knowledge consolidation

[ajschumacher]

How to Make the World Add Up introduces Cochrane and Campbell, two neat collaborations that focus on trying to figure out the truth by looking at multiple studies.

Some similar things, like What Works Clearinghouse, are also neat but have a bad reputation (at least on Wikipedia).

[ajschumacher]

Also from HtMtWAU:

"Daniel Kahneman himself dramatically raised the profile of the issue [of reproducibility/reliability of research] when he wrote an open letter to psychologists in the field warning them of a looming 'train wreck' if they could not improve the credibility of their research." (page 127)

This line of reasoning is a sort of argument for working with foundations rather than the delicate branches at the boundaries of research... Why chase such small effects? Why are we so concerned with effects that are so small? Isn't it better to make sure we have things really in order in the foundations?

[ajschumacher]

"Remarkably often, Meehl found, experts faired poorly when compared with simple checklists. Meehl described his Clinical vs. Statistical Prediction as 'my disturbing little book'."

This connects neatly with Gawande's modern checklist advocacy. It's also a kind of condensed knowledge: automate the things we know how to do into a checklist.

[ajschumacher]

https://twitter.com/planarrowspace/status/1375500733404082183

"By improving test output readability, you automate the understanding of feedback." Great way of thinking about any kind of explanation that helps "automate understanding"... Neat insight from @racheltbaum on page 483 of Software Engineering at Google!

[ajschumacher]

connect with #185

[ajschumacher]

"Individuals die, but the wisdom they have obtained in their lives does not die with them. Mankind keeps all this wisdom, and a person uses the wisdom of those who lived before him. The education of mankind reminds me of the creation of the ancient pyramids, in that everyone who lives puts another stone in the foundation."

From March 26 entry (page 98) of Tolstoy's Calendar of Wisdom. The section is marked "After Giuseppe Mazzini"

This is neat but the metaphor doesn't easily allow corrections/improvements to the foundation...

[ajschumacher]

"I absorbed what I call the fundamental full attribution ethos of hard science. And that was enormously useful to me. Let me explain that ethos.

"Under this ethos, you’ve got to know all the big ideas in all the disciplines less fundamental than your own. You can never make any explanation, which can be made in a more fundamental way, in any other way than the most fundamental way. And you always take with full attribution to the most fundamental ideas that you are required to use. When you’re using physics, you say you’re using physics. When you’re using biology, you say you’re using biology. And so on and so on. I could early on see that that ethos would act as a fine organizing system for my thoughts. And I strongly suspected that it would work really well in the soft sciences as well as the hard sciences, so I just grabbed it and used it all through my life in soft science as well as hard science. That was a very lucky idea for me.

"Let me explain how extreme that ethos is in hard science. There is a constant, one of the fundamental constants in physics, known as Boltzmann’s constant. You probably all know it very well. And the interesting thing about Boltzmann’s constant is that Boltzmann didn’t discover it. So why is Boltzmann’s constant now named for Boltzmann? Well, the answer was that Boltzmann derived that constant from basic physics in a more fundamental way than the poor forgotten fellow who found the constant in the first place in some less fundamental way. The ethos of hard science is so strong in favor of reductionism to the more fundamental body of knowledge that you can wash the discoverer right out of history when somebody else handles his discovery in a more fundamental way. I think that is correct. I think Boltzmann’s constant should be named for Boltzmann."

https://fs.blog/great-talks/academic-economics-charlie-munger/

[ajschumacher]

https://distill.pub/2017/research-debt/ on explaining/unifying disparate articles, ideas, etc.

[ajschumacher]

or: Wells's https://en.wikipedia.org/wiki/The_Outline_of_History or Russell's History of Western Philosophy

[ajschumacher]

see also #253

[ajschumacher]

the idea of building/improving/expanding "Common Knowledge"