This year's Cebit IT exhibition in Germany is expected to play host to 6,115 exhibitors showcasing technologies that facilitate new approaches to viewing data and greater data mobility, among other things. Fraunhofer Institute for Computer Graphics researchers will detail an augmented reality technology that replaces travel guides or city plans with downloadable 3D models of sites related to a person's location that can be viewed on a handheld; the institute has also devised the AR Telescope, a viewer that overlays digital details on the surroundings. The Personal Computer Memory Card Association will unveil ExpressCard notebook expansion cards, which are designed to succeed the PCcard standard and designate interfaces for removable expansion cards with flash memory or communications and multimedia interfaces. Novell, in conjunction with the University of Erlangen-Nuremberg, will spotlight an online Linux assessment tool that diagnoses a company's future IT strategy in five questions, and suggests a list of studies appropriate to that strategy. Microsoft will demonstrate its Office Live Collaboration Platform, which supports collective data viewing by office workers, while the beta version of its Istanbul universal instant messaging client will also make its debut. Meanwhile, RSA Security will coordinate a discussion panel on computer security from a psychological perspective, and IBM's research and development division will spotlight a prototype storage medium capable of holding 1 million MB of data within a 1-inch-square chip. Big Blue will also run scientific applications that tap the Barcelona Supercomputing Center's 40-Tflop MareNostrum supercomputer.
Click Here to View Full Article

RAND Graduate School science and technology policy professor Steven Popper and colleagues have developed the CARs computer program to expedite decision-making among groups of people with diverse backgrounds by removing the need to first agree on a set of assumptions about real-world behavior. The modeling software uses computers' ability to process millions of calculations in a very short time to determine the possible outcome under each set of assumptions. University of Southern California visiting associate professor Desmond Saunders-Newton employed CARs while supervising a team of experts from the State and Defense departments and the U.S. Agency for International Development (AID) that was trying to ascertain the most likely unstable countries in the hopes of avoiding American military action later on. Saunders-Newton reported that AID and State workers tend to view models with suspicion, but CARs found favor with the entire team because of its ability to accommodate a broad spectrum of assumptions. Popper's program has also been used by the Defense Advanced Research Projects Agency, the National Defense University, and Volvo Car. Popper says the CARs model could benefit decision-makers with differing opinions about how economic growth and environmental protection in the San Francisco Bay area should be balanced.
Click Here to View Full Article

ACM's SIGGRAPH 2005 Conference will have filmmaker George Lucas as its keynote speaker. Lucas, the creator of the Star Wars saga and the Indiana Jones series, will deliver the keynote address, "George Lucas: A Keynote Q&A With the Father of Digital Cinema," on Aug. 1, 2005, during the 32nd International Conference on Computer Graphics & Interactive Techniques. "Lucas' remarkable storytelling and cinematic technical achievements have awed and inspired the computer graphics community for more than 30 years," says James L. Mohler, SIGGRAPH 2005 Conference Chair from Purdue University. "Drawing from his three decades of experience making movies, Lucas' insights promise to intrigue and engage conference attendees across all industries and interests." Lucas, who launched Industrial Light & Magic as his visual effects company, shot "Star Wars: Episode II Attack of the Clones" entirely digitally, which was a first for a major live-action motion picture. SIGGRAPH 2005 is expected to draw more than 25,000 computer graphics and interactive technology professionals for its presentations, technical program, and special events from July 31-Aug. 4, 2005 in Los Angeles. The ACM SIGGRAPH sponsored event will also offer three days of exhibitions, from Aug. 2-4, 2005, featuring the industry's latest products and services.
Click Here to View Full Article
For more information on SIGGRAPH 2005, visit http://www.siggraph.org/s2005/

