ACM TechNews (HTML) Read the TechNews Online at: http://technews.acm.org

ACM TechNews
June 18, 2008

Learn about ACM's more than 3,000 online courses and 1,100 online books
MemberNet
CareerNews
Unsubscribe

Welcome to the June 18, 2008 edition of ACM TechNews, providing timely information for IT professionals three times a week.


HEADLINES AT A GLANCE:

 

Tartalo the Robot Is Knocking on Your Door
Basque Research (06/18/08) Bulegoa, Prentsa

University of the Basque Country researchers are developing a robot called Tartalo capable of finding its way around by itself by being able to identify different places and ask permission before going to new places. The university's Autonomous Robotics and Systems Research Team is working to develop a robot capable of walking without help and able to make decisions for itself in order to increase the autonomy of robots so that they are more capable of carrying out a wider variety of tasks on their own. Tartalo, a 1.5 meter high robot, is equipped with the ability to sidestep any obstacle in its path, which it detects using sensors and a laser that measures the distance of the robot to any object within a radius of 180 degrees. The sensors and the robot's programming allow Tartalo to wander without problems, but what the researchers truly want to achieve is a robot capable of going anywhere it is told. The researchers are using biomimetic systems as a basis for developing such capabilities. The result would be that Tartalo would do the same thing a person would do in a new environment--explore the terrain and note points of interest. Tartalo has be programmed to recognize common structures in buildings, including rooms, corridors, a front hall, and a "junction." In new environments, Tartalo runs an auto-location process where the robot moves around the area to find and memorize the location of these structures, creating a topological map that the owner could use to teach the robot new places.
Click Here to View Full Article
to the top


Southern Utah University Brings Supercomputing Into the Classroom with Star-P
Sun Herald (Mississippi) (06/17/08)

Southern Utah University (SUU) believes that high performance computing (HPC) should be more than a tool available only to elite research labs and should be a common feature in a variety of undergraduate studies. It is working with Interactive Supercomputing (ISC) to make HPC easy and accessible for students and faculty at all levels as part of a program developed by SUU's College of Computing, Integrated Engineering, and Technology (CCIET) to create an integrated, interdisciplinary curriculum that combines math, engineering, and computer science. SUU is using ISC's Star-P software to provide students and faculty with access to the school's powerful Dell 128-node parallel cluster. The platform allows SUU users to easily program models and algorithms using familiar desktop languages by automatically transforming the application to run on the parallel clusters, while also eliminating the need to re-program the applications in complex languages such as C and FORTRAN or the need to use a message passing interface to run in parallel. "Our mission is to help students achieve their academic goals and to compete on a global level for careers in government, industry, secondary education, and acceptance to graduate school," says CCIET dean and professor of mathematics Mikhail Bouniaev. "Supercomputing is increasingly playing a critical role in those career paths."
Click Here to View Full Article
to the top


Man vs Machine Poker Challenge Announced
Online Casino News (06/17/08)

After losing a close contest during last year's inaugural Man Versus Machine Poker Championship, a redesigned Polaris artificial intelligence program will once again challenge some of the best poker players in the world. The Polaris 2 was developed by researchers from the University of Alberta's Computer Poker Research Group, and the program will compete against professional poker players Nick Grudzien, IJay Palansky, and Matt Hawrilenko at the second Man vs. Machine Championship, to be held during the 2008 Gaming Expo at the Rio All-Suite Hotel and Casino from July 3 to July 6. Poker coach Bryce Paradis says the Polaris 2 team has made significant improvements to the program since last year's match, with perhaps the most incredible improvement being the program's ability to learn from and adapt to its opponents as play progresses. "This year's Man Versus Machine match is going to push our team to their limit," says Paradis. Last year, Polaris 1 played professional poker players Phil Laak and Ali Eslami in Vancouver, which ended with a final score of two wins, one loss, and one statistical tie in favor of the human players. Each round in the match consisted of a 500 duplicate hand matches where the same series of cards was dealt to both players with each one playing opposite hands in the game while the computer took the other side. At the end of the match the total number of chips won or lost by each team determined the winner. Organizers will be using the same format this year in an effort to reduce the element of random luck to a minimum.
Click Here to View Full Article
to the top


