A September survey of future Internet trends conducted by the Pew Internet and American Life Project offers a wide mix of predictions from approximately 1,200 responding technology experts, scholars, and industry leaders. Sixty-six percent agreed that "at least one devastating attack" on the Internet's infrastructure or the U.S. power grid will take place in the next decade, although opinions differed on the attack's impact. Around 42 percent expected civic involvement to rise over the next 10 years as people go online to look for and join organizations, while about 30 percent did not anticipate such an increase. Thirty-two percent of those surveyed agreed that more than half of America's voters will vote online by 2014, compared to 35 percent who disagreed; 59 percent thought that government and businesses will use network devices to expand surveillance, which will lead to more arrests in both democratic and authoritarian nations; and 32 percent expected more political polarization to result from people using the Internet in such as way as to sift out data that challenges their opinions on political and social issues. The results of the poll validated many media watchers' expectation that the news media and publishing industries will undergo more changes than any other societal sector because of the Internet and the emergence of bloggers, while 50 percent of respondents thought that most industry users will continue to employ anonymous networks to freely swap digital content, despite copyright owners' aggressive attempts to stamp out digital piracy. The Pew Project and Elon University's Imagining the Internet database at www.elon.edu/predictions incorporates the projections provided by the survey respondents. "Every [survey participant], we know it's not going to pan out exactly the way we think," admits University of Toronto sociology professor Barry Wellman. "But it gets us focusing on what some of the alternatives could be."
Click Here to View Full Article
(Articles published within 7 days can be accessed free of charge on this site. After 7 days, a pay-per-article option is available. First-time visitors will need to register.)

The prevalent Turing machine theory limits computers to relate all observations to a single program source, similar to how paranoid human patients relate all observations to an intent to harm them, writes documentary filmmaker Peter Krieg, who is incubating a new technology that frees computing from Turing-imposed limits. Turing machines are restricted to problems that can be solved through clerical labor by following fixed rules; such machines have no understanding of the context of the problem because they cannot integrate different logic domains. Such computers are stuck on the track of classical mechanical logic, where representations are defined by the program and not the data, and the machine can only be in one state at a time even though it can assume an indeterminate number of sequential steps. If computers are to simulate the entire spectrum of human thinking, they need to exist in many simultaneous states so they can integrate different observations and synthesize new logic domains, Krieg explains. Instead of observing the world holistically through a deductive process, computers need to connect partial observations as humans do. Independent inventor Erez Elul has created a Pile system approach that builds on the "polycontextural" logic architecture created by Gotthard Gunther: The Pile technology has been implemented on standard von-Neumann hardware and runs under a standard operating system, and is ready for practical testing in 2005. The so-called Elul machine works off of relations where the system has many possible beginning points. Since new information is represented in the paths, the Elul machine does not require this information to be physically stored, and scales linearly; theoretical arguments have already been put forward to test whether to Pile technology can solve problems not previously solvable by Turing-machine computers.
Click Here to View Full Article

Sony Computer Science Laboratory (CSL) Paris director Luc Steels describes the facility as "a scientific lab, but not all innovation is based on science and not all science leads to innovation." The lab was inspired by the Xerox Palo Alto Research Center in its establishment as a place where researchers can exercise their creativity with an unusual degree of freedom, rather than being forced to rigidly adhere to a corporate mandate to develop strictly commercial products. "[It's] probably for prestige and advertising purposes, but nevertheless, it's unique and great that a big corporation will put some of their profit to identify people they consider the smartest in the field and let them come up with new ideas," notes Emory University professor Philippe Rochat. Steels says Paris is an ideal locale for the lab, given the richness of its culture as well as its reputation as a hub of scientific research that is attractive to potential employees as well as visiting academics. Steels' previous experience includes the directorship of the University of Brussels' Artificial Intelligence Laboratory and his work with emotional speech synthesis technology that is now used in Sony's QRIO humanoid robot. Among the facility's six researchers is Atau Tanaka, who develops musical systems that make listening a vehicle for social interaction by integrating mobile technology and peer-to-peer networking. Another project of note at Sony CSL Paris focuses on unique search engines for digital music libraries.
Click Here to View Full Article

