There’s another factor - days where thr earth is orbiting faster, eg on the closer side of the ellipse - are a different length midday to midday from when we are on the far side of the ellipse.

You can convince yourself of this when you consider that the area of the arc we traverse each day is the same (Kepler’s law). On the short side of our eliptical orbit, since the orbital distance is shorter, the arc must have a larger angle that we travel. That means the amount a point on the earth rotates to have the sun come back directly overhead must be different in different parts of the year.

This difference, summed day over day, results in a +/- 20 min movement of actual midday to 12pm. The ‘mean’ in Greenwich Mean Time refers to averaging this difference over the whole orbit.

Danny Boyd wrote an excellent video essay on A Knight’s Tale. I too always wonder why there’s always someone cutting onions when I choose to watch.

Remember the incident at the docks when the revolutionaries threw the T-Mobile imports into the harbour? Talk about high tariffs!

“asked”

Without him, America is a little greater again.

I’m guessing that exactly the same LLM model is used (somehow) on both sides - using different models or different weights would not work at all.

An LLM is (at core) an algorithm that takes a bunch of text as input and produces an output of a list of word/probabilities such that the sum of all probabilities adds to 1.0. You could place a wrapper on this that creates a list of words by probability. A specific word can be identified by the index in the list, i.e. first word, tenth word etc.

(Technically the system uses ‘tokens’ which represent either whole words or parts of words, but that’s not important here).

A document can be compressed by feeding in each word in turn, creating the list in the LLM, and searching for the new word in the list. If the LLM is good, the output will be a stream of small integers. If the LLM is a perfect predictor, the next word will always be the top of the list, i.e. a 1. A bad prediction will be a relatively large number in the thousands or millions.

Streams of small numbers are very well (even optimally) compressed using extant technology.

If you’d read the article you will see that this is a report from a network of church abuse survivors and the person speaking has first hand evidence from speaking to Prevost. If you discount them, you are basically saying that witness statements count for nothing. You are silencing the victims.

You will also know that Prevost blocked the ‘real investigation’ you claim you want and this is core to the point of the article.

Almost every survey will get 6-10% of people answering yes to the most extreme or batshit crazy option, no matter what.

Probably the main reason is that people are pissed off that they are being approached by survey takers and punish the survey for revenge.

And there are some batshit crazy people out there.