Thought Shrapnel

As explained in this post, Tom Watson and I are putting together an offer to help organisations with their digital sovereignty.

Sign up if you’re interested in joining our first cohort. No obligation.

You cannot manage what you cannot see. Most organisations operate in a fog of invisible reliance; you rely on platforms that obscure their workings behind slick interfaces and terms you never read.

TechFreedom helps you clear the air. We give you the tools to look past Big Tech’s version of the Cloud and build a digital infrastructure that is open, visible, and under your control.

Source: techfreedom.eu

Source: Are.na

Jo Hutchinson reshared this on LinkedIn a few days ago, and I wanted to save/share my comments on it. The original post was by Tricia Riddell, a Professor of Applied Neuroscience and includes a short video. In it, she talks about there being “a range of authentic selves” that people can bring to a situation rather than a single “authentic self.”

Riddell wonders whether this is what leads to imposter syndrome, which is usually suffered by high-achievers. “You cannot feel like an imposter,” she says, “unless there is something you believe you’re failing to live up to.”

My response, because it’s easier for me to re-find things here rather than on social media:

As someone who suffers from imposter syndrome most days of my life, it resonates.

I am, though, reminded of something that John Bayley wrote in one of the biographies of his wife, the philosopher Iris Murdoch. He said that Iris believed that she “didn’t have a strong sense of self.” I’ve thought about this over two decades, and I think it maps onto what’s being discussed here.

On the one hand, you could say that those with a strong sense of self want to bring a single “authentic” self to every situation. Which, as Tricia Riddell points out, would mean that they’re not bringing other stories into the situation.

Or, on the other hand, you might say that those without a strong sense of self have a fragmented view of themselves. The imposter syndrome is the price that is paid for an integrated sense of self.

I haven’t figured it out yet, but I find the whole thing fascinating.

Source: LinkedIn

Image: Amy Vann

Source: Are.na

Most of this post talks about the differences between Japan and Italy, which makes it well worth reading in full. What I like about it, though, is that it goes beyond “mental models” to talk about framings and how these are the unseen, and somewhat hidden things that make cultures different. Fascinating.

A mental model is a simulation of “how things might unfold”, and we all build and rebuild hundreds of mental models every day. A framing, on the other hand, is “what things exist in the first place”, and it is much more stable and subtle. Every mental model is based on some framing, but we tend to be oblivious to which framing we’re using most of the time […].

Framings are the basis of how we think and what we are even able to perceive, and they’re the most consequential thing that spreads through a population in what we call “culture”.

[…]

Each culture is made of shared framings—ontologies of things that are taken to exist and play a role in mental models—that arose in those same arbitrary but self-reinforcing ways. Anthropologist Joseph Henrich, in The Secret of Our Success, brings up several studies demonstrating the cultural differences in framings.

He mentions studies that estimated the average IQ of Americans in the early 1800’s to have been around 70—not because they were dumber, but because their culture at the time was much poorer in sophisticated concepts. Their framings had fewer and less-defined moving parts, which translated into poorer mental models. Other studies found that children in Western countries are brought up with very general and abstract categories for animals, like “fish” and “bird”, while children in small-scale societies tend to think in terms of more specific categories, such as “robin” and “jaguar”, leading to different ways to understand and interface with the world.

But framings affect more than understanding. They influence how we take in the information from the world around us.

Source: Aether Mug

Image: Maksym Tymchyk

Source: Notes by Praveen

This post on the blog of the American Philosophical Association (APA) talks about political epistemology. Samuel Bagg is Assistant Professor of Political Science at the University of South Carolina and argues that the reason presenting people with facts doesn’t change their political opinions is due to our social identities.

This resonates with me more than the usual framing where in which we’re told that we need to appeal to people’s emotions. Social identity makes much more sense, and so (as Bagg argues) the way out of this mess isn’t to “ensure every citizen employs optimal epistemic practices” but instead to strengthen our “broadly reliable collective practices.”

