Despite crippling levels of poverty and violence, Brazil is home to some of the world's most innovative technology, and plays host to some of the most sophisticated hackers. Brazil often finds itself the locus of international debates over intellectual property rights and private media controls, and though it does not have in place the infrastructure that other developing nations do, Brazil has made significant advances in open access technology that place it at the forefront of the Third World. Brazil received a major economic boost when Google acquired the native firm Akwan Information Technologies and established an office in Sao Paolo. There is still a wide gulf between rich and poor in Brazil, and while its 22 million-plus residents with Internet access rank it in the top 10 worldwide, that number still only represents 12 percent of the population. Piracy is also a major issue, as roughly 60 percent of the software and 70 percent of the hardware in use in Brazil infringes on copyright laws; Brazil is also a notorious haven for cyber criminals, as it is estimated that approximately 80 percent of the world's hackers are based in Brazil. The country's emerging IT industry has reached the $10 billion mark in annual sales. The spirit of unfettered access has led to the widespread implementation of the Linux platform in government and private industry, along with a host of other open-source applications. Throughout Brazil, open access movements are seeking to provide free Internet capability to computer users, and its vibrant open-source community draws on innovation from all over the country to maintain Web sites, provide tech support, and develop new technologies.
Click Here to View Full Article

Ambassador David Gross, the State Department's coordinator for international communications and information policy, called appeals for the United States to cede control of the computers that manage the Web to a U.N. group "unacceptable." He cited the issue as non-negotiable and "a matter of national policy." American governance of the Internet is an especially contentious topic for developing nations, which believe Internet first adopters such as the United States and European countries have hogged most of the available Web addresses. They also desire guarantees that later U.S. policy will not scuttle their plans to implement Web-based governmental and other services. One proposal suggests ICANN hand over its control of domain names to an intergovernmental organization. In Geneva for the last preparatory meeting before the U.N. World Summit on the Information Society in November, Gross expressed his country's disappointment with a European Union proposal for a "new cooperation model" in which governments would participate in the naming, numbering, and addressing on the Internet. Some negotiators from other countries reported a growing feeling that a compromise is necessary, and that the Internet should not be managed by any one nation. The goal of the summit is to make certain that the Internet is shared fairly for the entire world's benefit, but the conference could be defeated if a deadlock is reached over who should be the ultimate authority for Internet routing and addressing.
Click Here to View Full Article

As new digital technologies continue to develop, the specter of obsolete applications threatens to render much of the content produced today unreadable. Libraries cannot keep a working sample of each iteration of hardware, and the traditional method of transferring a file into the next generation's format introduces small imperfections with each copy, the sum of which could make it impossible to read a document over time. One possible solution to preserving digital data in its original form is a software-based experimental virtual computer system being tested at the National Library of the Netherlands. IBM is developing the Universal Virtual Computer, which runs programs to decode various document formats. Though the libraries of the future will have to create new programs to emulate the preceding generation's virtual computer, once that process is complete, the decoders will unlock all of the documents stored on the computer. "The decoder can be tested for correctness today, while the format is still readable," said IBM's Raymond van Diessen. IBM has already programmed decoders for the JPEG and GIF formats, and will address Adobe's PDF format next.
Click Here to View Full Article

Sun Microsystems co-founder Bill Joy said at Technology Review's Emerging Technology Conference that the tech industry is experiencing dramatic changes as wireless, Web-connected devices such as personal digital assistants and cell phones proliferate. The mobile device explosion is underlining the "here" Web movement, which Joy described as "here, because you access [the Internet] through a device you always carry." He predicted that companies such as Google will benefit enormously from the here Web phenomenon, because they envision the most interesting and critical software applications running atop the Web in the future. The development of the PC platform, in contrast, is languishing. Joy opined that hierarchical organizations stand to lose because they cannot adapt as quickly to the influx of non-hierarchical companies. Joy, who is now a partner at the Kleiner Perkins Caulfield & Byers venture capital firm, characterized voice over Internet Protocol as a warm but frustrating technology that needs significant improvement.
Click Here to View Full Article

