Association for Computing Machinery
Welcome to the April 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).


Pentagon Announces New Strategy for Cyberwarfare
The New York Times (04/23/15) David E. Sanger

During a speech at Stanford University on Thursday, U.S. Department of Defense Secretary Ashton B. Carter announced a new Pentagon strategy that for the first time explicitly discusses the circumstances under which the U.S. could use cyberweapons against attackers. The new strategy also explicitly names China, Russia, Iran, and North Korea as the countries most likely to pose a cyberthreat to the U.S. A previous strategy released in 2011 was less detailed, only alluding to cyberweapons and talking vaguely about potential adversaries. During his speech, Carter said the Pentagon had detected and repelled an intrusion into its unclassified networks by Russian hackers in recent months. The new strategy lays out a hierarchy for cyberattacks and who should respond to them and how, and says the U.S. will only conduct what it calls a "cyberspace operation" if it has exhausted all "network defense and law enforcement actions." However, it allows the president and the secretary of Defense to authorize operations "to disrupt an adversary's military related networks or infrastructure so that the U.S. military can protect U.S. interests in an area of operations." The phrasing of the strategy seems to leave open the possibility of the U.S. carrying out a preemptive cyberattack.
View Full Article - May Require Free Registration | Return to Headlines | Share Facebook  LinkedIn  Twitter 

Professor Dame Wendy Hall Receives Outstanding Contribution to ACM Award
University of Southampton (United Kingdom) (04/22/15)

University of Southampton professor and former ACM president Dame Wendy Hall has been awarded with the Outstanding Contribution to ACM Award for her work in transforming the way the world views computing. As ACM president, Hall initiated the establishment of ACM Councils in Europe, India, and China, greatly extending the organization's scope. Hall also did significant work on the education of new computer science generations, promoting gender diversity and nurturing talent in computing. "Hall provided leadership and inspiration at a time when the computing discipline exploded onto the international scene, promoting ACM as the foremost association of computing professionals worldwide," says current ACM president Alexander L. Wolf. As a professor at the University of Southampton, Hall was the founding director of the Web Science Research Initiative to promote the discipline of Web Science and foster research collaboration between the University of Southampton and the Massachusetts Institute of Technology. In addition, Hall was president of the British Computer Society, and has served as a commissioner for the Global Commission on Internet Governance since 2014.

RSA: Panel Calls NSA Access to Encryption Keys a Bad Idea
Network World (04/22/15) Tim Greene

Panelists at the RSA 2015 security conference's cryptography panel this week expressed skepticism about the encryption key escrow schemes being put forward by the U.S. government as a means to ensure it has access to encrypted communications data. Massachusetts Institute of Technology professor Ron Rivest, one of the inventors of the RSA public-key cryptosystem, says key escrow schemes, which would allow the government access to keys held by third parties it could use to decrypt data in certain scenarios, would create numerous ways encryption could be compromised, pointing out multiple governments would want to have access to keys. Adi Shamir, who teaches at the Weizmann Institute and also helped develop RSA (and who shared the 2002 ACM A.M. Turing Award with Rivest and Len Adelman), says he sees no privacy advantage in an escrow scheme, especially if different government agencies would be able to access the keys independently of each other. Public-key cryptography pioneer Whitfield Diffie proposed a different scheme in which cryptography systems are made strong enough that the U.S. National Security Agency would be able to crack select traffic, but not to decrypt everything wholesale. The panelists also discussed several other security issues, including the impending roll-out of chip-and-PIN payment card technology in the U.S. and the increasing use of ransomware by hackers.

Cloud Security Reaches Silicon
MIT News (04/23/15) Larry Hardesty

Two years ago, Massachusetts Institute of Technology researchers proposed a method for thwarting attacks in which criminals infer computer data based on the pattern in which the machine accesses its memory, and now they have begun to implement it in hardware. The researchers have developed the layout of a custom-built chip that would use their scheme, and are now moving the system toward fabrication. The scheme involves confusing adversaries by querying several memory addresses at once to hide the source of the data. However, this process requires transferring much more data between the chip and memory than would otherwise be necessary. In order to minimize the amount of extra data needed, the technique stores memory addresses in a tree data structure, and every address is randomly assigned to a path through the tree. When the chip requires the data stored in a particular address, it also requests data from all the other nodes on the same path. The researchers proved that pulling data from a single path was just as confusing to hackers as if the chip had pulled data from every memory address in use. The researchers gave the new chip an extra memory circuit, with storage slots that can be mapped onto the sequence of nodes in any path through the tree.

