ACM TechNews (HTML) Read the TechNews Online at: http://technews.acm.org
ACM TechNews
September 19, 2007

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

Welcome to the September 19, 2007 edition of ACM TechNews, providing timely information for IT professionals three times a week.


HEADLINES AT A GLANCE:

 

Big Future Beckons for Tiny Chips
BBC News (09/19/07) Fildes, Jonathan

Intel has displayed what the company says is the world's first working chips built with transistors with features only 32 billionths of a meter wide, meaning Moore's Law will remain intact for the foreseeable future. Intel co-founder Dr. Gordon Moore says that he expects the law named after him will continue for at least 10 years more. "Eventually, however, we're down approaching the dimensions of individual atoms and that's clearly as far as we can go down the path of shrinking dimensions," Moore says. The next generation of chips displayed by Intel, known as Penryn chips, contain transistors with features just 45 nanometers in size and pack 410 million transistors in an area the size of a postage stamp. Penryn's successor, called Nehalem, will be launched in 2008 and will nearly double the number of transistors on the chip. Such tiny chips, which are also being released by IBM and its partners Toshiba, Sony, and AMD, have poor performance in their gate dielectrics and allow currents passing through their transistors to leak, reducing the efficiency of the chip. To increase chip efficiency, Intel replaced silicon dioxide gate dielectrics with metal hafnium. Moore says the new materials' development and integration into working chips is "the biggest change in transistor technology" since the late 1960s. Hafnium is a high-K metal and is better able to store an electrical charge than silicon dioxide. Silicon manufactures have already planned beyond this next generation technology, as devices with features 22 nanometers in size are expected to be released in 2011. The industry does expect to reach a physical limitation eventually, though Moore does not think such limitations will stop technology from advancing.
Click Here to View Full Article
to the top


Group Says E-Voting Paper Trail Wouldn't Improve Security
IDG News Service (09/18/07) Gross, Grant

A report from the Information Technology and Innovation Foundation (ITIF) think tank concludes that requiring printouts as a back-up to electronic voting would not improve security and would increase the costs of U.S. voting systems. The ITIF says that voter-verified paper trail ballots used with e-voting machines would prevent the use of more innovative voting technology that providers better security, transparency, and reliability than paper-only voting systems. The ITIF report also notes that people are willing to trust computers with many other important functions such as banking, medicine, and aviation. Supporters of paper-trail ballots dispute the report's findings. "The argument that people trust computers in other places is specious--safety-critical systems have been developed in other contexts using rigorous standards that are not applied to voting machines," says Eugene Spafford, chairman of ACM's U.S. policy committee. ACM has not called for e-voting machines to be abandoned, but suggests that e-voting machines go through two levels of auditing, paper trails and random machine audits, says Spafford, who notes that beyond the hacking threat, "errors, bugs, and accidents can also result in problems unless there is an independent, durable audit trail." Meanwhile, VerifiedVoting.org President Pamela Smith disputes the report's suggestion that a growing technophobic movement is driving mistrust for e-voting. "The harshest critics of e-voting--in particular paperless e-voting--are computer technologists who are the literal opposite of technophobic," Smith says. For more information on ACM's e-voting activities, visit http://www.acm.org/usacm
Click Here to View Full Article - Web Link May Require Free Registration
to the top