Intel Develops Programming Language for Multi-Core Computers
InformationWeek (06/12/08) Gonsalves, Antone

Intel showed off a new programming language for multi-core computing, called Ct, at the Computer History Museum in Mountain View, Calif., on June 11. An extension of C/C++, the programming language automatically partitions code to run on specific cores. "With Ct, it's almost like you're writing to a single-core machine," said Intel researcher Mohan Rajagopalan during the open house for Intel labs. "You leave it to the compiler and runtime to parallelize." Intel developed the Ct compiler, which chops up the code to run on separate cores based on the type of data and the operation performed on the data, in addition to the runtime and an API for the compiler. Less than 5 percent of Ct is new, so C/C++ programmers will find it easy to use. Rajagopalan also noted that programs compiled in Ct can scale to the available number of cores. Intel is relatively close to bringing to market a product developers will be able to use to make financial analytics applications and software for processing images or decoding video.
Click Here to View Full Article
to the top


NASA Tests Robots for Manned Move to Mars
Computerworld (06/12/08) Gaudin, Sharon

When astronauts step foot on Mars sometime in the future, they may be greeted by robots and robotic rovers that were sent to Mars in advance. The prepositioned robots will be sent ahead to help their human counterparts on what will probably be NASA's most challenging mission to date. Although the mission is still years away, NASA is testing the robots now, and is evaluating the three robotic arms aboard the International Space Station and the results from space shuttle Discovery, one of the most robotically intense missions yet. NASA's Allard Beutel says NASA keeps increasing the complexity of the robotics work, and such work has become commonplace. Beutel says when astronauts get to Mars, they will use robots as part of their everyday existence, including possibly building a workstation or habitat structure on Mars. The crew of the space shuttle Discovery helped deliver a 33-foot-long, 1,716-pound Japanese-built robotic arm to the space station. The arm has six joints and is designed for use outside the Japanese Experiment Module, specifically for moving materials outside the airlock so scientists can see how they react when exposed to space, for example. Beutel says a second Japanese-built robot arm, which will be about six feet long and have a grapple on the end, is scheduled to be delivered to the space station next year. Another robotic arm on the space station, called Dextre, is designed to handle most of the exterior maintenance jobs on the space station, reducing the number of dangerous space walks the astronauts must make. On Mars, the Mars Lander is using a robotic arm to scoop soil and ice from the Martian northern pole and deliver the material to different analysis tools.
Click Here to View Full Article
to the top


National Conference on Sustaining the Science and Engineering Workforce to Be Held at UMass Amherst June 24
University of Massachusetts Amherst (06/16/08)

The University of Massachusetts Amherst will host the national conference "Best Practices for Science Education: Retaining Science and Engineering Undergraduates, Sustaining the Science and Engineering Workforce" on June 24. Dean George Langford of the UMass Amherst College of Natural Science and Mathematics says the conference is in response to a serious problem at the national level, specifically the large percentage of students who enter college interested in science and engineering disciplines but switch to other majors, compounded by the fact that job opportunities in these fields has quadrupled. "This disparity has led to outsourcing of talent, which undermines the U.S. leadership position in educating and training the next generation of innovators," says Langford. The conference's keynote address will be given by the National Science Foundation's Linda Slakey, who worked at UMass Amherst from 1973 until 2006 and served as dean of the College of Natural Sciences and Mathematics, dean of the Commonwealth College, chair of the department of biochemistry and molecular biology, and professor of biochemistry. The conference is intended to identify activities that increase the number of undergraduates in science and engineering disciplines, and highlight curriculum reform and best practices for retaining undergraduates. The ultimate goal is to expand the number of graduates entering the science and engineering workforce, and to build momentum and visibility for legislation changes and new funding opportunities. The conference will feature four breakout sessions: Two concurrent sessions will examine issues surrounding classroom effectiveness, with a focus on large enrollment classes and creating a community of scientists; another set of concurrent sessions will look at methods for building research into the curriculum, and more effective introductory lab courses and alternatives to labs.
Click Here to View Full Article
to the top


Space Station Could Beam Secret Quantum Codes by 2014
Scientific American (06/08) Minkel, J.R.

