ACM TechNews (HTML) Read the TechNews Online at: http://technews.acm.org
ACM TechNews
March 14, 2007

MemberNet
The ACM Professional Development Centre
Unsubscribe

Welcome to the March 14, 2007 edition of ACM TechNews, providing timely information for IT professionals three times a week.


HEADLINES AT A GLANCE:

 

Study Says Computers Give Big Boost to Productivity
New York Times (03/13/07) P. C4; Lohr, Steve

Money spent on computing technology produces higher gains in worker productivity than other investments by a factor of three to five, according to a new study, which also noted that despite this the IT sector is not likely to be a significant source of new jobs. The study, conducted by the Information Technology and Innovation Foundation, found that the power of computing to enhance other areas of the economy trumps the importance of the actual technology itself, and suggests that policies be established to quicken the pace of technology-supported change in business. Health care, electric utilities, and transportation are listed as areas that could most benefit from the increased implementation of computing. "Going forward," according to the report, "it is unlikely that the IT industry will be producing jobs gains out of line with its size. In part this is because productivity in the IT industry has been strong, allowing it to produce more output with fewer workers." However, job gains are expected in those industries that begin using IT effectively to grow themselves, especially in the service sector. Some economists question the report's claim of exceptional productivity gains from investment in computing, claiming the evidence is not sufficient to support the statement and that it is best to promote all investment so all sectors have a chance to flourish. The report is available online at www.itif.org
Click Here to View Full Article
to the top


Where Are the Programmers?
EE Times (03/12/07) Merritt, Rick

At a time when the amount of students pursuing computer science degrees is declining, researchers are struggling with the parallel programming that will run on the next generation of multicore processors. Microsoft is keenly interested in the technology, and unveiled two parallel programming projects at its recent TechFest 2007. "We are at a low point of interest in computer science," says Microsoft Research VP Rick Rashid. "Jobs will go begging in the next few years." In order to meet the needs of programming up to and beyond 32 cores, new approaches will be needed. Microsoft introduced a language that uses a database programming method of joining several processes into what is known as an "atomic transaction," in order to automate the process by which some parallel-software constructs are created. "The way people write parallel programs now, with threads and locks, is intellectually challenging," said Microsoft Research lab manager Roy Levin. "People get it wrong because parts of the program are not in sync or the synchronization is too coarse, leading to poor performance." Although the language could make programming easier, speed will suffer unless chips can be made to avoid collisions in complex transactions, possibly by tracking memory read-and-write operations. Microsoft has also teamed with UC Berkeley to create a chip simulator that tests parallel-programming concepts. Eventually, the company hopes to test the ability of several potential memory architectures to handle transactions developed using the new language. Although progress is being made, Levin admits that more of the problems with homogeneous cores must be resolved before heterogeneous cores can be seriously looked into.
Click Here to View Full Article
to the top


Languages for Supercomputing Get 'Suped' Up
Computerworld (03/12/07) Anthes, Gary

DARPA's High Productivity Computing Systems (HPCS) contractors have come up with languages that aim to make programming easier for multiprocessor computers and clusters. Cray, IBM, and Sun have each released a language that is intended to increase programmer productivity by a factor of 10 or greater while operating at least as efficiently as languages such as Fortran and C; work on a broad range of multicore machines, including business and some mainstream applications; and allow programmers to take advantage of parallelism and avoid chances for coding errors. All three languages are available as open source software. Sun's Fortress language, currently in the alpha stage, is designed for a relatively static environment, so dynamic code-loading or Web accessing would still be done in Java. Cray's Chapel language, also in the alpha stage, will be ideal for machines with low communications overhead and will feature a separation of algorithm specification from machine-dependent structural considerations making it possible for programmers to code and debug algorithms in simple programs before specifying how the data is to be laid out in the machine. IBM's X10, a parallel, distributed, object-oriented language created as an extension of Java, is intended for systems made from multicore symmetric multiprocessor chips. The language uses object orientation in Java for serial code and adds language constructs for parallel and distributed processing. DARPA does not plan to choose a winner from these languages, but recognizes the shift taking place from the realm of sequential processing to that of parallel processing, and that in a few years, a program that cannot handle parallelism will not be able to survive.
Click Here to View Full Article
to the top


IBM Scientists Develop Streaming Video for Visually Impaired
InformationWeek (03/13/07) Jones, K.C.