Next-generation embedded system software tools are expected to boost developer efficiency by forsaking bells and whistles in favor of greater support for debugging and troubleshooting, as well as platform independence. "To speed up the debugging, code integration, and optimization of such complex systems, developers often need visibility into the source code of underlying libraries and [operating system] components, even if they're using an off-the-shelf RTOS," says QNX Software Systems' Mark Roberts. His company promises that its new QNX Platform Core Source Kit will add visibility to the source code for practically all QNX Neutrino system libraries and OS facilities, thus easing analysis of system behavior for code integration, trackdown of causes of unanticipated software behavior, and refinement of OS elements to satisfy the particular needs of their application. John Fanelli of Wind River Systems notes that more and more systems have hard real-time and soft real-time requirements, and fulfilling those requirements often involves the use of devices comprised of a series of devices. This is leading device designers to contemplate systems that exploit multiple OSes. Communication between devices is critical regardless of what OS is ultimately used. Wind River's solution is to address complexity across multiple OSes and devices by supplying support for scaling features, footprint, and performance according to the target device. Furthermore, the company is offering device, processor, and peripheral support for power management in a multicore/multisystem setting that is interoperable with an environment composed of multiple OSes.
Click Here to View Full Article

The prospect of legislation that would protect large corporations from patent litigation has some concerned about the stifling effect on innovation such a move could have as a byproduct. The pending legislation introduced by Rep. Lamar Smith (R.-Texas) seeks to address the trend of large settlements having recently been awarded to small companies, derisively known as patent trolls, that purchase dormant patents and file suit against companies using similar technologies, along with a host of other patent issues. The competition between private inventors and large corporations for patents has grown fierce in recent years. Although large companies always crowd the list of top individual patent recipients, the most patents overall are awarded to individual innovators, educational institutions, and small companies. One component of the proposed legislation would unseat the precedent of awarding a patent to the first inventor dating back to 1790, revising it to grant the patent to the first to file, a move which alarms the smaller, typically more disorganized inventors. For their part, backers of the "first-to-file" notion contend that it simplifies the patent process by eliminating the often contentious uncertainty that shrouds the question of who developed an invention first. The legislation would also funnel more money to the patent office to facilitate more thorough patent reviews. On the issue of patent trolls, software companies voice the loudest concern, as high-tech devices often contain scores of patents, so a small company alleging encroachment on a given patent can effectively halt production on a massive scale until the matter is settled in the courts.
Click Here to View Full Article
(Access to this site is free; however, first-time visitors must register.)

Results of the semifinals for the second annual Defense Advanced Research Projects Agency (DARPA) Grand Challenge competition would seem to indicate that this year's entrants will perform significantly better than last year's. The inaugural Grand Challenge pitted robotic vehicles against each other in a 144-mile desert race, but none of the contestants was able to travel even 10 miles. This year's race will cover 175 miles, with the winner, if there is one, receiving a $2 million prize. Forty-three autonomous vehicles are participating in the eight-day semifinals at the California Speedway, which will test how well they navigate gates, tunnels, mountain switchbacks, and narrow roads; only 20 entries will qualify for the finals. There are signs that the unmanned vehicle technologies being tested this year are substantially upgraded from last year. For example, Carnegie Mellon University's robot Hummer, which traveled only 7.3 miles in the 2004 Grand Challenge, successfully drove 200 miles on a racecourse this summer. The goal of the DARPA Grand Challenge is to encourage U.S. universities and private-sector researchers to develop innovative artificial intelligence technologies.
Click Here to View Full Article