Researchers hope to be running an experiment on the International Space Station (ISS) by the middle of the next decade that would allow for transcontinental transmissions of secret messages encoded using the quantum property of entanglement, which is when two particles, such as photons, are created by the same event and can communicate instantaneously no matter how far apart they are. Transmitting entangled pairs of photons reliably is the foundation of quantum key distribution, a procedure that converts those pairs of photons into potentially unbreakable codes. Photons can travel maybe 100 miles on modern fiber-optic cables before their quantum character breaks down, but that limit disappears above ground. Last year, a team of researchers led by physicist Anton Zeilinger from the University of Vienna successfully transmitted quantum keys up to 89.5 miles between a pair of telescopes in Spain's Canary Islands. Now they want to send quantum keys hundreds of miles or even more. The group is leading an international project called Space-QUEST, or Quantum Entanglement for Space Experiments, with the intention to prove that a system for generating pairs of entangled photons can fit the constraints imposed by the ISS. Quantum keys distributed from the ISS could be transmitted to any two points within the station's line of sight, limited only by the ability of transmitters and receivers to maintain a tight lock on one another and isolate entangled photons from background light. Zeilinger and his colleagues demonstrated they could detect single photons reflects off a satellite 3,700 miles above Earth earlier this year. Space-QUEST hopes to build a prototype device and gather preliminary data in time for a meeting of the European Space Agency in November, where officials will decide what projects will receive funding and earn the chance to be run in space.
Click Here to View Full Article
to the top


Can Computer Scientist Dream Team Clean Up E-Voting?
Network World (06/10/08)

Electronic voting has become a source of concern and controversy, with many e-voting systems proving to be security black holes. A Center for Correct, Usable, Reliable, Auditable, and Transparent Elections (ACCURATE) has received a $7.5 million National Science Foundation award to bring the latest research, insights, and innovations from the lab to the voting booth and make e-voting systems more secure and accurate. ACCURATE unites computer experts from across the country and academic disciplines to find areas that need additional research and to determine how to apply existing technology and research findings to voting systems. ACCURATE members from Rice University have designed and implemented a system called "Auditorium," which forms the base of a voting system prototype called "VoteBox." Auditorium is a networked logging and auditing system built using timeline entanglement and broadcast messages. Auditorium allows anyone to audit the events, in the order that they occurred, with strong cryptographic guarantees to protect against tampering with the timeline. Additional research on secure logging is examining how log verification could be scaled for an entire election in real time. ACCURATE members at the University of California, Berkeley, are examining methods for building trustworthy audit logs in electronic voting systems. Their goal is to design a mechanism that records the entire user interaction between the voter and the voting machine to allow auditors to replay a "movie" of the interactions after the election. Challenges include ensuring that the audit log does not compromise ballot secrecy and that it is a trustworthy system.
Click Here to View Full Article
to the top


