The Department of Homeland Security's inspector general yesterday released a report citing poor coordination, communication shortcomings, and a lack of prioritization as the chief reasons why the department's efforts to shore up the nation's cyber-infrastructure have been less than satisfactory. "If we are at war, as Bush and [Homeland Security Secretary Tom] Ridge say we are...based on this report, we are clearly not on a war footing on cyber-security, or in DHS," stated Entrust CEO F. William Conner. Members of Congress and tech company executives are concerned that cyber-security remains a low priority, and that some senior department officials do not truly comprehend the scope and magnitude of the issue; these concerns stem not just from fears of a major cyber-terrorist attack, but from daily intrusions that add up to annual losses of hundreds of millions of dollars for U.S. businesses and computer users. Executives and security experts are particularly frustrated by the DHS' failure to adopt some of the cyber-security practices it urges government agencies, businesses, and organizations to institute. The report recommends that the National Cyber Security Division create a process for superintending federal, state, and local government programs to fortify their systems against cyber-attacks. Information Technology Association of America President Harris Miller said that his organization "continues to be concerned that DHS does not have adequate resources devoted to cyber-security and that the cyber-security head does not have adequate visibility within the bureaucracy." Several industry groups are backing a bill co-sponsored by Reps. Zoe Lofgren (D-Calif.) and William M. Thornberry (R-Texas) that would promote National Cyber Security Division director Amit Yoran to the position of assistant secretary.
Click Here to View Full Article
(Access to this site is free; however, first-time visitors must register.)

The Democratic Party's 2004 platform echoes the Republican position that broadband services should be made universally available, although it differs with the Republicans as to how this goal should be reached. Republicans wish to eliminate telecommunication and broadband regulation, while the Democrats want to encourage the rollout of broadband in underserved areas by offering tax credits and other incentives. Universal broadband could inject $500 billion into the American economy, create 1.2 million new jobs, and have a transformative effect on work and education, according to the platform. The 41-page platform never mentions IT outsourcing by name, but says, "[U.S.] companies can keep and create jobs in America without sacrificing competitiveness." The 2004 platform contends that the Bush administration supports policies that are eroding U.S. jobs and hurting America's competitiveness: "Instead of meeting the challenge of globalization by strengthening our workers' ability to compete and win, this administration uses globalization as an excuse not to fight for American jobs," the platform posits. Rep. Rick Boucher (D-Va.) reports that Democrats and Republicans disagree over how outsourcing should be handled, and he favors the establishment of incentives to encourage U.S. companies to keep jobs at home as well as the institution of penalties for offshoring. The Business Software Alliance's Robert Cresanti argues that such penalties will inhibit job growth, although he says the Democratic platform's views on universal broadband and a permanent R&D tax credit gibe with the tech industry's goals. Cresanti adds that the 2004 platform has less of a focus on technology issues than recent platforms, and cites no apparent mention of cybersecurity in its security sections.
Click Here to View Full Article

The Senate Judiciary Committee will hear testimony today on the Induce Act, which aims to ban technologies that enable copyright infringement and allow civil penalties for users that intentionally assist a third person in violating copyright. Although the Inducing Infringement of Copyrights Act is aimed specifically at changing the behavior of 60 million Americans who have participated in unauthorized file-sharing, it is so broad in its potential application that it makes basic technology components suspect. Not only would this law undermine the landmark 1984 "Betamax case" that provides for reasonable recording and archiving, but it also threatens to stifle technological innovation. Peer-to-peer file-sharing companies would be the direct targets of the Induce Act because they offer the interface software people use to easily share files on the Kazaa and Grokster networks. Last year, a federal court ruled these software makers cannot be responsible for the illegal activities of their users because of the way they are designed; moreover, a previous federal court ruling allowed new digital technologies such as the MP3 player because they had "substantial non-infringing uses." The Motion Picture Association of America and Recording Industry Association of America (RIAA) say the Induce Act does not target normal technology, or "neutral technology," in the words of the RIAA's Mitch Bainwol--yet no technology is neutral, especially when it is as powerful and enabling as networked digital technologies are. When users have the opportunity to use alternative file-sharing technologies such as Gnutella, ICQ, FreeNet, and BitTorrent, they will do so. Unless authorities and industry officials are willing to re-architect the entire system to disallow this misbehavior, interfering policy such as the Induce Act will fail, writes Siva Vaidhyanathan, New York University assistant professor of culture and communication.
Click Here to View Full Article
(Access to this article is available to paid subscribers only.)