Anti-spyware software is expected to transition from threat-specific technologies to Host-based Intrusion Prevention Systems (HIPS) as vendors deploy proactive solutions that block new and unknown spyware programs from PCs. Such solutions are likely to be increasingly compelling for security architects as the development of spyware continues without respite and end users continue to install spyware-laden programs despite repeated warnings. Most anti-spyware programs use signatures and are only effective against programs that are already defined in the threat database, while the increasing difficulty of removing spyware once installed makes proactive prevention all the more urgent. Some vendors offer behavior-based spyware detection technologies that can thwart the installation of spyware on enterprise desktops without the use of signatures, although such solutions carry with them the risk of false positives. "The market is warming up to the notion that existing signature-based solutions aren't providing adequate malware prevention," says Finjan's Nick Sears. "Customers are looking to alternative solutions." Other anti-spyware options deliver protection at the network gateway by scanning incoming Web traffic for spyware and adware, preventing spyware on a PC from linking to a remote server on the Internet, and stopping end users from surfing to established sites for spyware or adware. However, none of the gateway products can protect mobile users outside the corporate environment.
Click Here to View Full Article

Audio networking has been eclipsed by high-profile wireless communication devices such as infrared or Bluetooth, even though it supports reliable data transmission with less complexity and power consumption by exploiting existing programmable smart phone interfaces. Modern smart phones use the Bluetooth object exchange protocol to swap contact details (phone numbers, home addresses, or Internet URLs) as mobile users interact spontaneously, although this process takes a long time and is susceptible to error. Audio networking is simpler: The desired contact information is encoded into an audible message by the smart phone, while the low-amplitude transmission of messages is facilitated by holding the two phones in close proximity. Attackers who do no have the advantage of sophisticated listening equipment would have no other choice but to get very close to eavesdrop, while adjusting the output volume enables two parties to guarantee an appropriate level of privacy for their exchange. The researchers opted for social-flow control between call participants instead of using a media access control layer for their audible packets. Permitting participants to manually schedule audio data transmission as part of their normal conversation eliminated surprise caused by the sudden insertion of data packets. An Internet rendezvous service was developed and tied into a telephone-conferencing application designed to supplement exchanges with a list of all conference participants and a real-time display of who is currently speaking.
Click Here to View Full Article

One in five public sector employees will retire in the next five years in over half the U.S. states, according to a report from the Government Performance Project. Their imminent retirement not only constitutes a loss of personnel, but the potential loss of institutional knowledge, particularly of IT systems. Other issues include a widening gap between required and actual IT skills, a dearth of younger workers, a lack of understanding among older IT workers of the younger generation's IT needs, and a shortage of in-house IT talent capable of assuming bigger duties. In addition, many legislatures are resistant to the idea of replacing legacy IT platforms with modern systems, given the costs involved. Government agencies and public institutions are therefore getting saddled with older systems that do not necessarily fulfill the requirements of the contemporary world. Experts suggest the federal government address such problems by hiring employees midcareer rather than recruiting them at just entry and senior levels. State human resource agencies have also started to concentrate on workforce transition and succession planning, and are working on strategies to convince potential retirees to stay on in return for higher pay or other incentives. Each state will eventually have to contend with the challenge of making e-government services accessible and convenient to younger, Web-savvy citizens.
Click Here to View Full Article

The proliferation of new information forms, coupled with its sheer volume, is a pressing challenge confronting today's librarians. Even as Stanford partnered with Google to digitize its collections, the university's Victoria Reich was developing the open source LOCKSS (Lots of Copies Keep Stuff Safe) project, which uses a decentralized method to create a permanent cache of data. The system mines the Web to collect content, then preserves the material by saving it in its original form, and disseminates a link to the original site to every URL that ever carried the content. Each LOCKSS box on campuses across the country can hold 3,000 years worth of journal articles. LOCKSS focuses on archiving journals, while Harvard is launching a digitization effort that targets the video medium. VideoTools codes, shares, and publishes all of the school's video content, assigning each piece with a URL that can be accessed via the Web. The system contains strict access controls, but once the user is authenticated, VideoTools streams video at 1.2 Mbps, roughly equivalent to an actual DVD. The system is also designed to scale with the growing body of content. The California Digital Library (CDL) is a collaborative project that culls together material from a multitude of institutions through initiatives such as the American West program. Rich with metadata, the program receives and normalizes content from disparate systems. Another CDL project, the Digital Preservation Repository (DPR), uses simple object access protocol (SOAP) to create long-term storage for a variety of content. MIT's DSpace project makes a duplicate copy of all content, and assigns each piece with a URL that lasts for its life. DSpace centers on flexibility, allowing any institution to submit content at any time, which carries the added benefit of drawing on users to refine the open source program's functionality.
Click Here to View Full Article