CS Profs and the DOD
Computing Research Association (09/18/07) Harsha, Peter

Recent policy changes at the Defense Advanced Research Projects Agency (DARPA) have reduced university participation rates in DARPA-funded computer science research projects. Between fiscal years 2001 and 2004, the amount of funding from DARPA to U.S. universities for computer science research fell by half, and evidence suggests that funding for universities is currently lower still, writes the Computer Research Association's Peter Harsha. Diminished support for university computer science not only creates a gap in federal IT research and development, but also weakens the "DARPA model" of research support. Since the early 1960s, the country has benefited from the two different approaches to research that the NSF and DARPA have taken. While the NSF focused primarily on small grants for individual researchers, DARPA worked to identify key problems of interest and to create and support communities of research to solve the problems. DARPA-supported research in computer science over the past four decades has established the U.S. economy and military as the most dominant forces in the world, Harsha says. Reducing support for academic computer science means some of the brightest computer scientists in the country are no longer working on defense-related problems. Many experienced computer science researchers say there is an entire generation of young researchers who have no experience working on DARPA and Defense Department projects. The Computer Science Study Group, managed by the Institute for Defense Analysis for DARPA, focuses on introducing researchers to the needs and priorities of the Defense Department by running workshops, mentoring, and hosting tours of DOD facilities, but the group does little to bring DARPA interests back into university research.
Click Here to View Full Article
to the top


House Committee Chair Wants Info on Cancelled DHS Data-Mining Programs
Computerworld (09/18/07) Vijayan, Jaikumar

House Committee on Homeland Security Chairman Bennie Thompson (D-Miss.) asked Department of Homeland Security secretary Michael Chertoff to produce a detailed list of all IT programs that were canceled, discontinued, or modified due to privacy concerns. Thompson also requested information on how much money the DHS spent on each program, the names of contractors involved in the projects, and information on how privacy issues are now being addressed. Thompson's inquiry follows the cancellation of DHS' Analysis Dissemination Visualization Insight and Semantic Insight (ADVISE) data-mining program, after $42 million had been spent on the project but privacy concerns had not been properly addressed. In a letter to Chertoff, Thompson expressed his concern over the "apparent litany" of DHS programs that have been canceled or modified after millions of dollars had been spent because of a failure to address privacy concerns early in the process. Thompson notes that a report by the DHS' inspector general showed that the department's privacy office was unaware that an ADVISE pilot program was using real data. In addition to the ADVISE program, Thompson has requested information on the $100 million Computer Assisted Passenger Prescreening system, the $8 million Multi-State Anti-Terrorism Information Exchange pilot project, and the $140 million Transportation Security Administration's Secure Flight effort, all of which were cancelled due to privacy concerns or security vulnerabilities.
Click Here to View Full Article
to the top


Clock to Tick Down U.S. Privacy
Washington Times (09/18/07) P. A3; Hudson, Audrey

The American Civil Liberties Union's "Surveillance Society Clock" is counting down until the U.S. government stops spying on private citizens as part of the war on terror, and the clock is quickly approaching midnight. "The extinction of privacy is a real possibility," says Barry Steinhardt, director of the ACLU's Technology and Liberty Project. "We believe that privacy is not yet dead--it is a patient on life support." The ACLU clock is modeled after the "Doomsday Clock," created by the Bulletin of the Atomic Scientists in 1974 to warn against the possibility of a nuclear holocaust. Steinhardt says rapid advancements in technology and data mining are leading to the possibility of a "1984-style surveillance society" and creating a false sense of security. "The false security of a surveillance society threatens to turn our country into a place where individuals are constantly susceptible to being trapped by data errors or misinterpretations, illegal use of information by rogue government workers, abuses by political leaders--or perhaps most insidiously, expanded legal uses of information for all kinds of purposes," says a new ACLU report on mass surveillance by the government. The clock is currently set at six minutes before midnight, and will be updated as events warrant moving the time closer to or further away from midnight. "With a flood of new technologies that expand the potential for centralized monitoring, a president who believes he can unilaterally sweep aside the laws that restrain government spying ... we confront the possibility of a dark future where our every move, our every transaction, our every communication is recorded, compiled and stored away, ready for access by the authorities whenever they want," the report says.
Click Here to View Full Article
to the top


Penn Engineers Design Electronic Computer Memory in Nanoscale Form That Retrieves Data 1,000 Times Faster
University of Pennsylvania (09/17/07) Reese, Jordan