The U.K. government has awarded 1.8 million pounds to the University of the West of England (UWE) and a quartet of research partners to investigate biologically-inspired computers, which is at the center of five nationally funded projects. UWE will play a major role in two of these initiatives: One of the projects is a collaborative venture between UWE biologists, chemists, and computer scientists and the universities of Sussex and Leeds to develop silicon chip alternatives through experiments that utilize actual neural cells and networks of chemical reactions. The project will integrate machine learning, cell culturing, experimental chemistry, and neurobiology techniques, according to project leader Dr. Larry Bull with UWE's Computing, Engineering and Mathematical Sciences department. He says the experiment uses neuron-like cells that are configured into networks via electrical stimulation, and the electrical signals generated by the network in response to the stimuli may be construed as a computation. A second project, a joint effort between UWE, the University of Sheffield, and Edinburgh University, will study the human brain's ability to control and stabilize bodily movement to see how such mechanisms can be extended to robotic systems, an example being small walking machines that traverse hostile and uneven terrain and can be employed in search-and-rescue missions. Dr. Tony Pipe says the government grant will be used to develop a six-legged robot with a motorized vision system and an artificial cerebellum to direct its movements. He predicts that "The research will bring about a significant advance in our understanding of how the cerebellum controls body stability and how this can be mimicked by robots."
Click Here to View Full Article

The National Science Foundation (NSF) has allocated $70 million to 13 U.S. universities to make their labs and equipment available to companies and individuals who are developing nanotechnology, which NSF senior engineering advisor Lawrence Goldberg believes will play an essential role in computing. One microtechnology project funded by the U.S. Army Space and Missile Defense Command is Georgia Tech Research Institute scientist Jud Ready's carbon nanotube-based electrochemical double-layer capacitor, which is designed to be more powerful, denser, and longer-lived than conventional batteries and electrical capacitors. Professor Ari Glezer and associates at Georgia Tech's School of Mechanical Engineering have developed and patented "SynJets," a cooling system that utilizes flexible, synthetic jets small enough to be incorporated into devices that cannot support conventional fans. The SynJets move considerably less air than comparably-sized fans, but they efficiently mix ambient air and dissolve thermal boundary layers by generating turbulent air flow. "You get a much higher heat transfer coefficient with synthetic jets, so you do away with the major cooling bottleneck seen in conventional systems," notes Georgia Tech research engineer Raghav Mahalingam. Meanwhile, MIT researchers Vladimir Bulovic and Moungi Bawendi have created a self-assembling nanoscale memory cell through a combination of organic chemistry and quantum dot technology. Another MIT scientist, Mark Baldo, has taken a step toward the creation of biocomputers that use organic/inorganic microprocessors by harnessing the properties of a newly developed organic molecule.
Click Here to View Full Article

Johns Hopkins University computer science professor and outspoken e-voting critic Aviel Rubin testified before the House Subcommittee on Technology, Information Policy, Intergovernmental Relations, and the Census on July 21 that policymakers foolishly neglected to consult with security experts about e-voting technologies, and warned that using the systems in the November elections without correcting security vulnerabilities beforehand is "irresponsible." Rubin upheld the position of several recent studies that labeled major security flaws in current e-voting machines, and expressed disappointment that the policy community failed to authorize timely security audits. He co-authored one of the studies, which outlined vulnerabilities in e-voting system software from Diebold, and his conclusions were confirmed by three other independent studies that faulted the reliability of e-voting systems in general. "Not only have [e-voting machine] vendors not implemented security safeguards that are possible, they have not even correctly implemented the ones that are easy," Rubin told Congress. The computer science professor expressed puzzlement that he and other computer security experts are regularly dismissed as "Luddites" and "conspiracy theorists" despite solid documentation of e-voting machine bugs. Missouri has only certified e-voting systems that offer voter-verifiable paper trails because of the public's worries about the election process' integrity and security, noted Terry Jarrett, general counsel to Missouri Secretary of State Matt Blunt. Randolph C. Hite of the Government Accountability Office said the reliability of voting systems is more likely to be impacted by poll worker training and voter education than technological changes between now and November. Hratch Semerjian of the National Institute for Standards and Technology stated that his organization is busy devising e-voting security standards and best practices.
Click Here to View Full Article

