House of the Dragon

They have exploded on to the scene this week, the dragonflies.

Not just appeared. Exploded. Last week there were a few smaller iridescent blue things about the pond, which I now discover may be damselflies, because nature has rules and I have a coffee cup. 

This week the big ones have arrived. Proper dragonflies. Large greenish, metallic, four-engine jobs, holding the air as if they own it.

The pond water is noticeably warmer after the little heatwave, and that seems to have been the signal. Somewhere under the lilies and among the pond weed, a lot of rather grim aquatic larvae have apparently decided that now is the moment to stop being underwater monsters and become aircraft.

It is a ridiculous process when you actually look at it.

They climb out of the water, grip a reed or some convenient bit of vegetation, split open the back of the old body, and haul a completely different creature out of it. Then they hang there while the wings expand and harden. That is not a minor upgrade. That is not fitting a better carburettor. That is a submarine becoming an aircraft.

Some do not make it, like the chap below. One must have lost its grip or fallen back into the water while trying to emerge. At that stage they are helpless. Soft, damp, half-assembled, and with no opportunity to ring the AA. If the wings do not set properly, or they fall, that is it. Nature does not do customer service.

But the ones that do make it are magnificent. The empty cases cling to the reeds like little abandoned diving suits. You look at them and realise that something has gone. Something that lived in the pond for years as a small underwater predator has climbed out through its own back and flown away.

There is something properly wonderful about that. Not sweet, not cosy, not the twee version of nature with a soundtrack and a slow-motion bee. This is real nature. Mud, predation, risk, transformation and then, suddenly, beauty with wings.

And all of it is happening in the pond as if it is perfectly ordinary.

Yesterday it was fish, blanket weed, lilies and a few newts minding their own business. Today it is a small air force. Tomorrow, no doubt, they will be mating on the reeds while I stand there pretending to be David Attenborough in gardening shoes.

You can also see why JK Rowling used Chasers in Quidditch. These things do not merely fly about. They pursue, intercept, turn on nothing, stop in mid-air, and then accelerate off as if some invisible schoolboy has just shouted something in Latin at them. They're almost impossible to capture on camera as they never stay still long enough.

Metamorphosis really is one of those things we slightly under-react to because it happens all the time. A creature lives underwater, climbs out, opens itself up, becomes something else, inflates wings and flies away.

And I shall probably still spend half the afternoon worrying about the blanket weed.

Chatty Petey and the Missing 10mm Socket

There is a curious thing that happens when you start using artificial intelligence for engineering jobs. You begin, quite innocently, by asking it a question about a GT6, a Ford Galaxy, a ride-on mower, or one of the boys’ motorcycles, and before long you realise you have not summoned the Oracle of Delphi.

You have summoned Dave.

Or, as Hayley has now christened it, Chatty Petey, which is perhaps even worse, because it sounds like a bloke in a fleece who has opinions about fuel pressure, has once owned a soldering iron, and is absolutely certain your problem is an earth fault.

Not useless Dave. Not pub bore Dave. More like helpful Dave from down the road, who has once changed a clutch cable, owns a multimeter, and says things like, “Have you checked the earth?” with the calm authority of a man who has absolutely not checked the earth.

And, to be fair, ChatGPT is often extremely useful. It can explain the principle, suggest a logical sequence, remind you not to set fire to yourself, and point out that perhaps before stripping half the front end off a mower you might first check whether there is a circlip under several decades of congealed grease.

This is good advice. This is the sort of advice that prevents a man in late middle age from lying on the gravel muttering darkly at a machine designed by someone who clearly hated both gardeners and access panels.

But ChatGPT is not always right.

Sometimes it is confidently wrong, which is worse than being uncertain. Uncertain is at least honest. Confidently wrong is the chap who says, “That’ll just pull off,” three seconds before you discover that the thing in question is held on by a hidden grub screw, a tapered spline, two spring clips, and the accumulated spite of a previous owner.