An IBM researcher who has been blind since the age of 14 has developed a tool that helps the visually impaired access streaming video and animation on the Internet. Screen-reading software and self-talking browsers cannot help those who cannot see buttons on a screen, but the IBM tool uses smart keys or keyboard shortcuts to adjust volume and playback. Users can increase audio speed, since audio from streaming video can seem very slow to the visually impaired. By identifying sound sources, users can single out items such as screen readers. The software also allows for metadata that allows users to hear what is happening on the screen, provided the content creator includes a voice narrative. "The new tool sits on top, so-to-speak, of a normal media player," explains IBM Worldwide Accessibility Center director Frances West. The tool is the first of its kind, and IBM plans to introduce it to the company's Open Source Consortium. IBM has also developed a talking browser and programs that allow the visually impaired to adjust font size and color contrast on Web pages.
Click Here to View Full Article
to the top


Software Developer Growth Slows in North America
IDG News Service (03/13/07) Mullins, Robert

More software developer jobs will be created in Asian than in North America by 2010, predicts Evans Data. The worldwide software developer population is expected to increase from 14.5 million in 2007 to 19.5 million in 2010, but the U.S. is only expected to account for 18 percent of these jobs, down from 23 percent today; the Asian-Pacific (APAC) share of the developer community is expected to make up 45 percent of the developer jobs in 2010, up from 37 percent today. Growth rates for developer jobs over the next three to five years are expected to be 15 percent in APAC, 8 to 10 percent in Europe, the Middle East, and Africa (EMEA), and only 3 or 4 percent in North America. However, the United States is still the strongest software development community in many respects, such as the use of the Ajax programming language. Despite the country having fewer Ajax programmers than China and other APAC countries, "There is a much more sophisticated developer in the U.S. versus the sophistication of the developers in those emerging countries, simply because of experience," says Evans Data President John Andrews. "Ajax is a deep development tool, you can either go shallow or you can go deep. The tendency in North America is to go deeper and leverage its full capabilities." Although some code development and testing jobs are being outsourced, most fundamental software development building new applications takes place largely in the U.S. "A lot of the ideas come from here and a lot of people internationally come to the U.S.," says Krugle VP Laura Merling.
Click Here to View Full Article
to the top


Mobile Web Searches Using Pictures
Technology Review (03/13/07) Greene, Kate

A software prototype from Microsoft, known as Lincoln, allows users to take pictures of objects they see and search for related clips, product information, or other material. "The main thing we want to do is connect real-world objects with the Web using pictures," says Microsoft researcher Larry Zitnick. "[Lincoln] is a way of finding information on the Web using images instead of keywords." The system works best when users take pictures of two-dimensional objects such as DVD covers or movie posters. Users can add pictures and links to the database, meaning the effectiveness of the system would partially rely on its ability to attract users. Although this technology is being used in other projects linking the real world to the virtual world, Microsoft claims that Lincoln is able to search through millions of images quicker than other applications, thanks to an algorithm that creates a signature from a given picture using a small amount of data. The signature describes the relative position of the pixels and the intensity of a certain feature. Data triplets created from three features make the information highly searchable. Once a picture is taken, these data sets are quickly established and compared with those in the database. The whole process of searching using an image takes about 10 seconds, a large part of which is the uploading of an image to the server and downloading the Web page.
Click Here to View Full Article
to the top