University of Melbourne Department of Electrical and Electronics Engineering associate professor M. Palaniswami believes intelligent sensor network technology could help prevent natural disasters such as those caused by tsunami waves. In comments to news agencies covering the second International Conference on Intelligent Sensing and Information Processing (ICISIP'05), Palaniswami said more research into intelligent sensor network technology and proper deployment of intelligent sensor networks would be needed to make it effective in recognizing natural threats or even hostile developments. "The early detection of such threats through advanced sensor technology will enable rapid responses from the appropriate authorities preventing potentially disastrous situations," said Palaniswami. The technology makes use of sensors that are the size of dust particles and are able to collect accurate information from the surrounding environment. More companies are showing interest in intelligent sensor networks, which will help bring down the cost of the technology. ICISIP'05, taking place this week in Chennai, India, is an interdisciplinary conference that integrates several advanced research themes such as intelligent sensing and adaptive learning with a view towards solving problems in smart systems. The conference, chaired by Palaniswami, provides a forum for researchers and academicians in the fields of intelligent sensors, sensor fusion, neural networks, support vector machines, fuzzy systems, evolutionary computers, biomimetics, bioinformatics, biomedical, radar signal processing, and intelligent multimedia processing.
Click Here to View Full Article

Digital Dunhuang is an initiative to protect the cultural treasures of Dunhuang in China's Gansu Province by constructing a database of artifact images as well as a digital archive of historical documents and research material related to Dunhuang, according to Dunhuang Academy researcher Liu Gang. Among the institutions contributing to the project is America's nonprofit Mellon Foundation, which has been funding an effort to digitally capture the sculptures and murals of Dunhuang's Mogao Grottoes using the latest digital photographic equipment. The treasures in the caves are threatened by natural and artificial elements that include long-term exposure, moisture from the breath of visiting tourists, and smoke from fires set by visitors as well as local residents. Liu says the technology used in the project can capture details concealed by the grottoes' structure or hidden in natural light. Besides preservation, the Digital Dunhuang project was started out of demands for scholarly access to the treasures and historical records; experts think this access will open up new scholarship opportunities in such fields as archeological linguistics, art history, and religion. The Chinese government has authorized the Dunhuang Academy to be the exclusive overseer of the grottoes' treasures, and the institution has declared that it retains all rights to images of the artifacts under China's intellectual property rights (IPR) laws. The academy has signed contracts with the Mellon Foundation that firmly establish the IPR of all the digital cave images: The contracts certify the academy as the sole owner of the IPR of the digital images and film negatives of Dunhuang; disputes will be resolved by both sides according to Chinese law and regulations, while mediation will be administered by the China International Economic and Trade Arbitration Committee. In addition, pay-per-view technology, click-through barriers, and digital watermarks have been deployed to further safeguard copyrighted material.
Click Here to View Full Article

The explosive growth of Wikipedia, the Wikimedia Foundation's free online encyclopedia, has generated skepticism about the resource's credibility, its status in the eyes of academia, and how it can maintain the spirit of openness and collaboration as it expands. Most of these problems stem from Wikipedia's systems for constructing and reviewing articles, and whether the scope of its coverage includes enough specialized topics to establish the archive as a scholarly, single-reference source. Wikipedia's entries are submitted and vetted by members of the Wikipedia community, but the articles are open to continuous editing, which means that there is no final authority to ensure their accuracy. University of California at Berkeley graduate student Danah Boyd notes that most entries are authored by only a few people with no credentials, and this undermines Wikipedia's value as a reference tool; she reports that many students are using Wikipedia to find information on ancient history, but ancient-history specialists' participation in the archive is paltry. Former Wikipedia developer Leonard Sanger writes in an article posted on Kuro5hin.org that some members of the Wikipedia community show little respect for academics and specialists. "If someone is made to defend his or her contribution by some crank, or a troll, the rest of the community, generally speaking, will not come to the defense of the expert," he reports. Wikimedia Foundation President Jimmy Wales insists that project managers are fiercely committed to ensuring the accuracy of articles by continuously reviewing the management of the community as well as the entry-writing process. However, he cautions that Wikipedia is envisioned as a reference source for general readers rather than specialists.
Click Here to View Full Article

