The question is which parts of the spectrum can the FCC auction off with minimal negative impact? If NASA, NOAA and the Commerce Department are correct, it is not the weather data communications segment. The Washington Post obtained a letter to the FCC at the end of February in which NASA Administrator Jim Bridenstine and Commerce Secretary Wilbur Ross urged the FCC to remove this auction proposal from their website. In their letter, they stated the proposal “would have a significant negative impact on the transmission of critical Earth science data — an American taxpayer investment spanning decades and billions of dollars with data supporting public safety, natural disaster and weather forecasting.”
The letter went on to ask the FCC to take down the proposal “immediately” before a meeting scheduled for next Monday in light of the purpose of the meeting with NASA being “to continue the long-standing interagency reconciliation process on this important topic.”
Eos; Vanessa Lucieer, Craig Johnson, and Neville Barrett
from
Seamap Australia integrates seafloor maps with information on plant and animal habitats, environmental stressors, and resource management to create a first-of-its-kind resource.
Last year, F5 Networks found that 63 percent of its BIG-IP customers’ virtual servers were HTTP/S. According to F5’s State of Application Services Report 2019, the percentage of survey respondents deploying HTTP/2 gateways has increased 27 percent in 2017 to 47 percent 2019, with another 25 percent expecting to use this service within the next 12 months. With its huge installed base of servers and advanced capabilities, it is no wonder F5 Networks sought to acquire NGINX as a way to bolster this area of growing demand.
According to data from Netcraft and Web Technology Surveys, Ngnix and Apache web servers are used on approximately the same percentage of computers and lead in adoption worldwide. By itself, Ngnix is a mature technology, but its use as a reverse proxy puts it at the center of microservice architectures.
How lonely are people today? That’s what Duke University’s Center for Advanced Hindsight hopes to discover, and if isolation and digital media are part of the problem or can be part of the solution.
The study, which will run through the Wisdo app, a social connection tool that links people together based on shared experiences, and lets users chat one to one, or take part in online groups. Duke University will be able to tap into Wisdo’s user base, asking them about different experiences in their life from divorce to parenting, depression to finding your dream job.
Incoming and current faculty members at the School of Engineering and Applied Sciences are now required to report their involvement in promoting diversity and inclusion, which could then affect the assignment of faculty bonuses, Dean of SEAS Francis J. Doyle III said in a February interview.
Based on an ever increasing amount of data and powerful novel computer architectures, learning algorithms appear to reach human capabilities, sometimes even excelling beyond. The issue: so far it often remains unknown to users, how exactly AI systems reach their conclusions. Therefore it may often remain unclear, whether the AI’s decision making behavior is truly ‘intelligent’ or whether the procedures are just averagely successful.
Researchers from TU Berlin, Fraunhofer Heinrich Hertz Institute HHI and Singapore University of Technology and Design (SUTD) have tackled this question and have provided a glimpse into the diverse “intelligence” spectrum observed in current AI systems, specifically analyzing these AI systems with a novel technology that allows automatized analysis and quantification.
San Jose Mercury News, Bay Area News Group, Ethan Baron
from
Amid a worldwide race for supremacy in artificial intelligence, Stanford University on Monday will unveil a new institute dedicated to using AI to build the best-possible future.
The Stanford Institute for Human-Centered Artificial Intelligence is co-directed by Fei-Fei Li, a former chief scientist for AI at Google, now a Stanford computer science professor.
The job market for U.S. science and engineering Ph.D.s is about to pass a long-anticipated milestone. For decades, educational institutions have been the largest employer of Ph.D.s. In 1997, for instance, they eclipsed private sector employment by 11 percentage points, according to the U.S. National Science Foundation’s (NSF’s) biennial Survey of Doctorate Recipients. But the academic job market has not kept pace with the supply of graduates, and the equivalent data for 2017—released last month—reveals a very different picture: For the first time, private sector employment (42%) is now nearly on par with educational institutions (43%).
The trend is particularly striking in the life and health sciences.
A spokesperson for Google said the company would never ask someone to write an anonymous piece. “We are transparent about our policy work—we disclose all significant affiliations and grants on our transparency page and we require all people whose work or research we fund to disclose that fact,” the spokesperson said in a statement.
Still, the email offers insight into how Google, a shrewd Washington player, has shifted into overdrive and adapted its approach as calls to regulate Big Tech have grown louder.
In the latest sign of the more hostile environment for tech, Democratic presidential candidate Elizabeth Warren last week issued a far-reaching plan to break up dominant tech platforms; she suggested Google be required to unwind its acquisitions of Waze, Nest, and DoubleClick.
A month ago Donald Trump issued an executive order designed to maintain America’s greatness on artificial intelligence. I reviewed it unfavorably for its lack of specifics, absence of funding details, and preponderance of management-consultant fluff.