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

Distinguishing between authentic digital photos and doctored images is critical for law enforcement, the military, and newspapers and magazines, to name just a few affected areas. Dartmouth College computer science professor Hany Farid has developed algorithms that can tell the difference between genuine and altered digital images: They can, for example, detect fine details that are signs of spliced or rotated images. Tests on several hundred doctored pictures showed that the method was practically infallible, provided the picture was of sufficient quality. Farid says that files created in the highly popular JPEG format were harder to grade accurately. Meanwhile, State University of New York at Binghamton professor Jessica Fridrich has embarked on an Air Force-funded project to design a camera that takes two pictures simultaneously; one of the photos would be of the photographer's iris, and this image would be compressed, encrypted, and buried within the larger photo taken through the camera lens, along with the time and place the photo was snapped. The photo's authenticity would be called into question if the encoded data is missing or tampered with: "It establishes a connection between the person who took the image, the camera used, and the digital image itself," Fridrich explains. Farid does not think digital watermarks are a secure enough solution, and acknowledges that preventing the doctoring of digital images may ultimately be a futile goal. He states, "At the end of the day, the person doing the tampering has the easier job. And they'll win."
Click Here to View Full Article
(Access to this site is free; however, first-time visitors must register.)

The COMIC project funded by the Information Society Technologies program has developed a prototype multimodal user interface stemming from the project's combination of fundamental cognitive research and technology to make computers capable of comprehending and generating cognitive behavior. "We need services that are able to actively assist and support users in solving their problems, rather than passively wait for the users to express their intentions in a language that is reminiscent of programming," notes COMIC coordinator Dr. Els den Os at the Max Planck Institute for Psycholinguistics. The prototype interface was devised to help consumers design their idealized bathroom, and the system required the melding of diverse input and output modules such as speech, pen gestures, graphics, text, and facial expressions. The system can integrate two input actions, such as pen and speech, into a single command, and the project conducted research into distinguishing between different types of pen gestures, such as drawing, pointing, or writing measurements. COMIC uses a "talking head" to support the concurrent generation of speech and facial expressions. "Several user studies...show users are able to use the combined speech and pen input modes to input shape and dimensions of rooms and that users can recognize emotions from synthesized facial expression," says den Os. At the same time, he admits that multimodal interaction architectures and technologies need to be significantly improved.
Click Here to View Full Article

Robotic technology has generally come up short as a tool for enhancing people's daily lives, but researchers at Spain's University Jaime I are developing a mechanized librarian that can find and retrieve books using a combination of sensors, cameras, and gripping appendages. Professor Angel del Pobil says safety issues are the key obstacle to taking robots out of industrial and research facilities and placing them in populous, unstructured environments, and he thinks libraries are the best public space to initially deploy robots, or at least demonstrate the technology's viability. "We think [a library] is a good environment in which service robots are out there, working in a human environment, but it is still a controlled one," he explains. The wheeled robot uses voice recognition software to match the titles of requested books with classification codes in a database, which only yields the volume's approximate location. The robot uses an infrared and laser guidance system to find its way to the bookshelf, and scans books within a radius of four meters. The robot's seven-jointed, two-fingered gripper is equipped with a pair of wrist micro-cameras to capture images of the bookshelf, and four tactile sensors that measure the force it applies when grasping to prevent damage to the book. Image processing and optical character recognition software enable the device to read the labels and the position of the book, notes Pobil. The professor believes that teams of robot librarians could conduct searching and retrieval tasks in approximately five years, and he raises the possibility of a collaboration with Japanese researchers currently working on an automated librarian that can be controlled remotely over the Web.
Click Here to View Full Article