University of Pennsylvania scientists have developed nanowires capable of storing computer data for 100,000 years and are also more energy efficient and 1,000 times faster than existing portable memory devices such as Flash memory and micro-drives. The researchers developed a self-assembling nanowire made of germanium antimony telluride, a phase-changing material that can switch between amorphous and crystalline structures, the key to the computer memory system. The nanoscale devices, about 100 atoms in diameter, were fabricated without conventional lithography. Instead, a self-assembly process was used in which chemical reactants crystallize at lower temperatures mediated by nanoscale metal catalysts to spontaneously form nanowires that were 30-50 nanometers in diameter and 10 micrometers in length. Memory devices were then fabricated on silicon substrates. Tests of the device shows extremely low power consumption during data encoding, only 0.7mW per bit, and data writing, erasing, and retrieval took only 50 nanoseconds, 1,000 times faster than conventional memory devices. "This new form of memory has the potential to revolutionize the way we share information, transfer data, and even download entertainment as consumers," says Ritesh Agarwal, one of the developers and an assistant professor in the Department of Materials Science and Engineering.
Click Here to View Full Article
to the top


The Singular Question of Human vs. Machine Has a Spiritual Side
Wall Street Journal (09/19/07) P. B1; Gomes, Lee

There are people who believe there will one day be a point of "singularity" when human intelligence is overtaken by machine intelligence, and they speculate that a new, super-intelligent organism cross-bred from man and machine could be one of the monumental developments this singularity could bring about. Lee Gomes writes that singularity advocates talk at length about the need for Artificial General Intelligence (AGI), which is seen as a key singularity milestone. Yet he says AI researchers have been trying, so far unsuccessfully, to achieve this goal since the 1950s. "There is a schism between the AGI and the AI worlds," Gomes notes. "The AGI faction thinks AI researchers have sold out, abandoning their early dreams of 'general' intelligence to concentrate on more attainable (and more lucrative) projects." Gomes agrees with this assessment, but while AI researchers insist that the revision of their approach was unavoidable given the naivete of their earlier ambitions, singularists are undaunted in their belief that new approaches will yield AGI breakthroughs. Gomes entertains the notion "that the discussion of singularity involves a sublimated spiritual yearning for some form of eternal life and an all-powerful being, but one articulated by way of technical, secular discourse," and he perceives significant intersection between singularists and proponents of "life extension." He adds that the popularity of the Search for Extraterrestrial Intelligence program among singularists reflects a desire for a messianic figure from space, which seems to again indicate that the need for spiritual enlightenment through advanced technology is a running theme among the singularity set.
Click Here to View Full Article
to the top


Reverse-Engineering the Brain for Better Computers
EE Times (09/18/07) Maxfield, Clive

University of Texas at San Antonio researchers are trying to build a better computer by reverse-engineering brain neurons. UTSA biology researchers are using Interactive Supercomputing's Star-P software to run biologically-realistic simulations of molecular diffusion in neurons with the hope that understanding how neurons process chemical signals when a person learns and retains information will result in more reliable computers that have stochastic computing components. Stochastic computing is a type of artificial intelligence that uses probabilistic methods to solve problems. The human brain has trillions of different types of neurons, each with complicated branching dendrites, so running the complex simulations to model even a single neuron requires massive amounts of computational performance and memory resources. To meet these requirements the researchers used a Star-P license to link their desktop computers to an eight-processor parallel cluster, and the team will soon be able to use a 120-processor cluster thanks to an additional license from ISC. In addition to advancing computing techniques, UTSA's research could lead to other neurobiological research breakthroughs, particularly in sensory acquisition, motor learning, and neurological diseases such as Alzheimer's, Parkinson's, schizophrenia, and epilepsy.
Click Here to View Full Article
to the top


New Research Seeks to Enhance Quality and Security of Wireless Telemedicine
Rochester Institute of Technology (09/17/07) Dube, Will

