The percentage of female computer systems analysts and scientists, programmers, and postsecondary computer science teachers declined from 30.5 percent in 1983 to 27.2 percent in 2002, according to an Oct. 13 report from the Commission on Professionals in Science and Technology. The study also finds that the presence of women has risen in all natural science professions, while females accounted for 14 percent of all engineering jobs in 2002, up from 10 percent in 1983. However, the portion of female math and computer scientists, programmers, and postsecondary math and computer science teachers fell by 0.8 of a percentile to 29.9 percent between 1983 and 2002. Meanwhile, the percentage of all U.S. jobs held by women has increased from 44 percent to 47 percent in the last 20 years, and the proportion of women in scientific, engineering, mathematical, and technological professions has expanded from between 16 percent and 19 percent in 1983 to between 23 percent and 26 percent in 2002, depending on how such jobs are defined. In a related story, an Oct. 13 Meta Group report projects that IT workers will experience as much as a 15 percent increase in salary in the next three years, while expected economic improvements over the next 12 months will spur key IT professionals to seek "greener pastures" in the form of more development opportunities and higher-paying jobs. The creation of new jobs in technology-related services categories and increased hiring by tech services is tempered by declining confidence in the job market among IT workers, as indicated in a poll. Meta advises CIOs to concentrate harder on human resources management programs to retain important IT employees.
Click Here to View Full Article

An Oct. 12 pre-election test of touch-screen voting terminals in Palm Beach County, Fla., had to be rescheduled to Oct. 15 because heat-related problems caused a computer server that indexes data from the machines to crash. Exiting county elections supervisor Theresa LePore says no data was altered or deleted as a result of the crash, and promised that the information would be retained even in the event of a system-wide outage. She suspects that excessively high temperatures in the room containing the server caused it to malfunction, possibly due to a disruption of air conditioning and electricity by Hurricane Jeanne. Current technical standards for e-voting systems mandate a tolerance for storage temperatures between -4 and 140 degrees Fahrenheit, and operation within "natural" conditions and temperatures between 50 and 95 degrees. Firmware design engineer Vincent Lipsio argues that most "mission critical" equipment should be able to function in temperatures of 180 degrees or higher, and he is currently helping draft e-voting hardware standards for the IEEE. Lipsio is concerned that touch-screen voting terminals could be vulnerable to lightning storms, heat waves, severe low-pressure systems, and other extreme weather phenomena. E-voting critics are citing the Palm Beach County crash as proof of the technology's fallibility, and an example of the need for paper ballots. Lipsio agrees that a paper backup system should be available, barring the installation of electrical generators at polling places.
Click Here to View Full Article

Most people prefer to maintain their emotional connection to others via low-bandwidth communications systems such as instant messaging, and researchers think new technology could augment these modes of communication to facilitate seamless interpersonal interaction. This is the goal of the IST Astra project, which designs and tests awareness systems that automatically capture data about users and their daily activities before passing it on to others. Astra project coordinator Panos Markopoulos notes that awareness systems are characterized by being easy to use and always on, and says a system similar to the prototype his researchers developed "would allow users to simply glance at a computer display to see someone's status or availability." Markopoulos compares awareness systems to instant messaging, in that they furnish limited data and an "image" of correspondents' current actions, as well as supplementary information such as their accessibility. A quartet of families tested the Astra system using network services and Internet-enabled mobile phones: The system let groups of home-based users maintain communications with mobile users, who took photos of themselves and their locations and sent them as email attachments to a server through their phones; the messages were then forwarded to the users at home via their wireless, PC-linked touch screens. "Astra focused on the concept of connectedness, developing a scale to measure just how connected people feel to others when using awareness systems such as our own," says Markopoulos, who notes that this concept is an essential step toward ambient intelligence.
Click Here to View Full Article

