Association for Computing Machinery
Welcome to the March 10, 2010 edition of ACM TechNews, providing timely information for IT professionals three times a week.


Computing Prize Winner Did Not Rest on His Laurels
The Wall Street Journal (03/09/10) Clark, Don

ACM has named computer science pioneer Charles P. Thacker the recipient of the 2009 A.M. Turing Award for his many contributions, which includes the Alto, a machine developed more than 30 years ago at Xerox's Palo Alto Research Center (PARC) that is often credited as the world's first personal computer. Thacker says the inspiration for the Alto came from PARC manager Bob Taylor's vision of computers as machines capable of transforming documents and other communications media. He says innovations such as dynamic random access memory chips enabled the PARC team to outfit the Alto with bit-mapped computer displays. ACM also cites Thacker's contributions to the invention of Ethernet and his work on tablet-style computers since becoming a researcher at Microsoft in the 1990s. Thacker "is a real genius," says former PARC researcher and fellow A.M. Turing Award winner Alan Kay. "We don't like to sling that word around in our field, but he is one. He is magic."

Machine-Learning Revolutionizes Software Development
ICT Results (03/09/10)

European researchers from the Milepost project used machine-language technology to develop a method for automatically optimizing software compilers for re-configurable embedded processors. The technology can learn how to get the best performance from the hardware, which enables the software to run faster and use less energy. The system learns to predict the optimal compiler solution for any program by analyzing the execution time of different compiler options. "If you can run things faster and more energy efficiently, you may be able to choose a different piece of hardware than before--perhaps a cheaper option for the same performance," says University of Edinburgh professor and Milepost project coordinator Michael O'Boyle. The Milepost team also launched a code-tuning Web site. Developers can upload their software code to the site and receive input on how to manipulate their code so it works faster. "We can use machine-learning technologies to look at multicore and heterogeneous platforms and we will be looking at dynamic online adaptation," O'Boyle says.

Could This Be The Robot Servant Who Will Serve You Breakfast In Bed?
London Daily Mail (03/08/10) Bates, Claire

Tokyo University researchers have created Kojiro, a humanoid robot that is learning to mimic how people walk. Kojiro has a skeletal structure similar to that of humans, which allows it to move in a more natural way, as it can bend and twist by bending and twisting using its artificial spine. The spine's design will enable scientists to develop lighter and more flexible robots in the future, says Tokyo professor and Kojiro researcher Yuto Nakanishi. Kojiro uses lightweight motors to pull cables attached to specific locations on the body, which simulates how muscles and tendons contract and relax when people move. Kojiro has 60 degrees of free movement using a system of about 100 cable-tendons. Sensors are used to keep track of Kojiro's various positions, and an accelerometer and two gyroscopes help the robot maintain balance. The researchers say more work is needed on the algorithms that control the robot's movements to enable it to handle complex actions.

A Little Black Box to Jog Failing Memory
The New York Times (03/09/10) P. D6; Bhattacharjee, Yudhijit

Microsoft researchers have developed Sensecam, a system for creating digital archives of a person's experiences that could help people suffering from memory disorders. Sensecam features a small black box containing a digital camera and an accelerometer to measure movement. Carnegie Mellon University (CMU) researchers had one subject go on three excursions with a Sensecam, a voice recorder, and a global positioning system (GPS) unit. The researchers found that the best way to help the subject remember the experiences was to focus on a few key images that might unlock the memories related to it. For a location-based experience, Sensecam uses data provided by the GPS and the accelerometer to determine which images might be the most salient. "The design is intended to give the patient the ability to engage actively with the experience instead of simply flipping through some pictures," says CMU's Matthew Lee. At Dublin City University, Alan Smeaton compares Sensecam images to categorize them by activity. At the University of Toronto, Ronald Baecker is studying the usefulness of complementing Sensecam images with an audio narrative created by a loved one.