Shu Ishiguro of Osaka's Robot Laboratory says his consortium has pledged to develop robots that will wrest the World Cup from the reigning human soccer champions by 2050. The precursor technology of such systems is VisiON, a 38-centimeter-tall machine that operates independently of humans and boasts a 360-degree vision sensor. The robot can identify, approach, and kick the ball, as well as identify an opponent and guard the ball. The consortium's decision to develop a soccer-playing robot was partly inspired by the triumph of Japanese robots at the most recent RoboCup championships, which involved the participation of 346 teams from 37 nations. Osaka roboticists are also designing robots for security, caregiver, and domestic applications: The 2005 World Expo will feature humanoid robots that provide directions to visitors and assistance in child-care facilities through their ability to recall names and recognize faces. Ishiguro reports that robotics researchers are not enthusiastic to apply their knowledge to the creation of robot armies, and downplays the possibility of a robot rebellion. "Everyone who saw the RoboCup could see the advantages of technology and, as long as its development is kept open to the public, there is going to be no danger," he insists. "But developments in the future we must discuss at that time, as a society."
Click Here to View Full Article

Molecular electronics promises far more powerful electronic devices and continued performance gains after the viability of traditional silicon computing has ended. Improved research equipment has pushed molecular electronics far forward in the past five years, according to UCLA chemist Amar Flood, who calls molecular electronics "pre-shrunk" because it leverages molecules' existing functionality. Molecules such as catenanes and rotaxanes are being studied by researchers and will provide a roadmap for what is possible with future molecular electronics materials, says University of California at Santa Barbara materials scientist David Steuerman. Flood says only a small percentage of possible molecular electronic scenarios has been tested, and a number of large industrial and government research organizations are working in the field alongside a handful of promising startup firms; Zettacore says it is harnessing bistable rotaxane molecules to create a 16-kilobit memory circuit at densities that are many generations beyond what traditional silicon circuitry is capable of. Besides performing traditional computing functions, molecular electronics will also augment computing by interfacing with biology: Flood says molecular electronics could connect to the biological systems that allow sight, taste, and smell. Some possible scenarios include early cancer-detection devices, or interfaces for advanced prosthetics where molecular electronics could pass information to and from normal biological cells. Steuerman helped design a color-switching molecule that could be used in computer displays, and he says future molecular electronic fabrication plants would also be more environmentally friendly than current semiconductor fabrication facilities. Researchers still face a number of basic hurdles, such as how to control interactions with device surfaces and manage defects inherent in molecular self-assembly.
Click Here to View Full Article

Wi-Fi wireless networking may lead the pack when it comes to electronic connections, but other connection technologies are undergoing a resurgence, if their presence at last week's Consumer Electronics Show is any indication. Bluetooth short-range wireless connectivity has bounced back from near-obscurity thanks to the cell phone, which can be linked to other electronic systems through Bluetooth. Technologies based on copper wire are generating excitement as connectivity options that could support, for example, the transmission of video from a system in one room (say, a computer) to a system in another room (say, a television). HomePlug AV, which has 10 times the transmission speed of its predecessor, HomePlug 1.0, was demonstrated at CES as such a product. One connectivity technology whose success is in doubt is Ultra Wideband, since the industry committee responsible for standardizing it has yet to release unified specifications. A lack of consensus has yielded a pair of competing technical approaches, and products that use one or the other were on hand at CES. Some people might view the presence of Bluetooth and other alternative connectivity systems at the exhibition as a sign that the computer and consumer-electronic industries are starting to address the headache of tangled wires that is typical of their products.

