ACM TechNews (HTML) Read the TechNews Online at: http://technews.acm.org
ACM TechNews
June 1, 2007

Learn about ACM's 2,200 online courses and 1,100 online books
MemberNet
CareerNews
Unsubscribe

Welcome to the June 1, 2007 edition of ACM TechNews, providing timely information for IT professionals three times a week.


HEADLINES AT A GLANCE:

 

New Immigration Bill Amendment Could Help Keep Foreign Tech Workers in U.S.
InformationWeek (05/31/07) McGee, Marianne Kolbasuk

In a bi-partisan effort, a group of U.S. senators is expected to introduce an amendment to the immigration reform bill that establishes a dual green-card system that would allow a pool of 140,000 employer-sponsored green card foreign workers to remain in the United States. The amendment also calls for eliminating any limit on H-1B visas for foreign professionals with masters or doctoral degrees in science, technology, engineering, and math (STEM) fields. By retaining a certain number of employer-based green card workers each year, businesses would have better control over the talent kept in the United States, according to tech employers. Technology companies argue that the merit-based system, which awards points based on several factors, including a person's education and skills, gives government bureaucrats too much control over the type of talent available to employers. The amendment would allow employers to retain talented workers and expand permanent residency to foreigners with advanced STEM degrees. The amendment would also eliminate limits on H-1B visas granted to foreign students that have earned advanced degrees from U.S. universities. However, the amendment, as well as the entire immigration bill, faces strong opposition.
Click Here to View Full Article
to the top


RAW Talent Tackles Risk Analysis
Information Sciences Institute (05/29/07)

An Information Sciences Institute (ISI) research team, working with the support of the Department of Homeland Security's Center for Risk and Economic Analysis of Terrorism Events (CREATE), is developing a system that will make the process of quantifying risk estimates faster and more consistent. The analytical tool, called the Risk Analysis Workbench (RAW), will eventually be used by all eight Department of Homeland Security research centers, including CREATE. RAW uses artificial intelligence, information sharing, and Web resources to gather and distribute specific data necessary to perform tree analyses of "what-if" risk scenarios in a quick and uniform process. A major element of RAW is ISI's innovations in the systematic structuring of database information and making such information more accessible over the Internet. RAW will be used to create "decision trees" to provide risk analysis when comparing the cost of prevention against the cost of response, for example, to natural disasters, food born disease, or terrorist attacks. CREATE director Detlev von Winterfeldt, a professor in USC's Viterbi engineering school, performed one such risk assessment during CREATE's first few years, comparing the cost of equipping all airplanes with counter measures for man-portable anti-aircraft rockets against the repercussions of a successful attack. The analysis was praised as a significant achievement, but the intense effort required to perform such an analysis highlighted the need for a faster, more uniform process. RAW will provide an integrated, generalized toolbox designed to be used by risk analysts throughout the United States, providing a standardized but flexible format. RAW is still in development, but will be distributed in July for research testing.
Click Here to View Full Article
to the top


EDA's Future Lies Beyond Electronics
EE Times (05/28/07) Levitan, Steven

Professor of computer engineering at the University of Pittsburgh and chairman of this year's Design Automation Conference Steven Levitan writes that the field of electronic design automation is concerned with solving the key engineering problems of today. Levitan says the industry has played a critical role in the development of computing and communication systems that have improved the world over the last half century. These powerful and pervasive systems are used on a daily basis to solve important problems. Levitan says this year's DAC will help show how EDA's influence is not limited to the traditional chip environment. The event has an automotive theme, and there are specials sessions on "wild and crazy ideas," nanoscale circuits and architectures, synthetic biology, and design and manufacturing for emerging technologies. Professor Jan M. Rabaey will deliver his keynote, "Design Without Borders," on June 5. The author believes the techniques of abstraction, extraction, optimization, and synthesis are useful beyond electronic system design, and that over the next decade the tools will be applied to other fields to solve problems involving vehicular traffic, drug design, biology, and health care. For more information about DAC, visit http://www.dac.com/44th/index.html
Click Here to View Full Article
to the top


