Welcome to the October 28, 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
Old Internet Flaw Causes New Problems
The Wall Street Journal (10/27/15) Drew FitzGerald
An old Internet problem is directly tied to recent hacker breaches exploiting a vulnerability in the border gateway protocol (BGP), which network routers use to determine how to channel data to its destination. Most network operators can recommend routing changes with little confirmation, leaving traffic susceptible to false information. Security experts warn leveraging BGP flaws to manipulate traffic can support the theft of proprietary information, surveillance on confidential traffic, or making information unreachable. "There's good evidence that people are playing serious malicious games with the routing table," says Columbia University professor Steven Bellovin. "I think that the risks are very serious." BGP can be traced back to the 1980s, when early Internet users were mostly academic institutions that developed the language so machines could easily keep track of each other's networks. The system's founding assumption is that other parties are trustworthy, and thus few built-in checks are included. Engineers have been devising a remedy that would add a verification process to changes to BGP's route map, and encrypt each route revision so a router could validate that an alteration comes from a trusted source. Parsons Group's Sandra Murphy says time is of the essence for developing and implementing such a solution before someone can launch a massive and widespread network misrouting attack.
A Basis for All Cryptography
MIT News (10/28/15) Larry Hardesty
The inability to practically apply indistinguishability obfuscation (IO) to deliver secure cryptography may be mitigated with Massachusetts Institute of Technology (MIT) researchers' exploration of an efficient method for functional encryption. "If you want to do obfuscation and get all of crypto, everything that you can imagine, from standard assumptions, all that you have to do is solve this very specific problem, making functional encryption just a little bit more efficient," says MIT professor Vinod Vaikuntanathan. IO's requirement is that an adversary cannot ascertain which of two versions of an operation a program is performing, instead of making it impossible to determine what operations the program is executing. The process for building IO schemes from multilinear maps was an important step, but it is unclear any proposed map construction methods will ensure security. Meanwhile, functional encryption is a technique for executing some operation on an encrypted file that yields an intelligible result without leaking any further information about the file's contents. Although encryption time is proportional to the length of the file being encrypted with standard encryption, the efficiencies of the best functional-encryption schemes also include a factor proportional to the size of the operation's result. Nevertheless, Cornell University professor Rafael Pass says, "functional encryption is a significantly simpler-looking primitive, so [the MIT] work opens a new avenue for getting secure constructions of IO."
UC3M Researches Simulator of Human Behavior
Charles III University of Madrid (Spain) (10/26/15)
Researchers at the Charles III University of Madrid (UC3M) in Spain are working on the IBSEN project, which is investigating how to build a system that recreates human behavior. The researchers say the technology could be used to anticipate behavior in a socioeconomic crisis, create more human-like robots, or develop avatars of artificial intelligence. "We are going to lay the foundations to start a new way of doing social science for the problems that arise in a society that is very technologically connected," says IBSEN project leader Anxo Sanchez. The researchers say they are preparing experiments that will simultaneously present certain problems of cooperation, social problems, and economic games to thousands of people in order to discover hidden patterns in their decisions. The researchers will use this information to create a simulator of human behavior. "The greatest difficulty is to design a new experimental protocol that allows us to ensure that all the participants in the experiment are available at the same time and really interact, because you are not observing them in a laboratory," the researchers note. They say the goal is to obtain a repertoire of human conduct that makes it possible to simulate the behavior of a person and apply it to a robot.
Can Data Help Save MOOCs?
The Stanford Daily (10/26/15) Sean Cummings
When massively open online courses (MOOCs) were introduced a few years ago, proponents expected them to be a revolution in education, but the results were not immediately inspiring. Although the first rounds of MOOCs from major providers and several universities boasted huge enrollment numbers, their completion rates were very low. However, Stanford University professor Andreas Paepcke believes MOOCs still have a lot of potential, and he says data gathered by MOOC providers on their courses and students can help make MOOCs better. "We really haven't scratched the surface of what could be done," Paepcke says. He points out several of the weaknesses of the first wave of MOOCs already have been identified. One example is the fact that early MOOCs were designed by engineers and computer scientists with other engineers and computer scientists in mind. Paepcke believes creating a more neutral and inclusive pedagogy will help MOOCs be more broadly appealing. Another major challenge is user isolation. Paepcke and his team are working to incorporate the sort of close, small-group interactions that are key to success in many college courses into MOOCs. He says this process of honing MOOCs also will get better as MOOC providers learn how to do a better job of recording relevant course data.
