Read: March 2022
Inspiration: Referenced in “Thinking in Bets” by Annie Duke
Written with the help of ChatGPT, below is a brief summary to understand what is covered in the book.
“The Half-Life of Facts”, published in 2004 by author, scientist, and investor Samuel Arbesman, discusses the concept of the “half-life of facts,” or the idea that the validity of certain facts tends to decline over time as new information becomes available. The book explores the ways in which our understanding of various fields of knowledge changes over time and the implications of this for how we think about and use information. Arbesman argues that in a world where the rate of technological and scientific progress is accelerating, it is important for individuals to be aware of the changing nature of knowledge and to be open to new ideas and perspectives. He also discusses the role of technology in the dissemination of information and the importance of critical thinking in evaluating the reliability of sources. The Half-Life of Facts offers a thought-provoking exploration of the evolving nature of knowledge and its implications for how we think and learn.
Direct from my original book log, below are my unedited notes (abbreviations and misspellings included) to show how I take notes as I read.
Scientometrics is the quantitative study of science, how science/knowledge grows—exponential relationship measured in doubling time (varies by subject area), computers struggle with blurred words—are u a robot blurry word test on sites helps NYT digitize archives, Lazarus taxa: living things that are presumed extinct until contrary evidence discovered, scientometrics founded by derek de solla price, half life of facts best studied in libraries in 1970 as worry about capacity and which books can discard—humanities have longest half life, compsci faster than psychiatry, Moore’s law introduced 1965 based on 4 data points but hold true, logistic curve is a math function explaining how something can quickly begin to grow exponentially only to reach carrying capacity, science is about what we know about the world while tech is about what we can do, science and tech intertwined, iron more magnetic over as tech help purify, population growth and innovation go hand in hand, info shared first is sticky even if incorrect, info spreads through networks of strong and weak connections (weak build bridges but often too weak to build trust/stick so need medium too as true bridge), hidden public knowledge exists when paper 1 says A leads to B and paper 2 says B leads to C but no one read both to see A lead to C, iphone was a phase transition—one state of awareness to another (can feel like from future, used in science when material melted or change states etc, Ising Model refers to mathematical point after slow accumulation where phase changes (“jumps”), Mount Everest named after George Everest scientist led British Empires Great Trigonometric Survey of India once controlled continent in 1800–successor measured height of Himalayas and name after Everest, p value is percent due to chance (so published result could be), publication bias = only success gets published but not if same test had failed so no one aware, leads to decline effect once published and replicated and realize effect is less or not at all, can have stat sig result that is not true, replication tough to incentivize b/c not get clout if confirm others’ result but replication key to science, as age we learn more and more about less and less, language unique as lacking objective truth—evolves over time as enough adopt , idiolect=one’s own language/way of speaking, unique, errors lead us closer to facts, currently in phase of exponential oveturning of facts and shortening of half lives—usually generational overturning as parents learn what kids learn (don’t relearn til then)