Moving Web-Based Software Offline
New York Times (05/31/07) P. C3; Helft, Miguel

Google announced that it has released a set of tools, called Google Gears, that will allow any software programmer to enhance Google's Web applications for offline use so they suit the programmer's requirements. "The whole idea of extending browser capabilities to offline is something that a lot of people are going to get pretty excited about," said Gartner Research vice president David Mitchell Smith. Google has been one of the most enthusiastic supporters of Web-based software and technology, and Google executives say such practices will create faster innovation because Web applications can be created quickly by merging existing components created by other programmers. Google says that Gears is in its early stages of development and will currently only work with Reader, a blog and news source tracking service, but the company plans to use Gears to make other programs available offline including Gmail, Calendar, and most notably Docs and Spreadsheets. While the offline availability of these programs will make them more competitive with Microsoft's Office, many analysts say that online programs, unlike Microsoft's competing software, lack the support and versatility that most businesses need.
Click Here to View Full Article
to the top


Taking It to the Streets: UCLA Scientists Seek to Turn Cars Into a Mobile Communications Network
UCLA News (05/26/07) Abraham, Melissa

The automotive industry has already utilized computing technology to make cars more efficient, safer, and easier to drive, but UCLA computer science professor Mario Gerla and researcher Giovanni Pau want to expand the computing capacity of cars by using them as computer nodes in a mobile network. "We have all of these computer devices as integrated systems inside our cars," Gerla says. "It's time to extend that concept. Computers are already being installed in many vehicles, and wireless capability will soon follow, so a mobile network deployment would only require the relatively low-cost addition of sensors to the vehicle's roof and bumpers and configuring the computer with new 'mobile' applications." A team led by Gerla at UCLA Engineering's Network Research Lab aims to reinvent cars and networks based on the principles of a wireless, mobile ad-hoc networking platform known as MANET. The platform would allow moving vehicles within a range of 100 meters to 300 meters of each other to connect and create a network of cars. Pau says the system uses radio protocols such as Digital Short Range Communication combined with wireless LAN technology to create the network, which can be used to gather and distribute safety information and complex multimedia data such as video. "The most essential aspect of this network is that it is not subject to memory, processing, storage and energy limitations like traditional sensor networks. It relies on the resources of the vehicle itself, along with those vehicles around it," Pau says. Aside from monitoring traffic information, a mobile vehicle network could provide essential communication infrastructure during emergencies, such as when communication infrastructures were destroyed during Hurricane Katrina and the attacks on Sept. 11, 2001. Although such a network raises privacy concerns, Gerla says that "consumer demand will ultimately drive rapid adoption past the point of concern over privacy."
Click Here to View Full Article
to the top


FEST Winner's Research May Revolutionize Computer Design
UVA Today (University of Virginia) (05/31/07) Maki, Melissa

University of Virginia assistant professor of computer science Kim Hazelwood's work on computer design earned her a $50,000 award from the Fund for Excellence in Science and Technology (FEST) Distinguished Young Inventor Grant program. Hazelwood says the traditional approach to optimizing computer systems by changing the hardware is limited as it only focuses on a single layer of computer design and ignores the interface between software and hardware, which has traditionally been fixed. Her system, called Tortola, creates a middle layer between hardware and software that translates and communicates between the two, creating the possibility of cooperative problem solving. "This middle layer would allow software to adapt to the hardware it's running on, something engineers have not been able to do in the past," Hazelwood says. An example of where a system such as Tortola would be beneficial, according to Hazelwood, was the famous Intel mishap where the company had to recall a massive number of microprocessors that were distributed before a flaw in their fine mathematics function was detected. Tortola could prevent such expensive mishaps by compensating for the flaws in the hardware, giving designers the ability to release products sooner because any problems could be fixed later. "Even so early in her career, Kim has done creative, extensive and groundbreaking work in developing techniques to integrate hardware and software," says Mary Lou Soffa, Chair of the Department of Computer Science and Hazelwood's mentor on the project.
Click Here to View Full Article
to the top


Northeastern University Researchers Solve Rubik's Cube in 26 Moves
Northeastern University News (06/01/07)