New president of the Open Source Initiative (OSI) Russ Nelson proposed on March 2 three additional terms to the initiative's 10-point open-source definition, which OSI uses to certify open-source licenses. Some people are working to curtail the proliferation of licenses out of concern that their spread will lead to many islands of non-interoperable open-source code, significantly reducing the amount of work programmers can share. Nelson's proposal calls for licenses to be clearly written, simple, and comprehensible; to not be duplicative of already existing licenses; and to support reusability by transferring the names of specific individuals, initiatives, or groups into an accompanying attachment. OSI co-founder and Open Source Definition author Bruce Perens endorsed Nelson's proposal, although he thinks the new provisions should reside in an OSI certification guideline rather than in an open-source license definition. Joel West of San Jose State University's Silicon Valley Open Source Research Project opposes Nelson's non-duplicability requirement, saying that, "It would be like Moses adding, 'Thou shalt file thy taxes on time' to be the 11th commandment." Meanwhile, Ernest Prabhakar responded to the proposal with a suggestion that the open-source definition could stay as it is, while the new terms could be required separately and exclusively through the OSI license certification process. Nelson insisted that OSI has no intention to decertify already approved licenses, but he did propose an "OSI Gold" certification that could accomplish the same purpose.
Click Here to View Full Article

The development of GeoWall technology was motivated by complaints that 3D "cave" display systems were too elaborate, expensive, and immersive to be effective teaching tools, particularly for the field of geoscience. The National Science Foundation supported the creation of the GeoWall, a cheap system that can visualize geoscience information for large numbers of students through a combination of digital projection and advanced graphics card technology. The original GeoWall system uses commercially available components as well as open source and vendor software to build 3D images from data; a graphics card-equipped PC operates a pair of digital projectors that use polarized light to beam slightly offsetting pictures onto a silvered panel, while users wear glasses with polarizing filters to receive stereoscopic imaging. Other GeoWall applications include a 3D GeoWall for elementary school students in Oak Park, Ill., and the University of Illinois, Chicago's 15-screen GeoWall2 and four-screen Personal GeoWall2 desktop system. The immediate value of these systems for geoscientists is the high degree of resolution they provide, which is an enormous help for analyzing maps and other kinds of imagery. Over 400 GeoWall systems are currently in use at various institutions. GeoWall consortium member Jason Leigh with the University of Illinois, Chicago's Electronic Visualization Laboratory says the consortium is always on the lookout for better display technology that delivers higher levels of resolution and seamlessness.
Click Here to View Full Article
(Access to this site is free; however, first-time visitors must register.)

Companies that sell personal data such as ChoicePoint have successfully fought regulation of their industry for many years, using a combination of lobbying and industry-funded research groups. House and Senate disclosure forms show seven of the country's largest data sellers spent at least $2.4 million last year to lobby Congress and federal officials. In addition, those firms have backed studies that emphasize data brokers' role in national security and benefits to other industries, such as e-commerce. Groups such as the Center for Information Policy Leadership and the Center for Social and Legal Research are funded by consumer data firms; people involved in those groups testified in Congress last year, helping to prevent a bill proposed by Rep. E. Clay Shaw (R-Fla.) from restricting their ability to resell data with Social Security numbers. Indiana University law professor Fred Cate also gave testimony against the proposed legislation and says restricting data provision would disallow conveniences such as easy credit applications and e-commerce. Data sellers formed the Individual Reference Service Group in 1999 to soften legislation that would have had them seek permission from consumers to use their data; the law passed with an opt-out clause instead, which is still difficult for individuals to use. With the recent security breaches at ChoicePoint and Bank of America, which affected nearly all members of Congress, new federal legislation is being proposed that would put data sellers under purview of the Federal Trade Commission and require consumer notification when people's personal data may have been stolen; it is unclear how far these efforts will go because Congress has pursued the issue unsuccessfully for many years. Though identity theft costs roughly $52 billion each year and affects 10 million Americans, government agencies such as the FBI and Drug Enforcement Agency rely on private data brokers for personal information that would be illegal for the government to collect on its own.

