Hacker News Reader

Comment by IKantRead | Hacker News Reader

I've always found it fascinating that geophysicist and earlier advocate for Bayesian methods, Sir Harold Jeffreys, didn't believe in continental drift and plate tectonics because he felt there was no known source of energy on the Earth massive enough to explain this movement. [0]

He remained an opponent until death (at which point continental drift was widely accepted) which is both a testament to the literally unbelievable energy behind seismic activity and the importance of updating your Bayesian priors as you gain new information.

0. https://en.wikipedia.org/wiki/Harold_Jeffreys#Opposition_to_...

View on HN

Another great illustration of Planck’s law:

> Max Planck, surveying his own career in his Scientific Autobiography, sadly remarked that “a new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.“ -The Structure of Scientific Revolutions

This is actually such a great insight. That's why we also need new generations of politicians and management every so often to keep the wheel of progress rolling. And, incidentally, why I think that developing anti-aging technology is not a good idea.

I wonder if there's something physical to do with aging that makes people cognitively inflexible. Anti-aging tech would be great if you could prevent that negative aspect of aging too.

Actually the lesson I take away is the importance of not having a 0% (or 100%) prior, since it leaves no room to update with new information.

It is also suspicious because it means someone has assigned a 0% prior to their being insane or in some sort of Plato's-cave scenario - which is hard to justify.

The minimum possible Baysian prior is a base rate of "my senses are just not picking up reality and/or my memory is catastrophically compromised and/or I cannot process logic right now due to some reason" which while low is never going to be 0%. There are too many known ways for human brains to fall over. 0% priors are unjustifiable.

I've always found it fascinating that the guy who came up with the theory (Wegener) died in on the ice sheet in Greenland while attempting to resupply a research station. Having spent some time on the Greenland Ice Sheet at Summit Station, life on the Greenland ice sheet is much more cushy nowadays...

Reply on news.ycombinator.com