certain operating systems than on others. Moreover, the interfaces will lack many of the safety measures that eliminated crashes and weird defects, which means programmers have to code with more caution, says Linley Group analyst David Kanter. Still, the promised benefits are substantial, with Microsoft claiming "[game] developers will be able to create bigger maps, improved performance and graphics, and bigger multiplayer environments stretching across devices."


Programming Intelligent Underwater Robots
Manufacturing.net (06/22/15) Kaylie Duffy

Researchers at the Massachusetts Institute of Technology (MIT) are testing a new method of programming autonomous underwater vehicles (AUVs). Most current AUVs are programmed manually, with explicit instructions about where to go and what to do coded into them by researchers. The MIT group is pursuing a method that would enable researchers to provide an AUV with several high-level goals, such as mapping a specific area, as well as information about its surroundings, which would be enough for the AUV to determine how best to meet that goal. The new system is inspired by and named after Star Trek's Enterprise starship, with different aspects of the system acting like different members of the bridge crew; for example, one part of the system will act as a navigator, while another acts as captain, making higher-level decisions. The system grew out of work done by principle developer Brian Williams while he was working at the U.S. National Aeronautics and Space Administration (NASA) during the 1990s. Williams says Enterprise is much faster and smarter than the system he developed at NASA, and is capable of forming plans within a fraction of a second. In addition to AUVs, Williams' team is exploring other possible applications for Enterprise, including monitoring crops with airborne drones and operating a Mars rover.


Automated Vehicles: One Eye on the Road, Another on You
Technology Review (06/19/15) Will Knight

The importance of tracking motorist behavior will grow as more automated vehicles are rolled out, especially as it relates to the issue of assigning liability in accidents. In situations in which drivers are not controlling the car, it is vital to ensure they are not too distracted to regain control if necessary, and to ascertain who is at fault if something does go wrong. Making sure drivers do not become too relaxed is a key issue and the Mercedes S-Class, for example, can monitor driver attention via the steering wheel, detecting whether their hands have been removed, and sensing drift. There is evidence to suggest the introduction of increasingly sophisticated automation in motor vehicles will alter driving habits, with Google X's Astro Teller saying his company's latest prototype car eliminates pedals and steering wheels because drivers of earlier versions quickly became complacent. To be used in traffic jams next year is an automated driving system from Audi, which will watch the driver from a dashboard camera that tracks eye position and other behavior. Audi's Thomas Mueller says the company plans to meld digital distractions, such as smartphone alerts, with a vehicle's infotainment system, so they can be more easily blocked if a driver suddenly needs to resume driving.


Shifting Identity: The Fashion for Wearable Technology May Get Rid of the Need for Passwords
The Economist (06/20/15)

Body shifts every time someone's heart beats could potentially generate a new type of biometric for wearable technology. Researchers from the Massachusetts Institute of Technology (MIT) have tested whether accelerometers fitted to wearable devices could be used to detect the slight shifts in body movement. The details of ballistocardiographic shifts seem unique to individuals. MIT's Javier Hernandez and colleagues developed software to detect relevant data on volunteers wearing either a head-mounted or wrist-mounted device. During a recent conference, Hernandez reported the software recognized individual participants 94 percent of the time when they were lying down. However, the software got their identities correct 86 percent of the time when they were sitting and 72 percent of the time when they were standing. Still, the team says the approach can be refined, and could enable people to gain access to wearable devices by doing little more than being still in their presence.
View Full Article - May Require Free Registration | Return to Headlines | Share Facebook  LinkedIn  Twitter 


MIT Turns Hacking Medicine Program Into an Institute to Study Digital Health
The Wall Street Journal (06/19/15) Timothy Hay

The Massachusetts Institute of Technology (MIT) is spinning out its Hacking Medicine program into a non-profit institute that will convene healthcare stakeholders to develop techniques for ascertaining how digital-health products impact users' health. MIT lecturer Zen Chu notes many digital-health startups face the same challenge: "What's actually working, and how do you prove that?" Chu argues apps, connected medical devices, and other high-tech gadgets and features cannot simply be adopted by consumers, but also must be accepted and recommended by physicians and covered by major insurers. Rock Health's Malay Gandhi agrees, saying insurers are demanding evidence of the effectiveness of digital-health products and programs, which in this case means demonstrating they have caused significant behavioral change in patients. Gandhi stresses digital-health firms must provide data on user engagement and behavior, as well as showing they trim the cost of care administration if hospitals and insurers are to accept them. Chu says the Hacking Medicine Institute will solicit the participation of drug and medical-device companies, policymakers, hospital administrators, physicians, insurance companies, self-insured employers, and company founders from across the range of digital health. The result of such gatherings will be documents offering guidelines on evaluating digital-health programs, and assessment of digital products that seek to help treat major diseases.