Lip Reader Combines Audio, Video
Discovery Channel (03/13/07) Staedter, Tracy

Adding the ability to read lips to computer voice recognition systems could greatly improve their accuracy, especially in "noisy" situations. Since some sounds that are easily confused in the audio domain are easy to identify in the visual domain and some words look identical in the visual domain, combining audio and visual recognition would help fill the gaps in each technique. A group of researchers at the University of East Angia, in England, is beginning a three-year project they hope will produce a camera capable of recognizing simple words and phrases. The technology could be applied to both law enforcement and voice-command systems. To iron out ambiguities in lip-reading, the team will focus on extracting information from the lips using two approaches--modeling the shape and color of the lips as they move and measuring the size of the mouth opening. They will then experiment with techniques for matching visual cues to the words spoken, in order to differentiate between words that are produced by similar lip movements. "The fact is that it works and gives good results," says University of Cambridge technology professor Peter Robinson.
Click Here to View Full Article
to the top


Hackers Get a Bum Rap for Corporate America's Digital Delinquency
University of Washington News and Information (03/12/07) Lewis, Peter

University of Washington communications professor Phil Howard conducted a review of data-breach incidents reported in major U.S. news outlets between 1980 and 2006 and found that organizational flaws in businesses, not hackers, should receive the most blame. "The surprising part is how much of those violations are organizationally prompted--they're not about lone wolf hackers doing their thing with malicious intent," Howard says. His study revealed that malicious intrusions represent only 31 percent of 550 confirmed incidents, while mismanagement, such as missing or stolen hardware, insider abuse or theft, administrative errors, or accidental exposure of data online was responsible for 60 percent of the incidents reported. State laws that require companies to report breaches enabled the study to be done with greater accuracy. "We've actually been able to get a much better snapshot of the spectrum of privacy violations," says Howard. The study also found that while universities make up less than 1 percent of the total records lost, they make up 30 percent of the reported incidents. Corporate America claims that market forces should be allowed to solve the problem of data breaches and reporting them, but Howard believes that this strategy is not sufficient, especially since identity theft is the nation's fastest growing crime. He also believes that states seem more capable of passing laws on the matter than the federal government.
Click Here to View Full Article
to the top


Tech Firms Push to Use TV Airwaves for Internet
Washington Post (03/13/07) P. D1; Babington, Charles

By sending Internet signals through idle TV channels, a coalition of tech companies believes it can offer another high-speed Internet access alternative. Two key considerations for the FCC are the outcome of a multiple-month evaluation of a prototype device built by Microsoft, and whether or not the Internet signal could be kept from leaking outside of designated channels, or "white-space." If the device is approved, it could be in stores by 2009. Both the FCC and the consortium's members, Microsoft, Google, Dell, HP, Intel and Philips, believe the idea could "take the success of Wi-Fi phenomenon to another level," says FCC commissioner Jonathan S. Adelstein. Telecom officials and analysts agree the technology, if successful, would be a threat to existing ISPs and force them to lower prices. However, they also say the available white space might be too limited to have a significant impact. In rural areas where setting up phone or cable lines is costly, the white-space spectrum could be beamed to homes to connect them to an Internet provider; and in urban areas the TV Internet system could be combined with other Internet service to redirect Internet signals throughout a house, while keeping the existing provider. Google is interested in the project for the potential to use new platforms to transmit information, as well as the potential advantage it could give them if Internet providers begin charging Internet companies to carry their content at higher speeds.
Click Here to View Full Article
to the top


Virtual Reality for Virtual Eternity
University of Illinois at Chicago (UIC) (03/12/07)