A 4.2 Million (Pound) Grant Ensures a Sustainable Future for Software
University of Southampton (ECS) (03/09/10) Lewis, Joyce

Academics and software engineers from the universities of Edinburgh, Manchester, and Southampton have established the Software Sustainability Institute (SSI), which will partner with about 30 to 40 research communities across the United Kingdom to develop ways to keep their software current and to help them develop it to meet new requirements. SSI will optimize strategies for sustaining software and provide communities with best practices for improving it for future users. "The issue at the moment is that there are no coordinated ways of sustaining important research software once it comes to the end of its funding," says SSI director Neil Chue Hong. "The creation of the SSI will ensure that important software is sustained so that it can continue to contribute towards high quality research."

Mapping the Malicious Web
Technology Review (03/09/10) Lemos, Robert

Websense researchers have developed FireShark, software that automatically monitors malicious activity on Web sites. Websense researcher Stephan Chenette says the experimental system scans the Web, identifies the source of embedded content in Web pages, and determines whether any code on a site is acting maliciously. FireShark then creates a map of interconnected Web sites and looks for potentially malicious content. FireShark, which maps nearly one million Web sites and servers per day, decodes the HTML, Javascript, and other code embedded in each Web site, looking for the ultimate source of content. "When you graph multiple sites, you can see their communities of content," Chenette says. Websense researchers plan to release a plug-in for Firefox that will reveal the content hubs that a site is linked to.

Japan Baby-Robot Teaches Parenting Skills
Agence France Presse (03/10/10) de Freytas-Tamura, Kimiko

Tsukuba University engineering students have developed Yotaro, a baby humanoid robot designed to teach young people about parenting. Yotaro's face is made out of soft translucent silicon and is backlit by a projector connected to a computer to simulate sneezing, sleeping, and smiling, while a speaker emits sounds such as giggling or crying. Sensors detect physical contact and can change the robot's mood based on the frequency of the touches. Yotaro also simulates a runny nose with a water pump that releases body-temperature droplets of water through the nostrils. Meanwhile, the University of Osaka recently unveiled a robot that mimics a crawling baby as part of a research project to study the way humans learn to move and speak. The Osaka robot has 22 motors and 90 tactile sensors and microphones located near the eyes and ears. When told to move forward, the baby-bot will wave its legs and arms, gradually learning which movements will enable it to push itself up and crawl, says Osaka professor Minoru Asada.

NC State Research Advances Voice Security Technology
North Carolina State University (03/08/10) Shipman, Matt

A North Carolina State University (NCSU) research team led by professor Robert Rodman has developed a computer model that accelerates the voice identification process without sacrificing accuracy. Existing computer models take several seconds or longer to compare acoustic profiles and identify a speaker, which is too long for the technology to be widely used, according to Rodman. "In order for this technology to gain traction among users, the response time needs to improve without increasing the error rate," he says. The researchers modified existing computer models to make the authentication process work more efficiently. "This is part of the evolution of speech authentication software, and it moves us closer to making this technology a practical, secure tool," Rodman says.

Life and Death of Online Communities
University of Haifa (03/08/10)

An online chat channel is more likely to survive over time when the community is heterogeneous, or when it has turnover and new members continue to join the group, according to researchers at the University of Haifa and the New Jersey Institute of Technology. The team surveyed 282 chat channels begun on the same day over the course of six months. The survey found that the greater the turnover among members, the more likely a chat community will sustain itself over time. Moreover, chat groups are more likely to survive when they have a higher number of messages between members from the first day of activity through the end of its second week. The researchers also found that chat communities that have an irregular ratio between the number of messages and the number of members after two weeks are more likely to survive. "The present study shows that prediction of an online community's survival chances cannot be based on quantitative data relating to the size of the group or even to its growth rate alone," says Haifa's Daphne Raban. "A social predictor, on the other hand, can much better predict its chances."

