In a key case for the technology industry, San Francisco federal judge Susan Illston on Thursday questioned the legality of DVD copying software developed by 321 Studios. Judge Illston indicated that she may ban the distribution of the software, although she also questioned the constitutionality of the 1998 Digital Millennium Copyright Act, which bans the circumvention of digital copy controls. The Motion Picture Association of America wants such software outlawed to ostensibly curb movie piracy, but consumer-rights organizations and certain technology groups (including ACM) counter that such a policy tramples on the fair-use rights. Without such provisions, consumers have no choice but to buy a new DVD if an old one is damaged. DVD copying software is available online, both freely and for purchase from vendors such as 321 Studios, whose business would be threatened if Illston rules in favor of Hollywood. Illston cited two recent rulings against programs that attempt to bypass electronic locks as factors that have "substantially persuaded" her to believe that 321's software violates the DMCA; MPAA wants Illston to issue a pretrial order to stop the distribution of the programs, while 321 is seeking a jury trial. Daralyn Durie, 321's lawyer, argued that the DMCA is unconstitutional since it effectively provides copyright holders with permanent copyrights, while MPAA lawyer Russell Frackman argued that content creators are not obligated to offer unprotected copies of their works to the public. Illston said she will rule shortly on whether the case will go to trial.
Click Here to View Full Article
(Access to this site is free; however, first-time visitors must register.)

The FCC has changed a 40-year rule to allow radio spectrum owners to lease or trade access, a move that will boost the wireless communications market and foster more efficient use of the radio spectrum. Instead of having to own the license themselves, wireless carriers and other companies that want to use radio spectrum will be able to forego large investments and instead lease or trade access on an ad hoc basis. The rule preventing such third-party access was established in 1963 to ensure liability for interference and other infringing uses. Former FCC Chairman William E. Kennard proposed the change three years ago in response to complaints of a dearth of spectrum space and flexibility. With the new rules, wireless carriers will be able to inexpensively patch up their networks, covering dead spots and adding capacity in congested areas, which likely will promote the use wireless data transfers via handheld computers and other devices. Carriers will also be able to provide wireless services to underserved rural regions. The spread of wireless communications such as wireless Web-surfing is prompting more and more companies to demand more spectrum. The FCC set up a new rule that makes the license owner responsible for interference and other regulatory compliance issues except in the case of long-term transfers where the licensee would assume responsibility.
Click Here to View Full Article
(Access to this site is free; however, first-time visitors must register.)

Members of Congress brought in the heads of four key U.S. security agencies to discuss the progress of their cyber-defense research and development efforts on Wednesday, and concluded that the country is still unprepared for a cyberterrorist assault on its critical infrastructure, which is becoming more and more dependent on computerized systems. House Science Committee Chairman Rep. Sherwood Boehlert (R-N.Y.) said the agencies were not committing enough resources to cyber research: For instance, the Homeland Security Department's Science and Technology Directorate has only tagged $7 million out of an $803 million 2004 budget request for cybersecurity research. Earlier legislation allocated $110.25 million for cybersecurity research at the National Science Foundation (NSF), yet the agency has submitted a request for a mere $51 million. Meanwhile, the Defense Advanced Research Projects Agency's (DARPA) unclassified cybersecurity research budget fell from roughly $90 million to $30 million over the past three years, although agency director Tony Tether claims DARPA's 2004 cybersecurity budget will be in the $100 million range. He claimed his agency's problem is plenty of funds but not enough cybersecurity research proposals, while NSF director Rita Colwell said her agency faced too many proposals and insufficient funds. Despite such complaints, the agency heads declared that progress is being made, and noted that they have started to collaborate on projects. But Boehlert insisted that, "The nation quite simply has been under-investing woefully in cyber security R&D, and as a result we lack both the experts and the expertise we ought to have in a world that relies so heavily on computers and networks for the necessities of everyday life."
Click Here to View Full Article