I’ve often said that one of the reasons for Brexit was the [attack on expertise by Michael Gove]9www.ft.com/content/3…). Attacking institutions which are proxies through which people can know and understand the world is corrosive, divisive, and ultimately, a cynically populist move.

On the individual level, we are right to worry about the epistemic impact of motivated reasoning, cognitive biases, and “political ignorance.” And at the systemic level, we are right to lament the rise of talk radio, cable news, social media, and the attention economy, along with the corresponding decline of local news, professional journalism, content standards, and so on. Trends in our political economy have indeed conspired with underlying human frailties to undermine the epistemic foundations of a stable and functioning democracy—even in its most minimal form.

Yet this diagnosis is also crucially incomplete, in ways that distort our search for solutions. When we understand the problem in epistemic terms, we naturally seek epistemic answers: better fact-checking, more deliberation, better education, greater media balance. We aim to develop the “civic” and “epistemic virtues” of unbiased, fair-minded citizens. And indeed, such projects are surely worth pursuing. The epistemic failures they aim to address have deeper roots in our political psychology, however—and overcoming them requires grappling with those roots more directly.

In short, decades of research have demonstrated that our political beliefs and behavior are thoroughly motivated and mediated by our social identities: i.e., the many cross-cutting social groupings we feel affinity with. And as long as we do not account for this profound and pervasive dependence, our attempts to address the epistemic failures threatening contemporary democracies will inevitably fall short. More than any particular institutional, technological, or educational reform, promoting a healthier democracy requires reshaping the social identity landscape that ultimately anchors other democratic pathologies.

[…]

We do not generally decide which groups to identify with on the basis of their epistemic credentials regarding political questions. And even if we were to use such a standard, our judgments about which groups satisfy it would be inescapably colored by our pre-existing identities, formed on other bases or inherited from childhood. Developing certain “epistemic virtues” may mitigate some of our blind spots, finally, but it is hubris to think anyone can escape such biases entirely. Like it or not, it is a basic fact of human cognition that we think and act politically as members of social groups.

[…]

The trouble is not that reliable collective truth-making practices no longer exist, but that significant portions of the population no longer trust them—and that as a result, the truths they establish no longer constrain those in power.

[…]

What best explains contemporary epistemic failures… is not the decay of individual epistemic virtues, but the growing chasm between reliable collective truth-making practices and the social identities embraced by large numbers of people. And as a result, the only feasible way out of the mess we are in is not to ensure every citizen employs optimal epistemic practices, but to weaken their social identification with cranks and conspiracy theorists, and strengthen their identification with the broadly reliable collective practices of science, scholarship, journalism, law, and so on. The most visible manifestations of the problem might be epistemic, in other words, but its roots lie in social identity. And the most promising solutions will therefore tackle those roots.

Source: Blog of the APA

Image: Steve Johnson

In the wake of Big Tech propping up Trump’s increasingly authoritarian regime, European countries and organisations are taking seriously the issue of digital sovereignty.

This initiative from the French didn’t spring out of nowhere — their Open Source Docs offering was something I experimented with last year — but they’ve now got ‘La Suite’ which offers more. The site is, probably entirely intentionally only in French (of which I only have a schoolboy understanding) so I translated their About page.

It’s pretty awesome that they’re a co-op, that they see digital technology as political (which of course it is), and that they want to contribute to the digital commons 🤘

A political ambition

Digital technology is political. The choice of our creation and collaboration tools can either perpetuate the captive income of proprietary publishers or support open source alternatives that set us free. This choice comes at a price: most organisations do not have the means to design tailor-made tools or manage their own free software instances.

With LaSuite.coop, we are creating a modular offering to enable associations, cooperatives, social and solidarity economy enterprises, institutions and universities to equip themselves with the best open source solutions and be supported by dedicated teams. A world is dying, and this project is our contribution to the future that is being built thanks to the actions of all these organisations.