Researchers at Penn State University have presented a paper at the IEEE Consumer Communications & Networking Conference in Las Vegas on a new model for transmitting high-speed broadband over U.S. overhead electric power lines at bit rates that would surpass DSL or cable over similar spans. Mohsen Kavehrad, the W.L. Weiss professor of electrical engineering and director of the Center for Information and Communications Technology Research, and Pouyan Amirshahi, a doctoral candidate in electrical engineering, evaluated data rate handling capability for overhead medium voltage unshielded U.S. electrical power lines in the paper, "Transmission Channel Model and Capacity of Overhead Multi-Conductor Medium-Voltage Power-Lines for Broadband Communications." "We've run a computer simulation with our new power line model and found that, under ideal conditions, the maximum achievable bit rate was close to a gigabit per second per kilometer on an overhead medium voltage unshielded U.S. electrical power line that has been properly conditioned through impedance matching," said Kavehrad. Half a dozen homes in a neighborhood would be able share the gigabit and receive rates in the hundreds of megabits per second range, which is much higher than DSL and cable. "If you condition those power lines properly, they're an omni-present national treasure waiting to be tapped for broadband Internet service delivery, especially in rural areas where cable or DSL are unavailable," said Kavehrad. Although he does not question BPL as a technical alternative to DSL and cable, Kavehrad is still uncertain about its economic viability.
Click Here to View Full Article

Boeing and the University of Illinois at Urbana-Champaign's Information Trust Institute have teamed up to design trustworthy, reliable, and secure networked systems and software employed in critical infrastructures, with a focus on basic breakthroughs that can become practical and important commercial products within five to 10 years, according to institute director William H. Sanders. The Information Trust Institute will receive an undisclosed amount of funding from Boeing's Phantom Works unit over the next five years for the purpose of investigating "trusted software," and Phantom Works VP Gary Fitzmire says U. of I. was chosen on the strength of its trusted software research. The institute's mission includes setting up science and technology for creating trustworthy networked information systems, the development of methods for evaluating such systems' trustworthiness, and the administration of those methods to applications in systems including e-commerce, finance, emergency response, data and information processing, and aerospace. The institute has embarked on research projects that include misbehavior detection in wireless networks and a railcar health monitoring system, while the U. of I. last month solicited research project proposals based on the Boeing agreement. Submitted proposals included new software security and survivability techniques, and reliable and robust control of automated aerial vehicles.
Click Here to View Full Article

Smart phones deliver the performance requirements needed to give users a convenient, off-the-shelf mobile interface for a burgeoning population of intelligent embedded devices. Smart phones integrate cellular telephony, programmable information management, and Internet connectivity, enabling designers to lower the cost, size, and complexity of embedded devices. Designers can opt for short-range Bluetooth and 802.11 networks or longer-range cellular data transmission as the means for exchanging wireless information. Smart-phone software boasts built-in data communications and custom graphics, and can be updated remotely to fix glitches, add new features, or change characteristics. A smart handset device could also access remote servers and display large data items via an Internet link. There is a wide range of smart phone form factors and operating systems available, which can complicate integration for designers of embedded systems. Designers must combine communications hardware and software within an embedded device and perhaps devise a mobile application for the handset in order to fully exploit a smart-phone interface. A simple solution is to simultaneously incorporate Internet connectivity and device management through the addition of a commercially available Web-server module.
Click Here to View Full Article

