Welcome to the February 28, 2022, 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).
To view "Headlines At A Glance," hit the link labeled "Click here to view this online" found at the top of the page in the html version.
The online version now has a button at the top labeled "Show Headlines."
|
|
How Taiwan Used Simple Tech to Help Contain COVID-19
BBC News Erin Hale February 25, 2022
Taiwan has been able to contain COVID-19 through a relatively low-tech tracing system developed by the g0v collective of designers, coders, and activists. When COVID first hit Taiwan, g0v started crowdsourcing solutions to emerging problems like contact tracing and mask rationing, eventually devising a hybrid scheme. The system employs quick response (QR) codes and a corresponding 15-digit code that can be texted freely without a smartphone to the 1922 hotline at Taiwan's Central Epidemic Command Center. Businesses post QR codes that customers must scan whenever they enter, text the 1922 hotline, or fill in a pen-and-paper form in the event of a community outbreak. The QR codes have helped local health authorities retrace an individual's movements when a positive case is detected, while Taiwan's cellphone networks also were tapped to find and contact potentially exposed persons.
|
Robotic Cubes Shapeshift in Outer Space
MIT News Rachel Gordon February 23, 2022
ElectroVoxels developed by scientists at the Massachusetts Institute of Technology (MIT) and Canada's University of Calgary are robot cubes that use embedded electromagnets to shapeshift. Users can manipulate the cubes with a software planner that indicates which cube is to pivot in what direction; the software will compute the sequence and address of electromagnetic assignments to facilitate the action. Up to 1,000 cubes can be controlled with a few clicks, or with predefined scripts that encode multiple, consecutive spins. The researchers are using the ElectroVoxels to test applications in space exploration. The European Space Agency's Dario Izzo said, "ElectroVoxels show how to engineer a fully reconfigurable system, and exposes our scientific community to the challenges that need to be tackled to have a fully functional modular robotic system in orbit."
|
Uber Revamps Driver Pay Algorithm in U.S. Pilot to Attract Drivers
Reuters Tina Bellon February 25, 2022
Mobility service provider Uber Technologies is testing a new driver earnings algorithm in 24 U.S. cities that allows drivers to view pay and destinations before accepting trips. Its goal is to further incentivize drivers to take short rides, in an attempt to attract more drivers. "Gig work is very competitive, not just with Lyft but other platforms, and we think this feature really enhances our platform's competitiveness versus others," said Uber's Dennis Cinelli. The company said data from some cities with upfront pay have indicated 22% average growth in driver earnings for trips in which the distance to the pickup location is longer than the actual trip.
|
Flaws Discovered in Cisco's Network Operating System for Switches
The Hacker News Ravie Lakshmanan February 24, 2022
Technology conglomerate Cisco has issued software patches to correct four security flaws that hackers could exploit to commandeer affected systems. The most critical patch fixes a command injection flaw in the NX-API feature of Cisco NX-OS software, stemming from insufficient input validation of user-supplied data. Cisco warned, "A successful exploit could allow the attacker to execute arbitrary commands with root privileges on the underlying operating system." Other bugs the patches target include two high-severity denial-of-service (DoS) vulnerabilities in NX-OS in the Cisco Fabric Services Over IP and Bidirectional Forwarding Detection traffic functions. The fourth patch corrects a DoS flaw in the Cisco Discovery Protocol service of Cisco FXOS Software and Cisco NX-OS Software, which could "allow an unauthenticated, adjacent attacker to cause the service to restart, resulting in a denial of service condition."
|
Refocusing Dark Matter Search: Supercomputer Simulations Refine Axion Mass
SciTechDaily February 25, 2022
A multi-institutional team of researchers has produced a supercomputer simulation of the generation of axions shortly after the birth of the universe, which could shed new light in the search for dark matter. "We suspect [dark matter] is a new particle we don't know about, and the axion could be that particle," said the University of California, Berkeley's Benjamin Safdi. The team used code developed at the Lawrence Berkeley National Laboratory for an improved simulation technique called adaptive mesh refinement; a three-dimensional grid represents a small section of the expanding universe, over which equations are solved, with more computing power focused on areas of interest. The researchers used Berkeley's Cori supercomputer to run the simulation, and determined the axion's mass to be over twice as large as theorists and experimenters had believed.
|
Lab's Best Friend: 3D-Printed Dog Treat Dispenser Outperforms Predecessors
Nebraska Today Scott Schrage February 24, 2022
The University of Nebraska-Lincoln's Walker Arce and Jeffrey Stevens created a three-dimensionally (3D)-printed automatic dog treat dispenser. The device features a Raspberry Pi mini-computer, allowing it to be mated to a monitor, mouse, and/or keyboard. The dispenser holds up to 59 treats and costs under $200 to build. The researchers found after extensive testing that the device dispatched the correct number of treats in 96% of cases. In one setup, the researchers put a touchscreen on the dispenser, so when a dog nosed it, the Raspberry Pi displayed a selection between two collections of dots, recording which option the dog chose; the preprogrammed system then dispensed the number of treats corresponding with its choice.