The National Science Foundation's TeraGrid has moved into full production mode, providing distributed computing, visualization, and data resources to partners. The TeraGrid currently offers 40 teraflops of aggregate computing power, petabytes of storage, and operates over a 40 Gbps network, while more than a dozen major computing systems from nine resource partners are linked to the distributed grid. Argonne National Laboratory offers relatively modest computing capabilities coupled with high-resolution rendering and remote visualization, and the California Institute of Technology's Center for Advanced Computing Research provides data on high-energy physics, astronomy, and application expertise in computationally intensive science. Indiana University and Purdue University contribute more than 6 teraflops of computing power; storage, data, and visualization resources; and access to the Purdue Terrestrial Observatory and Indiana Genomics Initiative life sciences data stores. The National Center for Supercomputing Applications provides the largest computational contribution to the TeraGrid with 10 teraflops of processing power, as well as 2 petabytes of storage, while Oak Ridge National Laboratory's High Flux Isotope Reactor and Spallation Neutron Source will also be able to simultaneously tap TeraGrid resources. The Pittsburgh Supercomputing Center provides heavy computing capabilities with 6 teraflops in computing power offered in tandem with a unique 21-node visualization system. The San Diego Supercomputer Center will contribute just over 4 teraflops computing capacity from an IBM Linux cluster as well as a portion of another 10-teraflop system. The Texas Advanced Computing Center provides computing capabilities as well as geoscience and biological morphology data.
Click Here to View Full Article

Female computer game designers, programmers, and producers are as rare as female game players: About 10 percent of gaming industry professionals are women, and most of them hold jobs in customer service, quality assurance, and marketing, according to informal estimates. Ion Storm executive producer and longtime gamer Denise Fulton observes that a major obstacle to women's pursuit of gaming industry careers is gaming's reputation "as a boy thing." Harvey Mudd College computer science professor Elizabeth Sweedyk, who is designing a female-oriented game design course with a $200,000 grant from the National Science Foundation, thinks that women find computer games unappealing for the most part. At last month's Women's Game Conference in Austin, a common complaint against games in general was graphic violence and sexually suggestive depictions of female characters, while some attendees expressed a desire for games with more immersive story lines and more relatable characters. Programmer Nicky Robinson, also a game enthusiast, says she felt obliged to improve gaming's appeal to women by designing less cluttered, more intuitive user interfaces. Sony Online Entertainment's Sheri Graner Ray says the first step to getting women interested in careers in game design is to get them interested in playing games by raising awareness. "As we do that, and get more women into the industry, the games they make will have much broader appeal," she remarks. Robinson points out that some game companies are overwhelmingly male-oriented, which can be a further discouragement to women; she notes, for instance, that employees often discuss business in terms of sports metaphors, while upper management harbors a macho attitude.
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.)

Some experts are convinced that U.S. software programmers will die out in the next few years as more American companies offshore programming to low-wage countries and give domestic programming positions to foreign immigrants. The computer-related U.S. job market grew by 27,000 new positions between 2001 and 2003, but Programmers Guild expert John Miano reckons that nearly 180,000 new foreign H-1B workers entered the United States in that period. "This suggests any gain of jobs have been taken by H-1B workers," he remarks. The H-1B visa cap, which currently stands at 65,000, was already reached by Oct. 1 of last year, and U.S. businesses are now lobbying Capitol Hill for additional visas. But the Programmers Guild and similar organizations argue that the existing H-1B quota is already unfairly excessive, given that over 100,000 U.S. programmers are unemployed and an even greater number are underemployed. Criticism has also been leveled at loopholes in the H-1B program that allow employers to hire H-1Bs without first hiring domestic workers, as well as pay H-1B holders less than the prevailing wage, even though it is required. American programmers are mobilizing to fight H-1B and L-1 visas in Congress, but Miano warns that business groups have the advantage in terms of organization, funding, and political clout. The Sphere Institute estimates that almost 25 percent of California's technology workforce has moved to non-tech careers since the dotcom bubble burst three years ago, while another 28 percent have become unemployed, opted for self-employment, or left the state.
Click Here to View Full Article