Researchers at the University of Illinois at Chicago and the University of Central Florida have been awarded a three-year grant to develop a system for creating virtual representations of real people using artificial intelligence and natural language processing software. The goal of the project is to enable people to interact with these virtual representations as if they were the actual person, complete with the ability to understand and answer questions. "The goal is to combine artificial intelligence with the latest advanced graphics and video game-type technology to enable us to create historical archives of people beyond what can be achieved using traditional technologies such as text, audio and video footage," said UIC Electronic Visualization Laboratory (EVL) director and principal researcher Jason Leigh. EVL will construct a motion-capture studio to digitalize the images and movements of people that will be made into avatars. Knowledge will be stored in databases, voices will be analyzed to create life-like synthetic voices, and mannerisms will be analyzed and used to develop 3D avatars. The first human subject to have his virtual representation created will be an NSF program manager. Over a few months, a student will make video and audio recordings of him, including interviews to glean the wealth of institutional information he possesses. The avatar could then be consulted on topics related to the subject's expertise. Leigh expects the increasing power of computers to make these avatars more "naturalistic" in the future, and allow their use to become limitless.
Click Here to View Full Article
to the top


File-Sharing Lawsuit Worries Techies
Investor's Business Daily (03/13/07) P. A4; Deagon, Brian

An upcoming U.S. District Court case will pit the entertainment industry against StreamCast, maker of the Morpheus file-sharing software and co-defendant in the MGM v Grokster case, to determine the measures file-sharing companies must take to comply with copyright laws. In the June 2005 Grokster case, the Supreme Court ruled that manufacturers of file-sharing software could be sued under copyright laws for the illegal music trading of their users, but did not explain what the manufacturers must do to comply with copyright laws. The new case will begin with a March 26 hearing and could result in U.S. District Court Judge Stephen Wilson mandating the type of filtering technology that must be used. Many are concerned about the possibility of a judge, rather than the market, being responsible for such a decision. "Putting courts in the business of redesigning software is a dangerous precedent to set," says Electronic Frontier Foundation attorney Fred von Lohmann. Last September, Wilson ruled in favor of copyright liability, establishing the "inducement doctrine," which states that the vendor of a product designed to infringe copyright is responsible for breaches committed using the product, but StreamCast has made it clear that it will force the courts to tell the company exactly what it must do in order to comply with the law. Other file-sharing companies suggest that StreamCast could use industry standard software filters, but StreamCast claims it could still be sued if the filter did not "exhaustively" stop illegal trading. The company maintains that it needs specific information from the recording industry concerning the artists and songs that must be filtered, but the industry says this information is too valuable to be released. Until a compromise is reached, Wilson will responsible for the decision on how copyrights must be protected.
Click Here to View Full Article
to the top


Animation Tool Puts You in the Frame, or the Game
New Scientist (03/12/07) Simonite, Tom

Researchers from the Max Planck Institute for Computer Science in Saarbrucken, Germany, have developed a 3D animation technique that makes it easier for animators to animate the movements of actors. Faster than traditional methods, the technique would allow animators to scan the motion of one person, and paste the high-resolution scan onto the movements of another person. Usually, a skeleton must be designed for a character to make sure movements impact the outer surface in a realistic manner. "We wanted to be able to go directly from the desired motion to animating your character," says Christian Theobalt, who headed the research effort. The team has created a video of the technique, and says an inexperienced user should be able to produce such animations in less than 15 minutes by using about 60 markers for corresponding points on the scan and the source of motion. The technique could be used to generate 3D video, or even to create an avatar for a computer or virtual world. According to Andy Lomas, who worked on the film "Matrix Revolutions," the approach is better suited for when animators have the exact motion, but not for repositioning. "Their method could be useful for getting those kinds of deformations right if you don't have lots of camera angles to reconstruct a scene in detail," says Lomas, head of computer graphics for U.K. firm Framestore CFC.
Click Here to View Full Article
to the top


The Promise, the Limits, the Beauty of Software
British Computer Society (03/08/07) Booch, Grady