Neill Hopkins of the Computer Technology Industry Association (CompTIA) observes that a wide gap exists between the IT skills employers need and the skills workers actually have, and he thinks the solution is the National IT Apprenticeship System (Nitas) CompTIA kicked off in March, which combines on-the-job training and classroom instruction. CompTIA's Steve Ostrowski explains that IT pros expect more company support for career development, especially when businesses acquire emerging technologies. Nitas can help companies complaining that networking and security experts and software developers are in short supply, as well as training companies that want corporate and government financial aid for wannabe IT staff who do not have the money to pay for training. Nitas organizes a "lifelong path" for college students and IT pros, who are to be tested on separate skillsets before being certified on industry-wide standards. The apprenticeship system establishes courses that will be offered by community colleges or authorized commercial learning organizations collaborating with local companies, who guarantee that the apprentices' training fulfills their skills requirements. Employees are selected to participate in the program by their companies, who register as sponsors, and their progress is tracked by a Web-based system. The sponsors pay for training and provide a minimum salary, while participants who lack sponsors can sign up for the program as "pre-apprentices." Attempts will be made to land sponsors for pre-apprentices during their training, but many may fail to get aid from the local boards.

Discussions at ICANN's sixth annual meeting in Kuala Lumpur have centered on multilingual domain names. Finding a technical way to accommodate non-Roman characters in Web browser spaces is a daunting task, Internet experts agree. Despite the enormous difficulties foreseen in developing such a solution, ICANN general manager Kieran Baker expressed some optimism, saying that the annual meeting would provide "some conclusions on how we can move forward" on developing a solution. U.S. Internet expert John Klensin is not as optimistic, pointing out that languages such as Arabic are written right to left, which poses a serious technical problem. Likewise, languages such as Chinese have regional variations that also pose problems, says Klensin. VeriSign and other domain name registration services accept non-ASCII domain names, but the waters have been muddied by the Chinese government's announcement that it will create its own register for Chinese character domain names. Ultimately, the prospects for multilingual domain names depend upon whether engineers will be able to mesh the ASCII-based Domain Name System with the Unicode coding system.
Click Here to View Full Article

The creation and duplication of sculpted objects is being transformed by technology such as scanners, computer-aided design (CAD) software, and automated milling equipment. Studio Roc, for instance, can create near-perfect copies of items by laser-scanning them and having the milling machine carve the facsimile out of styrofoam, stone, or other material. But controversy is brewing over whether the end products represent genuine art or merely knock-offs. Some artists viewed the technology with contempt when it was first introduced, but others have embraced it as a new tool for creating or enhancing artwork. Studio Roc CEO Kenneth Kai Chang explains that his company does not intend to compete with the art world, while Los Angeles architect William Hablinski expects computerized milling to become especially attractive to the home construction industry: "You're basically asking a robot to do what you would normally have to pay a master stone carver something on the order of between $70 and $180 per hour," he notes. Computerized milling is an offshoot of rapid prototyping technology, in which physical mockups of products are built using CAD software for the purpose of testing and refinement prior to actual production. The technology is also finding use in civic and federal initiatives to preserve a record of artwork, landmarks, and monuments so they can be restored if they suffer damage or deterioration.
Click Here to View Full Article
(Access to this site is free; however, first-time visitors must register.)