This is where Bard comes in. Or Gemini, or whatever Google is calling it this week, though it remains Bard on my browser tab because I annotated it that way and I refuse to be bullied by a rebrand.

Bard is, in my experience, slightly better at diagnosing mechanical problems. Not always. Let’s not get carried away. It is not Fred Dibnah in a server farm. But it often has a slightly different instinct. It will say, “That sounds more like fuel starvation than ignition,” or “Before replacing the regulator, test voltage at the battery and at the rectifier output,” and suddenly one begins to feel less like a man randomly buying parts and more like a proper troubleshooter.

Unfortunately, if you ask Bard for a diagram, you may as well ask the dog.

ChatGPT, by contrast, will have a decent stab at describing one. Sometimes it will even organise the parts in a way that makes sense, which is helpful when you are trying to understand why a 1990s Mercedes fuel system has apparently been designed as a joint venture between Stuttgart engineering and a medieval puzzle box.

So the trick, I have discovered, is not to treat either of them as an expert. The trick is to treat them as two useful mates standing in the garage.

One is leaning over the wing, saying, “I reckon it’s the return feed.”

The other is sitting on an upturned bucket, saying, “No, no, look at the symptoms. It’s more likely a blocked breather or a collapsing hose.”

And I am in the middle, holding a spanner, slightly oily, quietly wondering whether either of them has noticed the small pipe hanging loose behind the carb.

That is the useful bit. Not the answer itself, but the argument. ChatGPT says one thing, Bard says another, and I feed each answer into the other like a small domestic version of the Chilcot Inquiry, only with more jubilee clips. One spots a weakness. The other revises the theory. I ask a sharper question. One of them remembers a common fault. The other points out an assumption. Gradually, through a process of digital bickering, something approaching a sensible diagnosis emerges.

Occasionally I take the output from one, feed it straight into the other, and ask for an assessment. Then I take that assessment and feed it back again. It becomes a little feedback loop, which is something I used to do in my programming days, when the computer would either do exactly what I had told it to do rather than what I meant, or sit there producing error messages with the emotional warmth of a parking ticket.

There is something pleasingly old-fashioned about it. Not the technology, obviously. The method. Thesis, criticism, revision, further criticism, slightly better thesis. It is basically a garage argument, but without anyone saying, “My brother-in-law had one of those.”

This is not artificial intelligence replacing human judgement. This is artificial intelligence requiring it.

AI does not remove the need to think. It increases the number of thoughts available, some of which are useful, some of which are twaddle, and some of which sound useful until you look at the actual machine and realise the component it is describing is not there, was never there, and would only be there if the mower had been designed by NASA.

The advantage, of course, is that these two mates are available at any hour and do not require tea. They do not suck air through their teeth and say, “That’s going to cost you.” They do not wander off halfway through the job because their wife has texted. They do not stand there telling you about the time they had a Capri with exactly the same problem, which turns out not to be the same problem, a Capri, or relevant.

But they do share one vital quality with human garage advisers. They need managing.

You have to know when to listen, when to question, and when to say, “Hang on, that makes no sense.” You have to spot when a theory explains one symptom but ignores three others. You have to ask whether the proposed fix matches the physical layout in front of you. You have to remember that a wiring diagram is not the same as the wiring on a 20-year-old motorcycle that has been previously visited by someone with insulation tape and spiritual confidence.

In other words, you still have to be the grown-up in the room.

That is fine by me. I rather enjoy it. There is something oddly satisfying about playing one machine off against another, like a tiny workshop version of the House of Lords. ChatGPT proposes. Bard objects. ChatGPT amends. Bard concedes a point but raises a further issue. I sit there with dirty hands and the faint air of a man chairing a select committee on why the bloody thing still will not start.

And now, of course, I want No.1 Son to make it worse.