IBM Fellow Grady Booch's Turing Lecture focuses on the important role software plays in everyday life, and stresses that developers should make an effort to increase software's simplicity rather than its complexity. Booch cites C++ designer Bjarne Stroustrup's contention that software is the driving force of civilization--one that influences government, communication, work, leisure, and even sleeping and eating, to which Booch adds that our extensive dependence on software makes software designers among the most important people in the world. The IBM fellow makes the argument that civilization's reliance on software will probably become total "within the next decade," and maintains that although software can augment human intelligence, it cannot supplant human judgment or knowledge itself. He points to a "significant beauty" in software engineering, and calls software intense systems as possibly the most intellectually sophisticated things people have created, despite the fact that most are invisible. Booch calls the best software "simple, elegant and full of drama, manifest in cunning patterns that form their structures and command their behavior," and he recommends that developers commit 10 percent of their time to simplifying software; Booch notes that simpler software raises productivity. The retention of digital data is a source of concern to Booch, who cautions that although the velocity of communication can be boosted via email and other software, our aging digital archives are in such a sorry state that the preservation of history is endangered. He says, "Behind the preservation of classic software there is a need to have a museum of software, not just of old PCs but the software that ran on them. There is a need to establish the source code for Microsoft within this museum for future generations."
Click Here to View Full Article
to the top


Chinese Hackers Seek U.S. Access
USA Today (03/12/07) P. 3B; Swartz, Jon

The recent cyberattack on a U.S. military computer system highlights the weaknesses in Internet security and the Internet's infrastructure. Lt. Cmdr. Doug Gabos with the Navy Cyber Defense Operations Command in Norfolk, Va., said Chinese hackers were probably responsible for the November intrusion that disabled the Naval War College's network and forced it to disconnect from the Internet for several weeks. The hackers were probably looking for information on war games being developed at the naval college. The attack was part of an ongoing campaign by Chinese hackers to penetrate government computers. Chinese hackers primarily use targeted email attacks called "spear phishing" that try to trick the user into thinking the email is from the recipient's organization, but they are also using traditional attacks, such as viruses and worms, in very sophisticated ways. Hackers are exploiting the side doors of private networks that connect to military and government computers as well as trying to break in directly. The Chinese attacks point out the flaws in American cybersecurity, and emphasize the need for the government to develop policies that define responsibilities between the public and private sectors to protect against hackers and cyberterrorists. Part of the problem is that it is difficult to locate the perpetrators of international cyberattacks, and almost impossible to prosecute them. Jody Westby, a cybersecurity consultant in Washington, said there are 243 countries connected to the Internet, an estimated 100 countries planning cyberwarfare capabilities, and a great number of countries that have no cybercrime laws.
Click Here to View Full Article
to the top


Why Google Put a Research Lab in Poland
Christian Science Monitor (03/13/07) Woodard, Colin

Western companies are beginning to notice the value of moving operations to Eastern Europe, where the schools are producing many computer scientists that will work for one-third of the pay of those in Western Europe. The curriculum in many former Soviet-bloc countries is heavily grounded in math, given the former Soviet Union's emphasis on industrial and military production. Last year's TopCoder collegiate challenge drew 21,000 participants from around the globe, but half of the finalists were from former Soviet-bloc countries. IBM, Google, and Motorola have all recently opened research labs in Krakow, Poland. Successful programmers, such as Tomasz Czajka, who won three TopCoder contests between 2004 and 2005, are national icons that provide inspiration for children to pursue computing. Many university computer science departments are concerned that higher-paying western companies and universities could lure away professors, leaving them without the capacity to continue producing computer scientists and engineers. Some small cities that experienced rapid growth in the IT industry have found themselves on the verge of being without programmers, while others have been able to prosper by boosting salaries and collaborating to help local schools increase their capacity to retrain engineers and others. "We said why fight over the same 200 graduates each year?," says Jozef Ondas a CEO of a company in one such area. "Let's invest and create an educational system that can produce 500 specialists each year." Investment in universities and research in important areas will attract students to these areas, says Google's Kannan Pasupathy. "It's a nice circle which ultimately benefits everybody."
Click Here to View Full Article
to the top


Internet Name System in Growing Danger: U.N. Agency
Reuters (03/12/07) Evans, Roberts