Federal and state law enforcement officials demanded the worldwide closure of open relays that spammers exploit at a Thursday event in Dallas. Open relays are mail servers that will forward mail to anyone online; mass emailers use them to send huge amounts of spam because the relays do not restrict connections to those from authenticated users or people on their local networks. Representatives of the FTC, the U.S. Securities and Exchange Commission, the U.S. Postal Inspection Service, and other agencies said they had submitted letters to operators of more than 1,000 global email servers, warning them that open relays cause problems for consumers and law enforcement, not to mention the servers themselves. The agencies also declared that they had launched 45 legal actions against spammers using the Internet to send deceptive emails or advertise untrustworthy products and services. The exploitation of open relays by spammers has led to universities, companies, and ISPs putting relays on blacklists such as the Open Relay Database. However, the costs of banning open relays would counteract the advantages they offer and could encourage ISPs to impose even more limitations, according to critics such as Electronic Frontier Foundation co-founder John Gilmore. FTC staff attorney Marc Groman said the agencies sent the letters to make operators aware of the problems inherent in open relays. The letter was signed by U.S. agencies as well as representatives from Japan, Canada, and Australia.
http://news.com.com/2100-1028_3-1001868.html

Electronic voting machines have gotten a boost since the controversial 2000 presidential elections, but the momentum is being countered by a group of technologists who warn that such systems are less secure than paper-based ones. Stanford University computer scientist David Dill is leading a group of more than 100 academic and commercial experts, including ACM, who say more needs to be done to ensure electronic voting systems, including backing up the results with paper records, which can be stored for future verification. Part of the problem, say opponents of electronic systems, is that the technology and process is opaque and non-verifiable for the public. Sequoia Voting Systems' Paul Terwilliger says anyone can view Sequoia source code, provided that they sign a nondisclosure agreement. Some election officials and makers of electronic voting systems argue that the dissenters do not really understand the underlying election process. Studies at the University of Georgia and surveys from Kansas jurisdictions show increased voter confidence and satisfaction with touch-screen systems. The Federal Election Commission says 19.6 percent of nationwide votes were collected using touch-screen technology in 2002, while 31.6 percent of votes were counted with optical scanning equipment similar to that used in standardized school tests. Although many election officials are aware of the technology's possible drawbacks, they say the machines are generally reliable and voters say they are more confident that their vote is being counted when using them. Still, technologists warn of the risks of election tampering with electronic voting systems as well as the possibility of machine breakdowns. Computer scientist Rebecca Mercuri says, "We're concerned. These machines are showing huge defects." She also notes that none of the systems have met even the lowest level of government computer security standards.
Click Here to View Full Article
(Access to this site is free; however, first-time visitors must register.)

For more information on ACM's concerns about e-voting, visit http://www.acm.org/usacm.

Charles McQueary of the Department of Homeland Security told the House Science Committee at a Wednesday hearing that his department is reorganizing in order to make it clear that cybersecurity is a priority. The hearing was characterized by legislative criticism of the Homeland Security Department, the Defense Advanced Research Projects Agency (DARPA), the National Institute of Standards and Technology (NIST), and the National Science Foundation's apparent failure to commit enough resources to cybersecurity research and development. DARPA director Anthony Tether countered that the bulk of his agency's resources are committed toward classified initiatives that are not common to the private sector. One such project is to enable military networks for wireless and peer-to-peer communication in order to beef up security from hackers, while another is a prototype 400-node network for running attack simulations. Committee chairman Rep. Sherwood Boehlert (R-N.Y.) declared that the research community has long complained that the Homeland Security Department, which was handed cybersecurity duties when President Bush dissolved the Critical Infrastructure Protection Board, is not devoting enough attention to plugging network security holes. "It's impossible to conclude that far more needs to be done," Boehlert said. Federal research officials told the committee that industry indifference was the real culprit behind cybersecurity's low priority. NIST director Arden Bement argued that private companies "don't really see themselves as a target."
http://www.eweek.com/article2/0,3959,1090046,00.asp

Meaningful Machines founder Eli Abir is developing an innovative translation method that could help computers comprehend context in human language. "The man literally has figured out the way the brain learns things," declares Meaningful Machines CEO Steve Klein. "On a theoretical level, his insight basically is this: Understanding a concept is nothing more than viewing a concept from two different perspectives." Abir's technique takes a cue from a statistics-based method in which a computer carries out word-to-word comparisons in already translated text, and refers to those comparisons in order to determine likely definitions when the same words appear in untranslated text. The main difference is that Abir's software uses phrase-to-phrase analysis. The algorithm scans large portions of already translated text and splits up matching sentences in different languages into phrases that are recorded into a database; also recorded are words that often appear either near the beginning or the end of those phrases. Abir has filed patents on these techniques--two of which are known as Automated Cross Language Database Builder and N-gram connector--but such methods are designed to be applied to other corporate products. If the language context conundrum can be solved through innovations such as Abir's, computers will one day be able to automate customer service and other labor-intensive chores and boost security by enabling government agents to study large amounts of data written in foreign languages.
Click Here to View Full Article