House Bill Slashes Research Critical to Cybersecurity
Computerworld (04/22/15) Patrick Thibodeau

A new U.S. House of Representatives bill setting the nation's basic research agenda for the next two years is being criticized for significantly cutting research that can be vital to cybersecurity. The Competes Act, sponsored by Rep. Lamar Smith (R-TX), chair of the Science, Space, and Technology Committee, sets the U.S. National Science Foundation funding levels for the 2016 and 2017 fiscal years. Although the bill increases funding for computer and information sciences and engineering by 14 percent to $1.05 billion a year, it cuts funding in other areas that inform cybersecurity, in particular the social, behavioral, and economic sciences. The bill would slash funding for research in these areas by 45 percent, from $272 million to $150 million. On Tuesday, the Computing Research Association (CRA) strongly criticized the new funding bill, because of its failure to substantially increase overall research funding and its cuts to specific areas such as the social sciences. CRA chair J Strother Moore says social science research is "critical to understanding how best to design and implement hardware and software systems that are more secure and easier to use." Social science research can be especially important in helping to combat attacks such as phishing campaigns that have a strongly social component.

From Text to Virtual World: The MUSE Promise
CORDIS News (04/22/15)

The European Union-funded Machine Understanding for interactive StorytElling (MUSE) project is developing a translation system capable of converting text into three-dimensional (3D) virtual worlds. The researchers say in the near future, it may be possible for computers to understand a text and convert it automatically into characters, situations, actions, and objects in a 3D virtual world; doing so could enable passive readers to become active story participants. The MUSE team has evaluated its text-to-virtual-world translation system using children's stories and patient education materials. KU Leuven professor Marie-Francine Moens says the idea is to translate actions, actors, and objects recognized in a text into visuals. The researchers developed natural-language processing components for the semantic processing of the texts. Moens says the components include the recognition of semantic roles in sentences, spatial relations between objects, and chronology of events. The researchers also provided annotated datasets for training the recognition algorithms, based on a Bayesian network framework. The system tries to find the most probable interpretation of a sentence in light of the evidence obtained from the text and from background knowledge. Moens says the technology is primarily being geared toward the game and e-learning markets.

Wrangler Data-Intensive System Opens to Scientists
Texas Advanced Computing Center (04/22/15) Faith Singer-Villalobos

The Texas Advanced Computing Center (TACC) at the University of Texas at Austin (UT Austin), Indiana University, and the University of Chicago announced that Wrangler, a new data analysis and management system based at TACC, is now being made available to the scientific community. Wrangler, which is supported by a grant from the U.S. National Science Foundation, is designed to meet the big data storage, analysis, and access needs of science and engineering researchers. The system uses DSSD rack-scale flash technology and features tens of petabytes worth of data storage and more than 3,000 processor cores. In addition to hosting part of the system's replicated storage, Indiana University is providing operations and training to help users optimize their connections to Wrangler. The Computation Institute, a joint initiative of the University of Chicago and Argonne National Laboratory, is integrating its Globus Online service within Wrangler to make it easier and faster to transfer data to and from Wrangler. The new system already is being used by a pair of evolutionary biology projects at UT Austin. One project is using Wrangler to help model the evolution of monogamous behavior in vertebrate animals, while the other is using the system to speed its analysis and categorization of genes in plants that produce medically important alkaloids.

Disney Video-Dubbing Software Makes People Speak Gibberish
BBC News (04/22/15)

Disney Research has developed software that automatically redubs video clips with new words that fit the speaker's lip movements. The researchers say the technology can "literally put plausible words" into a person's mouth. The program tracks variations in the shape of a subject's mouth and jaw, and then searches a pronunciation dictionary to find alternative words that match the motions. For example, the short phrase "salary and tenure" was replaced with such expressions as "outside we nestle" and "is not raised as a tutor." The system appears to have limitations because the generated phrases largely made no sense. Disney researchers say the key application of speech redubbing would be to translate movies and TV shows from one language to another, but it seems unlikely the system would have sufficient suitable matching phrases to re-voice a complete movie or show seamlessly. The technology likely has more potential to be used for comedic effect. The software works by ranking strings of words according to the likelihood they would appear in that order in normal speech, and a voice synthesizer generates highly ranked phrases to ensure each bit of sound, or phoneme, coincides with the appropriate lip movements.