Manipulating Faces From Afar in Real Time
The New York Times (10/26/15) James Gorman
Computer scientists from Stanford University and the University of Erlangen-Nuremberg in Germany say they have created a process dubbed "live facial re-enactment" that can instantaneously transfer facial expressions. The researchers say with the new technique, a person's smile would appear seamlessly on live video of another person's face, even if the second person is not smiling. The team used a camera that captures gestures in three dimensions, such as the Microsoft Kinect, and developed software to map each pixel on both faces and then transfer the expression. The transfer took just 30 milliseconds, which is partly due to the software running simultaneously on several computer processors. The researchers say live facial re-enactment could be used to improve dubbing in movies, to make video in virtual reality more realistic, and to provide instantaneous translation. For example, if an English speaker was talking to a Mandarin speaker, each would appear to be speaking the other's language. The researchers will present their work next week at SIGGRAPH Asia 2015 in Japan.
Groovy Programming Language Thrives Under Apache
InfoWorld (10/26/15) Paul Krill
The number of monthly downloads of the Groovy language have doubled to 1.34 million since it was picked up by the Apache Software Foundation six months ago. Project leader Guillaume Laforge attributes the increase to Apache's stamp of approval and strong marketing. Pivotal orphaned the dynamic language for the Java Virtual Machine in March, but under Apache's jurisdiction it is now independent of any one company. "By putting the project at the Apache Foundation, it's a stamp saying that the project is here to stay for the long run," Laforge says. Groovy is a well-established language, but all new Apache projects require incubation. Laforge notes the language is set to graduate from Apache Incubator project status. Groovy has niches in areas such as scripting for automating tasks on servers and building domain-specific languages for business rules, and an upgrade, version 2.5, offering faster compilation, is planned for the end of this year. "Mobile developers have the choice to use Groovy instead of Java for developing mobile applications on the Android platform," Laforge says. "We are definitely seeing quite some usage there."
STEM Definition Expanded to Include Computer Science
SD Times (10/26/15) Christina Mulligan
President Barack Obama has officially signed the Science, Technology, Engineering, and Math (STEM) Education Act of 2015 into law, expanding the STEM definition to include computer science. Officially including computer science in the definition of STEM signals the importance of a computer science education, says Della Cronin, who handles federal affairs for Code.org. Although the act does not bring any additional funding for computer science, it does permit schools running programs backed by a STEM fund to include computer science programs as part of that funding. "The thought here was looping this in as part of STEM funding gave the maximum amount of flexibility for individual school districts to tailor programs to meet the needs of their local communications," says Gartner analyst Kelly Calhoun. Cronin says computer science education is important even for those who do not plan on becoming computer science professionals because it can help make sense of consumer technology. "There are logic principles that are developed as part of studying computer science, and for a lot of kids this could be a great doorway into learning new ways of thinking, reasoning, and problem-solving," Calhoun says. The new law also amends the U.S. National Science Foundation's Robert Noyce Master Teaching Fellowship program to enable individuals pursuing a master's degree to participate in the program.
The Computer System That Won Jeopardy Could Soon Help NASA With Research
NASA News (10/27/15) Joe Atkinson
Rob High, chief technical officer for IBM's Watson cognitive computing system, was at the U.S. National Aeronautics and Space Administration's (NASA) Langley Research Center recently to discuss Watson and the role it could play in furthering NASA research. NASA researchers primarily are interested in using Watson Discovery Advisor, one of the branches of the larger Watson project that focuses on helping researchers improve their awareness of scientific and technical literature. Discovery Advisor is designed to quickly read through massive amounts of technical documents, papers, and publications to provide researchers with a deeper knowledge of their field. High says the goal is to "amplify human cognition, [and] one technique for doing that is to help humans navigate vast quantities of information, discovering along the way associations in that literature that may be inspirational to people, that may allow people to consider new ideas they hadn't considered before." Other NASA facilities are interested in making use of Watson. Both the Ames Research Center and the Armstrong Flight Research Center want to use a Watson system to serve as a flight operations advisor, reading aviation documents, and providing real-time advice to both flight crews and air operations centers.
Squeezed Light and Quantum Clockspeeds
Technology Review (10/23/15)
One of the many conundrums impacting the development of quantum computers is finding an effective way to compare different quantum computation methods. Quantum computing involves manipulating quantum bits (qubits), but there are different ways of carrying out calculations with qubits and comparing them has proven problematic. However, an international research team says they have developed a technique for evaluating the performance of quantum computation methods using a single parameter, which functions as a type of quantum clockspeed. The technique rests on the phenomenon known as quantum squeezing, a way of manipulating entangled photons to reduce the background noise associated with them. The researchers say the technique is based on the Heisenberg uncertainty principle, which states it is possible to accurately measure either a photon's momentum or its position, but not both at the same time. In quantum squeezing, physicists reduce the noise associated with entangled photons by measuring some of their properties. To test quantum computations, the researchers replaced ordinary photons with squeezed photons and quantified the "quantumness" involved in each computation based on the amount of squeezing needed to carry it out. "This introduces a new perspective in which to think about hierarchies in quantum algorithms," the researchers say.
