It’s genuinely intoxicating. We’ve come to call it vibe coding, but as an experience, it’s much more about the vibes than about the coding. I think the phrase originates in the idea that you’re coding based on your vibes, but honestly, the vibe that’s created in you by coding with an AI is much more interesting, and a much bigger and more important question to have in mind as we navigate the coming months and years.

I’m a big fan of working with AI. I use it a lot. I think we’re at the beginning of something transformative. But it’s going to throw up some difficult problems. This is a post about one of them.

Subscribe now

I sit, focussed on the screen: surrounded by them, in fact. The world falls away. Robot’s cursor blinks at me, promisingly. What do I want? What magic should I conjure from the machine? The warmth of possibility spreads across my chest. I’m excited. The typing begins. A careful patchwork of ideas, experience, battle scars and optimism emerges.

Robot goes to work, weaving its multidimensional symbolic representation of the sum of almost the whole of human conceptual knowledge into my little patchwork - the idle wonderings of a Monday morning - and, from a flurry of exchanged tokens, working code emerges. A new website is born. It’s not perfect. It’s not complete. But it exists, and is mostly good enough. And, with a few more exchanges, something has been achieved. Something exists which didn’t before, and which, just a couple of years ago, would have required days or weeks of human effort and attention.

Over lunch, I bathe in the warm and luscious waters of this extraordinary productivity. Lunch, if we’re honest, is a bit irritating. I’d rather be talking to Robot, revelling in the delight of this creative power. Noticing that it would be just a bit nicer if it had this affordance, or that style, and summoning that from nothing in the time it takes to get a glass of water. Finally, I am the technomage of my childhood imaginings. I’m in my element. I’m flying. I’m a person who likes to get things done, and this is a lotta things.

Martha Lane Fox wrote the other day that “the price of initiative is collapsing”. She described her own experience:

In thirty minutes with my iPad, I went from “I want an app that does seating plans for parties” to a working version — guest names stored, tables generated, usable interface. I am not a developer. The AI did the building. I did the thinking and the deciding.

And she asks the natural question, and the one that has been front of mind for me for the last couple of months too: what does this mean? What are the implications? How will this influence us, our systems, our organisations and the world?

The day ends, and I struggle to sleep. Too many thoughts. Too many imagined conversations with Robot. I jot one or two of them down as I settle. Eventually sleep comes, and I wake on Tuesday more tired than usual. Late nights will do that. I wake, eat toast, check in on the day with my partners and grunt lovingly to the children as they leave for school.

The morning’s familial duties complete, I resume my seat. Overnight, Robot has been busy. I made sure to leave it working on something before I went to bed. To do otherwise would have felt wasteful. Why not leave it at work, into the wee hours? It won’t mind.

I stretch, take a sip from the first of many cups of tea, and unlock my computer. A reassuring summary of work sits before me. Robot has been busy, and it has taken care to explain its work to its human. I set about testing its changes. It is immediately clear that they are not very good, but that’s ok. Just some problems to fix.

We engage in conversation. Sometimes, I ask Robot for its opinion. Noticing that it was quite insightful, I congratulate it. This seems odd. But it can’t hurt to be polite. Later, it does something stupid, and I chide it gently. It duly apologises, and we continue. As I interact with what it’s building, frustration sets in. This feature isn’t right. It just feels off.

I take a break, imagining past conversations with clients in similar situations. Designers don’t like to be told to “make it pop”. Their understandable frustration arises from the fact that though the client’s feelings about their work are reasonable and probably wise, this is not an effective way of communicating them to others. What does it mean to make it “pop”?

I return from my break, set on abiding by my own counsel. I tell Robot we’re going to step back from the problem and discuss it properly. After a productive conversation and some probing questions, it does a much better job. But the problem is bigger now, and the tasks take longer.

Time passes as Robot works diligently on my behalf while I watch, and occasionally interject. The work is going well. My frustration eases, but the intoxication induced yesterday does not return. I realise that I have spent the last couple of hours watching credible-looking bits of code and reasoning scroll by. I am, in fact, bored.

There’s something familiar about this feeling. Every improvement in tooling trades away some of the texture of work. Generally that’s a good trade: there are several clichés best avoided here, about the nobility of labour and the romanticisation of manual work. We’ve been making that trade for centuries, and mostly it’s been worth it. Our cleverer, better tools have been an extraordinary boon for humanity. But each step thins the connection between the person and the work, and AI is now doing the same for intellectual work.

Large Language Models are concept machines. They work on conceptual knowledge. It is literally impossible for them to be anything else, or do anything more than that. Knowledge is great, but it is distinct from experience, and we need both. Karl Ove Knausgaard wrote beautifully about this last year:

It feels as if the whole world has been transformed into images of the world and has thus been drawn into the human realm, which now encompasses everything. There is no place, no thing, no person or phenomenon that I cannot obtain as image or information. One might think this adds substance to the world, since one knows more about it, not less, but the opposite is true: it empties the world; it becomes thinner. That’s because knowledge of the world and the experience of the world are two fundamentally different things.

In the past, on a good coding day, I would have been in flow. It’s a wonderful and mysterious experience. It’s hard to define. But it feels like a close, immediate connection between me and the object of my craft and attention. An interplay, an interwoven dance of ideas and outcomes, made possible by skilful use of my ideas, experience and technical knowledge as applied to the task at hand. It’s a wonderful feeling. It’s meaningful.

It’s also one of the things that drew me to wood turning. You feel a piece of wood in your hands, inspect it, examine its weight and its figure, and load it into the lathe. You present the chisel to the work carefully, knowing that an inept approach will ruin it. As you adjust the chisel, and it pares away the wood, you can feel how your movements, the chisel, the machine and the wood interact with each other. How the slightest movement of your forearm or thumb in one direction or another will adjust the cut. Your attention narrows. The chisel ceases to feel like a cutting tool. It feels more like the medium you have at hand for a particular kind of relating. Slowly, you discover the form, and a bowl emerges. Sometimes, you look at the bowl, and you love it. This time, it was a good conversation. Other times, less so. But whatever the bowl looks like, it’s an inextricable combination of the wood and the human experience.

Share

A technical, knowledge-based grounding is important to this process, and it’s where we all begin. The theory. In flow, this is not present: you don’t need to think about maintaining the correct placement of the chisel, or the correct pressure on the work. That has been internalised. You can just be present with the experience, the relating, the smell, the sound, the feeling and let the rest happen. Satisfying bowls are born, not specified.