|
Perovskite Artificial Retina Can Read Handwritten Numbers
New Atlas Michael Irving February 23, 2022
Researchers at Saudi Arabia's King Abdullah University of Science and Technology (KAUST) have constructed an artificial retina from perovskite materials. Perovskite can absorb light in a manner similar to the human eye, and the researchers embedded perovskite nanocrystals onto a polymer, then sandwiched that layer between aluminum and indium tin oxide electrodes, to admit light and generate a photoreceptor array. The array was fabricated on a flexible polyimide substrate, and attached to a complementary metal-oxide semiconductor (CMOS) sensor and a neural network. When illuminated by variable-hued light-emitting diodes, the artificial retina responded similarly to the human eye, and was especially sensitive to green light. The system also could recognize handwritten numbers with 72% accuracy.
|
People Are Bad at Spotting Fake LinkedIn Profiles Generated by AI
New Scientist Chris Stokel-Walker February 21, 2022
Scientists at the University of Illinois at Urbana-Champaign (UIUC) and Santa Clara University found fake profiles produced by artificial intelligence (AI) on the business social network LinkedIn can easily deceive people. Nearly 300 participants of the study examined three profiles each, two of which contained either a deepfake profile picture or AI-generated text. Participants accepted friend requests from 90% of the deepfake profiles that were consistent, and between 79% and 85% of those with obvious errors. UIUC's Jaron Mink said they apparently did not notice age differences, and were less suspicious of grammatical errors than of image glitches. When informed that deepfake profiles have previously been used to trick people, acceptance levels for inconsistent profiles fell by up to 43%. UIUC's Gang Wang observed that although trained users were better at spotting fake profiles, "The overall stats show they're still not super-good at them."
|
Amazon EcoVistas to Guide Sustainable Dam Siting
The Engineer (U.K.) February 18, 2022
Researchers at the nonprofit Cary Institute of Ecosystem Studies, an organization dedicated to understanding how ecosystems work, have developed a computational tool to guide sustainable hydroelectric dam siting in the Amazon River basin. The researchers developed the novel Amazon EcoVistas framework to analyze proposed dam projects collectively, for energy generation and environmental impact, by analyzing river flow, river connectivity, sediment transport, fish biodiversity, and greenhouse gas emissions. The algorithm was able to ascertain the "Pareto-optimal frontier," the combination of hydropower projects that minimizes environmental harm for any given level of aggregate hydropower output, for 158 current and 351 proposed dams.
|
For Better Immersion in VR: Flying Joysticks
Informationsdienst Wissenschaft February 22, 2022
At Germany's University of Duisburg-Essen, researchers have developed a toolkit for flying control elements to make virtual reality (VR) more immersive. The goal is to make VR controllers look and feel how they appear in virtual space. The researchers produced five common user interface elements from the virtual world—button, knob, joystick, slider, and three-dimensional (3D) mouse—using 3D printers; they attached each device to a quadrocopter drone. When a virtual control element is visible in the virtual world, a flyable with the matching element steers to the person operating it. Although researchers found that the flyables allowed for more realistic interactions with virtual content, the prototypes were not yet on par with modern controllers in terms of input precision.
|
Using ML to Understand How Brain Cells Work
University of Wisconsin-Madison News Charlene N. Rivera-Bonet February 17, 2022
The University of Wisconsin-Madison (UW)'s Daifeng Wang and colleagues used machine learning (ML) and artificial intelligence to better understand how interacting traits influence brain cells' functions. The researchers applied manifold learning to predict neuronal features by aligning gene expression and electrophysiological data for about 3,000 neurons in the mouse brain; both neuronal cell features manifested high values in the same group of cells, but low values in the remainder, and exhibited a relationship to one another that described their manifold shape. The researchers then used cell clusters to unveil connections between electrophysiological features and specific genes governing the expression of other genes. This informed the development of deepManReg, a new manifold learning model that enhances the prediction of neuronal traits based on gene expression and electrophysiology. "Basically, [we can study] how those genes are regulated to affect the electrophysiology or behaviors in diseased cells," Wang said.
|
Computer Security Researchers Aim to Prevent Tech Abuse
Cornell University Chronicle Adam Conner-Simons February 24, 2022
A model developed by Cornell University researchers aims to help domestic abuse survivors prevent assailants from hacking into their devices and social media. With a focus on "continuity of care," the model matches survivors of such abuse with a volunteer consultant who understands their needs and provides a seamless relationship over time, giving them multiple ways to communicate with their consultant safely, and securely storing their tech abuse history and concerns. Cornell's Emily Tseng said, "In an ideal world, the people on the 'Geek Squad' would be able to treat tech abuse with the sensitivity of a social worker."
|
|