He is a programmer, and understands AI properly, which is dangerous because it means he can probably build what I am now imagining. What I want is a little system that automates this whole process. ChatGPT says something. Bard assesses it. ChatGPT assesses Bard’s assessment. Bard questions the revised answer. Round and round they go, without me having to copy and paste things like a retired man operating a very small intelligence agency from the kitchen table.

In theory, this would be marvellous. A self-improving diagnostic loop. A pair of digital mechanics arguing endlessly until they converge on a well-reasoned answer. In practice, I suspect it would either produce something extremely good mechanically, or design a GT6 that does 300 mph and needs three software updates before you can open the bonnet.

Still, that is progress, apparently.

And often, between them, they get me there. Not because either one is infallible. They are not. But because two fallible perspectives, properly interrogated, can be more useful than one confident answer. Especially when the third participant is the idiot actually looking at the machine.

Which is me, obviously.

And that, I think, is the real lesson. AI is not a magic expert. It is not a replacement for experience, judgement, or noticing that the tyre bead is still stuck on the wrong side of the rim. It is a pair of clever, flawed, tireless, slightly overconfident mates in the garage.

Very useful. Occasionally brilliant. Sometimes wrong. Best kept under supervision.

And still no help at all when the 10mm socket has vanished.

Carol Vorderman and the Sovereignty Problem

There is a peculiar contradiction in the way some men attack Carol Vorderman.

They spend half their time telling us she is terrible, irrelevant, smug, woke, anti-Brexit, overexposed and wrong about everything from politics to public morality. Then, the moment their language slips its lead, what comes out is not simple contempt at all. It is sexual attention.

They are not merely furious because Vorderman is politically outspoken. They are furious because she is politically outspoken and attractive - and because the attractive bit does not make her compliant.

Many men, I am sure, find Carol Vorderman attractive. That is hardly a hanging offence. People find other people attractive all the time, and civilisation has somehow staggered on. The difference is that most people do not express it in the public square with the delicacy of a blocked drain.

The Robert Kenyon incident is useful because it shows the mechanism clearly. As reported, he did not write the original crude sexual comment about Vorderman. He endorsed it. He added the little public nod, the emojis, and the “he’s only saying what we’re all thinking” routine, as if he had been elected shop steward for the entire male subconscious.

Speak for yourself, mate.

That is the telling part. He did not merely approve a leering remark. He tried to universalise it. Not “I think this”, but “we all think this”. The rest of mankind was apparently to be hauled into the gutter with him, whether it had put its shoes on or not.

Even if many men did think something vaguely along those lines, the whole point of being an adult is that not every private thought needs to be released into the world like a ferret in a village hall. Private attraction is one thing. Public sexualised crudity about a named woman is another. You do not need a seminar room and a tray of herbal tea to grasp the distinction.

One wonders what Kenyon’s wife had to say about it. Not because she is responsible for him, obviously. But it must be a curious domestic moment when your husband explains that “what we’re all thinking” apparently included him, Carol Vorderman, and a level of public crassness normally associated with motorway service-station graffiti.

This is where “banter” becomes such a shabby defence. We are told this is locker room banter, as if Facebook, X, or any other public platform were some steamy changing room with wet towels on the floor and a broken shower in the corner. They are not. They are public squares. What you say there is not muttered to three mates while tying your boots. It is published. That is the word. Published.

Once it is published, the “private bloke having a laugh” defence collapses. You cannot stand in the middle of the market place shouting something crude about a named woman and then claim it was just dressing-room chatter because you happened to be wearing trainers at the time.

And this matters because Vorderman is not being attacked as a decorative celebrity. She is being attacked because she is a political nuisance. She keeps appearing with figures, arguments, receipts and a maddening reluctance to accept that blokes with flags in their profile pictures are the natural custodians of public reason.

For a certain sort of man, that seems to blow a fuse.