A model for contributing to the commons

To compete in the long term with proprietary software financed by venture capital and its quest for hegemony and profitability, we want every LaSuite.coop application to be based on a true digital commons. That is to say, free software developed, financed and governed in an open and transparent manner by a plurality of public and private actors. The Decidim project and the French government’s digital suite are among our main sources of inspiration.

With LaSuite.coop, we donate a portion of our revenue to the communities that maintain and develop the software we offer. In this way, we contribute directly to a virtuous and resilient circle of mutualisation and collective decision-making.

A cooperative venture

The organisations behind LaSuite.coop have been committed to promoting free and democratic digital technology since their inception. We pool our years of complementary experience in hosting, development, technical project management, training and strategic consulting.

Source: La Suite

A rare cross-post for me from my Open Thinkering blog. I’m really rather pleased with this and would like you to read it.

Source: Open Thinkering

I’ve been using Claude Opus 4.6 over the last couple of weeks, and let me tell you: it feels like digging with a SHOVEL.

Source: Are.na

While I expected studying Philosophy as an undergraduate to be personally useful and indirectly useful to my professional career, I didn’t forsee how relevant it would be to our increasingly AI-infused world.

In this post, Matt Mandel, after “[coming] to realize that using LLMs is pushing all of us to more closely examine our philosophical assumptions” has sketched out “a rough attempt at laying out what in philosophy is most relevant for AI.”

I’ve taken the list — covering things with which I’m familiar and those things I’d like to follow-up on — and added links. The “What is agency?" section is particularly timely, as I’ve been thinking about this a lot recently but need to do more reading.

Part 1: Philosophical Concepts

• What is a mind?

  • Nagel, What is it like to be a bat
  • Chalmers, Facing up to the problem of consciousness
  • Putnam, The Nature of Mental States
  • Searle, Minds, Brains, and Programs

• What is agency?

  • Anscombe, Intention
  • Dennett, The Intentional Stance
  • Bratman, Shared Cooperative Activity
  • Applbaum, Legitimacy (group agents)

• Who has moral status?

  • Kant, The Metaphysics of Morals (indirect duties)
  • Singer, Animal Liberation
  • Korsgaard, Fellow Creatures

• What are reasons and values?

  • Hume, Enquiry Concerning the Principles of Morals
  • Parfit, On What Matters (Volume II, Part 6)
  • Mackie, Ethics: Inventing Right and Wrong
  • Korsgaard, The Sources of Normativity

Part 2: How should we build AI

• How might AIs come to implement our values?

  • Kripke, Wittgenstein on Rules and Private Language
  • Bostrom, The Superintelligent Will
  • Plato, The Republic (Book I)
  • Anthropic, Emergent Misalignment

• How should AI be involved in governance?

  • Applbaum, Legitimacy (“domination by a stone”)
  • Séb Krier, Coasean Bargaining at Scale
  • Newsom, Snell v United Specialty Insurance Company (concurrence)
  • Cass Sunstein and Vermeule, Law and Leviathan

• How might AI impact human flourishing?

  • Aristotle, Nicomachean Ethics
  • Mill, On Liberty
  • Frankfurt, Freedom of the Will and the Concept of a Person
  • Brendan McCord, Live by the Claude, Die by the Claude

• How do we find meaning in a post-AGI world?

  • Camus, The Myth of Sisyphus
  • Nozick, Philosophical Explanations (Philosophy and the Meaning of Life)
  • Bostrom, Deep Utopia
  • Mandel, The Top Three AGI-Proof Careers and What They Reveal About Our Humanity and The Future

In terms of Part 1, to the What is a mind? section it’s definitely worth adding Turing’s foundational paper from 1950 asking “can machines think?” It’s the one that introduces the “Turing test” and directly sets up the questions that Searle and Nagel are responding to.

I’d also add a section entitled What is knowledge? and include:

• Dreyfus, What Computers Can’t Do — which argues that embodied, situated knowledge resists formalisation.
• Wittgenstein, Philosophical Investigations — not the easiest of reads, but it’s where we get notions of the impossibility of a private language and the difficulty of defining terms such as “game”.