3D Vision: UTSC Computer Scientists Develop Better Way to Visualize Molecules
U of T News (06/22/15) Patchen Barss

University of Toronto Scarborough (UTSC) researchers say they have developed a faster, less expensive, and more reliable way to determine the three-dimensional (3D) shapes of biological molecules such as proteins and viruses. The researchers say simplifying the modeling process makes it easier to understand these molecules' function and behavior, which is important for basic science and cutting-edge development of new drugs and medicines. The researchers first used a technique called cryo-electron microscopy to create thousands of two-dimensional (2D) images of a molecule. "There are many challenges in estimating the 3D structure, but the big one for us is that we don't know from which direction those 2D images were taken," says UTSC postdoctoral fellow Marcus Brubaker. The new method involves a time-consuming evaluation for every possible orientation for each 2D image. However, as the structure takes shape, more and more possibilities can be rejected as statistically unlikely. In addition, the process gets faster with each iteration, enabling a single computer to achieve in a day what previously would have taken an array of hundreds of computers several weeks to complete.


Middle School Girls Get Head Start in Computer Science With Help of Rutgers Students
Rutgers Today (06/22/15) Carl Blesch

Rutgers University's Douglass-DIMACS Computing Corps is a campus club co-sponsored by the Center for Discrete Mathematics and Theoretical Computer Science (DIMACS) and the Douglass Project for Rutgers Women in Math, Science, and Engineering. The club consists of women computer science majors that aim to inspire young girls to follow them into science, technology, engineering, and mathematics fields. The Rutgers students coach middle school girls in programming, game design, robotics, and the fundamentals of computing. The goal is to show girls computer technology can be understandable and fun. The mentors create visual examples of how programming works, and the middle school girls then develop computer games from scratch and program robots. The students were initially confused by what was taught, "but when you show them what they can apply it to, a game online or a robot, they start to get excited," says Rutgers mentor Helen See. "My high school didn't have anything tech-related. "I wouldn't have gone into computer science unless my classmates started a club. They taught me how to code." The DIMACS-Douglass Computing Corps ran a similar after-school tutoring program for a co-ed group at a local elementary school.


New Project Will Develop Virtual Reality Counterterror Training
Homeland Security Today (06/18/15) Kylie Bull

An innovative, collaborative training program could better prepare security personnel in Europe to respond to physical threats and cyberattacks on critical infrastructure. AUtomated serious Game scenario Generator for MixED reality training (AUGGMED) aims to develop a multimodal virtual reality (VR) and mixed-reality platform over the next 36 months. Trainers will be able to present police, security forces, counterterrorism units, and first responders with a wide range of scenarios in different VR environments using mixed-reality techniques. Trainers will be able to set learning objectives for individuals and teams, define scenarios, monitor progress, alter scenario parameters, provide real-time feedback, and assess trainee performance. The platform will be able to be used anywhere via devices such as smartphones, tablets, high-end PCs with multiple monitors, and head-mounted displays, and trainees will be able to initiate sessions at their convenience. AUGGMED is part of the European Commission's Horizon 2020 program. "It's vital with any training that it is fit for purpose," says AUGGMED project coordinator Christos Giachritsis. "It is for this reason that the end users...play a critical role throughout the development process, providing their knowledge and expertise in relation to the definition of training requirements and subsequent evaluation of the AUGGMED platform."


This Ancient Paper Art Makes Flexible, Super Strong Electronics
Popular Science (06/22/15) Dave Gershgorn

Teams of researchers around the U.S. are using Kirigami, the art of paper cutting, as inspiration for developing flexible and stretchable electronics. For example, researchers at the University of Michigan have applied Kirigami techniques at the microscopic level to conductors, making small incisions into sheets of conductive material to turn them into flexible meshes that can spread stress over a much larger, predictable area than the sheets themselves. The researchers say their method yields composite sheets that can withstand up to 370-percent strain, while the conductor in its normal state can only withstand 4-percent strain. The researchers think the meshes could eventually be overlapped to form the basis for flexible displays in consumer devices. Meanwhile, a team at Arizona State University (ASU) is applying Kirigami to batteries. ASU professor Hanqing Jiang says the process his team is developing is like making a chain of paper dolls. The goal is to create a flexible battery that could be embedded in the band of a smartwatch. Jiang says the technology would both increase the device's battery life and free up space in the watch, enabling designers to make the device thinner.


Abstract News © Copyright 2015 INFORMATION, INC.
Powered by Information, Inc.


To submit feedback about ACM TechNews, contact: technews@hq.acm.org
Current ACM Members: Unsubscribe/Change your email subscription by logging in at myACM.
Non-Members: Unsubscribe