Doubling Laptop Battery Life
Technology Review (06/13/08) Greene, Kate

Intel researchers believe they have discovered a technique that can double a laptop battery's life without changing the battery. The method optimizes power management across the system, including the operating system, screen, mouse, motherboard chips, and USB port devices. Manufacturers and researchers have been exploring a variety of ways to make mobile devices more energy efficient, including operating systems that deploy power-saving screen savers and put the system to sleep if the user has been away for a while, and Intel's upcoming Atom microprocessor for mobile Internet devices has six different levels of sleep depending on the tasks being performed. Intel's prototype power-management system, called advanced platform power management, is aware of the power being used by all parts of the laptop, as well as the power requirements of the user's current task, and it shuts down operations accordingly, according to Intel's Greg Allison. Allison says, for example, that when a person reads a static email, the screen still refreshes 60 times a second with traditional systems, and peripherals such as the keyboard, mouse, and USB devices continue to drain power while waiting for instructions. In such a situation, Intel's system would save power by essentially taking a snapshot of the screen that the user is reading and saving it to buffer memory, so instead of refreshing, the screen would maintain an image until the user tapped a button on the keyboard or moved the mouse, both of which also stay asleep until activated. Meanwhile, the operating system would monitor the use of other applications, and limit operations for applications not actively being used. Energy-monitoring circuits on the chips will also put unnecessary parts of the microprocessor to sleep. Allison says it takes only 50 milliseconds for the entire system to wake up, an imperceptible time to users.
Click Here to View Full Article
to the top


'Electron Turbine' Could Print Designer Molecules
NewScientistTech (06/11/08) McAlpine, Kate

Researchers at Lancaster University in the United Kingdom have tested the design of a nanomotor using advanced computer simulations. However, Adrian Bachtold from the Catalan Institute for Nanotechnology plans to build a carbon nanotube that spins in a current of electrons, similar to a wind turbine in a breeze. The design consists of three carbon nanotube 10 nanometers long and 1 nm wide, one suspended between the others, with its ends nested inside them to form a rotating joint. The central carbon nanotube spins around when a direct current is passed along the tubes. The electrons bounce off the spiral carbon rings of the nanotube turbine, which redirects them into a spiral flow that causes the tube to move in the opposite direction. The Lancaster team plans to design smoother nanotubes to prevent friction. Such an electron turbine has the potential to be used to serve as a tiny printer or to make computer memory smaller. The June issue of Physical Review Letters will feature a paper on the tiny, electron windmills.
Click Here to View Full Article
to the top


Whole Proves to Be Mightier Than the Parts
ICT Results (06/11/08)

Researchers in Europe have developed a strategy that will make it easier for research and education networks to connect with each other, and also to commercial telecommunications networks. The MUPBED project used an automated control plane to cut down vertically through the layers a network encounters, and established an automatic link between networks when a user requests a connection of bandwidth. The networks communicate with each other, provide the solution, then inform the parties at both ends and the operators in between what has happened over the links. Moreover, network resources are optimized, and capacity is only used when it is needed for a specific task. "We developed a network solution which allows multi-domain networking, and working with standards bodies tested it against emerging standards," says Jan Spaeth, coordinator of the EU-funded project. "We were able to influence the standards bodies, who had not previously been aware of the research networks' requirements and had no input from that source." The team believes commercial networks will be able to use its research to develop more advanced services.
Click Here to View Full Article
to the top


In Congress, H-1B Issue Pits Tech Workers Against Farm Groups
Computerworld (06/12/08) Thibodeau, Patrick

The tech industry's effort to make it easier for skilled foreign workers to remain in the United States came under criticism during a U.S. House hearing on June 12, 2008. Rep. Zoe Lofgren (D-Calif.), chair of the Subcommittee on Immigration, Citizenship and Refugees, Border Security, and International Law, made a strong case for expanding the cap on the number of H-1B workers, but committee member Rep. Luis Gutierrez (D-Ill.) said a narrow approach to the issue of immigration would be taken. "I think we should give the high-tech industry the innovators they need," said Gutierrez, who added that something would have to be done about other foreign workers, such as farmers. Immigration has become an all-or-nothing battle for many lawmakers after last year's failed bid at reform, and uncertainty surrounds the efforts of Lofgren, who has introduced three bills related to hiring foreign tech workers. Groups such as the Institute of Electrical and Electronics Engineers, the Semiconductor Industry Association, and the Association of International Educators have expressed support for improving the H-1B visa program. Sens. Barbara Boxer (D-Calif.) and Judd Gregg (R-N.H.) introduced a companion bill in the Senate earlier in the month.
Click Here to View Full Article
to the top


