Association for Computing Machinery
Welcome to the July 26, 2013 edition of ACM TechNews, providing timely information for IT professionals three times a week.

ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).

HEADLINES AT A GLANCE


President's Council of Advisors on Science and Technology Discusses Big Data as it Relates to Smart Cities
CCC Blog (07/25/13) Ann Drobnis

Big data was on the agenda of a recent meeting of the U.S. President's Council of Advisors on Science and Technology (PCAST). The conversation focused on big data as it relates to smart cities, and PCAST's United Kingdom counterpart, the Council for Science and Technology, participated in the discussion. Steven Koonin, founding director of New York University's Center for Urban Science and Progress, described an effort at the university to use New York City as both a laboratory and classroom. He discussed the emerging field of urban informatics and provided suggestions for a national program. Koonin also called for encouraging data sharing across government functions and with the private sector, defining data standards, furthering privacy research and regulation, and funding. In addition, he said partnerships must be formed, cross-disciplinary training in undergraduate and graduate programs must occur, and urban informatics research needs a "home." University College London professor Sir Alan Wilson addressed the Future of Cities Project, and said he hopes to learn what makes a successful city as he works to develop theories that would benefit all cities.


A Summer of Data Hacking Social Problems
New York Times (07/25/13) Steve Lohr

Rayid Ghani, chief scientist for President Barack Obama’s re-election campaign, recently created the Data Science for Social Good fellowship program at the University of Chicago to give young data scientists an opportunity to address real-world social problems. The program was financed by the Schmidt Family Foundation, an educational charity led by Google executive chairman Eric Schmidt and his wife, Wendy. "This is a small way to help that might have a bigger impact than we think," says Schmidt, who notes that urban social problems are a fresh challenge for sophisticated data tools. "Almost all of the interesting jobs will be related to big data in the future," he says. Ghani says the 36 summer fellows, chosen from 550 applicants, are working on complex urban and social problems. One project is helping the Cook County Land Bank develop a Netflix-style recommendation engine to identify the most promising properties for redevelopment, mostly in blighted neighborhoods on Chicago's South and West Sides. Another project involves developing quantitative tools for measuring the national effectiveness of the Nurse-Family Partnership, a community health program that is currently active in 42 states. "Social problems are an area that is under-served by computer science," says Data Science for Social Good fellow Skyler Whorton.


Six Months of Computing Time Generates Detailed Portrait of Cloth Behavior for Video Games
Carnegie Mellon News (PA) (07/23/13) Byron Spice

Researchers from Carnegie Mellon University (CMU) and the University of California, Berkeley have developed a data-driven technique that could improve real-time animation of complex phenomena. The researchers say they simulated almost every important way a piece of cloth might shift, fold, and drape over a moving human figure, after six months of computation. Taking advantage of the computing power of the cloud, the team used 4,554 central processing unit hours to generate 33 gigabytes of data. The researchers developed an iterative technique that continuously samples the cloth motions, automatically detecting areas where data was lacking or where errors occurred. CMU professor Adrien Treuille says their approach is a new paradigm for computer graphics. "The criticism of data-driven techniques has always been that you can't pre-compute everything," Treuille says. "Well, that may have been true 10 years ago, but that's not the way the world is anymore." He notes that it will be possible to provide real-time simulation for almost any complex phenomenon. "I believe our approach generates the most beautiful and realistic cloth of any real-time technique," Treuille says.


Twitter Can Tell Whether Your Community Is Happy or Not
Atlantic Cities (07/22/13) Emily Badger

Researchers at the University of Pennsylvania and Michigan State University recently released a study on whether a community's sense of happiness could be determined from communications on Twitter. The study examined 82 million tweets from about 1,300 U.S. counties between June 2009 and March 2010, with a minimum of 30,000 words geotagged to each county. The study demonstrates that Twitter reveals the level of well-being at the community as well as the individual level. The study used a model of language created by tweets that were indicative of community-level well-being, measured against more traditional survey results. By combining socioeconomic data with their Twitter language model, the researchers say they developed a highly effective tool for predicting well-being, without using formal surveys. Tweets about exercise and the outdoors correlated positively with happiness, possibly due to exercise lowering depression risk, the researchers say. In addition, tweets about problem-solving and engagement in activities are linked to well-being, while tweets about stress and boredom are tied to a low-level of well-being.


