Microsoft Manual Deskterity

Microsoft researchers have developed Manual Deskterity, a computer interface that combines touch input with the precision of a pen. The prototype drafting application, designed for the Microsoft Surface tabletop touchscreen, enables users to perform touch actions such as zooming in and out and manipulating images, but they also can use a pen to draw or annotate those images. Manual Deskterity also allows users to touch an image onscreen with one hand while using the pen in the other hand to take notes or perform other actions that pertain to that object. Users need to learn more tricks to use Manual Deskterity, but the natural user interface should ease the learning curve by engaging muscle memory. "This idea that people just walk up with an expectation of how a [natural user interface] should work is a myth," says Microsoft research scientist Ken Hinckley. The researchers also plan to adapt the interface for use with mobile devices. Incorporating only touch input into devices is a mistake, according to Hinckley, who believes that pen and touch interactions can complement each other.

Using Twitter to Predict Event Success

HP Labs researchers have developed a way to use Twitter to gauge real-time interest in movies and accurately predict how they will perform at the box office on opening weekend. HP Labs' Sitaram Asur and Bernardo Huberman developed computational formulas that analyze Twitter feeds and use the rate at which movies are mentioned in Twitter updates to predict the first-weekend returns. The research also showed Twitter could be used to predict other events, such as how major products will be received and the outcomes of elections, according to Huberman. HP Labs studied nearly 3 million Twitter updates that mentioned 24 major movie releases over the course of three months. The researchers factored in the release date and the number of theaters the movie would be shown in, to predict the opening weekend box office performance with 97.3 percent accuracy. They also developed a system that evaluates the sentiments of Twitter updates as positive, negative, or neutral, to predict the following weekend's returns with 94 percent accuracy.

Microsoft OfficeTalk

Microsoft is preparing a small-scale pilot for OfficeTalk, an experimental microblogging service for business users developed by its OfficeLabs researchers. The pilot will allow Microsoft to study how businesses use OfficeTalk, which enables employees to share information in short messages similar to Twitter. "This concept test applies the base capabilities of microblogging to a business environment, enabling employees to post their thoughts, activities, and potentially valuable information to anyone who might be interested," according to a company blog post. OfficeTalk was one of the most popular concepts in OfficeLabs' internal tests. Use of the microblogging service quickly spread across informal networks, offering a unique and efficient collaboration experience, according to the blog post. "OfficeTalk isn't a product--it's a research project focused on learning how people might use social networking tools at work and in what ways both people and organizations realize their value," the blog post says.

Mobile Learning

Peruvian and Belgian researchers have developed an open source mobile learning application that enables health-care workers to connect to the free learning platform Moodle with their iPhone or iPod. The application was tested by health-care workers engaged in 20 clinics throughout Peru. The three-month pilot program used multimedia, three-dimensional animations, group discussions, policy documents, and peer-reviewed literature. The researchers are now finalizing the code before making it available under a Create Commons GNU license. Once the application is completed, the researchers say that institutions, nongovernmental organizations, and companies will be a

Multilingual Search Engine

Universidad Politecnica de Madrid researchers have developed a multilingual search engine that can query a data repository written in Interlingua using questions formulated in any language and provide an answer in that same language. The search engine requires an information base that is written in the Universal Networking Language, the only general-purpose Interlingua. The search engine works by deducing the answer from the question instead of just finding the answer. First, the system searches the text corpus for statements that could contain the answer. Second, it determines which statements actually contain the answer. Finally, it generates the answer in the same language the question was formulated in. The researchers used a biographical encyclopedia to test the system, and reported that 82 percent of the 75 questions they posed were correct.

China & Russia Tops Intl Programming Competition

The top 10 rankings of the 2010 ACM International Collegiate Programming Contest (ACM ICPC) were dominated by five Russian teams and four Chinese teams. In first, second, third, and fourth place were Shanghai Jiaotong University, Moscow State University, National Taiwan University, and Taras Shevchenko Kiev National University, respectively. The University of Warsaw claimed eighth place, making it the only non-Russian or non-Chinese team to make the top 10. ACM President Dame Wendy Hall described the ICPC's global nature as an exceptional instance of the association's recent efforts to extend its technical activities, conferences, and services for the computing profession, and to acknowledge computing achievement in international areas. "By strengthening ACM's ties in multiple regions throughout the world and raising awareness of its many benefits and resources with the public and in-country decision-makers, we can play an active role in the critical technical, educational, and social issues that surround the computing community," she said. Hall also stressed the importance of computer science education in the international economy, citing ACM's initiatives to help high school students, teachers, and parents better comprehend the kinds of careers that studying computer science facilitates.

Digital Doomsday

In the event of a disaster that destroys the vast majority of the world, humanity's legacy will largely reside on data stored on hard drives. However, hard drives were never meant for long-term storage and no one can be sure how long they will last. The Canadian Conservation Institute's (CCI's) Joe Iraci says that although the most important data is backed up on magnetic tapes or optical discs, these formats cannot be trusted to last even five years. Iraci has conducted accelerated aging tests by exposing different forms of media to high heat and humidity. The tests found that the most reliable data storage devices are recordable CDs with a reflective layer of gold and a phthalocyanine dye layer. Many experts believe that after a major catastrophe only information that is written on paper will survive. "Even the worst kind of paper can last more than 100 years," says the CCI's Season Tse. Proposals to make a paper format that can store digital data for centuries using a system similar to bar codes have been slowed due to a lack of commercial interest. Another option is the Rosetta Disk, which holds descriptions and texts of 1,000 languages. The Rosetta Disk is made out of nickel, etched with text that is only readable at 1,000 times magnification. Each disk holds about 30,000 pages of text or images.