The World Intellectual Property Organization (WIPO) is warning that domain-name tasting, the harvesting of expired domains, and other new techniques being used by cybersquatters are threatening to undermine domain dispute procedures that allow patent holders to go after cybersquatters. "Domain names used to be primarily specific identifiers of businesses and other Internet users, but many names nowadays are mere commodities for speculative gain," said Francis Gurry, a top official with WIPO. These speculative tactics could make it virtually impossible for even the biggest companies to protect their name patents on the Internet, as domains have become "moving targets for rights holders," Gurry said. As a result, it would also become harder for consumers to find genuine Web sites. WIPO has overseen some 10,200 domain dispute cases over the last eight years, and the number of cases continues to grow. For example, WIPO heard 1,823 cases in 2006, the most since 2000.
Click Here to View Full Article
to the top


Sign Language at Your Fingertips, Anytime, Anywhere
University of Bristol News (03/06/07)

Mobile phone users will be able to access sign language signs and communicate with deaf people by using a new sign language dictionary that has been developed by the University of Bristol's Center for Deaf Studies. The first of its kind, Mobilesign.org features more than 5,000 British Sign Language signs, and an interface that simply requires the user to type in the word they want to translate or choose from a list of words that are presented in alphabetical order. The video dictionary is a free service. "The need was for a very simple interface to allow easy access and to compress the video files so they would play well on mobile displays and at the same time be small enough to download with virtually no cost," says Chris John, technical director at the Center. Mobile Sign will work with a mobile, on any network, that has a video player, or a 3G phone. "Apart from the obvious use to access signs when you need to meet a deaf person, it will be of great value to students of sign language and to parents--who just need that sign at any time," adds Linda Day, a sign language lecturer at the Center. The staff plans to provide greater support for interaction by building a phrasebook.
Click Here to View Full Article
to the top


Chip Revolution Poses Problems for Programmers
New Scientist (03/10/07) Vol. 193, No. 2594, P. 26; Biever, Celeste

Increasing the density of multiprocessors or cores on a computer chip will facilitate computing performance upgrades, but this sets up the challenge of programming the multicore chips to exploit the gains in performance. What is required is the division of programs into smaller tasks that different cores can execute, and so far the determination of an optimal methodology for this process lacks consensus. "From an industry perspective, this is pretty scary," notes University of California, Berkeley computer scientist and former ACM President David Patterson. Unless a simple, inexpensive multicore chip programming technique is found, innovation within the market will dry up, warns MIT electrical engineer Saman Amarasinghe. Researchers and chipmakers are attempting to avoid this scenario by exploring ways to increase accessibility to programming in parallel through the establishment of programming languages and tools and the redesign of the core configuration. A team led by IBM Research's Vivek Sarkar has developed an open-source language called X10 to address the deadlock bug, which is what happens when a pair of tasks that should be running in parallel are in stalemate because neither can continue without an output from the other. Meanwhile, Stanford University computer scientist Christos Kozyrakis is investigating transactional memory (TM) as a method for simplifying the programming of multicore chips, with an emphasis on solving the "race bug" program. Both X10 and TM allow programmers to particularize tasks that do not generally interfere with each other and so will probably work well in parallel, and then determine when these tasks will need the same fragment of memory and lock those fragments at only the times necessitated by each task. Among the hardware advances being probed as tools to improve the ease of multicore chip programming is enhanced core-to-core communication through chip design modifications.
Click Here to View Full Article - Web Link to Publication Homepage
to the top


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

To unsubscribe from the ACM TechNews Early Alert Service: Please send a separate email to [email protected] with the line

signoff technews

in the body of your message.

Please note that replying directly to this message does not automatically unsubscribe you from the TechNews list.

ACM may have a different email address on file for you, so if you're unable to "unsubscribe" yourself, please direct your request to: technews-request@ acm.org

We will remove your name from the TechNews list on your behalf.

For help with technical problems, including problems with leaving the list, please write to: [email protected]

to the top

News Abstracts © 2007 Information, Inc.


© 2007 ACM, Inc. All rights reserved. ACM Privacy Policy.