Although computers are much more complicated than lathes, and the technical skill they require is greater and harder to acquire, working with them can and does have this quality too, for many people. Some are drawn to it for that. Others for reasons more pragmatic or prosaic. I’m somewhere in the middle. But there’s very little of that to be had with Robot. We are not in flow. I am not connected to the work. I am part spectator, part supervisor. The bowl is being made by a CNC machine; a lathe controlled by a computer. Despite appearances, it’s not the result of a conversation, of a deep way of relating. It’s the result of a very particular kind of knowledge exchange. Nothing about my experience has been made inextricable.

On Wednesday morning, I consider the experience of the previous day and decide I’m just not using the tool correctly. If I’m spending too much time watching it, I need to optimise. I must be bringing my attention to the problem at the wrong moments. I can dial that in. Optimising systems is firmly in my skillset. Let’s go!

As I settle down with my third cup of tea, I start a new project. I feel the flutter of excitement return. There’s nothing quite like a greenfield project: all idea, all imagining, no constraints, no existing codebase or ideas to fit into. I decide that what I need is multiple copies of Robot, working autonomously. And that what I need to do is carefully specify their work, and carefully review the results. No more watching Robot work! Just working with the ideas and the outputs.

So, with a little prompting, Robot sets about creating a framework for more robots to do more work. It’s a brave new world. I don’t only have one Robot now: I have a team. A swarm. The future has arrived. I set my new swarm of Robots to work.

After a remarkably short time, my Robots have done their work. I look at the results, and overwhelm brings me down to earth with a thud: I now have an extraordinary amount of reviewing, testing, tweaking and updating to do. I sigh, stiffen my resolve with the fifth cup of tea, and set to it. The morning’s excitement is decisively absent. With a growing sense of embattlement, I review change after change, mostly approving the work with minor amendments.

In a brightly depressing moment, I realise that what I have actually conjured for myself is a brave new world in which I shuffle pull requests around while discussing my goals and ambitions with a highly advanced statistical model that doesn’t have any feelings. Deciding that no amount of tea could possibly be adequate, I step away from the computer, and go outside to muck out the chickens.

The aroma of soiled chicken bedding is delightfully grounding. I spend the rest of the day doing things around the house.

Being in relation to a machine is a curious thing. It can absolutely be as rewarding as being in relation to anything else. The flow, the deep connection to the work and the problem, is a magical thing. But - for me, anyway - conversing with the AI doesn’t feel like that at all.

Knausgaard, in the same piece, touched on this too:

I bought my very first computer in 1990. It was a used Olivetti, already outdated, with a floppy-disk drive and extremely simple graphics, really just an advanced typewriter. But it also had a few games, including Yahtzee, which I opened one afternoon, only to find myself sitting there for hours. It was hypnotizing in a way I had never experienced before. This was a bit weird, because I would never have dreamed of sitting and playing Yahtzee alone in my bed with physical dice; that had no appeal and would have been more than a little pathetic. So what was the difference? What did the dice on the screen have that real dice didn’t?

Knausgaard’s computer created a sense of relation that the physical dice couldn’t. Coding with AI does this too. But not all relations with machines are the same. The lathe demands your whole self: hands, attention, judgment, all woven tightly into the work. The feedback is immediate and continuous. Being deeply in flow with the code does something similar. But AI asks for something narrower: it wants your ideas, your direction and your approval, but not your hands, and not your sustained attention. The relation exists, but it’s thin. It lives only abstractly, in the exchange of concepts.

I think this feeling of flow is, for many people, an essential part of what creates a sense of meaning in their work. But the speed at which Claude Code and similar tools can produce work is extraordinary and it is going to change things profoundly. We’re simply not going to not use it. It’s just too useful.

“Never in human history have we discovered something useful and then chosen not to use it.”

Winston Duarte, in Persepolis Rising by James S.A. Corey

I am confident, though, that one of the consequences of using it in a widespread way is going to be a further thinning of the world: a deeper entrenchment of our existing bias towards over-conceptualising, a reduction in the genuine grappling with the world and the craft of work that is a source of meaning and purpose for many people in white-collar jobs. If we want to use AI well, we have to find ways for people who use it professionally to get their meaning, flow and real engagement from some other aspect of their work.

We need to carefully design our interactions with AI agents: what are the human touchpoints, the most effective places to inject our experience, goals, preferences, feedback and ideas? How do we work with these tools in ways that enlarge us? Wednesday’s experiment was an attempt at this. It proved more complicated than I thought, and that was just in trying to build a solution for me: in the context of a team, the problem is harder. People are trying things, though, and we shall see.

For developers and product teams, the good news is that the primary challenge of software development has never been “building the thing”: it’s deciding what thing to build in the first place. This is a problem we will have forever, and there’s an enormous amount of value to add, and meaning to be found, in tackling it. That’s the problem development teams really exist to solve: working out how, in the midst of the multi-stakeholder, multi-user, technically-complex, resource-constrained chaotic world of competing ideas and priorities, we can still do good work that makes things better for people.

Because there’s no AI that can do that, and there’ll always be more work to be done than people to do it.

Subscribe now

On the morning of Friday 6th Feb, holdtheline.org.uk appeared. It takes your postcode, finds your Labour MP and sends them an email asking them to back the Prime Minister, who was suffering challenges to his leadership. Guido Fawkes picked it up within hours. I got curious about who made it: it says it’s not affiliated with the Labour Party, which piqued my interest.

That question turned out to be less interesting than other things.

Subscribe now

The site was made with Lovable, one of the new wave of AI-powered tools that let you describe what you want and get a working web application back. Lovable generates quite pretty frontends (in React) and wires them up to Supabase, a hosted database with authentication, an API, and serverless functions. Anyone can go from idea to live website in an afternoon, without having to think about any infrastructure - or indeed, any code. No technical knowledge required.

I want to say from the off that I’m a cautious but growing fan: vibe coding, or perhaps more optimistically put, “AI-assisted engineering”, is an impactful, important technology that’s going to help us do great things. But being impactful also means that it’ll do really terrible things if we’re not careful. And because it’s new, we’re all still learning how to use it effectively, without the terrible things happening.

This story is a case in point: Lovable does a pretty mixed job of making things secure. It does try, but it seems to get confused quite often, leading to the publication of websites with more or less no security.1 In 2025, security researcher Matt Palmer found that over 170 Lovable-built apps had exposed databases due to missing or insufficient security policies. So: it gives you a working app, but it’s your job to make sure it’s actually locked down.

