Commonplace book entries: climate, slop, defence
Those American candy gift shops, that one usually assumes are some sort of money laundering scheme, are in fact multilevel tax dodges.
Valerie Aurora writes about how to start an internet resiliency club.
There's a nice map of Digital Public Infrastructure at https://dpimap.org/
Thanks to Full Fact for drawing attention to how Google is making web content harder to trust:
In a change buried in a little-known technical blog, Google has announced that it will abandon a key element of its tagging system for fact checks, effectively de-prioritising this content and making it harder for people to access reliable information.
Stuart Schechter writes a powerful reflection:
Not everyone will miss Ross Anderson.
Ross was not afraid to speak truths that made people uncomfortable. His ideas and arguments threatened the beliefs, status, power, ego, and bank balances of others — often those with power. His writings and talks undermined proponents of hardware attestation, chip-and-pin authentication, and surveillance, to name just a few. Many of those threatened by what Ross had to say professed to be faithfully working to serve the public. Many likely believed they were.
... We can honor Ross by embracing discomfort and questioning our own beliefs that we, individually and as a community, are using the powers we possess to make the world a better place.
Examining this question should indeed make us uncomfortable. I first met Ross in 2002, while I was a graduate student, at a time when the study of security was just starting to receive the attention it deserved. It was an exciting time, and many of us had high aspirations and ambitious goals. Yet, it is hard to imagine a metric of success for serving the public – real living people – for which our field can claim anything but defeat.
A brutal post-mortem of the period follows. In conclusion, Stuart writes:
One of the underlying problems for those of us who work in security, privacy, trust, and safety is the ambiguity over who we are most obligated to protect. Many of us in public-facing security roles purport that protecting the public (users) is our paramount goal when, in fact, our first obligation is to protect our employers.
Ambiguity and deception about our motives and obligations underlie many of the failures we need to own up to...
In addition to codes of nonmaleficence, we need to build a culture where we expect more from each other and help each other anticipate and make hard choices. We need to ask hard questions of each other, and we need to expect our peers to ask hard questions of us.
... Those most likely to be angered off will be people in authority who fear having their power undermined. We have to be prepared for them to brush us off with indignant responses, such as “Who died and made you the asshole who thinks it's okay to ask that kind of question?”
Funny you should ask. It was Ross Anderson.
Or, you know, you could create AI companions and monetise them. I knew this sort of thing was probably happening but it was still startling to see an example.
Who is free software for? Some interesting thoughts from Jürgen Geuter:
People within the Open* movements have done the impossible, have created whole encyclopedias, the most successful and most used kernel on the planet and a metric fuckton of custom, optimized operating systems, software libraries, and user facing programs. Have contributed to the commons to a degree that wouldn’t even have been credible within science fiction stories. Some of these systems – and I am not kidding here – should be considered the digital wonders of the world.
... Linux has a desktop market share of about 4%. Which is a lot given how hard the main competitors Microsoft and Apple are making the life of Linux developers and distribution builders. But still not even close to being an actually relevant player in the market for most purposes.
... So much work so many technical communities put into making the path towards becoming a member easier, more inclusive, fairer, etc. Those activities are fundamentally about fulfilling the Open Source promise: To give everyone the ability to have control over the software they use and the tool to build upon what’s already there. Mission … not accomplished but on its way, right?
But what are we doing? What are we trying to help “everyone” with?
We are trying to give more people the opportunity to “become hackers”. So they can profit off of all this stuff built by hackers for hackers. This isn’t a project to free all of us, it’s a project to give everyone a degree of freedom if they join our club.
... We need to get out of our comfort zones and modes of operation. Need to move beyond the seemingly apolitical cyberspace of free licenses. We need to reshape our thinking towards more political goals and values.
Quite the most interesting thoughts I've come across around big tech platform taxation, the ad-funded internet - and the Global South as a potential force for change. Juan Ortiz Freuler writes at Open Democracy about coordinated international action on tax.
Arvind Narayanan and Sayash Kapoor have a fairly comprehensive dive into AI and how it is evaluated and considered, and propose we might do better to view it as a normal technology (via data&society).
Nate Angell suggests we should talk about mirages rather than AI hallucinations. (as well as the blog post linked there, there's a proper paper.)
It took me six hours of listening to people with differing points of view discuss AI and copyright at a workshop, organized by the Sussex Centre for Law and Technology at the Sussex Humanities Lab (SHL), to come up with a question that seemed to me significant: what is all this talk about who “wins the AI race”? The US won the “space race” in 1969, and then for 50 years nothing happened.
Fretting about the “AI race”, an argument at least one participant used to oppose restrictions on using copyrighted data for training AI models, is buying into several ideas that are convenient for Big Tech.
One: there is a verifiable endpoint everyone’s trying to reach. That isn’t anything like today’s “AI”, which is a pile of math and statistics predicting the most likely answers to prompts. Instead, they mean artificial general intelligence, which would be as much like generative AI as I am like a mushroom.