Making information encoding and transfer safer through the application of quantum physics principles to computing is the goal of the newly-launched Institute for Quantum Information Science (IQIS) at the University of Calgary. IQIS director Dr. Barry Sanders says that quantum computing overcomes conventional information exchange's speed and security limitations by transmitting information in much smaller packages. "Quantum physics gives us a different basis, which is kind of like having all possible information coexisting at the same time," he notes, adding that quantum computing could facilitate data transmission without fear of eavesdropping, thus benefiting nearly every kind of computer-based transaction. Researchers expect IQIS will help Canada develop a homegrown information security infrastructure and end its reliance on the United States. IQIS scientists say their research could eventually yield a practical method for constructing a quantum computer, which would be able to crack any conventional data encryption scheme and effectively dismantle the foundation of e-commerce security, according to Sanders. "We are working on quantum computing more to study the threat of it than to create a quantum computer," the IQIS director notes. The province of Alberta could benefit economically from the institute, which has forged alliances with various companies with an interest in quantum information technology.
Click Here to View Full Article

Portland State University (PSU) computer science professor Bart Massey serves as faculty advisor to the Portland State Aerospace Society (PSAS), whose goal is to launch nano-satellites into orbit using computer-controlled rockets. Massey says PSAS' rockets, unlike those of other amateur rocketry groups, use computers to control both their ascent and descent. The PSU professor has earned a $21,000 Innovation Grant from IBM in recognition of his breakthrough work with on-board rocket control systems; the grant will go toward the purchase of three IBM Power PC processors to improve the speed and performance of the PSAS' LV-2 rocket. Massey says the rocket's current flight computer, an Advanced Microsystems Device running Linux, has reached its limit. He has also received a hefty NASA Small Business Initiative Research (SBIR) Grant to develop and test vehicle health maintenance software based on artificial intelligence. Massey says the AI software uses simulator-based machine learning, and researchers are training the software to carry out real-time, in-flight vehicle repair operations. Massey says IBM's interest in the PSU program partly stems from the learning opportunities it offers to students. The current phase of the SBIR project involves running simulated mission failures and teaching the AI to perform repairs by itself, while the next phase will involve the use of the PSAS AI software in NASA rockets.
Click Here to View Full Article

The IST-funded MobiHealth project has developed a mobile health care system in which patients' vital signs are remotely monitored by wearable, wireless sensors that form a body area network (BAN) linked to a mobile base unit that sends the data to the doctor or health care center via UMTS or GPRS. Project coordinator Rainer Herzog with Ericsson Enterprise explains that "different sensors can be added to monitor different body functions depending on the patient's illness." European MobiHealth pilot programs included a Swedish implementation to monitor respiration and physical activity, a deployment in Germany to keep track of cardiac patients, a Dutch trial involving pregnant women and trauma patients, and a Spanish test involving rehabilitation and home care. Herzog says the system was well received by almost every trial participant: Patients said monitoring eased their minds and encouraged them to be more active, and doctors said the systems could lower the amount of work they have to do while raising the quality of care. Herzog says the MobiHealth technology is being commercialized for four initial areas: The pharmaceutical industry, chronic disease patients, at-risk patients, and patients hospitalized for surgery. He notes that MobiHealth could also be used to monitor the disabled and elderly, since the BAN can trigger alarms if the patient falls and does not get up. Mobile health schemes will be especially beneficial to health care providers by helping them lower the costs and maintain the standards of European public health systems. The MobiHealth partners are currently developing an analytical tool for processing the raw data the system receives and rendering it in a simple format for health care workers.
Click Here to View Full Article