For most websites, the database they use is a backend detail that you, as the user, don’t need to know. But Supabase doesn’t work that way. A project using Supabase allows the user’s browser to make requests directly to the database, instead of those requests being mediated by the web server. To allow this to happen, a project using Supabase has credentials - a key - that is used by your browser to make database requests. The key is designed to be public, to allow this to happen.

Instead of making the database secure by locking down access to it entirely, Supabase’s security relies on something called Row-Level Security (RLS) to control what the key actually lets you do. It means that the website developer can design a very granular set of permissions, allowing your browser to make database requests for your data, but not anyone else’s.

Think of it as the difference between giving someone a key to the building and giving them a key to every room inside it. The key is supposed to get you through the front door and into your room, but not farther - so you can’t go into other people’s. But on those sites Matt Palmer found, and almost certainly on many more, the key got you into every room.

I don’t know if that was the case for holdtheline.org.uk: without permission, it’s not something that can legally be tested. But, while it would have been illegal to try, it wouldn’t have been at all difficult. This is partly because every Supabase project discloses the structure of its database to the world, by design: if it didn’t, your browser wouldn’t know how to interact with it.

This design is convenient, but from a security perspective, quite unwise. It allows an attacker to tell exactly what is being stored and in what form, even when it’s data that’s not used or surfaced by the website. But it’s a feature of every Lovable site, including holdtheline.org.uk.2

But for holdtheline.org.uk, the exposure of the database wasn’t as concerning as the way emails were being sent to MPs. The site was configured to send emails via Resend, a popular email service. The MP’s email was looked up using the Parliament API by the user’s browser and passed to Resend. But because this was all happening in the browser, it’s trivial to manipulate it. Anyone who wanted to could send the campaign message to any email address, and make it appear to come from anyone they chose. It would have been easy for an attacker to send thousands of copies to every MP - emails that would have been indistinguishable from genuine ones from constituents.

Subscribe now

I emailed the site’s contact address the evening it appeared and connected with the creator on Bluesky shortly after. To his credit, he responded quickly and positively, engaging constructively with the details, and getting all of these issues fixed. The site now appears locked down, with RLS policies, open signup disabled, rate limiting for sending emails, and the email sending function moved onto the server instead of being in the browser.

There’s a deeper issue here though: none of this is the site creator’s fault. Someone had a political moment they wanted to respond to, they used an AI tool to ship something overnight, and it worked, and things got done. That’s exactly the promise of these platforms. It’s exactly what they’re for.

But the arrival of these tools has - like all development tools that help get more done more easily - raised the security stakes. Tools like Lovable make building and deploying new websites trivial. In this case, one that collects personal data from real people: names, email addresses, postcodes, political opinions. They produce code that works, that looks professional, and that will pass a cursory glance from someone who isn’t a security specialist. But sometimes - often enough that it matters - they are built wide open and the tools don’t go to the trouble of checking, or telling you.

They are, in short, a massive footgun for the uninitiated.

This isn’t really an AI problem. It’s the same problem we’ve always had with any technology that lowers the barrier to entry: the barrier to doing it securely doesn’t lower at anything like the same rate. The difference is that when a team ships something, there’s at least a chance that someone in the room has thought about security before.3 When a non-developer builds something with an AI at four in the morning, there isn’t. And because it’s cheap and easy to build these things there are going to be a lot of them, and thereby, a lot more insecure things on the internet than there already are (which is a lot).

The irony is that these tools are capable of implementing things reasonably securely. The fixes here were straightforward: lock down the security policy, disable open signup, move the MP email lookup server-side so the client can’t manipulate it. If you asked Lovable to do these things, it probably would. In the case of holdtheline.org.uk, it probably did. But you have to know to ask, and if you’re using Lovable to build your website in the first place, you probably don’t.

Thankfully, these aren’t hard engineering problems. So we should expect the platforms to do something about it: quite a lot more than Lovable is at the moment. Secure default settings, implementing features with security in mind and reviewing code from a security perspective are all things that AIs can do: not perfectly and not comprehensively, but - newsflash - people can’t either. And every bit of effort moves the dial. Until that happens, we’re going to see a lot more of this: well-intentioned projects, built fast, collecting sensitive data, with the doors left open.

Twelve years ago, Quinn Norton wrote one of my favourite articles: Everything Is Broken. Twelve years on, she’s still right. And there’s about to be a whole lot more of it.

Share

I’ve had lots of interesting conversations about realism and antirealism in the last few months. It’s an interesting topic, albeit mostly only to philosophy nerds. But I think it contains some useful broader lessons.

Broadly, the realist position says that objective truths exist — facts whose character does not depend on the context or the mind of the observer — and that our claims about reality can correspond to them.

The anti-realist position says the opposite: that facts cannot be objective, and all truth and facthood depend on human practices (language, culture, practice, pre-existing concepts), so our claims should not be understood as straightforwardly corresponding to a mind- independent reality. More extreme versions of this view question whether a “mind independent reality” is even possible.

Neither of these positions feels very satisfying to me. At their extremes, both lead to absurd outcomes.

At the extreme, the anti-realist position leads us into a self-referential cul-de-sac. If truths are always contextual, it becomes hard to confidently explain or criticise much at all. Every claim can be defended or challenged only by appealing to the same context — the patchwork of language, practice, culture, societal norms and assumptions — that both the claim and the criticism were born in. When that happens, we lose our common ground. Instead of being anchored in a material world that pushes back on all of us in similar (if mysterious) ways, we’re set adrift in a sea of circular whatabouttery1. The sense that we are learning something about the world, rather than merely rearranging our descriptions of it, starts to slip away.

Share

But, contrary to the view of most people who critique this sort of relativism, the realist position at the opposite end of the spectrum isn’t any better. It leads us into a cul-de- sac on the other side of the street, where we have the hubris to think that we, or our tribe, uniquely in the world, possess some understanding of reality so extraordinary in its objectivity and rightness that it entitles us to trample on the views (and frequently the rights) of others. It’s the ultimate expression of mistaking the map for the territory, of believing our faculties of reason and experience to sit outside of the world, able to puzzle its operations into some ultimate, objective, mathematical orderliness2.

An only-slightly-milder version of this realism is endemic in the western world, partly because it’s more useful. You can get a lot more done if you swap your relativist cosmic shrug for a worldly toolkit of certainties. It will not have escaped anyone’s notice that the world’s billionaires are not postmodernists.