So the response is not “your argument is wrong”. It becomes “your body is available for public assessment”. Not debate, but demotion: a way of dragging a woman out of politics and back into an arena where some men feel more comfortable - judgement, ownership, appetite and humiliation.

There is also a certain amount of brass neck in this coming from the Reform-adjacent world, where people are forever announcing their deep concern for “our women and children”. Apparently women must be protected from outsiders, migrants, liberals, judges, human rights lawyers, small boats, drag queens, foreign courts and anything else that can be turned into a campaign leaflet. But protecting women from crude public sexual objectification by one of their own candidates seems to fall mysteriously outside the perimeter fence.

That is the hypocrisy. “Our women” are invoked as sacred symbols when useful, but an actual woman, with an actual name, actual politics and actual opinions, can still be reduced to a sexual target in the public square. The concern is not really for women as autonomous human beings. It is for women as tribal property. Protected from them, available to us.

The political irony is rather good. The same crowd who talk endlessly about freedom, sovereignty and standing up to elites seem oddly troubled by a woman exercising sovereignty over herself. Vorderman says what she thinks, uses her platform, annoys the people she wants to annoy, and refuses to be politely ornamental.

Which is why “speak for yourself” remains the neatest answer. Not all men think like that. Not all men need to turn disagreement into domination. Not all men see an attractive woman with opinions and immediately reach for the mental toolkit marked “put her back in her place”.

Some of us just think: there is Carol Vorderman, being politically irritating to exactly the right people again - and, yes, she is very attractive.

And then we put the kettle on.

Doomscrolling

The great thing about getting older is discovering, usually by accident, that half the things you have complained about for years were not really principles at all. They were just preferences with a hat on.

This occurred to me in the back of a car on the way to BCA Blackbushe, where Hayley and I were being taken to collect three cars. Two teenagers were in the front, chatting away. Hayley and I were in the back, both on our phones.

For years, people of our age have been told, and have often told each other, that teenagers no longer talk. They just stare at their phones. They grunt. They scroll. They communicate in fragments, symbols and little bursts of digital nonsense. Conversation, apparently, died somewhere between the invention of the smartphone and the disappearance of the Ford Cortina.

Except the teenagers in the front were talking perfectly well.

Not merely talking, either. They were properly engaged. Animated, alert, full of opinions. It just happened to be teenage car chat, which is a specialised branch of human speech best understood by those whose insurance quotes still look like punishment beatings.

Every engine, trim level, alloy wheel, acceleration figure and faintly ridiculous modification was apparently worth discussing. They had the tone of two people dealing with weighty matters. You half expected one of them to say, “Of course, the real issue with the 1.6 is torque delivery in the mid-range,” while the other nodded gravely, as if this settled the matter for all civilised people.

Meanwhile, in the back, Hayley and I were on our phones.

Not doomscrolling, I should add. That would be crude and slightly too convenient. I was researching the connection between periods of great art and periods of national wealth, because apparently I now do Renaissance economics between car collections. Hayley was arranging work appointments, because some people remain useful to society.

But from the outside, of course, we looked exactly like the thing our generation likes to complain about. Two older people, silent in the back, gazing at glowing rectangles while the young people in the front carried the conversation.

That is the funny bit. The old accusation had quietly reversed itself. The young were being sociable. The older generation were apparently lost to screens, even if one of them was studying art, wealth and patronage while being driven to a car auction, which may be the most middle-class sentence I have written this month.

It was not that they could not talk. It was that we did not particularly want to listen.

This is where a lot of generational complaint begins to look a bit suspect. When people say the young do not communicate, what they often mean is that the young do not communicate in a way that flatters the older listener. They talk to each other. They laugh at jokes we do not understand. They care deeply about things we consider absurd, which is a bit rich, frankly, given that we once had serious opinions about cassette decks, carburettors and whether a vinyl roof improved a car.

We forget this. We imagine our youthful conversations were full of wit, insight and promise. In reality, quite a lot of them were probably drivel with better hair.