MIT researchers have created new optoelectronic fibers that can be woven together into new types of light-sensitive computer interfaces and multifunctional textile fabrics. "The technique we developed allowed us to bring together two disparate technologies: Those involved in creating optical fibers and those for electronic components," said team leader Yoel Fink about the work that appeared in the Oct.14 issue of the journal Nature. The fibers allow researchers to control the interactions between photons and electrons in a woven fabric framework. There are two types of fibers showcased in the team's research: One conducts electrons and photons simultaneously by surrounding a hollow, mirror-lined optical core with metal microwires, while the second fiber is a type of photodetector that reacts to external light; when light shines on the fiber, it increases the flow of electrons in a cylindrical semiconductor core surrounded by four metal microwires, which in turn is encased in an optical cavity structure. Fink says this photodetector fiber could be integrated into large computer displays that sense user input via a laser pointer or other light beam. The fibers were created using optical fiber-type pulling methods: First, a macroscopic cylinder was created with the same structure as the end product, but without the functionality since the components were not shrunk to touch each other. When the cylinder was heated and drawn in a furnace, the conductor, amorphous semiconductor, and high-glass transition thermoplastic insulator all scaled appropriately in geometry so that the desired contact and properties are exhibited. The research was funded by the Defense Department and National Science Foundation.
Click Here to View Full Article

A team of Southampton University researchers led by computer science professor Nick Jennings and Ph.D. student Gopal Ramchurn has triumphed over 222 competitors in the 2004 Iterated Prisoner's Dilemma contest. The Prisoner's Dilemma, a game theory puzzle for two players, is described by Jennings as "a canonical problem of how to get cooperation to emerge from selfish agents" in a scenario whereby two accomplices are arrested and separately interrogated by the police, who offer them the choice of confessing or keeping quiet. The Iterated Prisoner's Dilemma is a variant of the traditional game in which the choice is repeated over and over, enabling players to evolve a cooperative strategy by recalling their previous moves. The long-standing champion of the competition was the Tit for Tat strategy, in which a player consistently cooperates with other players in its first move, and subsequently copies the other players' actions. The Southampton team, which specializes in software agent research, submitted 60 strategy programs designed to carry out an established sequence of five to 10 moves by which they could identify each other; once identification was confirmed between two Southampton players, one player would sacrifice itself so the other could win repeatedly. But if an opponent was identified as a non-Southampton player, the Southampton player would defect. The original Prisoner's Dilemma banned communication between players, but since the Iterated Prisoner's Dilemma contest permitted noise, the Southampton players were programmed to signal their intentions to each other in Morse Code. "Our initial results tell us that ours is an evolutionarily stable strategy--if we start off with a reasonable number of our colluders in the system, in the end everyone will be a colluder like ours," explains Jennings.
Click Here to View Full Article

Software startup iSpheres announced a new programming language that will allow easy development of event-driven enterprise applications, where information is updated to users immediately. The California Institute of Technology spinoff says event programming language (EPL) is meant for applications that detect fraud, enable real-time trading, guard networks, or monitor incoming RFID data. Analyst Roy Schulte says such event-driven applications will become a business imperative in the coming years because they inform decision-makers about the most relevant and up-to-date company information. EPL is based on Defense Department research done for command and control systems, and is compatible with existing middleware and development tools. Companies will have to set up a special event-processing server that will use iSpheres' server version EPL, for which it will charge a royalty; the EPL development language is royalty-free and the company plans to submit it to a standards body by the end of the year. ISpheres' Gary Ebersole says EPL can do in 10 or 20 lines of code what it would take 100 lines of Java code to accomplish. Forrester Research has said enterprise development is more difficult than it should be, and IBM and Tibco Software are working on event-processing software and languages that will address growing needs: IBM's common event infrastructure has already been submitted to the Organization for the Advancement of Structured Information Standards (OASIS). Meanwhile, Schulte says some event-driven language will be paired with the developing business process execution language (BPEL) to dramatically impact the industry.
Click Here to View Full Article

The University of Idaho's Microelectronics Research and Communications Institute (MRCI) has expanded from a $50,000-a-year initiative nine years ago to a facility with an annual budget of about $4 million under the leadership of NASA and Boeing veteran Touraj Assefi. "My goal is to empower researchers and labs to work together, collaborate with other agencies and leverage funding to do greater work," the MRCI director explains. The institute conducts research in microelectronics, avionics, nanotechnology, computer security, intrusion detection, autonomous vehicles, intelligent control systems, real-time software, and electromagnetics to develop technologies for industries and government agencies such as the National Science Foundation, NASA's Jet Propulsion Laboratory, and Hewlett-Packard. Current MRCI projects include an Office of Naval Research (ONS) initiative to develop a fleet of unmanned, synchronized robotic vehicles that can operate on land, in the air, and underwater for surveillance, mine sweeping, and other security and intelligence-gathering missions. The ONS is also underwriting MRCI's Advanced Microwave Ferrite Research program, whose goal is to devise wideband, high-frequency iron oxide devices on a single, multifunction system-on-a-chip to accommodate large data streams as an application for military radar and communications gear. Another notable MRCI project is a "biomimic artificial neuron" capable of learning; networks of these neurons may be applied to military strategy, automated traffic and emergency dispatching, smart cars, and prosthetics. Meanwhile, the MILS software architecture project for high-assurance embedded systems has yielded new mathematical models, sophisticated systems components, and verification and validation procedures that will speed up and lower the cost of designing secure aircraft computer systems.
Click Here to View Full Article