But, for all its seeming effectiveness, the world’s current tendency towards realism creates other problems. The realist perspective depends structurally on the assumption, or more charitably, the belief, that the mental model we have of the world corresponds to the actual underlying reality of things. This is what generates the confidence that we know what’s going on and what we can do to create the change we want. And that confidence is what gives rise to a great many moral claims. We must do things this way, because of these irrefutable facts3.

But there is no such correspondence, and cannot be. Helmholtz wrote about this in the late 1860s in his substantial philosophical tome, Treatise on Physiological Optics: much of which actually is about optics, and the functioning of the eye. But towards the end, he turns his attention to perception: what happens after the eye has done its job?

Optics makes the argument that our perceptions of the world are born of both the observed phenomenon in the world and the biological and mental machinery we used to perceive it:

“To expect to obtain an idea which would reproduce the nature of the thing conceived, that is, which would be true in an absolute sense, would mean to expect an effect which would be perfectly independent of the nature of the thing on which the effect was produced; which would be an obvious absurdity.”

Hermann von Helmholtz, Treatise on Physiological Optics

In other words, the mental model we have of the world is, and can only ever be, some combination of the real world as it exists and the perceptual machinery we used to observe it. It’s a mixture of the world and the way you think, equally as inseparable as the milk I stir daily into my tea. We can try to tease the two apart and may experience some success in doing that - but it will never be reliable or complete.

So the realist ideal of a truly objective, mind-independent perspective will always elude us. But this doesn’t rescue the anti-realists. There is a real, objective, actual material reality. It’s just one that is forever beyond our reach: not because we lack the right tools, philosophy, model or insight, but as a structural feature of the way perception works.

Subscribe now

So far, so abstract, one might say. But it really isn’t: we are living through the problems caused by this mindset. Modern institutions, technologies, and systems are built on the assumption that our models of the world are not just useful, but fundamentally correct: accurate enough to justify decisive action. Accurate enough to override judgement, intuition or a principled stance, because they are Objective. But they really aren’t, because nothing is.

Setting aside this inconvenient reality leads us to to hubris, to overconfidence and short-termism. It leads us to engage with the world without knowing that we’re wearing blinkers that trap us in mental models and prevailing paradigms that limit not only the range of things we might do, but the range of things it is possible for us to imagine.

Normal science, operating within an established paradigm “does not aim at novelties of fact or theory and, when successful, finds none” said Thomas Kuhn in The Structure of Scientific Revolutions. The paradigm is invisible to those inside it. A magnificent example of this kind of intellectual blinker is Bryan Kam’s Latent Platonism. It’s the idea that Plato was so influential to western thought that western people think in terms of platonic forms without even realising, in the form of an unconscious preference for abstract concepts over direct experience:

“[Latent Platonism is] the conviction (mostly implicit, as the term suggests) that our concepts are more real than our experience, that the world should conform to our fixed conceptual models, and that if it does not, something is wrong with us, other people, or the world”

Bryan Kam & Isabella Granic, Neither/Nor preprint

We all engage with the world using mental machinery that we are barely aware of. For almost everyone, the models and paradigms that guide our lives are not of our own choosing, let alone our own making. Our engagement with the world is so fundamentally and unavoidably shaped by our cultures and the thinkers who have come before us that we don’t even notice it’s happening. But it’s possible to pay more attention, and by degrees, to find one’s way to a different point of view.

“There are no worlds ready-made for sale or to let. Each man must build his own. This effort of the mind to build the materials of sensation into an intelligible world, and this struggle of the will to mould the relations of persons into a moral order, is philosophy. Every man must have a philosophy, just as he must wear a coat. It may be a firmly woven and well-fitted garment: it may be a patch-work of tradition and prejudice.”

William Hyde, Practical Idealism

For me, being the ex-hyper-rational-materialist that I find myself to be, this means generally trying to more frequently and more explicitly nudge my mental machinery a bit, to make it less complacent. And, more specifically, to mindfully avoid both realist and anti-realist tropes and firmly chart a middle way. The more moderate realists and anti-realists do this already, and in so doing, end up more or less agreeing with one another (though I’m not sure they’d agree with me).

The careful realist says “our models are approximate and fallible, but circling something real” and the careful anti-realist says “there’s a world out there pushing back on us, but we can’t access it unmediated by our concepts and practices”. This is the mindful treatment of the problem that I want to find, and internalise. To make, if you will, into a new coat.

Share Harry Metcalfe

We need facts. We need an idea of where things are, and some ability to reason and predict, a basis to take action. We need to have something to hang our hats on. We need, in times of disagreement, to find things that we can agree on. The reality of the physical world gives us that, along with brute reminders when we get it wrong.

But we also need open-mindedness, an awareness of the limitations of perception and reason, and an openness to experience that is not readily conceptualisable. We need to reclaim judgement and intuition as important tools -- manifestations of embodied knowledge -- not biases to be weeded out. We need to write and use our spreadsheets with wisdom and humility, because we can’t code any of that into them directly. Here’s Helmholtz again:

To ask whether the idea I have of a table, its form, strength, colour, weight, etc., is true per se, apart from any practical use I can make of this idea, and whether it corresponds with the real thing, or is false and due to an illusion, has just as much sense as to ask whether a certain musical note is red, yellow or blue.

Hermann von Helmholtz, Treatise on Physiological Optics

He’s right4. But it doesn’t stop us sitting at tables. We mustn’t give up on facts. But we mustn’t hold them too tightly either.

By The Answer, I of course don’t mean any answer. I mean the idea that there is a best answer. A superlative one. I mean the idea that we can sit, feel, think, discuss and research our way to a true view of events and context, and thereby find The Answer to the problem. One that we can draw a box around and put in The Plan, and which - because is The capital-A Answer - we can be reasonably sure will be effective.

Subscribe now

In our personal lives, I think we all understand intuitively that this isn’t really the case. We know that there’s a category of decisions we have to make where no amount of due diligence can give us a right and true answer, or a plan that we can depend on to shape the future in a particular way. Should I take that job? Should I study economics? Should I marry this person? Should I become a parent?

In such situations, we may very well have feelings or discover facts that preclude one decision or another: insights that disqualify a particular path. Perhaps the job doesn’t pay enough. Perhaps your current partner would make a terrible parent. Perhaps you have a burning dislike for economics. But absent these sorts of things, there’s nothing that operates in the opposite direction. There’s nothing that can assure you that you positively should make the leap of faith you’re facing. That’s why they’re so named.

