September was good to me. Reasonable amounts of work, starting of the Pirate Academy, prepping the wedding and passing some exams in my doctoral studies. Nothing to complain about, reasonable time to read. Decided to re-read some classics (Kesey, Foucault) and found a lot of new value there. Rereading is now officially good.

The Archeology of Knowledge

Michel Foucault

Re-read after quite some time: nothing much to add. Even if the writing sometimes gets in a way of the message and if the self-referencing moments really bothered me sometimes, still a fundamental book on that favourite “thinking about thinking” and the limits of knowledge and how they are set. Much more approachable than say Wittgenstein and would also argue that much more relevant, at least to me.

Blackfog Island

Jeffery Russell

My fianceé suggested this as our next “listen before sleeping”. I accepted the light hearted premise of “company of professional dungeon goers who use logic and cheese to sort out dungeons” and actively wanted to have fun.

I did not. All the characters were paper thin, there were dozens of them (and most of them just super actively trying to be funny, ending up obnoxious) and it all just felt so flat. Kamila liked the first one, but also felt that this one just did not do it. However, if you find characters overcoming classic fantasy obstacles with weird things funny, this could be a pick for you somehow I guess. Or if you really, really like vague descriptions of life on ships.

Noctuary & the Spectral Link

Thomas Ligotti

I probably overexposed myself to Ligotti. I loved his earlier stuff and I found quite a lot of interesting arguments in his Conspiracy against the Human Race. I did not agree with it, but the read was both philosophically stimulating on some level and most importantly aesthetically pleasing in a morbid way.  I thought that I would have a great time with his next horror collection.

And while the writing is still great, the atmosphere somber etc etc it is just too much of the same. Life does not make sense, things are awful, everything comes to some gruesome ending: and it is hard to care about it in any way, because the characters have the same lifeless, unsympathethic qualities as heroes of Lovecraft: sometimes they feel worse honestly, just spectators to their own inevitable end. 

In the end, it just felt so very bland and uninspired. There is hardly any struggle, fear, obsession, twists, reveals or actual plot beats: the stories just exist, adorned with lots of decor, but ultimately empty.

Smaženky a machři

Jiří Krejčik

CZ Only, sadly.

Banger! Krátký, úderný, v nejlepších chvílích připomíná moderní verzi Paula Therouxa, nikdy nezačne nudit a vždycky nahazuje “co se dá číst a zjistit dál”, nepřehání to s doslovností a vysvětlováním. A atmosféra je vynikající, stejně jako zpracování. Nikdy bych nečekal, že budu číst knížku “týpek žere divná jídla v nádražkách” s takovým zaujetím a prostou radostí ze života i z kusů poznání, které trousí. Krásný.

Leftover in China: The Women Shaping the World’s Next Superpower

Roseann Lake

I definitely learned a lot about the position of women in China - or I think I did. Sadly, there are quite a few issues with this one:

Many conflicted feelings.

a) the book is already a bit obsolete, sadly

b) the humor goes from “hehe” to “please no”

c) I am not sure if I learned much about the demography OR about the social and culture pressures. Everything just goes by so fast that it IS interesting, but…is it also true? Felt consistent with the other articles and books I read on China (see “Breakneck”), but I do not read Mandarin and that blocks me off from lot of understanding.

The biggest problem is however the allegation of erasing other scholar and appropriating her work. It’s hard for me to judge what happened, but it leaves a sour taste and feels unsolved. Based on this, I can´t recommend the book. Sadly I learned that AFTER reading.

One Flew Over the Cuckoo’s Nest

Ken Kesey

One of the classics and for good reason. What is easy to forget is how much is the supposed hero at the same time an anti-hero: he strives for freedom after being convicted of rape of minor, his sort of freedom the ultimate macho fantasy. The clash with the nurse is epic and dramatic also because there is no clear positive hero, just differently damaged people trying to enforce their will on the world.

Antimemetics: Why Some Ideas Resist Spreading

Nadia Asparouhova

This could have been a blog post. It should have been a blog post.

No, seriously, the basic ideas are of course relevant and somehow meaningful. However, the basic problem of the book comes up when you consider where the book gets its inspirations and ideas - and it’s mostly from the rat/post-rat ecosystem. The moment you write about discourses and narratives (which are crucial in the context of “memes and anti-memes”) and the main reference point is China Mieville or Scott Alexander, but Foucault or phenomenologists are not mentioned at all the results feel like trying to come up with the very same ideas again.

Still being worth a read probably, mostly due to being very short and as an illustration of type of thinking and its limitations.

We listened to this before sleeping with my fianceé.