Northeastern University computer science professor Gene Cooperman and graduate student Dan Kunkle have proven that any configuration of a Rubik's cube can be solved in 26 moves. The research from Cooperman and Kunkle improves upon the previous record by one move. The use of 7 TB of distributed disk as an extension to RAM for holding large tables and the development of faster computer moves using mathematical group theory were keys to their research efforts. Once all of the configurations of a Rubik's cube were placed in a family of sets of configurations, the researchers focused on applying a single move to the family of cosets at once. They used the new mathematical group theory technique to simulate the result on a computer at a rate of 100,000,000 times per second. Computers at Teragrid and Northeastern were used. "Search and enumeration is a large research area encompassing many researchers working in different disciplines--from artificial intelligence to operations," says Cooperman. "The Rubik's cube allows researchers from different disciplines to compare their methods on a single, well-known problem."
Click Here to View Full Article
to the top


Spammers' Use of AI Only Just Begun
InfoWorld (06/01/07) Hines, Matt

Image-based spam attacks have emerged as a new front in the battle against spam, but spammers' use of artificial intelligence in such attacks is only the beginning, concludes a new Forrester Research report. Forrester analysts believe that the use of AI in spamming attacks will only grow more sophisticated and argue that the only way to prevent more attacks is to abandon the current tactic of trying to filter out every type of spam that mass-mailers develop and instead stop spam at its source. Forrester analyst Chenxi Wang says spammers are applying the same principles used in CAPTCHA, the challenge-response test found on may online applications that ask users to input characters found in an misshapen and discolored images, to bypass anti-spam programs. "People have devised new filters that use technologies such as optical character recognition that has curtailed the spread of image spam," Wang says. "Unfortunately, image spam is only one type of AI problem, and spammers have many they will use in the future; this is only the beginning of an arms race." Wang warns that without a major advancement in AI research, there is no possibility of bridging the gap between the number of methods spammers can deploy and anti-spam defenses. One of the methods that spammers are already beginning to utilize involves sending distorted and obfuscated text images, graphic pictures, and audio and video files, all of which can bypass existing image-filtering tools. Wang says that instead of trying to counter each type of spam, customers and technology providers need to focus on catching messages and fundamental properties contained in each variation, such as links to malware sites that are contained in most of spam messages.
Click Here to View Full Article
to the top


HP Opens College Lab in China for Media Search
CNet (05/31/07) Kanellos, Michael

Researchers at China's Tsinghua University have developed an application that allows security cameras to identify individuals and is being used to expedite border crossings between Shenzhen and Hong Kong. Hewlett-Packard wants to bring the application to consumer PCs to improve searches for video files and is opening a lab at Tsinghua University to assist in the research. The facility, HP's first complete lab at a university, will initially work on developing pattern matching and search technologies for audio tracks, videos, or photos. One search application developed at Tsinghua lets users find songs by humming a few notes of the tune while another application finds songs using their rhythm characteristics, genre, or vocal track, rather than the tags associated with the music file. Tsinghua has also developed a photo-matching application that finds different photographs of the same person by matching contours in the person's face. While these programs are still experimental, they work fairly well, according to HP's chief technology officer of imaging and printing systems Patrick Scaglia. "We've searched the world, and it's the best one we've seen," Scaglia says. "It is not ready to ship, but it gives us confidence that the problem can be cracked." HP is also working with the university on a computing grid that connects 20 of the grids at major universities to a larger computing unit.
Click Here to View Full Article
to the top


Better Face Recognition Software
Technology Review (05/30/07) Williams, Mark

The results of the Face Recognition Grand Challenge showed that machine recognition of human individuals has improved tenfold since 2002 and a hundredfold since 1995, and today the best face-recognition algorithms are even more accurate than most humans. National Institute of Science and Technology program manager for tests Jonathon Phillips says the improvement in accuracy is due to the development of high-resolution still images, 3D face-recognition algorithms, and the recent availability of 3D sensors, which directly capture information about the shapes of faces. Current recognition software also focuses more on distinctive features of a human face's surface, such as the curves of the eye sockets, nose, and chin, where tissue and bone are most apparent and do not change over time. Carnegie Mellon Robotics Institute research Ralph Gross says 3D facial recognition can also recognize subjects from different viewing angles, up to 90 degrees, which was a problem before, possibly because most facial recognition technology was used for tasks involving ID cards and face scanners, which use full frontal faces of cooperative subjects under controlled lighting. High-resolution still images have also improved face-recognition technology with detailed skin-texture analysis. Any patch of skin, called a skin print, can be captured as a image, broken into small blocks that algorithms can then measure, recording lines, pores, and actual skin texture. Gross says skin-texture analysis is capable of identifying differences between identical twins, which is impossible using facial-recognition software alone.
Click Here to View Full Article
to the top