Rummaging Through the Internet
Economist Technology Quarterly (06/08) Vol. 387, No. 8583, P. 14

Web browsing promises to be transformed by new methods for navigating and collecting information online, and one such method is the freely available Hyperwords browser add-on, which turns every word or phrase on a page into a hyperlink. Meanwhile, the Cooliris startup has developed PicLens, free software that gathers and displays images retrieved from Google, Flickr, eBay, and other Web sites on a full-screen, 3D wall without any of the clutter on each image's Web page. Such applications hint at one possible future incarnation of Web browsing, in which users navigate through groupings of pages that appear to float in space, pushing undesirable ones away and organizing others in logical clusters. In late July, the Second Life 3D virtual environment will incorporate a feature that lets inhabitants post Web pages on walls, changing Web browsing from a solitary to a social experience because users roaming virtually through the environment can convene and chat next to Web pages, according to Linden Labs executive Joe Miller. Another social browsing tool is 3B's 3B browser, which arranges pictures of the results of product searches within the aisles of a virtual shop, where shoppers can gather to see better and chat through instant messaging with other shoppers looking for similar items. Carnegie Mellon University computer scientist Dave Farber predicts that the coolness of the visuals generated by such tools will eventually give way to the realization of the need for 3D navigation.
Click Here to View Full Article
to the top


Key Differences Between Web 1.0 and Web 2.0
First Monday (06/08) Vol. 13, No. 6, Cormode, Graham; Krishnamurthy, Balachander

Among Web 2.0's key attributes are the growth of social networks, bi-directional communication, diverse content types, and various "glue" technologies, and the authors note that while most of Web 2.0 shares the same substrate as Web 1.0, there are some significant differences. Features typical of Web 2.0 Web sites include users as first class entities in the system, with prominent profile pages; the ability to connect with users through links to other users who are "friends," membership in various types of "groups," and subscriptions or RSS feeds of "updates" from other users; the ability to post content in various media, including blogs, photos, videos, ratings, and tags; and more technical features, such as embedding of various rich content types, communication with other users through internal email or instant messaging systems, and a public API to permit third-party augmentations and mash-ups. Web 1.0 metrics of similar interest in Web 2.0 include the general portion of Internet traffic, numbers of users and servers, and portion of various protocols. About 500 million users reside in a few tens of social networks with the top few responsible for the bulk of the users and traffic, and traffic within a Web 2.0 site is more difficult to measure without help from the site itself. The challenges for streamlining popular sites for mobile users differ slightly between Web 1.0 and Web 2.0, in that instant notification to users through mobile devices can be facilitated because of the short or episodic nature of most Web 2.0 communications. Most communication in Web 2.0 is between users, so Web 2.0 sites have no easy way to select during overload; however, the sites apply varying restrictions to guarantee that overall load and latency is reasonably maintained. Some of the Web 2.0 sites are eager to maximize and retain members within an "electronic fence," which can facilitate balkanization, although total balkanization is likely to be prevented by a countercurrent stemming from the prevalent link-based nature of Web users continuously connecting to sites outside the fence. The authors point out that there are substantial challenges in permitting users to comprehend privacy implications and to simply represent usage policies for their personal data.
Click Here to View Full Article
to the top


Machine Translation for the Most Part Targets Internet and Technical Texts
Universidad Politecnica de Madrid (05/27/08)