Quantifying Cities' Emotional Effects
MIT News (07/24/13) Larry Hardesty

Massachusetts Institute of Technology (MIT) Media Lab researchers have created an online tool designed to facilitate the study of the impact that visual environments have on people living in cities. Harvard University researchers in the 1980s put forth the "broken-windows theory," which states that visible signs of neglect, such as broken windows, actually increase crime. Although many support the theory, it has come under recent criticism and is difficult to quantify because of the subjectivity of visual disorder. The MIT team showed online volunteers sets of images pulled at random from street-level Google Maps photographs, and volunteers chose the image that fit named attributes. Each attribute then received a geographical area score based on an algorithmic computation of the volunteer's comparisons. The researchers chose "safe," "upper-class," and "unique" as the three attributes for their experiment. Greater disparity between class and safety scores appeared in the study's U.S. cities of New York and Boston than for the Austrian cities of Linz and Salzburg. "By providing a very simple, intuitive-to-use tool, we gain, gather, and undertake sophisticated analysis," says Ohio State University professor Jennifer Cowley. "We can identify all of the possible variables that influence perception and test these relationships."


Perfecting Digital Imaging
Harvard University (07/22/13) Manny Morone

Harvard University researchers say they are developing computer graphics tools that narrow the gap between "virtual" and "real." One project tries to find better ways to mimic the appearance of a translucent object. The project examines how humans perceive and recognize real objects and how software can exploit the details of that process to make the most realistic computer-rendered images possible. The project's approach focuses on translucent materials’ phase function, part of a mathematical description of how light refracts or reflects inside an object, which determines how people see it. The researchers first rendered thousands of computer-generated images of one object with different computer-simulated phase functions. A program then compared each image’s pixel colors and brightness to another image in the space and determined how different the two images were. "This study, aiming to understand the appearance space of phase functions, is the tip of the iceberg for building computer-vision systems that can recognize materials," says Harvard researcher Todd Zickler. Another study investigates a type of screen hardware that displays different images when lit or viewed from different directions. The research demonstrates that interference effects can be exploited to control reflection from a screen at micron scales using well-known photolithographic techniques. A third project addressed the problem of color grading in digital film editing.


Google Searches Mined to Uncover Our True Opinions
New Scientist (07/25/13) Lisa Grossman

Google Trends is a Web tool designed for tracking what terms Google users are searching for in a given period. The anonymized, aggregated data is compiled from 100 billion searches every month, and can be broken down by geographic region. Researchers now are turning the Google lens on more difficult issues, some of which are invisible to conventional polls, says economist Seth Stephens-Davidowitz. For example, Stephens-Davidowitz analyzed Google searches and voting patterns in the United States to measure the extent to which racism hurt Barack Obama in the 2008 presidential election. He found that the states with the highest rates of racist-themed searches were ones in which Obama underperformed in the election, and concluded that racism alone cost Obama three to five per cent of the total vote. "This is about twice as high as is found by most surveys, presumably because many people do not want to admit this motivation," Stephens-Davidowitz says. He also used Google to test the theory that child abuse rates went down during the recent recession, a trend hinted at by declines in incidents reported to the authorities. However, searches for terms related to child abuse from concerned adults actually went up in regions where unemployment was higher or where social services budgets had been cut.


Lifelike Wobble Thanks to ETH Zurich Software
ETH Zurich (07/25/13) Angelika Jacobs

ETH Zurich researchers say they have developed software that makes the animation artist's job easier by proposing additional movements, which the artist can easily adapt afterward. Conventional animation programs that automatically add effects, such as the wobbling of a stomach, do not use the typical animator’s toolbox, known as the rig space, which is the sum of the levers the artists use to move a specific body part and consists of many individual dots. The researchers' program uses the levers in the rig space to mimic physical effects and make the wobbling look as realistic as possible. The program also enables artists to make changes at a later stage using the rig space levers to move entire body parts instead of shifting individual dots. The program calculates how it has to move the levers correctly much more quickly by incorporating the option of anticipating the movements for the next images in the animation sequences into the software. Therefore, the program does not test all the degrees of freedom of all levers for each of the 24 frames per second, but instead assumes that the movements calculated for one frame are also suitable for the next couple of frames.


Self-Driving Cars Could Create 1GB of Data a Second
Computerworld (07/23/13) Lucas Mearian

Self-driving cars will come with a wide range of sensors, creating machine-to-machine data at the rate of 1GB per second, according to BigData-Startups.com founder Mark van Rijmenam. He also notes the sensors will provide greater opportunities to spot mechanical problems before they happen. "With the amount of cars worldwide to surpass one billion, it is almost unimaginable how much data will be created when Google's self-driving car will become common on the streets," Rijmenam says. By 2020, there will be several autonomous vehicle offerings that consumers can buy at a reasonable price point, according to Gartner analyst Thilo Koslowski. If the 1GB per second prediction is accurate, self-driving cars would, on average, create about two petabytes of data per year. Future cars will have more infrared sensors, inexpensive video cameras, and laser-based radar to detect objects around them, according to Koslowski. "The self-driving car from Google already is a true data creator," Rijmenam says. "It uses all that data to know where to drive and how fast to drive. It can even detect a new cigarette butt thrown on the ground and it then knows that a person might appear all of a sudden from behind a corner or car."