At Some Schools, Facebook Evolves From Time Waster to Academic Study
Wall Street Journal (05/29/07) Lavallee, Andrew

Following a general obsession in students with social networking sites such as Facebook, schools are creating courses of study that focus on social networking, online communities, and user-contributed content, like the new graduate program in social computing at the University of Michigan. "There's a strong interest coming from computer scientists and from industry, based on the recognition that, increasingly, computing is indeed social," says Cornell University sociology professor Michael Macy. Cameron Marlow, a Yahoo researcher who received a Ph.D. from the Massachusetts Institute of Technology for his dissertation on how information spreads through blogs, says that since joining Yahoo in September 2005, there is a greater focus on hiring people with training in social computing. The Rochester Institute of Technology received a grant in April to develop a course in social media, and when the 90 students taking the class in the spring of 2008 study such social media as blogs and wikis, they will be studied by other communications researchers to see the effect social sites like Second Life have on work quality. Even schools that are not traditionally known for their technology programs have taken an interest in studying social computing. For the past year, Abilene Christian University in Abilene, Texas, conducted two research projects that used Facebook to examine student retention trends, partially because the school noticed how much time its students were spending on the site, according to K.B. Massingill, executive director of the division that funded the research.
Click Here to View Full Article
to the top


New Agency IARPA Develops Spy Tools
Associated Press (05/31/07) Shrader, Katherine

A new U.S. government agency called the Intelligence Advanced Research Project Activity (IARPA) has been established to develop ground-breaking technology for the U.S.'s 16 spy agencies. One technology under development is a "cloaking" technology that bends radar around an object, essentially making it undetectable. Others include smaller power sources using nanotechnology and faster code-breaking quantum computers, according to IARPA acting director Steve Nixon. IARPA has met with resistance, however, as members of Congress question the need for such a program. IARPA will be modeled after, though significantly smaller than, the Defense Advanced Research Projects Agency, which was created after the Russians launched Sputnik in 1957. Government agencies have had several technology development success stories in the past. The CIA developed lithium-iodine batteries, which are now used in pacemakers, as well as microdot cameras capable of creating images small enough to be hidden in the period at the end of a sentence. Nixon says that IARPA will not have labs and electron microscopes, but will sponsor research at universities, national labs, and other organizations. Nixon notes that IARPA will not be limited to hard sciences, but will also work on social-science problems such as finding tools for language research or analyzing the habits of other societies, as well as privacy protection.
Click Here to View Full Article
to the top


Software Learns When It Pays to Deceive
New Scientist (05/30/07) Merali, Zeeya

Computer scientists at the University of Witwatersrand in South Africa have developed a software agent that is able to bluff. Although some online poker players believe online poker sites use poker-playing bots to beat them, Witwatersrand's Evan Hurwitz suggests current computer agents would not be able to bluff convincingly because they are not programmed to deal with unexpected and illogical actions. Hurwitz teamed up with Tshilidzi Marwala to develop Aiden, an artificial intelligence bot based on a neural network algorithm for forecasting stock market movements. Initially, the researchers did not program Aiden with the rules for the card game lerpa, and the virtual player refused to play after it lost, so Hurwitz and Marwala forced Aiden to play the first 200 hands. Although Aiden was able to pick up the rules, the actions of opponents, and learn from its wins and losses, the virtual player still would not bluff until it was pitted against similarly trained bots. The computer agents began to develop playing styles on their own, and the aggressive bot Randy began to bluff after receiving bad hands and having to fold, while the cautious Aiden would sometimes fold even if it had a relatively strong hand. "This demonstrates that computers can learn this peculiarly human behavior," says Philippe de Wilde, a computer scientist at Heriot-Watt University in Edinburgh. "They generate the strategy from play, which is a very human way of learning."
Click Here to View Full Article
to the top


Robots Advance, Consumers Stall
Christian Science Monitor (05/31/07) P. 13; Peter, Tom A.