Though smaller, more compact consumer electronic devices are aesthetically pleasing, device manufacturers and human-machine interaction experts are concerned that continuous miniaturization could inhibit their usability and enjoyment. Hewlett-Packard's Ken Klestinec complains that mobile phones are becoming too small and increasingly difficult to hold and operate, noting for example that cell phone keypads are so minuscule and close together that misdialing is often a problem. Radio Shack's Stu Asimus reports that consumers are generally attracted to small products such as cell phones and cameras, but many--especially older people--display a preference for devices that have what they consider to be satisfactory size and weight. Asimus observes that customers who choose the single-piece "candy bar" cell phones over flip phones are often annoyed with the flip phones' small buttons. Motorola's Jim Wicks realizes that devices "need to be small, but they have to get back into this game of great usability." With that goal in mind, Motorola will soon release the RAZR V3, a cell phone that combines the thinness of a credit card with a full-size keypad equipped with backlit pressure plates instead of regular keys. Other companies following this design philosophy include Logitech, whose upcoming wireless, battery-powered computer mouse is smaller without sacrificing functionality because it has a removable receiver that plugs into a computer's USB ports. However, NASA Ames Research Center scientist Asaf Degani explains that it is not just a matter of shrinking electronic devices, but simplifying their usability, perhaps through the incorporation of artificial intelligence, for instance.
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.)

Cutting-edge robotics technology spotlighted at the 2004 Intelligent Robotics and Systems (IROS) conference in Sendai, Japan, included entertainment and therapeutic robots, as well as presentations on how robotics is influencing intelligence research. Entertainment bots on display included Fujitsu's HOAP-2, a commercially available device that can be programmed by a conventional PC and whose hardware and software is customizable; Sony's humanoid QRIO, which impressed attendees by negotiating an assault course, roller-skating, and dancing; and PARO, an artificial seal pup developed by Japan's National Institute of Advanced Industrial Science and Technology as a relaxation tool for recovering patients. The University of Tokyo's Max Lungarella said the creation of more intelligent robots is arousing the interest of neuroscientists, cognitive scientists, and behavioral psychologists. One example of this meeting of minds is IDIAP Research Institute's Jose del Millan's research into thought-controlled robots that could be used as neural prostheses for disabled and paralyzed people. IROS 2004 secretariat and Tohoku University robotics researcher Zhidong Wang cited nanotechnology and bio-mimicry as important research areas, noting that human-robot interaction is just as important as human control of robots. Gerard McKee with England's University of Reading said that networked and modular robots are also a critical development, while artificial intelligence scientist Claude Sammut with Australia's University of New South Wales said the most noticeable trend is the new opportunities emerging from fundamental hardware and software improvements.
Click Here to View Full Article

Several Louisiana universities have partnered on the development of UCoMS, a high-speed sensor and wireless network for monitoring offshore oil and gas production and exploration in the Gulf of Mexico. The University of Louisiana-Lafayette is heading the project, which received $1.2 million from the Department of Energy's Experimental Program to Stimulate Competitive Research; that amount was matched by the Board of Regents, while the collective pledge of the participating institutions is $1.5 million. UCoMS would be linked to a computing grid, supporting real-time sensing, computation, and data feedback operations at less cost than the monitoring technology currently employed in the Gulf, according to UL Lafayette computer science professor Hongyi Wu. The grid is driven by a 96-node cluster at UL Lafayette and the 1,024-processor SuperMike cluster at Louisiana State University. Wu notes that most oil and gas companies use microwave or satellite communications equipment that is expensive to set up, service, and maintain, whereas wireless mesh networks can be established "instantaneously." Computer science doctoral student Denvil Smith says the computing grid will function as the network's "brain" by evenly apportioning the data processing load among numerous computers. Fellow doctoral student Ravish Shah notes that one of the project's biggest challenges is programming the system to find alternate routes the data can take in the event a connection is overloaded or severed. UL Lafayette computer engineering professor Nian-Feng Tzeng wants the network to be operational in 2006, and the second research phase completed by the end of the decade.
Click Here to View Full Article