So there we were, rolling towards Blackbushe. Youth in the front, talking. Age in the back, researching Renaissance patronage and organising work appointments. A complete inversion, achieved without anyone making a speech, launching a government inquiry, or writing a stern column about the collapse of civilisation.

It was almost enough to make one reflect seriously on the unfairness of generational stereotypes.

Almost.

Then one of them started on about something with a remap and a pop-and-bang exhaust, and I returned gratefully to Renaissance patronage. Most unlike me, as I have an opinion on something like that.

Aldi Price Match

Every time I see one of those “Aldi Price Match” labels in Tesco, I can’t help feeling they’ve accidentally admitted the entire argument before you’ve even reached the vegetables.

Because the sign is not saying Tesco is cheaper than Aldi. It is saying, “Please don’t wander off and compare too much of the rest of the shop.”

You see the bright yellow labels attached to tins of tomatoes and packets of spaghetti like little retail distress beacons. Tesco desperately trying to reassure you that civilisation has not completely collapsed and that basic carbohydrates remain obtainable without taking out a second mortgage.

And it is always the same sort of products as well. Bread. Pasta. Rice. Milk. Bananas. The absolute foundations of human survival. They never stick “Aldi Price Match” on anything involving truffle oil, artisan crisps or olives marinated in the tears of a disappointed Tuscan grandmother.

Those remain mysteriously uncompetitive.

The thing is, the whole campaign only works because Tesco knows exactly what the public thinks already. Aldi has lodged itself in the national psyche as the place where things are simply cheaper. Not necessarily better. Not necessarily nicer. But cheaper in the same dependable way gravity remains cheaper than flying lessons.

So Tesco matches a carefully selected group of highly visible staples because supermarkets understand something governments never quite grasp. Human beings do not perform a detailed economic analysis while pushing a trolley. Most people remember about five prices. Milk. Bread. Coffee. Butter. Whatever it was they bought last week while muttering “bloody hell” under their breath. Once those few items look reasonable, the brain quietly relaxes and the rest of the spending becomes strangely theoretical.

That is how you enter Tesco intending to buy pasta and emerge £42 poorer with pistachios, a scented candle called Sicilian Orchard and something involving sourdough which appeared to have its own branding consultant.

To be fair, Tesco probably can work out cheaper for some people if you fully engage with the modern loyalty card system. But that now requires the sort of tactical preparation once associated with planning the Normandy landings. You need apps. Bonus offers. Personalised discounts. A digital coupon. Possibly retinal recognition and a small briefing from mission control.

Aldi by contrast still feels like retailing from a parallel universe. The products arrive in crates. The middle aisle contains kayaking equipment, tungsten drill bits and an inflatable canoe for reasons nobody entirely understands. The checkout process feels faintly adversarial. Yet somehow you leave with enough food for a week and the financial damage resembles the year 2014.

The deeper irony is that Tesco’s own signs are now doing Aldi’s advertising for them. Every yellow “Price Match” label is effectively a tiny in-store reminder that somebody, somewhere, is still making supermarkets feel slightly embarrassed about their prices.

And judging by the number of signs appearing lately, quite a lot of embarrassment is involved.

Leave Long and Prosper

The conference chamber aboard the USS Brussels was unusually tense.

Not tense by Klingon standards, obviously. Nobody had drawn a bat'leth or challenged anyone to ritual combat. It was more the particular sort of human tension involving folders, opinion polls and a man in a navy blazer insisting everybody else had lost touch with ordinary people on Deck 14.

Captain Jean-Luc Juncker sat at the head of the table, fingers steepled.

"So," he said carefully, "you wish to withdraw from the United Federation of Planets. Despite having full access to Federation trade routes, scientific cooperation, defence guarantees, medical exchange, replicator standards and freedom of movement across three quadrants."

Nigel Farage leaned back smugly.

"We want our sovereignty back."