San Diego is touted as the wireless capital of the world thanks to the formidable concentration of academic and commercial wireless projects and companies that have accumulated there. California Institute for Telecommunications and Information Technology founding director Larry Smarr believes that San Diego will be at the epicenter of what he terms the third big wave in technology, the wireless revolution, which could outstrip Silicon Valley's PC revolution because wireless devices have no limited market per person. "The per-capita number of devices that will communicate wirelessly could easily be 100 or more per person," Smarr postulates. He says, "The Internet now ends at the workstation...it needs to go out of the computer and do smaller things. You could have sensors deployed in rivers, buildings or cars and they could all be on the Internet, interacting with things on the Internet." The success of Qualcomm and its code-division multiple access technology has encouraged many smaller firms and leading mobile technology players to flock to San Diego to open offices or subsidiaries. Gateway VP Mark Steele reports that new levels of cross-fertilization among companies are being reached in San Diego, which is nurturing innovation. University research is also contributing greatly to San Diego's clout in the wireless sector. "We have a spirit of collaboration not found in other cities in this country," notes University of California at San Diego professor Leslie Lenert, who is leading the federally funded Wireless Internet Information System for Medical Response in Disasters (WIISARD) project. WIISARD has developed clip-on vital-sign scanners that can facilitate remote monitoring of injured people by wirelessly relaying their readings to the Internet.

Legislators and industry insiders are worried offshore software development could compromise the Department of Defense's (DOD) IT security, given its predilection for purchasing commercial off-the-shelf software. Software vendors save money by sending a lot of their software development overseas, and some officials are beginning to fear that foreign-based developers could deliberately insert vulnerabilities in the software. DOD officials have recently concentrated on purchasing proven commercial products, but the products still have vulnerabilities, and a lot of it was not meant to be subjected to the high threat level of the agency. To remedy this situation, the House version of the 2005 Defense Authorization bill includes up to $50 million in grants for DOD contractors to find alternatives to outsourcing jobs, such as plant upgrades and worker retraining programs. A Congressional Research Service report says that Congress must deal with the issue, but Congress also recognizes that commercial applications are cheaper and easier to use than customized ones. A Government Accountability Office report notes that DOD officials do not have enough control over software, and that current purchasing guidelines focus too much on external threats such as hacking and unauthorized access. Some companies are focusing on the concern, with Ounce Labs offering a tool that checks code for potential errors, security flaws, or holes. Open-source software is also a problem, and the traditional "many eyes" argument does not negate the possibility someone with malicious intent could sabotage open-source products. While experts admit the economic necessity of some offshore software development, they also say vendors must be held more accountable for their product's integrity.
Click Here to View Full Article

Graduate engineering enrollment in the United States reached an all-time high in 2002 with nearly half a million students, according to a National Science Foundation (NSF) report. Though the overall increase in college enrollment helps the trend, double-digit increases in fields such as electrical and biomedical engineering--up 10.7 percent and 20.3 percent for 2002, respectively--show that more students see future job opportunities in these areas. Postdoctoral appointments increased as well, growing 6 percent between 2001 and 2002 to total 32,100 engineering researchers. U.S. citizens and permanent residents increased first-time graduate enrollment nearly 14 percent in 2002, but foreign first-time graduate enrollment dropped 7.9 percent, or about 2,100 students. NSF senior analyst Jean Johnson said the trend of fewer foreign graduate engineering students was worrying because unofficial figures show that drop-off continued in 2003, and noted that foreign students may be put off by post-Sept. 11 visa restrictions or may be opting to study in Europe and Canada instead. Indeed, Canadian and European graduate schools are seeing an uptick in enrollment commiserate with the drop in U.S. enrollment, says Penn State University professor Mary Jane Irwin, who adds that these foreign graduate students strengthen the nation if they choose to live and work here after graduation. Johns Hopkins graduate engineering students Dat Truong and Landon Unninayar represent one of the best examples of the opportunity available in their academic field, having created a working prototype for a landmine-detection robot. Also remarkable is 14-year-old Ph.D. candidate Alia Sabur, who is beginning electrical engineering research at Drexel University; she says engineering should be differentiated from pure science in order to draw interest from more students, including girls, since it involves working with other people from diverse areas of knowledge.
Click Here to View Full Article

