Astral Codex Ten

Catalonia

5 min read

Catalonia

Article

Catalonia is a recurring place in the Astral Codex Ten archive, appearing 2 times across 2 issues between May 08, 2025 and July 26, 2025. The archive places it in contexts such as “a forest in Catalonia instead of Albania”; “since Catalonia isn’t going to do anything more than say”. It most often appears alongside Russia, 4chan, @DeepGuessr.

Metadata

  • Category: Places
  • Mention count: 2
  • Issue count: 2
  • First seen: May 08, 2025
  • Last seen: July 26, 2025

Appears In

Source Context

Recovered passages from the original issue text. When the raw archive preserved outbound links inside the source passage, they are listed directly under the quote.

Highlights From The Comments On AI Geoguessr
May 08, 2025 · Original source
It did okay on the next three. All relatively clean photos with at least a little evidence. It guessed a park in Barcelona instead of Rome, a forest in Catalonia instead of Albania, and Crete instead of the Parnasse mountains.
Your Review: The Astral Codex Ten Commentariat (“Why Do We Suck?”)
July 26, 2025 · Original source
The graph above shows the output of these two approaches. This is a really weird result, which defies easy explanation. Toxicity goes down over the whole SSC era, then starts ticking back up again from the ACX era. If you allow for a bit more variability in the simpler measure, the fancy neural network closely tracks the number of times we call each other ‘retards’ or ‘dumbasses’, which you would expect to track overall toxicity quite closely. This suggests the neural network is keying in on actual toxicity, rather than polite discussions which happen to involve contested or sensitive concepts. One caveat is that the ACX Commentariat is not very toxic to begin with, so the ‘toxicity’ metric may not be sensitive enough to capture the sort of politeness which the Commentariat values. In 2013, at peak toxicity, the toxicity score maxed out at 0.04 (the spike in October 2013 seems to be related to attracting some external neo-reactionaries (very roughly the precursor ideology to the modern alt-right) to comment on this post. In 2021, the lowest toxicity ever was reached at around 0.01. This means that a typical comment would be around 4% likely to be perceived as toxic by a human reader in 2013, but by 2021 this has fallen to 1%. Here is a snippet of a comment which is rated as having a 1% chance of being perceived as toxic by a human, written by John Schilling: The purpose of war is, roughly speaking, to settle the question of whose police get to enforce which laws in a region, and since Catalonia isn’t going to do anything more than say, “We’re going to make you look like Evil Meanies on TV and Youtube if you don’t pull back your policemen and let us have our own”, that point is moot. [Link] By contrast, if you promise to draw your fainting couch nearby, here is a snippet of a comment which is rated as having a 4% chance of being perceived as toxic by a human, written by Maximum Limelihood: Being fired means nothing about the speed you’re learning at. It means that the employer overestimated how much you *already* knew. …. Unfortunately, it doesn’t matter how great you’ll be at coding in a year when you’re costing me time and training effort today [Link] The most toxic the comments section has ever got (beyond the very early days) was on the post Gupta on Enlightenment. I feel like the comments section on this post should be part of the ACX main canon because it is so cosmically hilarious. It concerns a man name Vinay Gupta (founder of a blockchain-based dating website) and his claims to have reached enlightenment. Some people in the comments are sceptical that Vinay Gupta is indeed an enlightened being, citing that enlightened people don’t typically found blockchain-based dating websites. A new forum poster with the handle ‘Vinay Gupta’, claiming to be Vinay Gupta and writing in a very similar style to the actual Vinay Gupta, turns up and starts arguing with everyone in an extremely toxic way (in the objective sense that his comments score very highly on the toxic-bert scoring system), which provokes more merriment that a self-described enlightened being would deploy such classic internet tough-guy approaches as ‘I don’t think much of a four-on-one face off against untrained opponents’ (link) and ‘this board is filled with self-satisfied assholes who feel free to hold forth on whatever subject crosses their minds, with the absolute certainty that they’re the smartest people in the room’ (link, no further comment…). More prosaically, this is a great example of what I was discussing earlier – the comment section is usually so civilised that a single individual turning up and acting out of the Commentariat norms is enough to make it the most toxic discussion which has ever taken place. Of Scott’s classic posts, the most toxic the comment section has ever become was on Radicalising the Romanceless. The least toxic the comments section has ever been are the posts on Scott’s conworld, Raikoth (technically the Raikoth post on history and religion specifically, but the whole series is so good I’ve linked to the index). Complexity of thought
Inline links: post, Link, Link, Gupta on Enlightenment, link, link, Radicalising the Romanceless, Raikoth