The Eco-Race to Beat Congestion
Queensland University of Technology (07/22/13) Kate Haggman

Queensland University of Technology (QUT) researchers will use the school's interactive digital display space, known as the Cube, to collect data on eco-driving behavior. Working with Japan's National Institute of Informatics (NII), QUT has launched an eco-friendly car racing game on the Cube that will enable its researchers to better understand how people behave as eco-drivers. Eco-friendly driving involves common-sense actions such as accelerating more slowly, avoiding hard breaking, and timing the approach to traffic lights to stay above a certain speed. Eco-friendly driving is predicted to become more common as oil becomes scarcer and more expensive. NII's Helmut Prendinger built the game iCO2, which tests active decisions to conserve fuel. Prendinger says the game is a sophisticated three-dimensional model of part of Tokyo's road network. "ICO2 is unique in that it allows multiple users to share the same space and be aware of each other, which encourages social behaviors such as competition and collaboration," he says. QUT researchers will use the data to design the traffic infrastructure for the new eco-driving style.


The MOOC That Roared
Slate (07/23/13) Gabriel Kahn

Georgia Institute of Technology has announced plans to offer a low-cost version of its computer science master’s degree through massive open online courses (MOOCs), in a move that could transform the higher education landscape. The MOOC-based program will cost $6,600, while the on-campus computer science master’s degree costs about $45,000 in tuition for out-of-state students and $21,000 for Georgia residents. Top-tier universities have made courses available through MOOCs, but have been reluctant to offer diplomas for the programs. In addition, many universities offer online degree programs, but they are priced at the same level. Georgia Tech's program, offered in conjunction with Udacity, is intended to be as reputable as the university's traditional computer science degree program, which U.S News and World Report ranks among the nation's top 10. The program is likely to pressure universities, especially mid-tier schools, to reduce prices to compete. Zvi Galil, the head of Georgia Tech's school of computing, acknowledges that the computer science MOOC is "uncharted territory." However, he says if Georgia Tech does not make this move, another school will. "There is a revolution," Galil says. "I want to lead it, not follow it."


Desktop Printing at the Nano Level
Northwestern University Newscenter (07/19/13) Erin White

Northwestern University researchers say they have developed a low-cost, high-resolution tool that could revolutionize how nanotechnology is produced from the desktop. "With this breakthrough, we can construct very high-quality materials and devices, such as processing semiconductors over large areas, and we can do it with an instrument slightly larger than a printer," says Northwestern professor Chad A. Mirkin. The tool produces working devices and structures at the nanoscale level in a matter of hours, right at the point of use. The researchers say the tool can prototype a diverse range of functional structures, from gene chips to protein arrays to making electronic circuits. "Instead of needing to have access to millions of dollars, in some cases billions of dollars of instrumentation, you can begin to build devices that normally require that type of instrumentation right at the point of use," Mirkin says. The tool also enables researchers to rapidly process substrates coated with photosensitive materials called "resists" and generate structures that span the macro-, micro- and nanoscales, all in one experiment. "You just assign the beams of light to go in different places and tell the pens what pattern you want generated," Mirkin says.


Ajay Bhatt: Intel’s Rock-Star Inventor
IEEE Spectrum (07/23/13) Steven Cherry

IEEE Spectrum's Steven Cherry recently interviewed Intel engineer Ajay Bhatt, who invented the universal serial bus (USB) port about 25 years ago. Bhatt went on to develop USB 2.0 and 3.0, and to work on the Accelerated Graphics Port, PCI Express, and Intel’s desktop power-management architecture. Bhatt currently is analyzing PC architecture to provide portable computers with the power efficiency required to last all day. Over the past four years, Bhatt has "audited each part of the architecture and systematically gone in and tried to fix the issues in the system that would result in draining the power unnecessarily," he says. The goal is to adjust power consumption based on a user's tasks so that highly demanding applications receive the required level of processing power, while simpler tasks such as word processing are more power efficient. Interoperability is a consideration, because computers include parts from a variety of vendors, which must agree on a common set of power privileges. Bhatt says Intel is highly successful at bringing research and development ideas to market because of its investment in Intel Labs, partnerships with academia, and long-term view of technologies that will be needed in the future.


Abstract News © Copyright 2013 INFORMATION, INC.
Powered by Information, Inc.


To submit feedback about ACM TechNews, contact: [email protected]
Current ACM Members: Unsubscribe/Change your email subscription by logging in at myACM.
Non-Members: Unsubscribe