Humans faced with continuously advancing technology should learn to form symbiotic rather than adversarial relationships with computers, such as with Garry Kasparov's "Advanced Chess" matches where human-computer teams compete against each other, writes George Dvorsky. Advanced chess matches have resulted in landmark chess games and uncovered new insights into how the best human and computer chess programs play the game. Human and computer weaknesses have already been probed in matches pitting grandmasters against new chess programs, with humans becoming physically tired or making obvious errors and computers failing to recognize strategic gambits and adapt during the game. More interesting than the juxtaposition of human and computer chess playing is the combination of the two styles in Advanced Chess, the first official game of which was played between a Kasparov-Fritz 5 team and Veselin Topalov teamed with ChessBase 7.0 in 1998. That game ended in a 3-3 draw, but was groundbreaking in that it highlighted the powerful combination of human strategy and raw machine power. Advanced Chess games have been played annually in Spain since 1998, resulting in games completely free of obvious blunders and perfect tactical play. Kasparov's Advanced Chess concept provides a good framework for how humans should deal with evolving artificial intelligence systems, argues Dvorsky. Instead of fearing replacement by machines, humans should look for new opportunities where the combination of human and computer capabilities achieve more than any single component could alone.
Click Here to View Full Article

Attensity has developed software that can almost instantly parse electronic documents such as emails and chat room discussions and make unstructured data usable and relevant. Analyst Nick Patience estimates that about 80 percent of corporate or government information is unstructured data such as emails, instant messages, call logs, and memos, and he says the Attensity software takes everyday language "and [compiles] it in a way that a machine can use." Attensity co-founder David Bean says the software can work even when the sentences to be parsed contain errors or abnormalities such as slang, dangling participles, and misspellings, since the software distinguishes between subject and object through the constant reapplication of heuristics. In addition, the Attensity software can uncover meaning in sentences in the process of being parsed. Approximately 60 percent of Attensity's business is derived from government bodies such as the FBI, the Defense Intelligence Agency, and the National Security Agency, and Bean says the software helps federal analysts extract indications of criminal or terrorist activity in "the text from the dispatches around the world, the field reports, the newspaper articles and the chat rooms." Attensity's system, financed partly by the CIA's In-Q-Tel branch, differs from others that ferret out information from unstructured data in that it requires little training from users to understand relationships between words. The Attensity software currently works exclusively with English, while software from Inxight can parse documents in 31 languages because it converts grammatical relationships into mathematical formulas. Intelliseek software, meanwhile, can categorize blog entities by extracting proper names and places, and can also tag a document as positive or negative according to the words in the document.
Click Here to View Full Article
(Access to this site is free; however, first-time visitors must register.)

Speaking at this week's EclipseCon application programmer conference, Google VP of engineering and operations Urs Hoelzle detailed his company's data center infrastructure, which relies on cheap, simple commodity servers in order to build up redundancy, thus ensuring that an operation is not hamstrung by the malfunction of one machine. Hoelzle said a rate of one failure a day is a safe bet in a Google data center with thousands of PCs, so the sensible approach is to automate failure strategies to avoid service outages. The Google File System, developed internally, operates on the assumption that a malfunction can occur at any time, and supports the replication of data in three places; a "master" machine locates the duplicates if the original piece of data is unavailable. The thousands of Google PC servers run a streamlined version of Linux, while another system handles massive volumes of data and returns fast responses to queries by fragmenting the Web into millions of "shards" that are copied in the event of failure. Google possesses document servers that store copies of crawled and downloaded Web pages, and Hoelzle explained that his company has also engineered an uncomplicated way to write programs that operate across thousands of servers. Google keeps costs down by automating the program recovery process via the MapReduce programming tool, while its Global Work Queue software performs "batch" job scheduling. Another variable that Google's data center designs take into account is the cost of power, which Hoelzle described as "directly proportional" to operations' physical cost.
Click Here to View Full Article