Researchers at the Rochester Institute of Technology and the University of Alabama are working to advance the use of radio frequency identification (RFID) technology in cardiac sensor networks, a new wireless technology for telemedicine delivery. The researchers will work on improving the security of the systems, reducing the possibility of identity theft and cyber-terrorism. "Telemedicine technology can greatly increase the quality of medical care while also decreasing health care costs," says Rochester Institute of Technology assistant professor of computer engineering Fei Hu. "Through this project we hope to increase the integration of RFID into existing cardiac sensor networks, ensure the overall security of the system and promote the implementation of the technology in nursing homes and adult care facilities across the country." One of the major challenges of the project is concern over the security of wireless networks used in telemedicine delivery. Hu and University of Alabama computer science professor Yang Xiao will research the use of anti-interference technology to reduce radio distortion on the networks, and design and test new RFID security systems that will decrease the chance of information being stolen. "There are well known security challenges associated with cardiac sensor networks and RFID," Hu says. "It is my hope this research will assist in better protecting these systems and allow greater numbers of doctors and patients to take advantage of the benefits of telemedicine."
Click Here to View Full Article
to the top


Multi Robot Underground Pipeline Explorer Being Developed at Louisiana State University
Louisiana State University (09/15/07)

Robots will improve the maintenance of underground pipelines in the future, according to robotics and computer science experts at Louisiana State University. A team of researchers from LSU's Robotics Research Laboratory (RRL) and computer science department has developed a prototype miniature robot that is capable of easily maneuvering through pipelines four to six inches wide in diameter, and is testing it under different conditions. The team behind the Autonomous Pipeline Exploring Robots (APER) project believes that hundreds of inexpensive robots could be used to traverse Tees and elbows, discover problems, and provide data for mapping underground pipelines. "Most of the pipelines be it gas or sewage or water requires manual maintenance operations, so we are focusing on completely autonomous maintenance operations by employing these cheap tiny robots," says Jong Hoon Kim, chief architect of APER. "The basic theme of our research is to employ many robots to map the pipeline simultaneously, which will result in efficient mapping and faster search operations." The researchers say the Pipeline Explorer will lower costs, improve efficiency, and inspire other applications.
Click Here to View Full Article
to the top


Click Go the Votes, Click, Click, Click...
Sydney Morning Herald (Australia) (09/18/07) Timson, Lia

Australia is making progress in bringing its electoral system into the digital age, with the country's first electronically aided federal vote slated for this year. As many as 2,500 Australian Defense Force (ADF) personnel scattered throughout the world will be able to vote remotely using the ADF's satellite and ground-based telecom infrastructure. The votes will be encrypted and wired electronically to the tally room through the ADF intranet. The ADF test was recommended by the Joint Standing Committee into Electoral Matters, as was a machine-assisted voting test for the visually impaired scheduled to operate in 29 locations this year. Electronic polling and tabulation was carried out with a direct-recording electronic (DRE) voting machine and an electronic voting and counting system (eVACS) by the ACT in 2001 and 2004, and the tests determined that the ACT system was secure and reliable. But the Australian Electoral Commission balked at the cost of installing eVACS at all 7,700 national polling sites, while concerns over the duration of screen-based voting and the need to redesign the ballot paper for use with a DRE screen also led the joint committee to conclude that there was "no need to rush into the widespread implementation of DREs, especially when the costs may overwhelmingly outweigh the benefits." In addition, the ACT decided not to implement remote e-voting for Australians through the Internet, touch-tone phones, interactive digital TV, mobile SMS, and private intranet because such a move would eliminate polling place attendance, which was deemed to be a "key contributor to Australia's democracy." Australian software developer Alex Pollard warns that full security cannot be appended to e-votes, and says the paper-based system is trustworthy because scrutineers from the big political parties maintain a honest vote tally. Pollard says he would have fewer objections to e-voting systems if they provided an auditable paper trail and count.
Click Here to View Full Article
to the top


Japan Eyes Robots to Support Older Population
Reuters (09/12/07)