Patrick Andrews of British e-commerce consultancy Break-Step Productions claims to have developed software that imitates the processing pathway in people's upper visual cortex, thus giving machines the ability to perceive shapes in a wider range than conventional shape-recognition systems. The software, known as Foveola, distills shapes from a visual scene and designates a "mathematical signature" to each. These signatures allow the system to guess shapes that may seem unfamiliar due to distortion or noise. Foveola recognizes shapes after being trained like a neural net, but can identify shapes after seeing them only once. Andrews, a Cambridge University postgraduate who specialized in human vision, says several decades' worth of Cambridge research on shape recognition went into Foveola, which can identify approximately 3,000 specific shapes, icons, and letters. The majority of machine vision systems, in contrast, are designed to recognize text or industrial parts. Andrews believes Foveola could be used to boost the accuracy of handwriting recognition programs, enable robots to read signs and recognize faces, and even construct recognition systems for Asian languages or numerical notation. However, the Foveola software has yet to be tested or compared to other recognition technologies, and Gabriel Taubin of IBM's Thomas J. Watson Research Center says this can only be done through experimentation. Andrews says his company is currently in discussion with several Asian robot manufacturers about the possibility of enhancing numerous systems with Foveola.
http://www.wired.com/news/technology/0,1282,58846,00.html

Princeton University graduate student Sudhakar Govindavajhala presented a paper at the Institute of Electrical and Electronic Engineers (IEEE) Symposium on Security and Privacy on Tuesday, detailing a method he devised that exploits security flaws in virtual machines running on .Net and Java. Virtual machines enable software to operate on multiple platforms, while preventing applets from interfering with the computer's data by containing them in a software "sandbox"; Govindavajhala says his method allows that sandbox to be breached. Govindavajhala shined a light on a computer to heat up the chips and cause one or more memory bits to flip. His next step was to hack into the system by embedding his own code into memory and slipping the new code's address into the remaining free memory. This led to Govindavajhala's discovery that an arbitrary bit flip would allow the attack code to run over 70 percent of the time if he crammed 60 percent of memory with the addresses. Hackers could use this technique to circumvent smart card security, and Govindavajhala thinks that the increasing speed of computer memory and processors could boost the effectiveness of the method by reducing the amount of energy required to flip bits. "Here is a case where people thought they had thought of everything, but they hadn't," declared Burton Group analyst Fred Cohen, who does not think sandboxing untrustworthy applications will safeguard against such an attack. "If you let people run programs in your computer, then there is a chance they can do what they want."
http://news.com.com/2100-1009_3-1001406.html

University of Delaware researchers say they have built a cell phone-sized device, similar to a laser pointer, that makes it possible to tap into previously unattainable terahertz waves whose frequency exceeds that of microwaves a thousandfold. Whereas prior terahertz lasers were implanted with germanium, the new invention features a silicon-germanium nanostructure. James Kolodzey of the University of Delaware notes that the device adds portability to terahertz systems. The researchers also discovered that terahertz waves, or terawaves, can pass through solid matter like x-rays, but with no radiation or harmful side effects. Terawaves occupy the electromagnetic spectrum between infrared light and microwaves, and are not well suited for long-range communications because of atmospheric absorption. Ingrid Wilke of Rensselaer Polytechnic Institute's (RPI) Center for Terahertz Research reports that silicon-germanium may be more interoperable with silicon-based integrated circuits. RPI physics professor Michael Shur declares that terahertz technology "is expected to find numerous applications in imaging, radio astronomy, biology, medicine, atmospheric measurements from satellites, advanced short-range radar, short-range communications and antiterrorist measures."
http://www.newsfactor.com/perl/story/21518.html