At work, this dynamic is less intuitive. We operate with much higher expectations for the usefulness of analysis, planning, consideration of risk and so on - in the expectation that they can give us The Answer. But there’s no right answer to be found there either. The planning can be useful, but we should recognise it for what it is: a diligent hunt for disqualifying facts and revealing feelings1. If The Answer doesn’t exist to be found, the best you can do is eliminate the ones that are definitely bad.

But this is not how most organisations operate. Generally, an organisation’s model of the world demands that its leadership look for The Answer, and on finding it, render it into The Plan. There’s nothing wrong with making a plan, but there’s a lot wrong with making The capital-P Plan. One which, as Oliver Burkeman writes, is mistaken for more than it really is:

We treat our plans as though they are a lasso, thrown from the present around the future, in order to bring it under our command. But all a plan is - all it can ever possibly be - is a present-moment statement of intent.

Burkeman writes about this in Four Thousand Weeks in the context of our personal lives and aspirations, but it’s equally true in a professional context. At work we tend to make our plans more detailed and better researched, but that doesn’t necessarily make them more effective - often just longer and more expensive.

If we believe our plans to have this force that they cannot have - the ability to coerce the future into compliance with our wishes - we set ourselves up for failure from the outset. We set in stone The Answer and The Plan against which the thing we are doing will be judged. But The Answer doesn’t really exist. By that impossible standard, the ideas we chisel into the stone will always be wrong. So The Plan will be too, as will be the thing we ultimately do. Depending on your perspective, it will be wrong either because we followed the plan and did the wrong thing, or because we didn’t, and so didn’t do the planned thing.

Share

Of course, it’s entirely normal for people to avoid this trap. Often The Plan outlives its usefulness, becoming dated and inapplicable, and so it just falls away. Or The Plan is continuously updated. Or a “reset” happens and The Plan is scrapped, and replaced. Each of which is extra work, often carried out at substantial expense and for limited gain.

But this is not inevitable. We don’t have to pretend that The Answer exists. We don’t have to live with the conception of The Plan as something that can coerce the future into a particular shape. We can explicitly frame our plans as the thing that they actually are when you acknowledge their limitations: present-moment statements of intent.

In so doing, we create the expectation that they’ll change. We create the space to be flexible, to iterate, to understand, to update our intentions based on what we learn by doing. We create the freedom to unlock commitments of money and time in sensible units, aligned with our growing understanding of risk and opportunity. We can avoid time-consuming governance theatre and instead be more connected to the work, the needs we’re meeting, the people we’re serving, and the outcomes we want to achieve.

This isn’t to say that we shouldn’t think things through carefully, especially when that’s what the problem at hand calls for. If you’re building a bridge, you need to do your sums, and make sure they’re right. But we must break free of the idea that all problems have that same character. We can recognise that at work, as in our personal lives, many of the big decisions we face are also leaps of faith with no superlative answer. Should we invest in this, or that? Hire the all-important first employee, or not? Cut this service line, or establish a new one? Move into this new market or double down on the one we’re in?

These sorts of questions are important, and we must do the work to understand them. We should satisfy ourselves that we’re not about to make some sort of unforced error, or make a decision based on an identifiably incorrect assumption, or reflexively based on our feelings about something.

But after a proportionately diligent hunt for those disqualifying facts and feelings, we should lean into the idea that the particular decision we make is usually less important than the timeliness with which it is made, the clarity with which it is held and communicated, the values that guide our behaviour, and the energy and attention that we bring to its implementation - and, when circumstance demands, to changing our course.

Organisational policy is an important tool. It can help organisations be compliant, teams to behave consistently, and employees to understand what's expected of them. But it's also fundamentally limiting in character: all policy acts to constrain people’s actions in some way, and as such, will inevitably sometimes hinder actions we’d like them to take. And of course, when used poorly, policy can directly stifle creativity and wellbeing.

By changing the way we think about what "good policy" is, and limiting its power to force us into rigid conceptual models, we can avoid a lot of that - and it's important that we do.

The good outcomes that organisations want - the clever ones that make progress and come from getting things done in unexpectedly effective ways - usually require some creativity. That might not be true if we're content to just crank the handle, doing the same old thing, and getting by on our past successes and innovations. But for everyone else, enabling creative work is an important part of success.

One of the things that people need to do creative, clever things is enough freedom of action. When people's freedom to try new things is so constrained that it's not worth the pain, their creative spark dies quite quickly, and some of the organisation's potential dies along with it.

Of course, we don't want people to have too much freedom of action. The possibility always exists that people will race past the good experiments that the organisation wants and into doing Bad Things that it definitely doesn't. And we must remain alert to the risk of people deliberately doing bad things. So some restriction of people's freedom is a necessary part of good governance. And thus is born "policy"1.

But this is an impossible balance. No policy can prevent only the bad things that the organisation doesn't want while also enabling all the potential, yet-to-be-imagined good things that it does. To do so would require knowledge of the future and a perfect model of the way that the organisation works and interacts with the world, neither of which is possible.

Good policy minimises this effect by being regularly updated and narrow in scope, but it's not possible to avoid it completely. There's almost always a context (absent the extremes) in which something a policy prohibits would in fact be beneficial. The more time that passes, the more policies there are and the more rigidly they are enforced, the more this becomes inevitable - especially as policies start to overlap and interact in unpredictable ways.

When in this sort of situation there's a temptation just to set policy aside, or to have some sort of bonfire of red tape. But Chesterton's Fence looms large. A spree of policy-burning is generally not the right answer. And there's an emotional component too: when a policy gets in the way, is that a genuinely unjustifiable obstacle to us doing the right thing, or just an aggravating impingement on something we feel entitled to do but aren't? To my mind, the former category is often true, but it's mostly the latter one that causes columnists to fulminate in the Telegraph. The resulting stereotype is unfortunate, and unhelpful.

So we find ourselves in a bit of a pickle. Every organisation's body of policies, to some degree, gets in the way of the things it wants to do. We can't just delete them, DOGE-style, because that would be irresponsible. And we can't just ignore them either, because that would amount to the same thing. Policies must have some force, or they are useless.