I had to ask Perplexity for help with Part 2. I haven’t read any of these but apparently they’re relevant additions:

• How should we build AI?
  • Vallor, Technology and the Virtues — A virtue-ethics framework for emerging technology which Perplexity describes as “arguably the most important contemporary work bridging classical ethics and AI practice.” The original list focuses on alignment as a technical problem, whereas Vallor asks what kind of people we need to be to build good technology.
• How might AI impact human flourishing?
  • Crawford, Atlas of AI — I’ve had this book on my shelf for a while but haven’t read it yet. It’s a materialist analysis of AI’s supply chains, labour exploitation, and environmental costs, which sounded too depressing for me to read last year.
• What does AI mean for how we know things?
  • Floridi, The Ethics of Artificial Intelligence — A framework of five principles for ethical AI (beneficence, non-maleficence, autonomy, justice, and explicability) which apparently is “the standard reference in AI governance circles.”
  • Nguyen, Games: Agency as Art — Explores how gamification narrows our values into simplified metrics and therefore relevant to AI alignment. If we have to specify values precisely enough for machines to optimise, do we inevitably impoverish them?

Source: Substack Notes

Image: Hanna Barakat

Source: Substack Notes

As Jan Muehlig posted on LinkedIn recently, a machine-first internet looks quite different from a human-first one. For a start, when people are using AI agents to get things done — find and book holidays, compare and contrast political manifestos, various work-related things — they don’t need to be accessing pretty websites.

This post begins by talking about 13 Markdown files which apparently wiped $285 billion off the valuations of publicly-traded technology companies. Markdown files are just text files and, in this case, were part of a “knowledge work plugin” for Claude Code which just gives it more context. As I’ve already discussed artificial general intelligence is already here — it’s just in AI companies' best interests to point to it being some thing to worry about in future.

I’ve probably got more to write about this, specifically about how a future beyond the web browser and apps looks like accessing your digital world via two dominant AI companies. I’m definitely not saying that’s a good thing, that’s just the trajectory we’re on currently.

So what does agentic-first software look like? Initially I thought we would see people replace SaaS tools (intentionally or not) with their home grown versions. While that’s definitely true, the improvement in agentic harnesses and the underlying models have meant that I think there’s a whole new category ready to emerge.

Effectively, API first solutions for each vertical. These are products built from the ground up to allow programmatic access - instead of the other way round where the UI is the main feature and API access is a checkbox on their feature list.

This means really thinking through the most flexible way to offer access to data. It also means generous and fast API access to it, along with access and permissions to control and secure it at scale.

This isn’t actually a new concept - we’ve had so-called “headless” CMSs and ecommerce platforms before AI came along. But I now think we’ll see an explosion of them.

Source: Martin Alderson

Image: Deborah Lupton

There are two types of people in life: those who make binary statements, and those who don’t.

But seriously, the older I get the more it feels l like there are some people who I’m drawn to who share certain qualities. A lot those qualities are described this this this post by Dave Kang outlining his Octopus Manifesto. The “octopus” part comes from having different “tentacles” focused on different things.

It’s been over a year now that I’ve been living the Octopus Way, and my thinking around this has evolved somewhat. When I first began this journey post-sabbatical, an octopus was just a more fun way to describe a generalist, polymath, or multi-whatever person.

But lately I’ve been wanting to draw the circle a little tighter, as I think Octopus People are actually a smaller subset of this group. One of the fun things about making up your own identity metaphor is you get to define the term. So in this edition I present my updated, tighter definition of an Octopus Person. I’m interested in finding my unique tribe of people, and I hope clarifying and narrowing my definition will help me specifically find more people who are in the same boat as me.

Without quoting paragraphs and paragraphs, here’s the first sentence of each of the numbered points on the manifesto:

1. We walk our own life path.
2. We are creative.
3. We have unlimited curiosity.
4. We tend to fly solo.
5. We are octopreneurs.
6. We like meandering.
7. We are resourceful.
8. We think different.