Into the Wild

Erin Hunter

In the beginning, I was bored, disappointed and not very invested in any of the cat adventures. In the end, I was dragged in by the lore, conflicts, very solid plot twists and cliff hangers: with some ideas about purely cat society that as a person living with four cats can only confirm. Solid young adult adventure kind of thing.

Persuasive Games: The Expressive Power of Videogames

Ian Bogost

One of the seminal, academic books on the topic you can quote endlessly.

The important power here being “quote endlessly”, because it seems to be overly dense, super academic and needlessly complex on purpose: most of the thoughts are elaborated for four to five pages to be summed up then in three sentences. This would make sense for the crucial topics of the book, but when the majority of the things get the same treatment it’s not great. Procedural rhetoric can be complicated as a concept, but the book is not great in generating understanding.

The problem is also in selection bias for the games: we will meet mostly serious games, with some preference for the serious games done by the author. Is that a problem? Not really. It is a feature that can be interesting if kept in mind, it might be quite deceiving for people without the context.

House of Huawei: The Secret History of China’s Most Powerful Company

Eva Dou

It is a good summary with very low added value on top of that. If you had been following Huawei and its shenanigans, there is nothing new to learn here. Still, as a timeframe/ summary this is good and useful for people trying to get into the topic, no hate there. The “secret history” or “untold story” are just doing a lot of overexaggeration here.

Read

Dept. H Omnibus Volume 2

Matt Kindt

Slower, much more flashbacky and pseudo-philosophical than the first omnibus. Still quite entertaining and the little details and world building artifacts etc are cool enough for this to be a fine graphic novel. 

If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All

Eliezer Yudkowsky, Nate Soares

If Anyone Builds It, Everyone Dies by Eliezer Yudkowsky and Nate Soares does not aspire to be a detailed scientific paper or a systematic survey of current AI research. Instead, it positions itself as an alarmist manifesto: a deliberately urgent, almost polemical text meant to jolt the reader out of complacency about the risks posed by advanced artificial intelligence. The book focuses on a small set of core ideas—most notably the alignment problem, instrumental convergence, and the structural dynamics of competitive technological races—and uses them to argue that uncontrolled AI development could plausibly lead to global catastrophe.

At the level of basic principles, it is hard to disagree. The claim that sufficiently capable optimization systems will tend to seek power, resources, and control over their environment is well-established in theoretical discussions of AI risk. Likewise, the argument that even a single reckless actor can endanger everyone else in a competitive landscape is intuitively and historically compelling. In that sense, the book’s central warning is neither fringe nor absurd, and readers already familiar with existential risk debates will likely find much of it reasonable.

The problem lies not so much in what the authors argue, but in how they argue it.

The metaphors that structure the book—comparisons to nuclear weapons, runaway evolutionary processes, or almost demon-like optimizing entities—are vivid but ultimately thin. Rather than progressively deepening the reader’s understanding, they tend to circle back to the same images and claims, slightly rephrased. As a result, the text moves slowly despite covering a relatively narrow argumentative space. For readers with even moderate prior exposure to AI safety discussions, this repetition can feel more exhausting than illuminating. The book rarely refines its concepts, introduces sharper distinctions, or meaningfully advances beyond its initial premises.

Although the authors explicitly reject the goal of writing a technical or academic work, they also largely avoid engaging with the strongest counterarguments in a systematic way. Alternative visions of AI development—such as gradual capability scaling, institutional oversight, economic and political friction, or partial alignment successes—are often dismissed rhetorically rather than examined in detail. The result is a text that feels primarily addressed to readers who already share the authors’ worldview, aiming to emotionally reinforce their concerns rather than intellectually persuade skeptics.

A particularly problematic aspect is the book’s handling of uncertainty. The stakes are presented as existential and near-total, yet probabilities, timelines, and concrete failure pathways remain vague. This creates a tension at the heart of the argument: catastrophe is portrayed as both overwhelmingly serious and almost inevitable, while the analytical machinery needed to distinguish between levels of risk is largely absent. Critics have rightly noted that without such distinctions, the book risks sliding into an apocalyptic register that may alienate precisely the audiences it most needs to reach—policymakers, engineers, and cautious skeptics.

This raises a central question: not whether the authors are “right” in an abstract sense, but whether the chosen form is capable of changing minds or behavior. The book may well serve as a rallying cry within the AI safety community, where its assumptions are already largely shared. It is much less clear that it can persuade readers outside that bubble. If the goal is to warn the world, then the heavy reliance on categorical claims and emotionally charged metaphors may be more of a liability than a strength.

Perhaps this skepticism is unfair. Perhaps a sharply worded manifesto is exactly what the authors believe the moment demands, given what they perceive as a race against time. Still, the lingering impression after finishing the book is that it gestures in the right direction while choosing a path that limits how many people are willing—or able—to follow it.