> There are fewer and fewer organizations protecting civil rights without being dragged into left/right tribalism.

I would rather challenge this image that civilization is declining, independently of the political forces in power. This is a common motif in facism; I'm reading from your comment something along the lines of: "once we had noble organizations that were pure and didn't bother with ideology -- now things are worse, and in fact those guys are dirty for engaging in politics". What's really happening is that power in the US has been seized by fanatics and you fucks (respectfully) are letting them get away with it.

Who's organizing a fork, or is python back to having only shitty packaging available? :(

Accuracy and reliability are necessary to know real productivity. If you have produced code that doesn't work right, you haven't "produced" anything (except in the economic sense of managing to get someone to pay for it).

For example, if you produce 5x more code at 5% reliability, the net result is a -75% change in productivity (ignoring the overhead costs of detecting said reliability).

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

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

(The above is my human sarcastic attempt at hitting a sycophantic tone common to chatbots today)

  git commit --fixup 

  git rebase -i --autosquash 

[1] https://fle.github.io/git-tip-keep-your-branch-clean-with-fi...

Ads are 509 pixels wide coming in 3 ... 2 ... 1 ...

The canonical approach is to build a generative model with a parameter (or multiple for ~anova) that codes for the difference between groups and do inference on that parameter of interest. Most of the recipes taught in statistics classes can be modelled as a regression of some kind (this counts for frequentist stats too, see https://lindeloev.github.io/tests-as-linear/ ). Some advocate to do that inference with bayes factors. Others, like discussed elsewhere in this thread, advocate combining the resulting posterior with a cost/value function, but either way the lesson is that there is less focus on "t-test-vs-anova" because they're the same thing anyways.

The point is that while LLMs can solve the puzzle when the constraints are unchanged -- as you said, there are loads of examples of people asking and answering variations of this puzzle on the internet -- but when you change the constraints slightly ("you can open the door to look at the bulbs and use the switches all you want") it is unable to break out of the mold and keeps giving complicated answers, while a human would understand that under the new constraints, you could simply flip each switch and observe the changes in turn.

A similar example that language models used to get stuck on is this: "Which is heavier, a pound of feathers or two pounds of bricks?"

Languages are optimized firstly for the people that speak it as their first language. There are some suggestions that in some cases, a group of people that learned a language later in life can influence the development of the language too, but it's not the primary mover. When you are a L1 speaker, the grammatical gender doesn't really require any cognitive effort in most cases.

And there is an actual benefit in a conversation between two proficient speakers: redundancy. If you're in a bar that serves those three kinds of alcohol, if you were speaking a language without grammatical case, you'd only have to miss one word (the alcohol) to misunderstand the order. If there is also a gendered article, you'd be able to mostly understand the order even if you missed the alcohol.

They also let you communicate more complex ideas more easily but letting you use two or more pronouns in parallel without ambiguity. This sentence is confusing in English, but makes total sense in gendered languages: "Hand me the wine and the sangria. It's in the tall bottle, and it's behind that door"

Then I reinstalled windows 10 on the same computer a couple of years later and the installer couldn't even format my drive to install it, just threw bland "not working" errors until I manually formatted and installed the windows bootloader from a rescue drive.

What's your point?

Or, to answer more directly, I've recently played on linux via steam and can recommend:

- Total War Warhammer II (native, don't need the DLC unless you want to play specific factions)

- Frostpunk (proton)

- Don't Starve (native)

- Civilization VI (native)

- Factorio (native)

- Crusader Kings II / III (native)

- Valheim (native)

- Pathfinder: Kingmaker (native)

[1] https://gitlab.com/maltelau/org-lookup-dnd

I'm getting no sound throughout the whole video, and youtube lists it as having music from a particular Debussy recording, which is then also unavailable (https://www.youtube.com/watch?v=vhogGOTnT9I)

Muting the sound is the usual response to streams and videos that plays audio that someone claims to own the rights to. Is this another casualty to how easy youtube makes it for copyright trolls (or people indistinguishable to copyright trolls)?

Edit: This of course goes hand-in-hand with the claim that it is easier/faster to write R scripts. If you're not familiar with it, the tidyr and dplyr packages in particular (part of the tidyverse) are fantastic in the verbs they provide for thinking about data cleaning.

Example script:

  library(tidyverse)
  old = read_tsv("path/to/old/db.tsv")
  new = read_tsv("path/to/new/db.tsv")
  deleted   = anti_join(old, new)
  new       = anti_join(new, old, by = "id")
  unchanged = semi_join(old, new)
  changed   = semi_join(new, old, by = "id") %>%
                anti_join(unchanged)