Industry observers say the average American household will include one or two simple robots within the next 10 years, but first Americans will need to overcome their preconceived prejudices against robots, which include skepticism, unrealistic expectations, and the fear that robots are dangerous. Americans have little problem with their unknown interactions with less obvious robots and robotic elements incorporated into larger products, such as cruise control, but the idea of autonomous robots in the house is still viewed cautiously by the American consumer. For example, for the same cost as a regular vacuum, a consumer can purchase the Roomba robot, which automatically vacuums and even knows when to charge itself, but after four years on the market only 1 percent to 2 percent of American households felt compelled or comfortable enough to use a robot for a task as menial as vacuuming. The problem in many cases is that people question if the Roomba really works or is just an elaborate gimmick. Another problem is that Americans have an underlying cultural fear of robots due to movies involving robots that overthrow humans or hunt them down. As with many technological changes, a generational shift may be needed to bring mainstream acceptance of robots into American culture. Children are frequently the first to test new robotic innovations. Aside for acclimating the next generation to the presence of robots, robotic toys can also provide a realistic expectation of what robots can do. Sarjoun Skaff, cofounder of robotic toy company Bossa Nova Concepts, says that unrealistic expectation create distrust for machines like Roomba. "So far, our perception has been shaped by science-fiction movies," Skaff says. "And the public's expectation of what the robots can and should do far exceeds the technical ability of today's robots." Even when robotics manage to overcome the public's general distrust, robots will still need to be made simple enough so that even the least technologically inclined consumers can use robot products with ease.
Click Here to View Full Article
to the top


Xerox Develops New Way to Print Invisible Ink
Extreme Tech (05/30/07) Gardiner, Bryan

Xerox announced that its scientists have perfected a new way to print hidden fluorescent content on standard printing equipment. Xerox says the new technique will allow companies to add an additional layer of security to sensitive material such as checks, tickets, coupons, and high-value documents. Xerox Innovation Group research fellow and co-inventor of the patented process Reiner Eschbach says the hidden fluorescent print will only show up under ultraviolet light and does not require the use of special fluorescent inks. "That means a four-color digital printer can print everything it normally would, and it can simultaneously individualize a document with a fluorescent identifier," Eschbach says. The fluorescent technology is part of a larger research project Xerox is conducting to build security into documents based on a digital printer's ability to make any element on the page, such as lines, text, and images, unique to the recipient. Eschbach and a research team realized that most paper manufacturers already inject fluorescent brightening agents to enhance the "whiteness" in the paper. The team then focused on creating certain toner combinations that would allow the paper's fluorescent elements to shine through under ultraviolet light. The result is a font that uses inherent contrast to basically "write" hidden fluorescent content. Xerox expects that eventually this technology will be used to make personal checks with a printed fluorescent signature that the merchant or bank teller can compare to the actual signature.
Click Here to View Full Article
to the top


Nose Can Play Music on New Instrument
Discovery News (05/29/07) Staedter, Tracy

A recent graduate of Rensselaer Polytechnic Institute in New York State is behind an adaptive-use musical instrument that is designed to serve as a musical therapy tool for people with physical disabilities, such as cerebral palsy. Zane Van Dusen, a graduate of computer science and electronic media arts and communications, teamed up with Pauline Oliveros, a musician and arts professor at Rensselaer, to develop the interface, which relies on an affordable webcam and specialized computer software. The user, sitting in front of a computer, sees live video of their face through a webcam, and uses motion-tracking software to choose a point to track, such as the tip of their nose. Movements are tracked across an onscreen keyboard, with low notes to the left and high notes to the right, and sounds are also determined by the rectangle that appears on the screen around the face of the user. The percussive mode allows the user to produce a snare drum or cymbal sound by moving outside the rectangle. The tool helps children focus more on their movements, according to a pilot study at Rehabs Programs in Poughkeepsie. "The added benefit of all this is that the children are working on their head control," says Leaf Miller, a professional musician and an occupational therapist at Rehab Programs. "It can also be adapted for speech language pathologists to use for communication."
Click Here to View Full Article
to the top


Internet Governance Forum in November to Address Access, Security Issues, UN Official Says
eGov Monitor (05/24/07)