New 'Hearing' Maps Are Real Conversation Starters
Engineering & Physical Sciences Research Council (03/05/10)

Cardiff University researchers have developed software that creates audibility maps of proposed room designs. The maps show hotspots where conversations would be inaudible if the room was noisy, enabling architects to adjust their designs to eliminate them and maximize audibility. Cardiff University professor John Culling says the software is specifically designed to improve the acoustics of indoor spaces where a large number of people meet, chat, and interact. He says the software also produces results much faster than other acoustic software. The key to the software is a mathematical equation that is based on research examining how people take in sound through both ears as it travels around a room. Culling says the work will be useful in areas where audibility is important, such as rail and airport announcement waiting areas. He says the research also will help in the future development of hearing aids and cochlear implants.

Context Is Ev . . . Well, Something, Anyway
MIT News (03/05/10) Hardesty, Larry

Massachusetts Institute of Technology researchers have developed a way to improve object recognition systems by using information about their context. The researchers used a set of more than 4,000 images and 107 different types of objects and created an algorithm that sorted the images into a hierarchical map of the object categories. In the map, each object is connected to at most one object above it in the hierarchy, which greatly reduces the number of connections the system must consider. The connection between two objects is given a weight that describes how often the objects appear together in the image set. When the system analyzes a new image, it uses object recognition algorithms to create a list of candidate objects and a confidence score. In testing, the system had a success rate of about 25 percent. "We absolutely cannot afford to take our eye off the ball of the component recognition systems that need to feed these context engines," says University of California, San Diego professor Serge Belongie. To be useful, object recognition systems need to be much more precise than today's prototypes, Belongie says.

Scientists Devise Global Food Alert
Kingston University London (United Kingdom) (03/04/10) Coslett, Cara

Kingston University scientists have developed a computer tool for monitoring food products. Kingston professor Declan Naughton says the program, which analyzes the alerts of other food-monitoring programs around the world, can provide detailed information on which countries are trading contaminated food as well as detect contaminated food more quickly. "No other system can reflect the complexity of this information in a snapshot form," Naughton says. He worked with statistician Andrea Petroczi and computer programmer Tamas Nepusz to develop the system. "We'd like to develop the tool to create an international alert system that will provide real-time information about emerging patterns and problems," Naughton says. He also believes the technology could be used to help prepare for malicious attempts to contaminate food and even monitor other kinds of global health hazards.

Developing Web Technologies to Share Secure Information (03/02/10)

Semantic Web technology will be a key tool in the development of a standard policy language for sharing information between agencies, countries, and organizations. Research conducted by a team at the Massachusetts Institute of Technology (MIT) will help the U.S. Air Force exchange resources and data securely, as well as adapt to new or changed policy dynamically without modifying code. "We are creating for the first time a policy interchange language or Interlingua grounded in Semantic Web technologies that will enable a secure exchange of information between entities using different languages to express their security constraints," says MIT's Lalana Kagal. Interlingua will use translations to capture key functions and characteristics, and facilitate translations of commonly used policy languages. "We would like to see the interchange language become standardized so that there would be widespread use, development of more tools and protocols around it," Kagal says.

Assessing the State of U.S. Science and Engineering
ScienceNews (02/27/10) Vol. 177, No. 5, P. 32; Lanzerotti, Louis

New Jersey Institute of Technology physicist Louis Lanzerotti, chairman of the National Science Board's Science and Engineering Indicators committee, says the board's recent report to the White House on the state of science and engineering is generally positive. The United States is very robust in terms of research and development (R&D), but there are areas where it is falling behind the rest of the world, according to Lanzerotti. He notes that countries such as Japan and South Korea are surpassing the United States in the portion of their gross domestic product committed to R&D. Lanzerotti also notes a slight uptick in the performance of basic research by U.S. industry since around 2006 or 2007, a trend he says mirrors what is happening worldwide. Lanzerotti maintains that the public perception of science and engineering in the United States is very high, with 75 percent to 80 percent of the population viewing it in a positive light. "I think a very important point is that scientists rank as high in public respect as do firefighters," he says.

Abstract News © Copyright 2010 INFORMATION, INC.
Powered by Information, Inc.

To submit feedback about ACM TechNews, contact: [email protected]

Change your Email Address for TechNews (log into myACM)