Starting Oct. 11, Carnegie Mellon University's (CMU) Robotics Institute will celebrate its 25th anniversary with a four-day event that will put its past accomplishments into perspective and focus on the challenges and responsibilities that must be met in order to reach the ultimate goal of creating universal robots that can perform virtually any task. The institute has morphed from a $1 million robotics technology research program launched by now-retired Westinghouse Electric executive Thomas Murrin and CMU professors Angel Jordan and Raj Reddy into a powerhouse facility with almost 200 faculty and personnel and a yearly budget of $50 million. The Robotics Institute's original mission emphasized robotic manufacturing, but that objective has expanded in the last 25 years to include projects in nearly every conceivable sector. Robots developed by CMU researchers have performed tasks ranging from searching for life in arid environments to inspecting nuclear reactors to probing gas mains to kicking soccer balls to playing bagpipes to assisting in heart surgery to locating meteorites in the Antarctic to climbing down the walls of a volcanic crater to troop defense. Robot intelligence is improving thanks to faster and less expensive computing power and memory, but CMU robotics professor Reid Simmons explains that programming robots to make sense of their surroundings and respond appropriately in the same way people do is a major design challenge. Former Robotics Institute director Takeo Kanade notes that creating truly helpful robots involves intense study into human motivation and behavior. The center is the only other U.S. institution besides the University of Utah to offer a robotics graduate program.
Click Here to View Full Article

World Wide Web Consortium (W3C) members are working on a Web services constraints and capabilities framework that will allow organizations to communicate the terms of their service. Requirements for using HTTP or the ability to support GZIP compression, for example, need to be communicated in a standard manner using the Web Services Description Language (WSDL 2.0) specification, SOAP, or HTTP. W3C director Tim Berners-Lee said more standards were needed to support automated Web services. Participants at a two-day W3C workshop on Web services constraints and capabilities were required to write a position paper stating how they would preferably communicate constraints and capabilities in regards to reliable messaging protocol requirements, encryption using WS-Security or other security mechanisms, and an attached P3P privacy policy. Besides discussing how best to implement such constraints and capabilities requests and what vocabularies to use, the workshop participants also discussed their framework's impact on and relation to other W3C protocols and Web technologies. Upcoming W3C recommendations include WSDL 2.0, which is currently in the "last call" stage and will be extended by the constraints and capabilities framework. SOAP 1.2 is also nearing completion, having advanced to candidate recommendation status in August 2004.
Click Here to View Full Article

Redwood Neuroscience Institute (RNI) founder Jeff Hawkins believes that intelligent machines could be developed by understanding the fundamentals of human intelligence. In his latest book, "On Intelligence," Hawkins posits that intelligence, perception, creativity, and consciousness are underpinned by the brain's use of stored memories to make accurate predictions about the world. Going by Hawkins' definition, intelligent machines will not necessarily resemble humans physically, but rather perceive and understand the world using the same mechanisms people do. Hawkins explains that such machines will "have senses tuned for their data world" so that they can monitor and measure phenomena--weather, for instance--beyond the observational capabilities of humans. He theorizes that it will be fairly simple to design machines with "exotic senses" by modeling them after the human cortex, whose flexibility is demonstrated in its ability to allow both a blind person and a deaf person to build identical models of their environment, for example. A greater challenge lies in devising an effective mode of human-machine communication, which Hawkins thinks could be met by identifying environmental elements common to both humans and machines--perhaps language or graphical representations. The RNI founder foresees the emergence of a major industry based on his ideas within a decade, if he can push people to start breaking ground fast enough. Hawkins notes that his theories about machine intelligence have had an impact on the design of traditional computing technology: One example of their influence is evident in his decision to design interfaces that support only one route for getting a specific dialog box, so as to avoid user confusion.
Click Here to View Full Article