The second U.N. Internet Governance Forum, to be held in Rio de Janeiro Nov. 12-15, will concentrate on issues such as access, security, and openness, according to U.N. officials who attended a May 23 meeting about potential topics for the forum. Around 200 Internet stakeholders participated in the May 23 meeting, and some participants opined that the November forum should focus more on internationalized domain names and other Internet resources, said Markus Kummer, executive coordinator of the forum's secretariat. Participants also felt that the November forum should go beyond the scope of the first forum in Athens. Nitin Desai, the secretary-general's special adviser on Internet governance, said the shape of the forum could evolve in years to come, explaining that "we are experimenting with a multi-stakeholder open-ended process without a fixed membership." Desai announced that India and Egypt would host the 2008 and 2009 forums, respectively.
Click Here to View Full Article
to the top


Read ID, Real Debate
Washington Technology (05/28/07) Vol. 22, No. 9, P. 24; Lipowicz, Alice

Security professionals, vendors, and trade groups continue to argue over the feasibility and effectiveness of the 2005 Real ID Act, which would standardize driver's licenses nationwide. Under the act, states would gather and electronically house millions of individuals' personal information, and the states' databases would link together. The concept was developed by the 9/11 Commission to close gaps in the current system, but critics contend that the act's mandates would put people at risk for identity theft, racial profiling, and other threats to civil liberties. Eugene Spafford, ACM's U.S. policy committee chairman, asserted that Real ID sets up the possibility of identity theft "on an unprecedented scale," and voiced concerns that states will establish insufficient privacy protections as they hurry to meet Real ID deadlines. Spafford cited audit trails, strong data access controls, and employee background checks as types of protections that should be employed, as well as a paper trail for the system. The DHS Data Privacy and Integrity Advisory Committee declined to sanction the program, as panelists felt that concerns about privacy, data security, and cost, among others, had yet to be resolved. Still, the committee noted that the American Association of Motor Vehicle Administrators' (AAMVA) database system could be a potential prototype for Real ID. ACM also characterized AAMVA's system as "effective," and said that its system design could create a national database, if expanded in scope. For more information on ACM's Real-ID activities, visit http://www.acm.org/usacm
Click Here to View Full Article
to the top


Melding Mind and Machine
ANU Reporter (05/07)

Australian National University Ph.D. student James Sheridan dreams of making a device that can sense its user's mood and make suggestions to help the person feel better. "It's an idea I've had for a little virtual brain in a personal assistant that you carry around with you," said Sheridan. "Maybe there are sensors melded into your hat, or parts of your cloths, that are able to monitor your brainwaves or heart rate. The device will know that you're quite stressed. It will know from your schedule that you've got an hour free. The device may have a GPS, so it will know where you are and what's around you. You still have control, but the computer's suggestions are going to be much more contextualized to your life." Sheridan's desire to create a more user-sensitive device came from his need to create a research project that would fit his needs, particularly his struggle with Attention Deficit Disorder. Sheridan negotiated a doctorate that split his study between the Computer Science and Information Technology department and the Centre for New Media Arts at ANU, allowing for part of his assessment to include elements of a performance and an installation. While creating a form of art that could "tune" itself to human thought, Sheridan started developing what he calls a "mental Zen garden," which will allow the interaction of humans and machines through sensory equipment and software. A user is placed in a virtual reality laboratory that can track human vision and movements while creating 3D graphics. Head gear containing electroencephalography sensors read electrical activity in the user's brain, feeding information about thought processes into an equation. The result is an evolving garden with animated branches and environmental sounds based off of the users mental thoughts. Sheridan's project has spurred several other graduate student projects that are helping to create the software to track movements and thought processes, and could someday result in emotionally responsive PDAs and computers.
Click Here to View Full Article
to the top


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

To unsubscribe from the ACM TechNews Early Alert Service: Please send a separate email to [email protected] with the line

signoff technews

in the body of your message.

Please note that replying directly to this message does not automatically unsubscribe you from the TechNews list.

ACM may have a different email address on file for you, so if you're unable to "unsubscribe" yourself, please direct your request to: technews-request@ acm.org

We will remove your name from the TechNews list on your behalf.

For help with technical problems, including problems with leaving the list, please write to: [email protected]

to the top

News Abstracts © 2007 Information, Inc.


© 2007 ACM, Inc. All rights reserved. ACM Privacy Policy.