recentpopularlog in


« earlier   
Data — from objects to assets - Nature Oct 2019
"How did data get so big? Through political, social and economic interests, shows Sabina Leonelli ..."

"Over the past 150 years, ideas have shifted drastically as to what counts as data, which data are reliable and who owns them. . . This explicit designation as assets has become possible only through a complex web of institutional, technological and economic developments."

Until the start of the 19th C: "... the myth of the heroic theoretician, mining order from the chaos of observations. This individualistic view was tied to an understanding of data as fundamentally private — their scientific value residing in conceptual interpretation."

"The nineteenth century marked a shift. Data, as we now recognize them, became institutionalized as social commodities."

"From the 1970s, almost every scientific field was building global, digitalized infrastructures for data sharing. . . . Increasingly, data were seen as sharable assets for repurposing, the value of which could change depending on their use.

". . . These reforms are temporary solutions to a large-scale crisis of the contemporary research system, rooted in the inability to reconcile the diverse social and scientific aspects of data."

e.g. issues of ownership

"As algorithms become ever more opaque, the transparency and accountability of techniques and tools used to interpret data are declining. "

". . . history shows that moving research data around is not so simple. Underpinning technical questions around integration and use are thorny social, ethical and semantic issues."
NatureJournal  data  research  history  BigData  * 
2 days ago by pierredv
How science has shifted our sense of identity - Nature Oct 2019, Nathaniel Comfort
I want to suggest that many of the worst chapters of this history result from scientism: the ideology that science is the only valid way to understand the world and solve social problems. Where science has often expanded and liberated our sense of self, scientism has constrained it.
NatureJournal  opinion 
6 days ago by pierredv
Make more digital twins - Nature comment, Sep 2019
"Virtual models boost smart manufacturing by simulating decisions and optimization, from design to operations, explain Fei Tao and Qinglin Qi. "

"There is much to be done to realize the potential of digital twins. . . . Here we set out the main problems and call for closer collaboration between industry and academia to solve them."
1. Data difficulties
2. Model challenges
3. Twin teams

Four bridges
"The following steps would make research and development of digital twins more coherent."
1. Unify data and model standards.
2. Share data and models.
3. Innovate on services.
4. Establish forums.
mirror-worlds  digital-twins  NatureJournal  opinion 
22 days ago by pierredv
Global 5G wireless networks threaten weather forecasts - Nature, Apr 2019
The US government has begun auctioning off blocks of wireless radio frequencies to be used for the next-generation mobile communications network known as 5G. But some of these frequencies lie close to those that satellites use for crucial Earth observations — and meteorologists are worried that 5G transmissions from cellphones and other equipment could interfere with their data collection.
5G  NatureJournal  Interference  meteorology  NASA  NOAA 
may 2019 by pierredv

Copy this bookmark:

to read