Penn State researchers have developed software for optimizing the allocation of anti-terrorism resources across competing projects by prioritizing resources according to objective standards and supplying a cost-benefit analysis for various implementations. Acceptable risks and losses in terms of money are also weighed. The researchers acknowledge the challenge inherent in deciding what, how, where, and when to allocate resources to shield vital infrastructure, given the high cost of anti-terrorist mitigations and the many factors that must be taken into consideration, such as mission importance, protecting people, replacement time and cost, and location of and access to backup facilities. "People need help in deciding which proposed projects to fund when each responds to a need and has its own benefits," explains lead researcher Steven Haynes with Penn State's School of Information Sciences and Technology. He says the cognitive-support system, which was developed with the backing of the U.S. Marine Corps and the Marine Corps Research University at Penn State, reduces the subjectivity of those decisions and helps more people understand the process and rationale of resource allocation decisions. The software's development was influenced by visits to Marine bases and discussions with personnel about what technologies could best enhance asset optimization, but the system can also be applied to the protection of government facilities, public sites, utilities, and commercial buildings. The software was detailed in the Journal of the American Society for Information Science and Technology.
Click Here to View Full Article

So-called patent trolls represent a huge cost of business for deep-pocketed companies and create a chilling effect for small startups that cannot afford to defend their use of a contested technology. Patent trolls are companies who do not market any product, but earn revenue through licensing questionable patents; the term was coined by former Intel patent counsel Peter Detkin. Intel claims to receive one letter from patent holders per week threatening legal action, and has a policy of fighting such lawsuits rather than settling out of court or paying requested fees. Many companies opt to pay licensing fees in order to avoid the average $4 million it requires to defend against patent claims. The National Academy of Sciences claims the current patent system requires companies to spend significant resources on licensing, royalty negotiations, and infringement protection. Business groups such as the Business Software Alliance and the Information Technology Industry Council have lobbied Congress to draft new legislation that would limit the ability of patent trolls to conduct business. One idea supported by semiconductor companies is to give the courts the option of fining the defendant instead of ordering an injunction that could stop production of computer chips, for instance. Such a change would take away a weapon from suing companies, but is controversial because it changes one of the core protections of patents. The U.S. Patent and Trademark Office is struggling to quickly ramp up and modernize its operations in order to keep up with growing patent litigation and disputes, and is considering a post-grant opposition system that would provide a less costly alternative to litigation. A noteworthy patent case involves Eolas Technologies' suit against Microsoft that resulted in a half-billion-dollar court reward in 2003, but also in a preliminary invalidation of the Eolas patent by the Patent Office. [A federal appeals court overturned the ruling against Microsoft on March 2 and ordered a retrial.]

A growing movement of academic institutions is offering their course material online for free and software tools for making that material accessible, thanks to startup funds from philanthropic groups such as the William and Flora Hewlett Foundation. The movement begun by MIT's OpenCourseWare project has so far brought in a number of other prestigious groups, such as the Harvard Law School's Berkman Center for Internet & Society and the Johns Hopkins University's Bloomberg School of Public Health. Participants in the open courseware movement, in addition to interested Chinese and European academics, gathered at MIT in February to discuss best practices and coordinate efforts. A major concern is making it easier and less costly to put courseware online and keep it updated. Utah State University offers open source software that automatically converts existing course Web pages into material that conforms to open courseware standards; the software was developed with a $915,000 grant from the Hewlett Foundation. Utah State instructional technology assistant professor David Wiley also led the creation of free chat-room software that allows open courseware users to have group-moderated discussions. The University of Michigan at Ann Arbor is also authoring free software to assist schools in publishing open courseware materials. All of those involved in open courseware see their efforts as helping to fulfill their outreach missions, but are seeing benefits as alumni refresh their knowledge and students inquire about programs. Only MIT has made the commitment to put all its course material online, while other universities fill in the gaps with their own signature courses or classes not taught at MIT.