Coding Bootcamp Grads Boost Their Salaries by 40 Percent on Average
Quartz (10/26/15) Alice Truong
Graduates of coding bootcamps found completing the programs helped boost their salaries by an average of 38 percent or $18,000, according to the Course Report. On average, participants paid $11,852 in tuition and 35 percent reported their schools guaranteed jobs for students after completion. However, 21 percent of 665 students graduating between 2013 and 2015 said they were unemployed. Still, Course Report co-founder Liz Eggleston reports 89 percent of students who were 120 days out of school obtained full-time jobs when employment rates are broken down by graduation date. There also are indications coding bootcamps are contributing to workforce diversification by encouraging people in their early- to mid-careers to become programmers, even though most students who attend bootcamps are mainly white and male. Thirty-six percent of bootcamp graduates were women--nearly double the number of women awarded computer science bachelor's degrees in 2013-2014, according to the survey. A slight increase was recorded in the proportion of African-American students, who comprise 5 percent of coding-school grads versus 3.2 percent of college grads awarded bachelor's degrees in computer science. Among the more sought-after technical skills by employers, as indicated by graduates' salaries, are proficiency in Python, C#, and Ruby.
Upgrading the Quantum Computer
University of Innsbruck (10/23/15)
Researchers at the Austrian Academy of Sciences' Institute for Quantum Optics and Quantum Information (IQOQI) say they have developed a scalable quantum computer architecture that overcomes the fundamental limitations of programmability in conventional approaches that aim to solve real-world general optimization problems by exploiting quantum mechanics. The new model involves detaching the logical quantum bit (qubit) from the physical implementation. Each physical qubit corresponds to one pair of logical qubits and can be tuned by local fields. "By using this approach we are not only avoiding the limitations posed by the hardware but we also make the technological implementation scalable," says (IQOQI) researcher Philipp Hauke. The researchers arranged the qubits in such a way that four physical qubits interact locally. "In this way we guarantee that only physical solutions are possible," and the solution of the problem is encoded redundantly in the qubits, says (IQOQI) researcher Wolfgang Lechner. He notes this redundancy also gives the model a high fault tolerance. The new architecture can be realized on various platforms ranging from superconducting circuits to ultracold gases in optical lattices. "Our approach allows for the application of technologies that have not been suitable for adiabatic quantum optimization until now," Lechner says.
The Future of Encryption
National Science Foundation (10/22/15) Amina Khan; Aaron Dubrow
U.S. National Science Foundation (NSF)-supported researchers are developing new methods to ensure the security of important data. One technique is called fully homomorphic encryption, developed by Craig Gentry in 2010 while he was a graduate student. His dissertation, in which he constructed the first fully homomorphic encryption scheme, was awarded the ACM 2009 Doctoral Dissertation Award, and then the ACM Grace Murray Hopper Award in 2010 for the same work. Fully homomorphic encryption was developed as a way to process data without ever decrypting it. As data and computation move to the cloud, fully homomorphic encryption would enable the data to be processed without ever having to give away access to it. Gentry, now a research scientist at IBM's Thomas J. Watson Research Center, is working to make it a practical reality. Another potential approach is called honey encryption, in which guesses of the key produce information that appears accurate but is not. Functional encryption involves restricted secret keys enabling a key holder to learn only about only a specific function of encrypted data and nothing else. In a third approach, called quantum key encryption, the quantum nature of atoms protects the data.
Three Scenarios for the Evolution of Exascale Computing
Scientific Computing (10/23/15) Erik DeBenedictis
The realization of exascale computing will require overcoming semiconductors' mounting energy consumption issues, and Sandia National Labs' Erik DeBenedictis outlines three scenarios for evolving supercomputers to the 1- to 50-exaflop range. The first case he cites involves finding a lower-voltage transistor or millivolt switch to enable continued reduction in energy per operation so existing software performance can be raised with minimal rewriting. However, DeBenedictis notes the electronics industry has little control over the time frame for the switch's discovery and commercialization. The second scenario he offers involves more three-dimensional (3D) integration of processors and memory, although it entails enough architectural alteration to require retuning existing code for performance. DeBenedictis sees a likely industry development of 3D technology for storage and mobile devices separate from supercomputers, but he cautions "industry will spend [research and development] funds on developing the technology and is likely to demand premium prices for new product." The third scenario calls for an application-specific architecture, which may be a challenge for software and algorithms and carries the pitfall of the supercomputer only being proficient at the function manufactured into the system. "This scenario is very likely to present the programmer with idiosyncratic architectures that require extensive experience by the programmer, and may lead to code that is not easily repurposed to other applications," DeBenedictis says.
Abstract News © Copyright 2015 INFORMATION, INC.
To submit feedback about ACM TechNews, contact: [email protected]
Current ACM Members: Unsubscribe/Change your email subscription by logging in at myACM.