Phishing email schemes, which utilize spam emails in order to send receivers to realistic but fake Web sites where they update personal information as supposedly requested by their bank or credit card company, exploded in 2004. Email security vendor MessageLabs blocked an estimated 18 million phishing email schemes this year with just 279 phishing email schemes blocked in 2003. Trojan horse programs, which infect Internet-connected computers and allow hackers to collect data, download and execute files, launch denial-of-service attacks, and remotely control Web cams, also exploded in 2004. Over a span of just nine months, 480 different versions of the Trojan were identified, according to antivirus company Sophos. With the increase of Trojans also comes the increase of "botnets," which compromise computers and use them to spam other users. Authorities made several high-profile arrests of cybercriminals in 2004, including the 18-year-old who released the Netsky and Sasser Internet worms, the 21-year-old who created Agobot and Phatbot Trojans, and several suspects believed to be involved in the theft of Half-Life 2 source code in 2003. Sophos senior security analyst Gregg Mastoras believes user authentication technology will enable banks, e-commerce companies, and consumers to protect themselves from hackers, and replacement of static passwords with improved security is also going to make progress against hackers next year. Hackers will face strong resistance to phishing scams and other activities if more people would adopt email sender authentication technology; Yahoo and Microsoft currently make two popular versions of the software, says Mastoras. Frontbridge Technologies technical product manager Jesse Villa believes progress against hackers in 2005 must begin with Internet service providers sharing more of their data on Internet attacks and compromised computers with authorities.
Click Here to View Full Article

TeraGrid is set to become the new national cyberinfrastructure, providing not only high-speed connectivity, but also active computing resources to scientific and academic users, according to Purdue University technologists. Unlike Internet2, which is primarily a networking infrastructure, TeraGrid offers a number of other resources, including supercomputing capacity; storage; and scientific resources such as imaging satellites, climate modeling data, and other tools. Purdue University is harnessing TeraGrid to bolster its nanotechnology community. Nano-scale devices require atomic-level physics and chemistry simulations, and researchers can easily tap necessary resources to run those simulations via TeraGrid, says TeraGrid Purdue site lead Sebastien Goasguen. While TeraGrid shares many of the conceptual elements of commercial grid computing efforts, it is distinct because it is government-funded and meant to provide a common infrastructure not tied to any institution or vendor. TeraGrid is still developing a mechanism to balance work loads between sites and allow for more seamless use of resources, says Purdue high performance computing technical architect Michael Shuey. For now, increased capacity means TeraGrid users have better reliability than if using standalone resources. Like most other TeraGrid participants, Purdue contributes supercomputing capacity, and built a 1,000-machine Linux cluster out of recycled computers previously used in instructional labs. The idea of TeraGrid is to provide a single environment where users do not focus on the TeraGrid technology, but rather the content. Purdue TeraGrid researchers envision making the infrastructure useful to students in the classroom, and believe most U.S. scientists will rely on TeraGrid for computing storage visualization in three years.
Click Here to View Full Article

Roboticist Hans Moravec, who has predicted future trends such as people routinely downloading their minds into robots as well as robots eventually overtaking human beings in terms of adaptability and intelligence, views the Seegrid robotics firm he founded as a vehicle for helping realize such visions. Moravec contends that the evolution of robotics "has to be driven forward by a lot of trial and error, and the only way to get enough is if you have an industry where one company is trying to outdo another." Seegrid targets the potentially massive "product handling" market with its focus on vision systems that let simple machines transport supplies around a warehouse without human intervention. The firm's prototype system uses commercially available CCD cameras and sonar and infrared sensors to collect information whose inconsistencies are smoothed out by software, thus yielding an accurate 3D map of the environment. With this system, wheeled carts can learn to navigate on their own by being walked through various routes. Colleagues such as Carnegie Mellon University's Raj Reddy confirm that Moravec's sometimes far-out ideas are balanced by his pragmatic approach to work. Moravec expects machines to evolve incrementally, noting that "We don't need a lot of Einsteins to do this; we need a lot of engineers working diligently to make little improvements and then test them out in the marketplace." He says this strategy will ensure the robots' ultimate ascension as the most adaptable and resourceful species on the planet.
Click Here to View Full Article