Researchers from the University of Chicago have developed a new grouping technique based on users' Web data requests. Although data requests have been analyzed before to gauge Web site popularity or plan caching schemes, this is the first time researchers used requests to segment users according to interest, according to University of Iowa assistant professor Filippo Menczer. The technique was discovered as part of a larger project meant to locate resources for distributed systems, especially grid computing networks. University of Chicago researcher Adriana Iamnitchi says her group's goal is to develop an efficient mechanism for locating storage, computing capacity, or data files within a grid computing community. She says, "The ultimate goal is to provide scalable, adaptive mechanisms [that are] able to deal with variations in resource participation." While studying scientists' use of a grid computing network for six months, the researchers noticed Web requests were a reliable way to group like-minded users. The technique for grouping users could be applicable when analyzing Web site logs, says Menczer. He notes that possible commercial applications include e-commerce, Web data caching, and Web broadcasting. Iamnitchi says her group also discovered through research that there is a high instance of collaboration within the grid computing community.
Click Here to View Full Article

More than 60 percent of computer-based crimes committed from January 2002 to March 2003 came from the United States, with the majority being hack attacks and scams, according to a report by the International Chamber of Commerce (ICC). ICC's Cybercrime Unit (CCU) senior analyst Steven Matz says hacking remains the largest security threat for financial companies and solo investors. But computer crimes also occur in such countries as Japan, New Zealand, and Romania, he says. In addition, fraud comes in a variety of modes, from "extortion and 'get rich quick' schemes, to attacks by disgruntled employees," says Matz. Identity theft is another major crime identified by CCU, but Matz says such crimes are underreported in many areas. For example, the Far East accounts for only 10 percent of total incidents, he says. Meanwhile, the CCU found that most actual arrests occurred in the area of hacking (34 percent) and only 3 percent in the area of "Current Employees Seeking Financial Gain." Matz says that corporate theft often remains undisclosed by companies for public relations purposes.
Click Here to View Full Article

Computer security researchers would be allowed to hack through copy protection schemes in order to look for security holes in the software being protected, under a proposed exception to the Digital Millennium Copyright Act (DMCA) being debated in official hearings this week. The DMCA's provision generally makes it unlawful for anyone to "circumvent a technological measure that effectively controls access" to DVD movies, digital music, e-books, computer programs, or any copyrighted work. To do so for commercial advantage or personal profit is a felony carrying up to five years in prison. But Congress built a safety-valve of sorts into the law, giving the U.S. Copyright Office the power to create exceptions to the DMCA to protect legitimate, non-infringing uses of copyrighted material. When the law took full force last October, the office carved out two exemptions: one allowing researchers to crack "censorware" applications to learn what Web sites they block, and a second for old programs and databases Rendered unusable by a defective or obsolete access control mechanism. To that list, the ACM would like to add an exemption permitting white hat hackers to crack copy protection schemes "that fail to permit access to recognize shortcomings in security systems, to defend patents and copyrights, to discover and fix dangerous bugs in code, or to conduct forms of desired educational activities." ACM's Barabra Simons said "I'm going to argue the [current] exemptions aren't sufficient, because we're having security people threatened." A recording industry standards group used the threat of a DMCA lawsuit last year to block Princeton professor Ed Felten from publishing a paper on weaknesses in a digital audio watermarking scheme. The group quickly retracted that threat, and similar cases are rare, but Simons says the DMCA still casts a subtle shadow over the academic security: "It's much harder to document what doesn't get written, what doesn't get published," she says. "But it's had a very chilling effect." The current exemptions expire in October, unless the Copyright Office chooses to reestablish them. The office held a final round of public hearings in Los Angeles this week. In testimony on Wednesday, Simons argued the computer security exemption as a homeland security issue: independent software security research is more important than ever, she says. "The bad guys aren't going to publish the results, they're just going to exploit them... We should be eliminating the laws that encourage insecurity."
http://www.securityfocus.com/news/4729

To review USACM testimony before the DMCA review panel and other DMCA activities, visit http://www.acm.org/usacm/Issues/DMCA.htm