About a month later the semiconductor giant Intel offered its own A.I. plan, a 13-page whitepaper of its own recommendations for a U.S. strategy on A.I. Intel sites the work of many management consultants, and it uses (in an accompanying fact sheet) some interestingly diplomatic language, like grouping together China, India, Japan, and the European Union as “global neighbors” to the U.S. (One president’s adversaries are a multinational corporation’s neighbors.)
On balance, though, the Intel proposal is a cut above the White House’s. It also refrains from offering specific dollar values, but Intel’s report does specify where money should be spent.
Fiber-optic engineers have set a new record for data-transmission efficiency in an experiment on a submarine cable with the highest data capacity of any yet in operation. Able to carry up to 20 terabits per second on each of eight fiber pairs, the MAREA cable runs 6605 kilometers between Virginia Beach, Virginia, and Bilboa, Spain. At the Optical Fiber Communications Conference cosponsored by IEEE—which ran from 3–7 March in San Diego—engineers from Infinera and Facebook reported that new transmitting and receiving equipment could squeeze 26.2 terabits per second through the same bandwidth on MAREA.
Trends in Cognitive Science; Jan R. Wessel, Krzysztof J. Gorgolewski, Pierre Belle
from
A laboratory’s programming language has wide-ranging implications. As demands towards scientific programming change and languages evolve, investigators may look to change their existing software stack. Following up on a recent online debate, we discuss key considerations and challenges in choosing and changing languages and suggest solutions for investigators looking to switch.
Recently, J.R.W. was faced with a dilemma: he was considering switching his laboratory’s existing software infrastructure from MATLAB, a proprietary scientific programming language, to another, cost-free ecosystem. Looking for advice on some anticipated challenges, he publicly solicited input from the scientific community on a popular social media platform (https://twitter.com/Wessel_Lab/status/1031208646263418880). An intense conversation ensued, illustrating many of the challenges and complexities associated with switching between scientific software. After being approached by the editor of this journal to summarize that conversation, J.R.W. solicited help from K.J.G., the lead developer of several open-source scientific software packages, and P.B., an open-source contributor who recently transitioned his laboratory from MATLAB to Python. Together, they assembled a series of concrete questions and recommendations around three key considerations: how do you decide on the most appropriate programming language, what challenges are associated with switching, and what solutions exist to ease the transition (Figure 1)?
New York, NY April 11-12. “This workshop uses a tabular dataset and teaches data cleaning, management, analysis and visualization. There are no pre-requisites, and the materials assume no prior knowledge about the tools.” [registration opens March 21]
Brussels, Belgium April 4, starting at 9 a.m., Press Club Brussels (Rue Froissart 9). “A discussion that will take stock of the progress achieved so far across member states; compare targets, priorities, and dimensions; and assess the extent to which these national strategies will support Europe’s goal of becoming a global leader in AI.” [registration required]
Detroit, MI September 27-30, David Whitney Building & Cranbrook Academy of Art. “For Sketching 2019, we will of course examine the changes in the design and use of tools for creative expression and design with digital hardware. We also aim to explore how the hardware tools used by designers, artists, educators and developers increasingly straddle–and erase–the line between what is and is not real, and when the distinction matters.” [save the date]
Atlanta, GA April 17, starting at 12:15 p.m., Georgia Institute of Technology Marcus Nanotechnology Building, Rooms 1116-1118. [free, registration required]
“In 2006, the European Association for Decision Making (www.eadm.eu) was pleased to announce the creation of the Jane Beattie Scientific Recognition Award to honor the memory of our late colleague, Jane Beattie. The award is to be made every two years and is intended for researchers who have recently completed the first stages of their careers – defined operationally by those who are 5 to 10 years post-PhD.” Deadline for applications is March 31.
“The National Socio-Environmental Synthesis Center (SESYNC) requests proposals for collaborative and interdisciplinary team-based research projects under two programs: Pursuits and Workshops.” Deadline for submissions is May 15.
One of the core problems in creating a data culture that aligns with your mission is the history we are fighting against. Data has been a tool for those in power to consolidate that power for centuries. For organizations working in the social good sector, this should feel problematic! If you’re deploying some tool or process, you need to be wary of any pieces that reinforce that history. They can cultivate the opposite of the empowerment, engagement, and ownership goals that are probably at the heart of your mission.
2018 was quite a year: Metadata 2020 was on the agenda at 17 conferences, workshops and meetings; community groups formed into six cross-stakeholder projects; the projects got to worked to better understand metadata challenges; and we held two end-of-year-one in-person workshops with 50 people in New York and London to share things out and discuss the 2019 agenda. Over 200 people participated in some way, whether through signing up for the mailing list, chatting on Slack, or attending online meetings and webinars. Our six project groups discussed and conducted research to learn more about how we all use metadata, and how we can make metadata more effective and efficient for these uses.
In this post, I’ll discuss some of the more advanced techniques for creating and manipulating Core ML models. Using an artistic style transfer model as an example,