Niall Murphy, author of "Front Panel: Designing Software for Embedded User Interfaces," says it is possible to create strong tests that catch actual glitches without angering the design team. He recommends that testers start out by checking the parts of the system responsible for checking and logging errors, and then try experimenting with the power cycle, for instance by quickly turning the unit on and off several times. Running the system without interruption for several months can help testers determine what elements will malfunction first, although some tests can be done faster simply by accelerating the calendar; Murphy writes that many location transitions can be tested by boosting the frequency of writes to model the passage of a greater amount of time. The author notes that many embedded systems contend with invisible phenomena, and he thinks invisible properties should be made detectable in software, one example being making original calculated values visible. Murphy calls for breakdown tests conducted by opening the box and disconnecting cables or mechanical interconnects, as well as coverage tests that display a list of each character string the system can present. Pressing all invalid keys in each state of the user interface is also suggested, as is deliberately failing system calibration. Murphy recommends decreasing the period of the watchdog timer to test whether the software is strobing the software at the proper rate, and notes his preference to lower the interval for all systems during development, and then reset it to the desired design interval closer to the release. Debugging hooks or backdoors installed for test and debug purposes must be removed as the final build approaches.
Click Here to View Full Article

In an online question-and-answer session, Linux experts and authors Marcel Gagne and Peter van der Linden discussed a wide range of topics concerning the Linux operating system. Questions regarding Linux's benefits prompted the experts to list such things as ease of use and installation, exceptional hardware and peripheral support, better security from malware, lower cost, and support for "fair use" of copyrighted digital content. One query broached the subject of how to choose from the many Linux distributions available, and Gagne responded with the observation that the main differentiator is the packaging rather than the actual distributions, which are very similar. Van der Linden said the Linspire distribution stands out in terms of fast and easy installation, as well as desktop experience for non-technical users. He also noted that many Linux distributions interact cooperatively with each other, and foresaw increased consolidation. Gagne called attention to the negative press Linux has received, specifically that it is too complicated for average users; he dismissed these allegations as "nonsense" and called for friendlier representation. Gagne said the biggest impediment to Linux's penetration of the desktop is Windows users' fear that the operating system is too complex, a concern reinforced by media-borne disinformation. Van der Linden predicted that the next "killer app" for Linux, whatever it may be, will come from Google, because Google generates new ideas rapidly.
Click Here to View Full Article

Making the Internet's educational and communications advantages accessible to children everywhere is the goal behind MIT Media Lab director Nicholas Negroponte's One Laptop Per Child initiative, which seeks to produce between 5 million and 15 million $100 laptops within a year. The device, which Negroponte detailed at Technology Review magazine's Emerging Technologies conference, features an AC adapter/carrying strap; a 500 MHz processor; a tight rubber casing to protect the unit from damage; a hand crank that supplies power when electricity is unavailable; a tint-adjustable display; four USB ports for multimedia and data storage; and flash memory. The laptops would save on software costs by running the open-source Linux operating system rather than Windows, and reduce the need for costly base stations by participating in "mesh" networks and connecting to Wi-Fi wireless networks. In the second year of the One Laptop Per Child project, Massachusetts Gov. Mitt Romney intends to start purchasing the $100 laptops for 500,000 middle and high-school students in his state, while Negroponte expects production to increase to between 100 million and 150 million units. The MIT team plans to continually push the cost of the laptops down by incorporating low-power enhancements such as "electronic ink" displays.
Click Here to View Full Article