Commander Spock raised an eyebrow approximately two millimetres. On Vulcan this was considered open mockery.

"Captain," Spock said, "the petition specifically objects to Federation regulations concerning Romulan ale bottle labelling, curvature standards for shuttle docking ports and the alleged overreach of the Federation Committee on Agricultural Replication."

"Exactly," said Farage triumphantly. "It's bureaucratic madness."

Data tilted his head.

"Yet according to the figures, the humans leading this movement possess the largest number of shuttle exemptions in the sector, receive disproportionate agricultural subsidies and account for 23% of all Federation procurement contracts."

"Project Fear," said Farage instantly.

Worf frowned.

"I do not understand. Your world voluntarily joined the Federation."

"Yes, but nobody explained there'd be Andorians."

An Andorian ambassador shifted awkwardly.

"We have been here for two centuries."

Farage pointed dramatically.

"Exactly. That's the problem. Ordinary humans can no longer recognise their own starports."

Counsellor Troi sighed softly.

"I am sensing confusion, resentment and nostalgia for a past that may not have entirely existed."

Farage smiled.

"That's because you're an elite."

At the rear of the chamber, Chief O'Brien muttered into his tea.

"This is starting to sound very familiar."

A large holographic chart appeared above the table.

"If humanity leaves the Federation," said Data, "there will likely be immediate disruption to supply chains, reduced scientific collaboration, labour shortages on lunar infrastructure projects and probable decline in gross planetary product."

Farage folded his arms.

"We'll be fine. The Alpha Centaurians need us more than we need them."

"They appear not to agree," said Spock.

"Experts said warp drive was impossible."

"No they didn't," replied Scotty. "That was literally the one thing they thought was possible."

The room descended into argument.

One admiral warned of economic damage. Another insisted it would all be worth it for blue passports. Nobody could explain why blue passports mattered in space.

Outside the viewport, the Federation continued functioning much as before. Starships came and went. Trade flowed across sectors. Scientific discoveries continued. Nobody in the wider galaxy entirely understood why one reasonably prosperous species had decided to make interstellar customs declarations vastly more complicated for itself.

Years later, aboard a somewhat underfunded human cargo vessel delayed outside the Vega Trade Zone due to paperwork irregularities, an exhausted customs officer looked up from a stack of forms.

"You voted for this," he muttered.

The captain stared into the middle distance.

"We were told there'd be less bureaucracy."

Normal People?

The other night I was watching one of those PBS documentaries Americans make rather well. Serious men in spectacles explaining how civilisation was built with clipboards, stopwatches and the ability to measure how long it took a man called Frank to carry a shovel across a factory floor in Ohio.

It was about scientific management in early 20th century America. Time and motion studies. Efficiency. Productivity. Industrial rationality. Humanity reduced to measurable output with the occasional sandwich break thrown in. And, as often happens when people become overexcited by systems, it drifted quietly from "how can we make factories run efficiently?" into "how can we make humans run efficiently?"

Which is usually the point where clever systems start wandering off into very dangerous territory.

There was a section on eugenics. Not the cartoon version people imagine now with instant goose-stepping and villainous music, but the respectable version. University people. Philanthropists. Public intellectuals. The sort of people who probably corrected your grammar while proposing forced sterilisation over soup.

One poster appeared on screen: "Every 50 seconds an American goes to jail. Normal people don't go to jail." And there it was. The entire intellectual emptiness of the thing in two sentences. Because the obvious question immediately presents itself: what is a normal person? And the answer, from their perspective, appears to have been: "A person who doesn't go to jail." Which is not science. It is a circular argument wearing a lab coat.

The more you think about it, the more absurd it becomes. A drunken idiot stealing a bicycle outside a diner in Milwaukee is deemed biologically defective because he got caught. Meanwhile the man quietly rigging markets, bribing officials or engineering fraudulent financial products from a walnut-panelled office remains perfectly "normal" because he has accountants, lawyers and a decent tailor.