Virtual engineering is an emerging field that aims to facilitate the virtual equivalent of physical engineering collaboration by blending together a wide array of engineering processes, while also embedding more data into the virtual engineering process than can be layered into any physical prototyping approach. At least five core elements must be integrated to fully realize virtual engineering: Virtual reality visualization methods, real-time analysis, numerical model coupling and integration, a remote collaboration infrastructure, and on-demand information handling. Current virtual engineering environments, which can incorporate desktop systems and specialized immersive virtual reality environments, ultimately seek to enable cross-system collaboration. Upgrades in computing and graphics power have significantly lowered these systems' costs, while the software required for the display and analysis of virtual engineering data in virtual environments is both commercially and academically available. Five benchmarks of virtual engineering development are contemplated between the starting point of high-fidelity model visualization and the end-point of a cross-industry virtual engineering standard, with each successive benchmark yielding more advantages to engineering design. The first milestone achieves an interaction between virtual reality and engineering data; the second milestone enables design teams to directly manipulate models in the virtual workspace; the third milestone provides new approaches to performing accurate on-the-fly model recalculation; the fourth milestone embeds models within larger systems; and the fifth milestone allows as-needed tracking of and interaction with field data and component studies throughout a product's life cycle.
Click Here to View Full Article

Most widely available tools are ill-suited for information retrieval: They either miss valuable material or return too much worthless content, while the useful data they do uncover is often rendered in a textual format that is hard to visualize or use in data mining or standard database processes. Information extraction (IE) technology is making headway against these problems and enhancing corporate data and knowledge management applications such as data mining, text search, text categorization, and visualization. IE helps decipher human language by closely analyzing it, accounting for vocabulary and grammar, and attempting to determine "who did what to whom" from a fragment of text while solely concentrating on items of specific importance to the user's subject area and application. An IE system's input is basically a constant flow of text documents, while output is a list of tags and where they appear in the documents. These tags typically identify one or more of three classes of data--entities, relationships, and events. IE products are distinguishable in a number of areas, including accuracy, pattern development efficiency, and speed and simplicity of integration. Many vendors and researchers are using IE solutions that combine machine learning and human analysis of patterns in text to find information: The first option has yet to prove itself as conclusively better than the second, and the mistakes a machine makes in learning a pattern requires adjustments to the learning method; human analysis, meanwhile, can boost accuracy, yet the increasing complexity of IE applications makes effective rule set organization and management more complicated. IE products' speed differs not just because of the products themselves, but because of the data and the number and intricacy of the extractions, which makes output formats, product-readable document formats, and access to the offsets of items tagged in the text important integration considerations.
Click Here to View Full Article

Enterprises can obtain a deeper understanding of their business processes with wireless sensor networks, even though the technology is still in an early stage of development. High costs, limited battery life, redundancy shortcomings, interference issues, an absence of standards and security measures, and the complexity of integrating a network with key business systems make sensor net deployments unaffordable for most small and medium-sized enterprises. However, there are economical ways for a company to gear up for the expected mesh networking revolution. CIOs can start automating asset tracking via relatively inexpensive starter sensor net kits, while deploying sensor nets as a preventive maintenance measure is another good point of entry. Analysts say the biggest challenge lies in incorporating sensor networks into the company's business system infrastructure, but the ZigBee Alliance promises to have standards that facilitate integration and security ready by year's end. Before purchasing sensor nets, CIOs are advised to discuss with business unit directors how the nets could improve operational efficiency or deliver deeper business analytics. Analysts are convinced that, even in an infant state of development, sensor nets can yield solid returns in the supply chain, asset monitoring, and HVAC system construction. BP's P.P. Darukhanavala argues that investing in mesh networking now is a sensible move, as the technology promises to "transform everything we touch, not just in business but our personal lives as well."
Click Here to View Full Article