Like all human endeavors, science is imperfect and often fails us. Yet still it is the best system we have for learning how the world works. And when it does work it is very powerful. Look at for instance the Michelson-Morley experiment to find the ether. When this experiment produced negative results it upended our understanding of electromagnetism and caused a revolution in science. Scientific progress can be slow, but learning from our failures is built into the system.

Imagine if religion had to adhere to the scientific method. All claims need to be peer reviewed and reproducible. Seems like a pretty big difference to me.

Also great is “The Case of the Unexplained” series by Mark Russinovich, author of Sysinternals, where he uses the various Sysinternals tools to solve real problems:

https://learn.microsoft.com/en-us/sysinternals/resources/webcasts

Is that more than the kernel? How much more?

Give me some stone knives and bear skins and I could construct a mnemonic memory circuit. ;-)

https://www.youtube.com/watch?v=F226oWBHvvI

Einstein was not sure if the universe was expanding, contracting or static. He famously had a constant he could use to change this:

"…Einstein’s cosmological constant, is a coefficient that Albert Einstein initially added to his field equations of general relativity.

Einstein introduced the constant in 1917 to counterbalance the effect of gravity and achieve a static universe, which was then assumed. Einstein’s cosmological constant was abandoned after Edwin Hubble confirmed that the universe was expanding."

https://en.wikipedia.org/wiki/Cosmological_constant

Ok I see a lot if discussion on this topic but no one seems to have mentioned the main feature of the spec that makes them phishing resistant: presence detection. This is what makes FIDO resistant to credential replay. The spec is not perfect but it prevents most common phishing attacks.

It is a little slow at times, but the fun bits are really awesome. And you can get a linux package to display system info as !Krells