The clever criminal becomes morally superior simply by avoiding detection.

In fact, if one follows the logic properly, intelligence itself almost becomes evidence of normality. A successful fraudster who avoids prison for thirty years is, by their standard, a better specimen than the impulsive fool nicking copper cable from a building site.

Which rather exposes what they actually meant by "normal". Not morally good. Not psychologically healthy. Economically functional.

The ideal citizen in that worldview was productive, compliant, orderly and useful to the machine. The definitions emerged during the great age of industrial capitalism, when everything was being measured, standardised and optimised. Factories were becoming astonishingly efficient, and a certain type of managerial mind started imagining society itself could be engineered along similar lines. Good stock and bad stock. Productive and unproductive. Fit and unfit.

Only humans are not bolts. A factory can reject defective rivets without moral consequence. Applying the same mindset to people leads fairly quickly to some extremely ugly places.

And the remarkable thing is that the instinct never really disappeared. It just modernised itself. Today we still categorise people endlessly through metrics, scoring systems and predictive models. Credit scores. Productivity tracking. Behavioural analytics. Risk profiling. Algorithms estimating reoffending rates. Schools reduced to performance tables. Workers monitored for output. Insurance premiums adjusted by postcode, habits and probability curves.

The language has changed, but the managerial temptation remains the same: reducing complicated human beings into measurable units for administrative convenience.

The old eugenicists would have absolutely loved spreadsheets.

And now, with AI looming over large parts of the economy, the old logic starts looking uncomfortably familiar again. If human worth is subconsciously tied to economic usefulness, what exactly becomes of the person made redundant by artificial intelligence?

For two centuries industrial capitalism always had an escape route. Workers displaced from farms moved to factories. Workers displaced from factories moved into offices and services. There was usually somewhere else for the surplus humans to go.

But AI threatens something slightly different. Not merely manual labour, but cognitive labour itself. Clerks, analysts, coders, designers, administrators, even parts of law, engineering and medicine now sit nervously watching software improve at unnerving speed.

The irony is rather delicious. For years the managerial classes tended to assume they themselves represented the secure and superior form of labour, while physical work remained economically vulnerable.

Now the spreadsheet has started coming for the spreadsheet people.

And once productivity becomes detached from human labour altogether, capitalism arrives at a slightly awkward philosophical question. If millions of people are no longer economically necessary to production, what exactly justifies their claim to income, dignity and social worth?

Because if human value is reduced entirely to measurable economic output, then the logically perfect society is one with no humans in it at all. No illness. No pensions. No lunch breaks. No trade unions. No awkward insistence on taking the caravan to Cornwall during August. Just machines optimising machines in perfect sterile efficiency.

Which is perhaps why societies built entirely around productivity eventually start becoming cold and faintly inhuman.

Humans have always justified their existence through things larger than efficiency alone. Family, loyalty, humour, pointless hobbies, strange obsessions, restoring old cars in cold garages, growing giant marrows, spending three years building model railways nobody asked for. The sort of gloriously inefficient behaviour that makes people human.

A man restoring an old Triumph GT6 in his garage contributes almost nothing measurable to GDP compared with an algorithmic trading platform. And yet one of those things feels profoundly more human than the other.

One suspects that if the original poster were redesigned today, it would not mention prison at all.

It would simply say: "Normal people maintain an acceptable credit score."

Preferably underneath a photograph of a smiling consultant holding a reusable coffee cup outside a glass office block in Connecticut.

Blair, Starmer, and the Missing Steering Wheel

The media have managed to turn Tony Blair’s essay into a leadership story, because of course they have. Westminster sees a former prime minister criticise a current one and immediately reaches for the ceremonial dagger. It is much easier than reading the argument.

But Blair did not really say “Starmer must go”. He said something more awkward, and rather more important. He said Labour has no coherent plan.