Vendors are planning to leverage the wide-scale implementation of 10 Gigabit Ethernet (10 GE) outside of the network core so that the technology is more cost-effective than Fibre Channel and similar links. There is only so much data that users can consume, and there does not appear to be a large data type beyond video to fuel the unquenchable thirst for bandwidth. Cost-reducing volume 10 GE applications remain an uncertainty, because of the lack of a driving application in the enterprise, consumer, or storage networks. Enterprise demands for high-quality, high-bandwidth video are rare if not nonexistent, which means there is no justification for constructing an infrastructure to transport it. Nevertheless, many vendors hope that video will be the key 10 GE driver. Greater bandwidth dispenses with issues of contention, but convenience requirements will in fact lower bandwidth needs to make applications such as remote and wireless connectivity viable and effective. Beyond the core, 10 GE might have a role in storage-area networks and data centers, for the purposes of aggregation and passive data use, simple database tweaks being an example of the latter. Data consumption grows more slowly than one might need to legitimize 10-Gbps speeds beyond the core. There is little doubt that 10 GE will serve in a niche capacity as an interswitch connection or link between campuses, though perhaps only in the biggest organizations.
Click Here to View Full Article

Four American universities--Stanford, MIT, the University of Michigan, and the University of Indiana--are leading the Sakai Project, a cooperative effort to develop and distribute free course-management software, the first version of which was issued on July 15. Sakai is an amalgam of computer code derived from existing course-management software created at the four lead universities and uPortal, a Web-based application patched together from open-source software components originating from several hundred institutions. "One thing that distinguishes us from any other open-source project is our commitment to eat our own dog food," notes Amitava Mitra of MIT, which has promised to start using the Sakai software by fall 2005; every professor will be asked to employ the software, so it is particularly important that MIT ensures that the product is solidly designed. Certain college officials think Sakai was developed partly so that universities could become less dependent on commercial software vendors such as WebCT and Blackboard, whose products are becoming more and more expensive--but Sakai's installation, customization, and maintenance is time-consuming. The Sakai Project has received hefty funding grants from the William and Flora Hewlett Foundation and the Andrew W. Mellon Foundation, and MIT, Michigan, Indiana, and Stanford have each committed roughly $1 million in staff time; the project expects to get enough colleges to informally use and develop the software to make Sakai self-sufficient within three years. Some officials believe colleges will not adopt Sakai unless it boasts compelling features, given the costs of migration and faculty retraining. Both Blackboard and WebCT have expressed interest in working with Sakai and making their products interoperable. Northwestern University's Bob Taylor hopes that Sakai will encourage commercial vendors to make their software more customizable.
Click Here to View Full Article

Some 643 entries were submitted to ACM's SIGGRAPH 2004's Computer Animation Festival (CAF), and of the 83 that passed muster, around 50 percent were the work of international animators and one-third were student projects. Diversity also extended to the festival jury, which included people with backgrounds in art and story. Members included Paul Debevec of the University of Southern California's Institute for Creative Technologies, Ars Electronica co-director Christine Schoepf, Ines Hardtke from the National Film Board of Canada, and Darin Grant of the U.S. visual effects company, Digital Domain. Grant explained that the studios were advised to supplement their submissions with behind-the-scenes images, which increased the odds of acceptance. Industrial Light & Magic showcased breakthroughs with virtual actor technology such as skin and hair modeling, cloth simulation, and motion capture with a reel featuring computer-animated creatures from "Harry Potter 3" and "Van Helsing." "Spider-Man 2" and "Shrek 2" stood out as examples of how filmmakers have been influenced by CG research: Debevec's trailblazing photogrammetry research at SIGGRAPH was employed in the "Spider-Man" movies, while state-of-the-art image rendering research and skin rendering methods detailed at SIGGRAPH were applied to "Shrek 2." CAF Chair Chris Bregler expects CG techniques currently relegated to academia will branch out to feature films this year, and he says the winning entries were distinguished by having compelling narratives in addition to quality animation. Computer game animation accepted into the CAF included "Ruby: The Doublecross," an ATI Technologies-RhinoFx production that spotlighted realtime rendering technology.
Click Here to View Full Article

For more information about ACM's SIGGRAPH 04, or to register, visit http://www.siggraph.org/s2004/.