Two: it’s a worthy goal. But is it? Why don’t we talk about the renewables race, the zero carbon race, or the sustainability race? All of those could be achievable. Why just this well-lobbied fantasy scenario?
Three: we should formulate public policy to eliminate “barriers” that might stop us from winning it. *This* is where we run up against copyright, a subject only a tiny minority used to care about, but that now affects everyone.
Anthony Finkelstein proposes a new doctrine, noting that democratic values should be strategic assets in defence. Also some interesting points about how defence views innovation and tech waves.
John Elkington is always good value on sustainability. This article reflects on London Climate Action Week and the good bits start when he goes into the 4 takeaways.
wealth divides are increasingly distorting our economies and societies.
That’s why yet another book called Abundance is now on my reading list this coming week. It asks the question why we are so singularly failing to build a better future. Its conclusion:
... our problems today are not the results of yesteryear’s villains. Rather, one generation’s solutions have become the next generation’s problems. Rules and regulations designed to solve the problems of the 1970s often prevent urban-density and green-energy projects that would help solve the problems of the 2020s. Laws meant to ensure that government considers the consequences of its actions have made it too difficult for government to act consequentially. In the last few decades, our capacity to see problems has sharpened while our ability to solve them has diminished.
... I genuinely believe we are only just getting to the end of the beginning in all of this. The challenge is not simply to reinvent our agenda but to invent ourselves and our platforms.
In hindsight, the exchange that sticks out most clearly in my mind from the Volans session came when Indy Johar of Dark Matter Labs declared that “sustainability is dead.” I respect Indy profoundly—and if we stick with the incumbent definitions of sustainability, I fear that he will prove to have been on the money.
But then, speaking from the sofas, Pavan Sukhdev of GIST Impact countered by saying that sustainability has scarcely been tried yet...
Louis VI ... One of his messages: “Rewild Yourself.”
But even the most sophisticated comms will only get us so far. We also must engage—and rewild—economists, policymakers, scientists, technologists, troublemakers and so endlessly on. Everyone who shapes our politics and markets.
... For my part, though, London Climate Action Week 2025 has left me determined to do several things simultaneously:
1. To try even harder to “think the unthinkable,”...
Jonathan Porritt, whose name feels like a blast from the past for me, asks if the IPCC can regain its credibility.
The IPCC is made up of tens of thousands of scientists in dozens of countries, covering multiple facets of “why” and “how” the climate is changing. It lays down the consensus-based line in great doorstopper reports every five or six years. ...
... if you're looking for an institutional manifestation of what good science looks like in practice, then the IPCC is seen by many as providing that gold standard.
... In January 2025, without any huge fanfare, the Institute of Faculty and Actuaries published its “Planetary Solvency: Finding our Balance in Nature” report, in partnership with scientists at the University of Exeter. It robustly critiques orthodox economic predictions, which estimate that the impact of an average temperature increase of 3°C by the end of the century would be around 2% of annual GDP. “These estimates are precisely wrong, rather than being roughly right, and do not recognize there is a risk of ruin.” The institute's risk management experts diligently reassessed risks associated with impacts such as fires, flooding, droughts, temperature increases, and rising sea levels through 2050 and on to the end of the century.
... Two billion prospective deaths by 2050. That's just 25 years away. And for the final kicker, bearing in mind that we're currently on a business-as-usual trajectory towards at least a 3.7°C temperature increase by 2100, the contraction in GDP then rises to 50% and the number of projected deaths rises to 4 billion.
The institute's definition of “planetary solvency” is fascinating:
Planetary Solvency” assesses the ongoing ability of the Earth system to support our human society and economy. In the same way that a solvent pension scheme is one that continues to be able to provide pensions, a solvent Earth system is one that continues to provide the natural services we rely on, support ongoing prosperity, and a safe and just future.
The problem is that the IPCC, on behalf of citizens of Planet Earth, does not assess risk in the same way that the managers of pension schemes assess risk for their clients.
Unmitigated climate change and nature-driven risks have been hugely underestimated. Global risk management practices for policy makers are inadequate, and we've accepted much higher levels of risk than is broadly understood.
... another equally devastating report from scientists at the Institute for Climate Risk at the University of New South Wales, confirming the actuaries’ hypothesis of underestimated financial risks—simply because the IPCC's Integrated Assessment Models on which it has depended for decades are incapable of capturing major risks—what actuaries describe as risks with “low probability but catastrophic impact.”
As Professor Andy Pitman, one of the co-authors of this report, put it:
It's in the extremes when the rubber hits the road. It isn't about average temperatures. In a hotter future, we can expect cascading supply chain disruptions triggered by extreme weather events worldwide.
An older article from Indy Johar discusses structural shifts in the world and some ways to think about them. One centres on defence; another on environmental nationalism and the right.