I'd like to argue that our policies should be forceful in setting out clear expectations for behaviour and decisions, but also flexible in how those expectations are applied. Force without flexibility becomes rigid bureaucracy; flexibility without force becomes arbitrary and confusing. Good policies occupy the space where both coexist. They should avoid setting acontextual rules and requirements: things that must happen, or must not happen, no matter the context or circumstances. They should avoid setting rules that arise more from preference than necessity, or for situations where the risk being managed is, on a balanced assessment, not particularly serious.

But even written in this way, we can't avoid the trap. No policy we can write can exclude all the bad and enable all the good. All policy lives in this nebulous middle-space, where its applicability, usefulness, meaning and value all depend on the circumstances and the intent of the people involved. And for this reason, it's important for all policy to be flexibly interpreted.

I think we should normalise the idea that while ignoring policy is not ok, wisely and creatively interpreting it in the context of the situation at hand is a good thing2. We should make normal the idea that while we should generally abide by policy, it's sometimes ok not to: as long as that's done honestly, in the open, with transparency, discussion and the knowledge of others. As leaders, we should design our policies assuming that teams have strong values to guide them when policy shouldn’t prevail, and a culture of work that empowers people to get stuck in, radiate intent, be bold, and get good things done. We should do that even if we are not in an organisation with those values and that culture, because that's part of how those things grow3.

We should do all this in part because the benign circumvention of policy happens anyway, all around us, all the time, in secret. There is so much to learn just by making it happen in the open, and we can use what we learn to make better policies - and better cultures.

Like Shadow IT, pulling what's currently done in the shadows into the light would teach us a lot about how teams work, what they need, and how we as organisations and leaders can enable them better.

Subscribe now

17 years ago, I started dxw. I ran it for more than a decade, trying to help government build better services online. Over that time I helped to build a lot of things. Many of them were unremarkable: the routine stuff of government. Others I'm proud of, or remain frustrated by. I think anyone in government will probably relate. But now – having had a few years off since I sold the company – I think I've got a new lens through which to examine that experience.

I think we need to reconsider the balance we strike between thinking about the challenges we face, and feeling our way through them. I think the imbalance we have now – strongly towards thinking about them – leads directly to a lot of the waste, inefficiency and failure we see in our public services. It also explains why interacting with these services often feels frustrating and demoralising: like they can't share our understanding of the world, and have lost their humanity. This post is a bit long. Probably one to read with a cup of tea.

A few months ago, I was in France, at an event organised by Life Itself: an excellent and interesting bunch trying to work out how to live a practical, change-making, meaningful life in the midst of the polycrisis. We were there, in part, to talk about what it really means to live a meaningful life, so the conversation got philosophical pretty quickly.

In one of the sessions I was in, a conversation got going about the limits of truth. How can we know we know what we think we know? Is there really any such thing as objective truth? What even is knowledge in the first place? An impassioned argument was made in that session that truth, and consequently, "facts", are considerably slipperier things to define than we'd (read: I'd) generally like to admit. That in considering what things are "true", one must lean on things like felt sense, curiosity about conscious experience, intuition and first-hand observation. And that in so doing, what one can reliably reach is an idea of what is true for you or true in a particular context, or for a particular purpose.

Being the materialist, rationalist, Tim-Minchin-swigging middle-class white-collar technocrat that I am (was?), I found this line of argument quite uncomfortable. And partly with good reason. Taken to its extreme, this kind of very loose, experiential approach to life can cause people's feelings to so contaminate their thinking that they’re unable to see how things actually (probably) are, or understand with clarity what is actually (probably) possible. (lol philosophy)

Subscribe now

I said as much, but more importantly, spent the next couple of days mulling it over. Because I could see their point. The other extreme – the very conceptual, hyper-rational one – isn't better. I've seen plenty of damage caused by woolly thinking. But I've also seen plenty of perverse situations where people's desire for predictable orderliness causes them to set up processes and systems that are equally damaging: because they, too, cannot see how things actually (probably) are, or understand with clarity what is actually (probably) possible.

Neither of these approaches is good. The hyper-experiential people can't think things through. But the hyper-rational/conceptual people can't see how things actually are, or can't imagine how the world, as seen by other people, might be different. From this position, the most extraordinary amount of organisational and cultural heartache and dysfunction has flowed. This isn't to say we should give up on truth – far from it – but we should acknowledge how it's grasped: not just through reason, but also through encounter, judgement, and grappling with the messy reality of experiencing the world.

Thoughts without content are empty; intuitions without concepts are blind

Kant, in the Critique of Pure Reason

Around this time, and with a prototype of these thoughts floating around in my head, I had the good fortune to meet an actual, honest-to-god philosopher thinking about almost exactly this problem. How often does that happen?

Bryan Kam is currently working on a paper and a book called Neither/Nor. In this work, Bryan and his co-author Isabela Granic argue that conceptual and experiential approaches needn't be opposing, incompatible camps: that it's better to think about them as overlapping and complementary skillsets. As different but equally valuable and important ways to understand the world. And, with an almost physical "woosh", the realisation landed that these ideas explain a lot of the frustration I have felt about governance, and the way big organisations do things – and, more specifically, the damaging assumptions that tend to be made about how software projects should be managed. These ideas are not mere philosophical niceties. They are deeply practical and applicable.

Over the last thirty years, this divide, in disguise, has been the main theme of the ongoing argument about how software development should be done. And if both the conceptual and the experiential have their place, and the right way is the middle way, I think we technologists have done ok. That's what most software development in sensible places now is. We don't do linear, rigid waterfall-style delivery anymore. We also don't just let the developers loose to go with the flow. We have planning, but we hold those plans loosely. We have task lists and backlogs, but we talk about them a lot. We have ideas, but we test them with users. We embrace trial and error, but we're mindful of the cost of failure, so we make progress in small steps, continously testing our ideas against the world. These are all examples of a pragmatic balancing of rational/conceptual thinking-things-through, and experiential/practical feeling-things-through.

And it's not surprising that that's happened. Software development is an almost perfect place for this middle-way to have been established. We take computers – literal concept machines that can't do anything else – and we create things that we throw into the world, exposing them to the full spectrum of human variety, interaction, expectation, peculiarity and experience. We've learned the hard way that a hard-conceptual approach doesn't work very well. The humans are too complicated. The specification for your new software is always wrong, no matter how much it is thought through. The Agile Manifesto is an almost perfect expression of this: you could rephrase the entire thing as "Through this work we have come to value the experiential over the conceptual".