Quantum computing enabled by solid state electronic components has come one step closer to reality with a new University of Maryland breakthrough. Researchers at the school's Center for Superconductivity Research used Josephson junctions--macroscopic superconductors separated by a thin insulating layer--to create entangled quantum bits. Quantum computing research has taken two tracks--one on the atomic level with scientists manipulating individual electrons and atoms, and the other on the macroscopic level with researchers looking to create quantum bits, or qubits, using larger and more easily integrated devices. The University of Maryland work is significant because it is the first instance where relatively larger electronics systems have produced entangled qubits. Professor Fred Wellstood and graduate student Andrew Berkley published their work in the journal Science. Because entangled qubits share all the possible on and off properties of each other, the computational power of the entangled bits grows exponentially as more are added. Whereas six binary bits carry just six pieces of information, six entangled qubits cover 64 pieces of information. The Josephson junctions used in the experiment created a quantum effect because the insulator film is so thin it allows electrons to pass through. The devices are built using the same technology as for integrated circuits.
Click Here to View Full Article

Grid computing is starting to take root in the commercial world, according to analysts. Companies will not be able to resist the tremendous speed and cost benefits of setting up in-house grid computing networks. Such systems allow IT managers to make better use of existing resources and finish projects faster since they do not have to build standalone supercomputers. Meta Group Asia Pacific research director Kevin McIsaac says the virtualization of enterprise computing means IT infrastructure can be treated more as a commodity. While the first deployments of grid computing are relegated to within firms, next-generation projects will link the IT resources of partnered companies for even greater flexibility and capacity. Finally, IT resources will be available on-demand through a utility model, a change that IBM Internet technology and strategy director Mike Nelson says will challenge IT managers; Nelson says IT resources will be aggregated on what he calls a "holy grid," and managers will be able to allocate needed resources quickly for particular projects. As appealing as this vision sounds, there are several serious barriers, including software that allows disparate systems to act as one virtual machine. Currently, the open-source Globus Project is preparing to release grid-computing standards. McIsaac says grid computing will develop in the same way the distribution of electricity did, with companies first producing their own power, then moving to the utility framework where they pay for resources as needed. Existing commercial grid deployments include Ford Motor's drive train design test system, which pools extraneous CPU cycles from company servers and workstations, and Motorola's workstation grid that taps unused capacity to finish cell phone designs faster.
Click Here to View Full Article

The EPA's Marianne Lamont Horinko told National Electronics Product Stewardship Initiative (NEPSI) stakeholders at the Electronic Industries Alliance's (EIA) Environmental Issues Council meeting on May 6 that they must devise a voluntary national electronic waste recycling solution "by no later than year's end," or face a crazy-quilt of state-approved recycling programs. "[NEPSI] must rally around one or two workable [financing] options soon, or it should close up shop," Horinko warned. NEPSI is debating several possible financing models: Instituting an advanced recycling fee that consumers would pay, having manufacturers keep e-waste collection and recycling costs partially in-house, or a combination of the two solutions. The first model is thought by many people to be the best, since it would encourage manufacturers to design more environmentally friendly products. However, EIA environmental affairs director Heather Bowman notes that a national recycling solution cannot be the sole responsibility of the electronics industry, and says the NEPSI dialogue must reach a compromise that consumers are willing to embrace. National Recycling Coalition executive director Karen Krebs says the dialogue allowed stakeholders to better understand the problem and communicate, and agree at the first NEPSI meeting that they should split responsibility. The dialogue was aided by the University of Tennessee's Center for Clean Products and Clean Technologies, which received between $200,000 and $250,000 in grants from the EPA.
http://www.wastenews.com/headlines2.html?id=1052512939

Moore's Law, along with the scarcity of a killer application, is ramping up the commoditization of information technology. Commoditization--and a subsequent decline in value--occurs when technology proliferates, becomes easy to understand, and homogenized; IT is unique among other technologies in that value decreases as products improve. The rapid commoditization of PCs was chiefly driven by IBM's decision in the early 1980s to outsource the parts for its first commercial PC, which enabled other vendors to manufacture PC clones. Servers are also being standardized thanks to penny-pinching IT departments--for example, Opsware's Marc Andreessen sees little point in buying an expensive server when commodity machines that offer better performance can be bought cheaper. Companies such as Google are using massive computer systems constructed out of commoditized electronics that run on commoditized software like the open-source Linux operating system. Transistors are also moving into commodities--in fact, Gilder Technology Report editor Nick Tredennick says, "In the future, what engineers do with transistors will be more important than how small they are." Intel is taking advantage of processor commoditization with such products as its Centrino chips, which serve as computing platforms that boast wireless technology and other features that are prized more than chip power. As electronics commoditization accelerates, the value of Moore's Law is likely to decline in favor of other semiconductor sector "laws," which will probably change the tech sector's governing economic architecture.
Click Here to View Full Article