Revisiting the STEM Workforce
National Science Foundation (04/21/15) Nadine Lymn

A new report by the National Science Board (NSB) examines the U.S. science, technology, engineering, and mathematics (STEM) workforce. The report says in 2010, 16.5 million individuals reported their job required at least a bachelor's degree-level of science and engineering expertise, representing about three times the number of individuals working in occupations classified as science and engineering. "New industries and the growing importance of STEM skills in jobs not traditionally thought of as STEM, means that we must revisit what we mean by a 'STEM worker,'" says NSB chairman Dan Arvizu. Depending on the definition used, today's STEM workforce includes employees across a wide swath of disciplines and job arenas, possessing everything from non-degree certifications to Ph.D.s. The report notes the workforce can even include individuals without a STEM degree who work in STEM jobs. What is typically called the STEM workforce is a complex aggregate of "sub-workforces," the report says. The report also notes career pathways tend not to be linear for individuals with STEM knowledge and skills. "Ensuring access to high-quality education and training experiences for all students at all levels and for all workers at all career stages, is absolutely essential," Arvizu says.

Carnegie Corporation of New York Grants Carnegie Mellon $1 Million to Support Activities of the Simon Initiative
Carnegie Mellon News (PA) (04/20/15) Ken Walters

Carnegie Mellon University (CMU) has received a two-year, $1-million grant from the Carnegie Corporation of New York to support its technology-enhanced learning (TEL) activities. CMU pioneered TEL courses that have proven to be faster and more effective than traditional curricula covering the same material. In addition, 18 months ago, CMU established the Simon Initiative, a strategic, university-wide commitment to harness learning science research and the latest in technology to improve learning outcomes for all students. As part of the initiative, a consortium of academic, industry, and nonprofit leaders chaired by CMU president Subra Suresh has begun to establish best practices for TEL and sharing learning data. The grant will enable CMU to begin implementing the recommendations of the Global Learning Council. "We have several decades of research in how to build and use TEL effectively, and this research has been extremely successful," says Simon Initiative faculty lead Richard Scheines. "But it is clear that there are sizable barriers to wide adoption of effective TEL, some rational and some not. This grant gives us the opportunity to take this work to the next level, making these breakthrough techniques available to more students here at CMU and to learners everywhere."

Developing a Robotic Therapist for Children
Carlos III University of Madrid (Spain) (04/20/15)

Carlos III University of Madrid (UC3M) researchers are developing an interactive autonomous social therapist robot that can perceive patients' reactions and determine if they are doing their exercises correctly. The goal of the project, called Therapist, is to develop a system that looks like a toy to avoid the kind of discouragement that often inhibits children's motor rehabilitation. The researchers tested the system with more than 100 patients and found the children, their parents, and the medical staff believe it makes rehabilitation more fun and attractive. The robot, called NAO, works with medical staff to define therapies for patients that can involve the robotic system. "The main benefit is that the children see the robot as a friend; they like playing with it and they become uninhibited," says UC3M's Jose Carlos Pulido. The researchers focused on planning tasks and automated learning, which gives the robot the ability to decide which actions it wants to complete at any given moment. The system's architecture, called Robocog, can be applied to other robot therapists, and the researchers plan to fine-tune the system's algorithms, enabling the robot to recognize the patient's gestures.

OSU Innovation Boosts Wi-Fi Bandwidth Tenfold
Oregon State University News (04/20/15) Rachel Robertson

Oregon State University (OSU) researchers have developed WiFO, a prototype technology that uses light-emitting diode (LED) lights to transmit information, increasing the bandwidth of Wi-Fi systems by a factor of 10. The technology could be integrated with an existing Wi-Fi system to reduce bandwidth bottlenecks in crowded locations. "In addition to improving the experience for users, the two big advantages of this system are that it uses inexpensive components, and it integrates with existing Wi-Fi systems," says OSU professor Thinh Nguyen. WiFO uses LEDs that are beyond the visual spectrum for humans and creates an invisible cone of light in which the data can be received. The researchers also created a hybrid system that can switch between several LED transmitters installed on a ceiling, and the existing Wi-Fi system. "I believe the WiFO system could be easily transformed into a marketable product, and we are currently looking for a company that is interested in further developing and licensing the technology," Nguyen says. The researchers note the new system has the potential to send data at up to 100 Mbps.

User Creativity Made YouTube the World's Biggest Music Service
Aalto University (04/21/15)

Aalto University researchers studied the popularity of music videos on YouTube and found one reason why video views and search activity is high is due to users' own videos. Alternative videos that re-use the music of popular artists can reach an audience of millions. These cover versions, parodies, and other user-appropriated videos are readily available and well promoted on the site. Moreover, users are willing to listen to music with rolling lyrics over a still photo, which ranks highly in YouTube search results. Aalto researcher Antti Salovaara reports the team found three primary music video types: traditional, user-appropriated, and derivative music videos. "Earlier studies ignored music's tremendous pull, even though it must have been obvious to everyone using YouTube," says Aalto researcher Lassi A. Liikkanen. He notes over the years, "the artists have changed, but music has remained on top of the charts."

Abstract News © Copyright 2015 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