As the interconnectedness central to the dream of the digital home rapidly becomes a reality, a host of security and privacy concerns arises. The same Web cams that alert users to suspicious activity within their homes can also be used by hackers seeking to break in to determine if anyone is home. Internet connectivity is being incorporated into a growing number of devices that have not yet evolved to carry the same level of security as PCs and desktops. As attacks on traditional hardware become more sparse, the added functionality in mobile phones makes them a more popular target. The number of reported malware threats menacing mobile devices has grown to 87, up from fewer than 10 at the beginning of last year. Symbian is the most popular operating system for mobile phones in the world, and its series 60 was the target of 82 of the reported viruses, though analysts are quick to point out that that proportion speaks more to the system's popularity than its vulnerability. Faster download speeds elevate the risk of a virus infecting and spreading throughout a mobile phone. It is projected that the threat against mobile devices will increase as more hackers recognize the potential vulnerabilities and turn their attention away from traditional attacks.
Click Here to View Full Article

The WebLabs project, sponsored by the IST program, has developed an approach to teaching young people math and science that employs virtual models of abstract ideas and Web-based collaboration. The project involved the participation of technologists and educationalists who worked with selected schools and approximately 200 students aged 10 to 14 in various European countries. Scientific and mathematical concepts participating students and educators focused on included sequences, infinity, collisions, and models, systems, and randomness. These concepts were virtually represented via a Web-based interface, and rendered as interactive animated characters through the ToonTalk programming language. ToonTalk was used to create some 50 tools that enable students to test theories while sparing them from the need to write complicated programs. Students can share working ToonTalk conceptual models through the WebReports collaboration system, which incorporates the models into reports as clickable images that link to the code objects. The system allows students to access other students' ToonTalk environments, where models can be manipulated, tweaked, and commented on. "This last point is crucial--rather than simply commenting on another's ideas, the student can rebuild the other student's attempts to model a given task or object," says professor Celia Hoyles of the London Knowledge Lab.
Click Here to View Full Article

Google and NASA announced an alliance to establish a research complex at NASA's Ames Research Center and collaborate on mutually beneficial projects at a Sept. 28 press conference. Research center director Scott Pace cited Google's ability to organize massive volumes of information for public consumption as vital to the space agency, which is inundated with data generated by its various missions and space probes. The partnership is a major boost for the NASA facility, which seeks to retool itself into a center for public-private research. As part of the deal, NASA will make the researchers who manage the Project Columbia supercomputer available to Google engineers to brainstorm new supercomputing designs that could help yield new Google products. In addition, Google will be able to access NASA's space data and imagery, which the search company could augment with additional information, according to Google's Peter Norvig. He mentioned that Google wishes to devise mapping applications for the Moon, Mars, and other extraterrestrial sites. The announcement of the NASA-Google alliance did not specify how extensive Google's presence at NASA would be, its impact on the surrounding community, or the precise nature of the collaborative projects.
Click Here to View Full Article

The latest initiative from the Open Geospatial Consortium (OGC), OWS-3, promises greater geospatial interoperability through the acquisition of geospatial data from sensors and the enabling of decision support by the definition of a common infrastructure, according to OGC executive director for interoperability architecture George Percivall. Sensors are made available as enterprise Web services through OWS-3's Sensor Web Enablement (SWE), and the OGC has been working with the Institute of Electrical and Electronics Engineers (IEEE) to achieve scalability for its ground-based sensors. Percivall says OWS-3 continues its work from a previous iteration of workflow automation by refining components universal among OGC's Web services, and treats them with an unprecedented level of flexibility. To enable data sharing, OWS-3 has made contributions to digital rights management for geospatial data (GeoDRM). OWS-3 is also offering visualization services to aid decision making, such as a variety of options for symbol selection. Percivall notes that while OWS-3 is still in its preliminary stages, it has already undergone one revision, and the team is preparing a demonstration to submit to the OGC Specification Program. OWS-3 will offer some specifications that will augment existing, mature specifications. OWS-3 has attracted a broad coalition of sponsors that have expressed interest in the project out of a desire to have their individual needs addressed in its development and to gain access to a platform on which to test their own work.
Click Here to View Full Article