Captology, a term conceived by B.J. Fogg of Stanford University's Persuasive Technology Lab, focuses on how computers are used to persuasively modify user behavior. Fogg predicts that captology will play an important role in business training, management instruction, and marketing in the near future: For example, software may incorporate behavioral science in order to improve managers' skills or boost employee productivity in call centers. Fogg also notes that an anti-virus program could be even more effective by reminding users how many times it has successfully blocked viruses just prior to upgrading. Captology applications will follow a reward system, such as praising a user who tries out applications or improves performance by following a computer-recommended course of action. Fogg illustrates the viability of captology by citing studies indicating that people more willingly carry out simple tasks for a computer when the computer provides them useful data. "The computer points to good work it's done for you, much as a good employee tells a boss, 'I've done this and this,' before asking for a raise," Fogg explains. Examples of captology technologies include RSIGuard software that keeps track of how much workers use the keyboard and mouse and recommends that they take breaks in order to avoid injuries caused by repetitive stress. Another instance is America's Army, an online military game that is designed to attract potential recruits.
Click Here to View Full Article

The Internet2 project continues to lead the development of new Internet technologies despite its focus on academic research applications. New software, middleware, and network engineering projects at Internet2 project sites provide a glimpse in tomorrow's commercial applications. Whereas commercial networks suffer from a money-losing bandwidth glut, the Internet2's American Abilene network is being upgraded to 10 Gbps, says Internet2 backbone network infrastructure director Steve Corbato. The group is also working with Multi-protocol Label Switching (MPLS) for use with VPNs deployed on Internet2. Corbato says the group is committed to the long-term development of IP technology and that its work in MPLS is solely to enable VPNs, though he does say MPLS will help integrate native IP and optical technologies over the next five years. Internet2 is also experimenting with larger packet sizes of 9,000 bytes, instead of the normal 15-byte packets, which will help keep relevant data close together as it travels over the network. Corbato says work continues on IPv6 deployment and that network performance monitoring capabilities have been greatly increased on Internet2 so that four times the number of metrics are now being monitored. Indiana University's Internet2 Research and Education Network Information Sharing Analysis Center (REN-ISAC), which provides real-time security threat monitoring for Internet2 members, is also helping in national cybersecurity; REN-ISAC recently joined efforts at the national Department of Homeland Security to share information and develop new defense tools. REN-ISAC acting director Mark Bruhn says universities are often at the forefront of cybersecurity because of the unique nature of educational networks, which he says host 15 percent of all Internet network addresses.
http://www.nwfusion.com/news/2003/0512specialfocus.html

The U.S. Defense Advanced Research Projects Agency is funding a half-dozen brain-machine interface projects for $24 million over two years, and program manager Alan Rudolph says these technologies could both restore and enhance cognitive functions, and have "transformational consequences for defense and society." A Duke University team led by Miguel Nicolelis is attempting to develop real-time, two-way mind-machine communication so that animals and later humans can perform sophisticated operations, while University of Michigan researchers guided by Daryl Kipke are implanting electrodes into rodent and primate brains and teaching the test animals to control six-legged robots via the interface. Such research could one day yield interfaces that allow people to control machines by thought while simultaneously receiving multisensory feedback. Meanwhile, Ted Berger of the University of Southern California is trying to build a computer chip that could be used to bring a damaged hippocampus back to full functionality, as well as augment memory in a healthy brain. Wake Forest University's Sam Deadwyler, a collaborator on Berger's project, believes that such technology could enable people to retain memories longer or remember more and more information. Tomaso Poggio and James DiCarlo of MIT are testing ways to tweak the sending, receiving, and processing of sensory input to enhance a person's communicative and perceptive faculties, perhaps to the point where one brain will be able to wirelessly communicate with another. A key challenge in brain-machine interface research is physically integrating electronics and brain cells in a sustainable way, according to John Chapin of the State University of New York Downstate Medical Center. Wide acceptance of such breakthroughs will depend on whether researchers can find a noninvasive technique to connect brain to machine.
http://www.technologyreview.com/articles/huang0503.asp
(Access to the full article is available to paid subscribers only.)