Source: Dave Kang’s Octopus Life

Image: Julia Kadel

Source: LinkedIn

Source: Are.na

A complex system, contrary to what people believe, does not require complicated systems and regulations and intricate policies. The simpler, the better. Complications lead to multiplicative chains of unanticipated effects.

(Nassim Nicholas Taleb, Antifragile)

I no doubt shared this when I first read it, but I had reason recently to re-find this excellent post from Milan Cvitkovic containing a list of “things you’re allowed to do.”

Brewing in my mind is a longer post over at my main blog about what it means to have agency, and how to develop it in yourself and others. It’s not a fixed state, but something which fluctuates over time. It needs feeding, sometimes by giving yourself permission to just do things.

This is a list of things you’re allowed to do that you thought you weren’t, or didn’t even know you could.

I haven’t tried everything on this list, mainly due to cost. But you’d be surprised how cheap most of the things on this list are (especially the free ones).

Note that you can replace “hire” or “buy” with “barter for” or “find a DIY guide to” nearly everywhere below. E.g. you can clean the bathroom in exchange for your housemate doing a couple hours’ research for you.

Source: milan.cvitkovic.net

Image: Are.na

I like experimenting with AI tools, but as I mentioned in a recent post about the Claude Constitution, I’m not a huge fan of the governance behind them. I also get for those in the creative arts, and especially for those for whom AI has arrived midway through their career, the whole “innovation” feels like armageddon.

This post from Andrew Sempere “artist, designer, developer and internet old” starts off with a heartfelt note about how he is desperately looking for work, which lends a appropriately melancholy vibe to what follows. Having really struggled this time last year, both in terms of work and life in general, I really feel for him.

It’s worth reading in its entirety, but here’s an extended excerpt, because he’s not wrong. But what can we salvage from the ruins? What can we create together?

To the extent that any of these AIs “think” at all, it’s because the models have strip-mined the last two decades of internet conversation. The conversation that we used to have in public. All of the blog posts, commentary, free software and detailed debates, RFCs and white papers. Free documentation and APIs and Stack Overflow posts. And now, as far as we can tell, this collective generative activity has almost completely ceased to be.

We have stopped talking to each other on the open channels.

We don’t “need” to anymore.

We definitely don’t want to anymore.

The current moment is so incredibly extractive, it has converted every ounce of human care, kindness and creativity into a “model,” which we then burn as much fossil fuel as possible to convert into a subscription service. We are being asked to pay for a chopped-and-screwed memory of the time we used to live in a semi-functional society.

We are, effectively, being fracked to death. And this isn’t an impending collapse, it’s already happened, we’re just living out the consequences right now.

It’s not the internet that’s dead, it’s the entire tech industry and it wasn’t an accident, it was a murder to try and cash in on the life insurance money. We’re not into late-stage capitalism, we’re fully in The Jackpot, and I fear nothing will ever get any better than it was about two years ago, when we ingested the entire program into itself and then just… gave up… trading all of our living friends for their ghosts.

It seems increasingly likely that we’re locked in a death-cult-groundhog-day timeloop now, forever, remixing only hit-parade records from 2023 again and again and again and again and again until we’re convinced they’re new.

I almost miss the era when there was a shitty startup every week.

[…]

I have memories. Vivid ones, of living in a different country, a different city, a different planet. I look at screenshots and photographs of life from as little as 18 months ago and it becomes difficult to believe that future ever existed. I don’t even feel right in my body, nothing seems to work, everything previously solid seems damaged, unstable, treacherous and unreliable. Things have glitched, gone sideways, completely wrong timeline. I’m told I’m wrong, that things are fine, that this is normal. We’ve always been this way.

So this might be just old-man shit, but I think the word I’m looking for is: bereft.

Source & image: Feral Research

My one week of rest to give my autonomic system a break turned into two as, predictably, I caught whatever lurgy my daughter brought home from school. Two weeks away from exercise provided a bit of a reset, and I felt much better.

It reminds me of this excellent podcast episode.

Source: Tumblr