President of the Association for Machine Translation in the Americas Mike Dillinger says in an interview that machine translation (MT) is primarily oriented around Internet and technical texts, so content creators must be trained to assure that documents are machine translatable. "The new approach [to MT] uses statistical techniques to identify qualitatively simpler rules" in a rapid, automatic, and scalable manner, says Dillinger. He describes MT systems as mature for industrial applications but immature for use by the general public, and says that people are almost always disillusioned by translation systems because they have unrealistic expectations about their capabilities. Dillinger outlines five steps in the MT process: Document preparation, translation system adaptation, document translation, translation verification, and document distribution. He says MT follows the exact same stages as human translation except in two key respects--translation systems can accommodate a large volume of translated documents, while wording must be handled very carefully because the systems, unlike human translators, do not possess the technical knowledge to perceive erroneous wording and take corrective action. Dillinger dismisses fears that MT systems will drive flesh-and-blood translators out of work, and attests that "MT takes the most routine work out of translators' hands so that they can apply their expertise to more difficult tasks." Variability of vocabulary usage can be a major impediment that commercial MT systems overcome by using the most common words to generate a core system and then adding 5,000 to 10,000 customer-specific words, although Dillinger says this method is not workable for Web applications. The development of systems to direct authoring of Web content is therefore necessary, he reasons.
Click Here to View Full Article
to the top


Information Accountability
Communications of the ACM (06/08) Vol. 51, No. 6, P. 82; Weitzner, Daniel J.; Abelson, Harold; Berners-Lee, Tim

Accountability for the misuse of personal information must be enforced by systems and statutes, as the openness of the information environment makes protection via encryption and access control impossible. "Information accountability means the use of information should be transparent so it is possible to determine whether a particular use is appropriate under a given set of rules and that the system enables individuals and institutions to be held accountable for misuse," write the authors. Rules are needed, both in the United States and internationally, to address the permissible use of certain types of information, in addition to simple access and collection restrictions. The authors say that the information-accountability framework is more reflective of the relationship between the law and human behavior than the various initiatives to enforce policy compliance via access control over information. Supporting information accountability requires a technical architecture that features policy-aware transaction logs, a common framework for representing policy rules, and policy-reasoning tools. "One possible approach to designing accountable systems is to place a series of accountable appliances throughout the system that communicate through Web-based protocols," the authors suggest. The authors conclude that perfect compliance should not be the standard for evaluating laws and systems that aid the enforcement of information accountability. "Rather we should ask how to build systems that encourage compliance and maximize the possibility of accountability for violations," they write.
Click Here to View Full Article - Web Link to Publication Homepage
to the top


Can Machines Be Conscious?
IEEE Spectrum (06/08) Vol. 45, No. 6, P. 55; Koch, Christof; Tononi, Giulio

Some people are convinced that a conscious machine could be constructed within a few decades, including Caltech professor Christof Koch and University of Wisconsin, Madison professor Giulio Tononi, who write that the emergence of an artificially created consciousness may not take the form of the most popular speculations. They note that consciousness requires neither sensory input nor motor output, as exemplified by the phenomenon of dreaming, and emotions are not a necessary component for consciousness, either. Koch and Tononi also cite clinical data to suggest that other traditional elements of consciousness--explicit or working memory, attention, self-reflection, language--may not be essential, while the necessary properties of consciousness depend on the amount of integrated information that an organism or machine can produce. The authors offer the integrated information theory of consciousness as a framework for measuring different neural architectures' effectiveness at generating integrated information and achieving consciousness, and this framework outlines what they describe as "a Turing Test for consciousness." One test would be to ask the machine to concisely describe a scene in a manner that efficiently differentiates the scene's key features from the vast spectrum of other possible scenes. Koch and Tononi suggest that the building of a conscious machine could involve the evolution of an abstracted mammal-like architecture into a conscious entity.
Click Here to View Full Article
to the top


To submit feedback about ACM TechNews, contact: [email protected]

To be removed from future issues of TechNews, please submit your email address where you are receiving Technews alerts, at:
http://optout.acm.org/listserv_index.cfm?ln=technews

To re-subscribe in the future, enter your email address at:
http://signup.acm.org/listserv_index.cfm?ln=technews

As an alternative, log in at myacm.acm.org with your ACM Web Account username and password, and follow the "Listservs" link to unsubscribe or to change the email where we should send future issues.

to the top

News Abstracts © 2008 Information, Inc.


© 2008 ACM, Inc. All rights reserved. ACM Privacy Policy.