Robots are rarely found outside of industrial sites in Japan, but the nation's researchers hope to change that as the population ages. Approximately 40 percent of the population will be over 65 by 2055, and robotics researchers want to provide assistance to the elderly by developing robots that are capable of operating in homes. The labor force will also take a hit due to its older population, and the researchers are eyeing smart robots for offices and other venues as well. Researchers at the University of Tokyo have teamed up with experts at Toyota, Fujitsu, Mitsubishi, and several other top Japanese firms to lay the foundation for the development of next-generation robots in the next 15 years. They plan to unveil prototypes capable of performing mundane tasks in 18 months. The new robots do not have to look human, says Isao Shimoyama, dean of Tokyo's Graduate School of Information Science and Technology. Currently, about 10 buildings use a vacuuming machine (a droid with wheels) to clean their floors, and about a handful of shopping malls and corporate sites use Enon, a guide and patrol robot that has a humanoid upper body but no legs. Two-legged humanoid robots are unlikely to find their way into homes for some time.
Click Here to View Full Article
to the top


Employers Back Initiative to Turn Youngsters on to IT
Computer Weekly (09/17/07) Thomson, Rebecca

E-Skills UK has signed up Microsoft, Cisco, Vodafone, BBC, LogicaCMG, and John Lewis for a new initiative that seeks to create a positive image of information technology in the minds of students. Since 2001, the number of students pursuing IT-related degrees has fallen about 50 percent, even though the sector has played a key role in the prosperity of the United Kingdom, says E-Skills UK CEO Karen Price. E-Skills UK expects to add more companies to its Revitalize IT scheme, which also includes schools and universities as participants. Launched Monday, the two-year pilot will have employers lead 50 workshops in London and the South East for 40,000 students, and assist universities in developing their curricula for the IT job market. "The employer involvement is a major boost for us," says Richard Pettinger, director of the information for business program at University College London. "It is no longer enough to teach computer science--you have to put it in context." There are plans to expand the program nationwide if it is successful.
Click Here to View Full Article
to the top


Copyright Office Chief: I'm a DMCA Supporter
CNet (09/17/07) Broache, Anne

The Digital Millennium Copyright Act is not perfect, but it is doing what it is supposed to do, says the U.S. Copyright Office chief. Speaking at the Future of Music Policy Summit in Washington, D.C., Marybeth Peters, Register of Copyrights, said she is not opposed to the controversial 1998 law. Consumer groups criticize the DMCA because of its impact on fair use rights, and programmers, open-source advocates, and some hackers have come out against the law because of its anticircumvention dictate. When asked about the anticircumvention rules by an audience member, Peters said: "I think that's a really important part of our copyright owners' quiver of arrows to defend themselves." Peters said she has come around on the issue of having the Copyright Office decide locked content disputes. "In hindsight, maybe that's not such a bad thing," she said. Peters did suggest the DMCA is not tough enough on hosted companies for legal liability.
Click Here to View Full Article
to the top


Business by Numbers
Economist (09/13/07) Vol. 384, No. 8546, P. 85

The application of creative computation to business complexity is reflective of the proliferation of algorithms throughout our daily life. An algorithm is basically a step-by-step technique for performing a task, and the kinds of problems businesses are applying algorithms to are generalized into two categories; one category is process improvement, while the other is data analysis. One reason algorithms are growing in importance is because businesses are seeking to harness data in order to deliver personalized service to customers. "Optimization" algorithms are being used by many companies, including mail services, logistics firms, and telecoms operators. Jeff Gordon with the Covergys call center operator warns that "if you get the algorithm wrong and put customers into the wrong hands you degrade the [call center] experience. No one likes being handed off to someone else." The most powerful algorithms are those that can manage continuous change, and operators are exploring ways to combine algorithms that find the shortest route through a network with those that control the speed of the data stream. Meanwhile, companies are using statistical algorithms to cope with complex datasets, while algorithms' affiliation with Internet search engines is widespread. The effectiveness of algorithms depends on the proper alignment of several elements, including the provision of an intuitive user interface and users with algorithm proficiency.
Click Here to View Full Article
to the top