Researchers at Lucent Technologies' Bell Labs facility have developed "nanograss," an adaptable material comprised of ultrathin silicon posts whose properties can be adjusted on the spot through temperature, electricity, ultrasound, or other means. "By adjusting the area of these posts and their density, and how you pattern them, you can engineer how...fluid interacts with the substrate," explains David Bishop with the New Jersey Nanotechnology Consortium (NJNC). "So you are not stuck with what nature gave you, and you can do lots of amazing things." The NJNC is commercializing nanograss in partnership with Bell Labs. Batteries could become cheaper and boast a 25-year shelf life thanks to nanograss, which would permit the battery's electrodes and electrolytes to remain separate until the battery is needed. Nanograss could also be incorporated into heat sinks for computer processors, and Bell Labs is about to forge a contract with a company to devise a "smart" heat sink that adjusts its cooling properties to changing requirements. The deployment of nanograss within switches, power splitters, multiplexers, filters, and other devices could enable the manipulation of light in ways that are beyond the capabilities of conventional techniques. Nanograss could also facilitate the construction of minuscule, inexpensive liquid lenses whose focal lengths and other characteristics can be altered rapidly by electrical currents, and Bishop expects this concept will lead to disposable cameras and zoom lenses within a year.
Click Here to View Full Article

Wireless sensor networks will be a huge market in the future, surpassing even cell phones and PCs in volume, according to Crossbow Technologies CEO Michael Horton at the IEEE Conference on Sensor and Ad Hoc Communications and Networks (Secon 2004). Better chip technology is one of the main factors pushing sensor network development, as chip makers are readying integrated devices that use less power and cost less. Companies such as Philips Lighting are already prototyping sensor networks that will automate building functions, while the government is studying possible applications for homeland defense. National Science Foundation Networking of Sensor Systems program director Guru Parulkar has become convinced there is a tremendous market opportunity for sensor networks and that researchers need to provide privacy and security solutions: Only a few of the proposals his group reviewed for funding addressed those areas, he says. Another area that needs work is the architecture, says Oak Ridge National Laboratory SensorNet program manager John Strand. Much of the current Department of Homeland Security research into sensor networks is focused on node functionality, but networking and interoperability issues will play a critical role as systems come together, he notes. Intel sensor net researcher Mark Yarvis, meanwhile, stresses application and ease of use, such as the ability of networks to self-configure. University of California-Riverside provided some headway on that front with their eBlocks system that allows people to quickly set up simple sensor networks using plug-and-play modules.
Click Here to View Full Article

Former ACM President Barbara Simons writes that enthusiasm for paperless electronic voting systems, originally touted as a solution to the hanging chad problem that muddled the last presidential election, is giving way to demands for voter-verifiable paper trails from lawmakers, public interest groups, and computing professionals. Such demands are being fueled by incidents and studies that all too clearly demonstrate security and reliability shortcomings in direct recording electronic systems (DREs). E-voting machines use proprietary software that is not required by the Federal Election Commission to be examined during certification testing, whose process and results are kept secret. The machine vendors pay three independent testing authorities to certify their products--but these private companies, despite being certified themselves by the National Association of State Election Directors, are not subject to government supervision. By a stroke of luck, Diebold voting machine software was discovered online and subsequently analyzed by academic researchers, who uncovered the software's susceptibility to tampering; worse, Diebold had been warned about this problem by a member of the Iowa Board of Examiners for Voting Machines and Electronic Voting Equipment several years earlier, yet had taken no remedial action. Another e-voting machine provider, ES&S, has courted controversy because its products' auditing mechanism was found to be vulnerable to a software glitch that rendered audit reports "unusable," according to an internal memo. This bug went unnoticed because the current certification process does not check for such flaws. Simons offers precautions election officials should take to minimize the security and reliability problems of paperless DREs, including: Rigorous testing of e-voting systems before, during, and after elections; secure storage of DREs between elections and at polling stations before and during Election Day; extensive testing of all the ballot definition files, and the public disclosure of test results and their archival in a central repository; the establishment of a national repository of DRE problems; and the provision of paper ballots at all polling places that use DREs.