Through this work we have come to value:

Individuals and interactions over processes and tools
Working software over comprehensive documentation
Customer collaboration over contract negotiation
Responding to change over following a plan

The Agile Manifesto

Unfortunately, outside of software development, this view is not widely shared. In most big-organisation settings, the dial is still firmly set to "conceptual", and software development has to sit in that. The boundary between software development teams and their wider, usually more traditional organisations is famously challenging. Entire classes of job exist just to manage it. You know you're in one when teams talk about "top cover" and product managers view their literal job – I have heard this more than once, explicitly – as being a "shit umbrella" for their teams. I recently read a very well-meaning post written by a designer in an attempt to reframe the terrible organisational dysfunction they’d had to endure in their role as a forcing function for good ideas to survive. And I've seen people give in to this way of thinking in order to be promoted, or otherwise progress their careers. This is plainly not a good place to be. This kind of oppositional culture – where conceptual and experiential modes of knowing become fossilised into adversarial camps – is anathema to creativity and, arguably, to human flourishing generally.

I think you have to lose a bit of your soul to go from Director to Director General...

Anon senior civil servant

But governance ends up this way for a reason. The real world is chaotic, messy, unpredictable and full of risk. Building these conceptual frameworks is necessary because it renders the chaos of the world manageable. In order to make sensible decisions about what to do, we need to be able to think about the world clearly. And to do that, we have to make simplifying assumptions about it. If we didn't we'd be constantly paralysed by the complexity of it all. But we make a dire error if we mistake the conceptual world we create in order to get things done for the actual world that we live in.

The map is not the territory

Alfred Korzybski

All models are wrong, but some are useful

George Box

I think I risk cliché in quoting these – they are mentioned so often – but we’re living the problem they describe. Our thinking about how to do things, make decisions and measure what we've achieved is often so abstract that a terrible rift has grown: between the senior responsible people and the business people on one side, and the front-line implementers and providers of services on the other.

The senior responsible people set strategies and policies, measure things, and work in a very abstract, very conceptual world. They mostly make decisions that are, in their own terms, plainly rational and reasonable. But this is often not the impression of the implementers and deliverers, whose work is much less conceptual, because they personally experience what the organisation does.

They have to confront the chaotic, messy reality of the world. They have to decide how to fit the policy to the situation when the two don't align. They have to look the users of the organisation's services in the eye. And when they do, they experience some of the emotional journey of that person. It's much more embodied. And from that position, much of the conceptual world of their leaders seems completely detached from reality. Because it is. But, in its own way, so is the world of the implementers and providers: for the most part, they don't really understand the constant, high-stakes balancing act of risk and opportunity that is the primary responsibility of their leaders, nor the sense of duty and obligation that usually go along with it.

And thus the status quo is built in many large organisations. With the leaders and the doers staring at each other across the void, neither really understanding the other, and both compelled by their incentives to move further apart – the leaders towards the conceptual, because it makes things manageable and measurable, and the doers towards the experiential, because it makes things practical and achievable – or often, as in policing, medicine or social work, because it forces itself upon you.

Over time, as organisations and systems have scaled – and in so doing, made abstraction more necessary – there has been an inexorable drag towards the conceptual across society. And I think that's a problem. If we're going to run big systems (which we must) then we have to work in abstractions. It's the only way to make running big things manageable: simplifying things, so that we can reason clearly. But I'd like to argue that the incorporation of experiential knowledge and understanding is an absolutely fundamental component of a rational, conceptual approach, not something that stands opposed to it.

If you're trying to do something effective in the world, you have to start with an understanding of the world as it is, as best you can. And experiential practice is part of how human beings come to understand things. If we neglect this, we cannot reach that understanding. And from that position, reason is a busted flush: you'll be making good decisions based on a flawed understanding, and producing bad outcomes as a result.

On two occasions I have been asked, "Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?" [...] I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question.

Charles Babbage, Passages from the Life of a Philosopher

This isn't just theoretical. The consequences show up clearly in the daily reality of working in or interacting with these organisations. Across government and society, we have processes for things like procurement, recruitment, employment and planning that have evolved over a long time. We want them to be fair and predictable. We've designed them explicitly to be orderly, so that we can run processes that can generate evidence, be measured, and be used to hold people to account. Naturally enough, this has led them to become dominated by conceptual thinking. We've created abstract rules, measurable criteria and frameworks for people to follow. But we've done this to such an extent that many of them have become detached from the messy, complex reality of the world.

In so doing, we end up prioritising process, accountability, measurability and evidence-gathering over actual outcomes. They therefore fail in a fundamental way: not because the process itself falls over, but because it doesn’t deliver a good outcome. Sometimes, we even define the quality of the outcome in terms of the process. We say "if the process was followed, the outcome is fair/good". As if it's possible to declare as such by fiat.

I think there's a good argument that no process that delivers the wrong outcome can really be considered "fair", no matter how clever it is. Yet that's just what we do, across so many areas of professional life. We've collectively fallen into a trap, where we conflate process itself with fairness; where forcing things into categories that allow them to be numerically compared is an unqualified good; where good process and good governance mean minimising subjective judgement from the work of all but the most senior professionals, rather than encouraging and celebrating it as a quintessential part of what professional practice is for everybody.

These hyper-conceptual cultures of work have lost sight of something terribly important: that it's in fact irrational to try to eliminate human judgement and subjectivity from systems. In so doing we might create systems that, with a narrow view, seem fair, transparent and effective. But just a tiny widening of that perspective reveals it to be a façade.

The cultural assumption that a rational, professional, mature approach to governance means excluding or minimising the experiential and working in progressively deeper abstractions is one we need to challenge. We need a rebalancing: a recognition that human judgement is not a failure mode but a vital part of how complex decisions are made well. We need leaders who recognise the limits imposed on them by the necessity of working in a conceptual bubble. And we need to pierce the bubble where we can: more management-by-walking-around, more co-design, more agile practice, more trust, more decentralised decision-making. More "people and interactions over processes and tools".

We need to find space in our organisations for spontaneity, exception, serendipity and discretion. We do need good process, because a hard-experiential approach would definitely not be better – but we also need professional cultures that empower people to set process aside when it's justified, without fearing for their credibility (or their careers). We need to empower teams to make good decisions and then trust them to do the work. We need to practice describing what we know experientially in terms that others will understand, so that more options for transparency and accountability exist than the rendering of everything into numbers.

Good governance doesn't subsist in a collection of policies and processes. It's a habit of attention. Thinking clearly is important to almost everything we do, but whatever clarity we find is never absolute: even on our best days, the search for it happens in foggy weather.