Scientists such as MIT researcher Thomas Knight are looking to "synthetic biology" to surpass the manufacturing limitations of silicon-based computers by exploring ways to build computers out of living cells. Shrinking circuits mean that silicon chip fabrication will inevitably hit a wall, while Knight observes that biochemistry offers a far more precise and sophisticated manufacturing scheme. Constructing a programmable organism is the goal behind the BioBricks project Knight's team is working on, a BioBrick being a snippet of DNA that can imitate conventional computer circuit behavior. BioBricks can execute very simple tasks individually, or be integrated to perform more complex operations. Knight doubts that BioBricks will ever surpass the millisecond-level performance of conventional computer systems, but he does note that BioBricks could generate carbon nanotubes and similar elements for building high-performance systems at the molecular level, or perhaps form biological factories for fabricating super-dense silicon chips. Princeton University professor Ron Weiss, who studied under Knight, reports that "eventually we might come up with an abstraction that allows you to program billions of little biological computing elements that are not robust at all and don't have a lot of resources, and that might be a useful paradigm for programming certain kinds of silicon-based computational devices." Knight notes that biological circuits' ability to self-replicate makes synthetic biology a powerful as well as dangerous area of research, with most of the danger stemming from the potential for abuse. He maintains that "Our best defense is our ability to do it faster, better, and cheaper than anyone else."
Click Here to View Full Article

Work has only just begun on software applications that take advantage of third- and fourth-generation wireless network technology. Wireless 3G communications technology that delivers data speeds of 300 Kbps to 500 Kbps and higher promises to revolutionize data access via applications that can be run from practically any location. Trina Seinfeld with Microsoft's MapPoint business unit expects the convergence of high-speed wireless Web access, handhelds, and application integration via Web services to result in the increased incorporation of location-based functionality within inventory and customer relationship management applications. Siebel's Bill Hou predicts that future wireless applications will feature automatic data aggregation, making multiple application log-ins unnecessary. Teresa Lunt with Xerox's Palo Alto Research Center foresees systems that know and provide the data users require for tasks at hand, and expects information to be relayed not just visually, but tactilely and audibly as well. Meanwhile, YellowPepper Wireless President Carol Erickson believes the development of mobile video conferencing for consumer and business applications and streaming video applications and content will be driven by next-generation, high-bandwidth networks; she also says digital-rights management (DRM) will be a more pressing concern, so carriers must therefore adopt a single DRM technology standard. A further challenge will be effectively managing the massive amount of data individuals will receive through wireless apps. Antenna Software founder Peter Semmelhack says these innovations will make workers capable of doing their jobs wherever they are; as a result, "Everyone is going to expect that wherever they are they will have access to information," he concludes.
Click Here to View Full Article

Virtual 3D prototypes are on track to deliver information at least equal to that provided by actual physical prototypes through a combination of enhanced imaging devices, sound, and tactile feedback. Numerous forms of virtual prototyping are employed by General Motors to expedite vehicle development cycles, including UGS' Teamcenter Concept and other tools for design, decision-making, and maintenance. IT manager of Strategic Planning and Benchmarking for GM Diane Jurgens says VR is used to visualize aesthetics, while stereo viewing is utilized for styling and design; in addition, "virtual rooms" allow virtual versions of alternative car designs to be displayed in diverse environments to enhance decision-making, while decisions related to auto interiors are arrived at through the use of actual vehicle seats and steering wheels with tracked 3D viewing, audio cues, and other simulation elements. Meanwhile, engineers at Lawrence Livermore National Laboratory's National Ignition Facility use PTC's Division Mockup tool to examine the highly complex subassemblies of the center's equipment to ensure that they will fit together within the assigned space prior to manufacture. UGS' Stu Johnson says recent improvements in graphic cards have been a tremendous boon for VP products. He notes, "We've all benefited a great deal from the video game developers; so much so, that we're starting to see PC clusters come into the virtual reality environment, replacing Unix clusters." Johnson also thinks the addition of haptic devices into the VR mix could help train workers in manufacturing processes, while PTC's Michael Rygol observes that VR use isa increasing as people realize how limited 2D prototypes are compared to 3D prototypes. He points out that today's VR systems can be used to help address ergonomic issues with the insertion of artificial humans into visualizations.
Click Here to View Full Article