If RSA Is Cracked, Here's Plan B
New Scientist (09/15/07) Vol. 195, No. 2621, P. 31; Graham-Rowe, Duncan

Two research groups, one from the University of Queensland in Brisbane, Australia, and the other from the University of Science and Technology of China, in Hefei, say they have successfully created quantum computers that can run a routine called Shor's algorithm, a development that could have a profound impact on cryptography and how we protect our banking, business, and e-commerce data. Some say that quantum computing is nowhere near developed enough for real-world code breaking, but others say that cryptography will have to develop beyond current prime-number-based encryption techniques. The ability to run Shor's algorithm indicates the quantum computers are capable of using quantum processes to factorize large prime numbers. Almost every strong encryption system relies on a regular computer's inability to factor such numbers in a reasonable amount of time, exactly what the two groups claim to have done. However, Jon Callas, head of technology at cryptographic software developer PGP, says the work done by the two groups is significantly behind current cryptography techniques. Callas says the researchers only used four qubits, whereas current cryptography uses about 4,000 bits, which would require a quantum computer with about 50 trillion qubits. Eventually, the number of qubits in quantum computing is expected to surpass the point where it can outperform traditional computers and the length encryption keys can reach, but that could be 50 years away. When that point is reached, however, there are other cryptographic systems that even quantum computing will have trouble with. Hash chains, for example, use a sequential encoding process, and there is currently no known way to break them using a quantum computer.
Click Here to View Full Article - Web Link to Publication Homepage
to the top


Peer to Patent Project Sees First Submissions
Software Development Times (09/01/07)No. 181, P. 8; Handy, Alex

The Peer to Patent Project, designed by the New York Law School's Institute for Information and Policy and the U.S. Patent and Trademark Office (USPTO), launched its pilot trail in mid June, giving computer-related patent seekers the opportunity to fast-track their patent applications. The Peer to Patent Project allows computer-related patent applicants to post their application online so the public can post links to prior art and vote on the validity of the application during an 18-week comment period. The most valid comments are sent to the USPTO with the applications, which can be quickly processed and thus skip the estimated 40-month waiting period for standard patent approval. Some big technology companies are already submitting patent applications to the Peer to Patent Project, including a cryptography application by IBM, a digital rights management application from Microsoft, and three patent applications from GE. New York Law School research fellow and Peer to Patent Project manager Christopher Wong says with support from major companies such as IBM and GE it is hard to argue against the validity of the project. Wong hopes when the project is completed next year that the USPTO will find a permanent place for the project. Wong says the project specifically focuses on prior art, the largest bottleneck in the approval process.
Click Here to View Full Article
to the top


Cultural Modeling in Real Time
Science (09/14/07) Vol. 317, No. 5844, P. 1509; Subrahmanian, V.S.

Policymakers' prediction of political, economic, and social groups' behavior can be aided through the use of computer models under development, according to V.S. Subrahmanian of the University of Maryland Institute for Advanced Computer Studies. Previous behavior models have required a sizable construction effort, involving surveys done by hand or through interviews, or statistics gathered through close observation of subjects. However, such models are inapplicable in politically turbulent regions such as modern-day Sudan or Iraq. Software development is only now starting to tackle the challenge of building behavioral models in real time. Among the tools that can be utilized to model political parties, terror organizations, companies, regulatory bodies, and other entities is the Cultural Reasoning Architecture (CARA), which uses stochastic modeling agents, and The Resource Description Framework Extractor (T-REX) program, which employs socio-cultural-political-economic-religious variables. The last step in the process is to predict the actions of the modeled group's members once a set of determining conditions has been established. "The ability to access real-time information on these topics, to rapidly analyze the possible actions that interested parties might engage in, and to determine how best (e.g., with methods of game theory) to respond, will provide a key tactical advantage to organizations that are entering foreign cultures with goals as diverse as stopping terrorism or improving corporate profits," Subrahmanian concludes.
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.