More recently -
 |
| https://bsky.app/profile/indyjohar.bsky.social/post/3ljkc5e3otc2i |
If you are in England, MedConfidential have made a map where you can find out where you can get a medical scan - here it is.
‘At its root, is the fact that “we don’t pay what food’s worth”.’ HT to Jonathan Schofield whose thread introduced me to both this article about food production in the UK, and also to the Fen Power project.“Fen Power is a research project exploring what people need to live in a good place – and how we might begin to reverse decades of managed decline in English regions.” There's a 20 minute video.
The food stuff is pretty grim.
The first crop of wheat I grew, I got £120 a tonne for it – when I was 18 and beer was 50p a pint. So I got 240 pints for every tonne going off the farm. Last year, I got £65 a tonne and beer has come up to £3. So now, I’m getting 20 pints a tonne.I had missed this from April -
In April 2025, the UK’s first farmers’ strike broke out, with some farmers refusing to load milling wheat needed for Britain’s bakeries.
While a one-off event, it may not be the last.
An archaeological excavation of AI-generated salmon fillets, and what these say about the narrative circulation of algorithmic folklore.
Drew Austin notes that 'slop' is the culmination of where online culture has been going for years.
Yesterday morning, I walked past the small grocery store on my block and heard REM’s “It’s the End of the World as We Know It” emanating from inside. I’d usually ignore that kind of thing but here it suddenly struck me as absurd—it was 9:30 am on a weekday and there was absolutely no reason for that song to be coming out of an empty grocery store. Not even annoying or dissonant, it was just the least appropriate accompaniment for the moment, in its own subtle way. And of course there was no reason for it—no person had chosen the song and the process that led to it playing then had no audience in mind. ... It was slop. Lately the idea of slop has fascinated me. I’ve started seeing it everywhere and grasping for a more expansive definition of it. Slop is probably the definitive term of our current digital experience; it increasingly functions as shorthand for all of our vague frustration with digital technology. Although most people wouldn’t call that REM song slop, its appearance in that context fits my working definition: content as environmental filler, a choice that’s not quite right (but also not quite wrong), which nobody really wanted, probably unnoticed, meant to only register at a subliminal level, and put in place to negate a worse alternative: silence.
... Today, “slop” implies AI more than anything else, and primarily refers to the AI-generated content that is flooding the internet. The subtext is that slop is being dumped on us against our will—that it’s something that happens to us—but that lets us off the hook far too easily. Most of the slop we see is still made and distributed by real people, often with no AI assistance. If AI is able to suddenly pump slop into our environment it’s only because we already turned on the faucets ourselves.
... The end result of this process, as Liz Pelly has described, is an opportunity for AI content creation, which barely registers because the human-made content with which it coexists has already become fungible.
Dan Sinker writes that there's slop everywhere and the real issue is that no one cares:
Earlier this week, it was discovered that the Chicago Sun-Times and the Philadelphia Inquirer had both published an externally-produced "special supplement" that contained facts, experts, and book titles entirely made up by an AI chatbot. .... But the thing that is most disheartening to me is how at every step along the way, nobody cared.
The writer didn't care. The supplement's editors didn't care. The biz people on both sides of the sale of the supplement didn't care. The production people didn't care. And, the fact that it took two days for anyone to discover this epic fuckup in print means that, ultimately, the reader didn't care either.
It's so emblematic of the moment we're in, the Who Cares Era, where completely disposable things are shoddily produced for people to mostly ignore.
Rob Horning writes about Mr Beast:
... the “form” Donaldson has mastered and made visible in some ideal, essential expression of itself — a Platonic form of “pure content,” as O’Connell describes it — is a “signature artifact of our time”: the video optimized for algorithmic promotion and consumption. In describing one of MrBeast’s works — a 13-minute video called “I Survived the 5 Deadliest Places on Earth” — O’Connell offers what could be taken as a definition of this form. Thirteen minutes, O’Connell reminds readers,
does not leave a lot of time for old-timey narrative conventions such as establishing a context, or making the viewer understand or care about why any of this is happening in the first place. It just is happening. That’s the thing with MrBeast: everything that happens is always just happening, and if you want a reason for it, it’s no more or less than the fact that you yourself are watching.
“Why anything is happening” is just because you, the individual, wants to see — other causes and reasons are obscured, made to seem irrelevant. History unfolds because of the universal demand for content. And the content exists because you will watch, and moreover, your anticipated watching is what the videos are all about, the implicit subject matter of all of them.
... “Maximum viewer stimulation” (i.e. measurable manipulation of the audience) replaces conventional ideas of what is “fun to watch” (which is some kind of unmeasurable reaction that takes place in a viewer’s consciousness).
Wikipedia x TikTok: https://wikitok.vercel.app/
Geoguessr x the future: https://futureguessr.fr/