I've made surprising observations after eating certain foods.

Examples.

• Keto mudcake
• Burger patties
• Ground meat
• Bolognese
• …

I want to develop a point that you should pay attention to the effects of each meal. Don't be blinded by theory.

I have had periods under a ketogenic diet when I felt absurdly healthy, and periods that felt bad. I'm 90% sure my food selections were largely behind it. But all selections were keto. You can conclude that keto is not the whole answer.

For some reason, when I eat a keto variant of some pastry or cake, I feel similar to as if I had eaten the regular carb-laden variant.

When I broke a fast with burger patties, cheese and bacon, I didn't feel nearly as good as I thought I would. It's much better to break it with whole meat, e.g. slow cooked chuck steak.

I don't understand why I had these effects, but seeking explanations like hidden carbs didn't have much success.

The lesson is maybe obvious, but: even if you don't understand why you feel bad, you can still adapt. Let go of your theories and go back to the bare observations. Refine your personal diet not only from theory but from your experiences.

Computing #emacs #tech

You may be averse to GUIs for Git. Because they don't have feature parity with the command-line version. They cover only a subset of the possible actions, and so at some point you have to "drop into the command line" anyway.

Magit is an exception.

• As of 2020 it's the only porcelain that boasts to be effectively complete, i.e. you never have to "drop into the command line".
• It's discoverable, so it requires you to search the web less for tricks. What you may still search for is the theory behind concepts like "rebase" or "stash", but not the how-to.
• It doesn't have a clunky UI as is typical for GUI apps, where hotkeys only exist for a subset of the things you can click. Instead, it's keyboard-driven, and bashing out a complex git command is usually faster with Magit than the command line.
• It does not overlay its own metaphors like an extra layer of abstraction, it exposes Git's own metaphors.
• For every command you're about to execute, you can reveal what exact Git command-line expression it maps to, in case you're curious.

A very layman understanding: Monte Carlo methods allow you to estimate the size of something – say, estimate the area under a curve – without knowing its true shape. You only need a way of telling whether a given random point is within the area or not. Then you generate lots of random points and see how large a fraction ended up within the area.

It was discovered by Stanislaw Ulam during World War II, and the US Government codenamed it Monte Carlo because it was codenaming everything coming out of Los Alamos just in case.