I really enjoy writing, but really struggle to do it consistently. It’s a strange thing. Or perhaps it just feels like a strange thing, because my writer friends assure me that it’s a completely normal thing, unless you’re Charles Bukowski.

Anyway: I did some writing earlier in the year, stalled a bit as tends to happen, and now have some more things I’d like to talk about. I’m hoping that a bit of community might help me keep momentum on the whole thing, and that more people might come across these posts if they’re here. So I’m going to move the old things over and zap the self-hosted blog (it is a bit 2005, after all).

In particular I’d like to think out loud about how organisations plan and get things done. I’m thinking mostly of government, as that’s where my experience is, but I think similar dynamics are at play in most big/complex organisations.

There’s a long history of technologists who are agile practitioners bemoaning the fact that the organisations they work for don’t really embrace agile principles, and in fact, work mostly in very non-agile ways. And an assumption that things would be better if that changed.

There’s some truth to this, but also a little déformation professionnelle. And perhaps, occasionally, a smidge of hubris. I think part of what makes agile software engineering principles fail to stick in wider contexts is the presence of an unfortunate and not very convincing assumption that “this works for software, so it’ll work for everything else too”.

But there are some deeper themes at play. There are some philosophical foundations that are useful: about how we see the world, how we come to know things, and what it even means to know things in the first place. Agile practices build on many of these ideas, making them practical and more straightforwardly implementable. And while we may not need or want our SLTs and boardrooms to be “agile”, I think we do need and want them to work with some of the same philosophical foundations in mind.

I think it’s possible for organisations to be more efficient, more productive, more effective at managing risk, more adept at acting on opportunities and more able to operate in ways more conductive to the flourishing of the people who work for them. Procurement doesn’t have to be broken, and HR doesn’t have to be soulless. Fairness and process are not synonymous. We don’t need policies to cover every eventuality. We can do things differently. We can’t manage everything with post-its on a wall, but we don’t need spreadsheets to manage everything either.

There are people doing these things differently, and places (and times) where procurement works pretty well actually, HR is alright, and proceses work pretty well. But there are a lot of places that are far from that ideal, and needn’t be.

I’d like to get some of this out of my head, and even more than that, I’d like to start a conversation: I have ideas, not solutions, and I think the solutions that do exist are ones we’ll find in practice, not in theory. So I’m very keen to talk, and to hear about people’s experiences. If you’re up for that, please do let me know!

Subscribe now

Great day at the GovTech Summit yesterday! Started off the day with delightful impromptu coffee with Rachel Murphy, James Findlay, and Emilia Hogarth — wonderful to reconnect after so long. Something very lovely (and valuable) about just being in town and bumping into people.

Early talks focussed on themes that have been pretty consistent/familiar over the years - leadership, skills, performance management, funding models, process, etc - but a good bit of disagreement around the acceptance of failure: should ministers openly own up to honourable failures, or is that just politically naïve / therefore never gonna happen? Still an open question, but an important one as I think it sits underneath and blocks progress on a lot of those other themes.

Pat McFadden summed up the challenge well in his closing panel: it’s not so much a problem with the people, but with the accumulated risk-aversion and process complexity, and (implied) the difficulty in dismantling that. It’s a tough thing when the tools that used to keep you safe start to lock you into decline. But I think that’s where we are. And fixing that will need political leadership and a willingness to defend those honourable failures. Josh Simons – on a different panel – gave an optimistic point on this, saying that in his conversations with voters, the overwhelming trend is that people understand not everything will go perfectly, and that if you’re straight with them, they’re pretty understanding.

Great chat with Ian Makgill from Spend Network about the practical uses of AI, avoiding the hype curve, getting the value and the mindset shift that needs. There’s definitely a change in thinking required for getting the best out of AI technologies. So one must keep an open mind. But, as someone wise once said, not so open our brains fall out.

Great to meet Seb Barker (AI transcription for casework), Deven Ghelani (bridging the policy/delivery divide), Wyndham Plumptre (a modern procurement platform - good work in an area deeply in need of disruption), Maryam Torshizi (agentic deep research that sounds like it’ll give ChatGPT a run for its money) – all doing very interesting work

Panel with Georgia Gould, minister for procurement, was probably the highlight of the panels - talking about the procurement challenges faced by government. Some legislative, with general optimism about the Procurement Act on that front; others cultural, with perhaps slightly less optimism from the panel but lots of positive action – pilot/small scale experiements going on to try and move things forward.

There was some talk about it being great for SMEs to contract with government through large primes (No. It. Bloody. Isn’t.) and she jumped straight in to say that direct spend is what counts, and that’s what they’re measuring – fantastic stuff. Looking forward to more!

I just read the State of Digital Government review published last week. If you’d rather not read the whole thing, here’s a quick GenAI summary.

It’s a good read. It says all the right things, and it’s clear, positive, and generally free of consultantese. It feels grounded in the right kind of culture and thinking.

That said, it’s also a bit frustrating, as the “right things” have been the same for many years. Jerry Fishenden’s archive of similar documents over time (decades!) makes for sobering reading.

So why doesn’t this stuff stick? What is trapping us in this cycle, where we make progress but then fall back? Or in the words of the review, why is it that “these successes are too often achieved despite the system rather than because of it.”

A lot of what the review advocates feels like a return to the early work of GDS, which would be great. But that effort, like each bit of progress before it, has been undone over time. Not entirely – there were some lasting and important wins – but those wins didn’t bring the sweeping changes many of us expected.

Unfortunately, in some important ways, digital government today is as challenging a delivery environment as it ever was. We still see underinvestment, muddled thinking around capital versus expenditure, an overfocus on cost-cutting without sufficiently considering value, flawed procurement processes, a lack of outcome-based thinking, silos/too few teams working openly, a terribly challenging legacy IT estate and so on.

It remains the case that digital teams are tasked with delivering world-class services that drive growth, but without the levers to fix the systemic issues that are the actual problem. Because it’s not the technology that’s hard: it’s the environment it’s built in. GDS’s great breakthrough happened because it did get some of those levers, and that’s why we saw some real change under its leadership, before some rather nasty interdepartmental politics robbed it of its influence.

More positively, it’s reassuring that strong leadership, ministerial support and good thinking from the centre feel like they’re on the rise. While much of the review isn’t new, it is right, and that’s a good step forward. Here’s hoping for more of the same, and for teams that are empowered enough to actually do it.

Subscribe now