That is not the same thing.

In fact, he specifically warned against removing Starmer before deciding the policy direction. Which, inconveniently, is the bit much of the coverage has treated as small print. The headline version is assassination. The actual argument is strategy.

And this is the point I keep coming back to. People talk as if Starmer can simply wake up one morning, stride into No 10, slap the desk, and announce a new direction for the country. As if the prime minister is the managing director of Britain plc, and the rest of the state is just waiting for a briskly worded memo.

It does not work like that.

He can change the slogan by lunchtime. He can have a new lectern by teatime. He can probably find a backdrop with the word “renewal” on it before anyone has finished microwaving soup in the Cabinet Office. But changing the actual strategic direction of a government is rather different.

That needs policy. It needs money. It needs the Treasury. It needs the Cabinet. It needs parliamentary discipline. It needs the party not to start chewing its own ankle in public. It needs a plausible electoral coalition, which is slightly harder now that Brexit has taken the old map, folded it into a hat, and thrown it into a canal.

And here is the bit people keep missing. Many of the levers Starmer would need have already been removed, weakened, rusted solid, or buried under events no British prime minister could simply wish away. Some of that was political choice. Some of it was public choice. Some of it was the brute arrival of crisis.

The Tories inherited a country still carrying the damage from the financial crash, then chose austerity as a governing doctrine. The electorate then endorsed Brexit, sold as control but delivered as friction across trade, labour supply, investment and the public mood. Covid was not chosen. Ukraine was not chosen. Global energy shocks were not chosen. But the condition of the country when those shocks arrived was not an act of God. It was the product of years of underinvestment, short-term politics and pretending that resilience was an optional extra.

So when people say “Starmer should just change direction”, one has to ask: with what?

The public wants better public services, lower taxes, controlled immigration, higher wages, cheaper housing, secure borders, faster growth, lower bills, less debt, and no visible disruption to anything they personally enjoy. Fair enough. I would also like a Triumph GT6 that does 60 mpg, never rusts, and comes with an E-Type parked inside it.

But politics is not a menu where you tick all the pleasant boxes and send the bill to someone unpopular. If you vote for fourteen years of managed decline, austerity, Brexit friction and performative sovereignty, then watch the world add Covid, war, energy shocks and inflation on top, you cannot be astonished that the next prime minister finds half the controls missing from the dashboard.

This is where Blair is both useful and limited. He is right that Labour lacks a coherent governing story. He is right that changing leader without deciding what Labour is actually for would be court politics dressed up as renewal. But he is also a man who won elections before Brexit smashed the landscape. His map is not useless, but it is not current either.

The centre he dominated no longer exists in quite the same form. Scotland changed. The Midlands and North changed. The graduate vote changed. The Brexit divide cut through old party loyalties like a badly supervised angle grinder. You cannot simply reboot New Labour in a country that has spent the last decade being reformatted by austerity, Brexit, Covid, Ukraine, energy shocks and the faint whiff of Boris Johnson’s decorating arrangements.

So yes, Blair has added something useful. He has pointed at the hole in the middle of Starmerism. But he has not filled it. He has mostly reminded us that there is a hole, that it is quite large, and that in his day holes were managed with more confidence and better tailoring.

The real question is not whether Starmer stays or goes. The real question is what Labour becomes next.

If it moves right, it risks losing its base and younger voters. If it moves left, it risks frightening business, the Treasury and half the press into clutching the furniture. If it moves closer to Europe, the Brexit wound reopens. If it avoids Europe, the economic drag remains. If it promises public service repair without serious tax reform, it is pretending. If it promises growth without explaining where it comes from, it is doing motivational speaking in a slightly better suit.

That is why the leadership story is so shallow. It treats politics as casting. New face, new energy, new beginning. Wonderful. But if the script is still unfinished and the plot makes no sense, changing the lead actor only gets you a different person looking worried in front of the same collapsing scenery.

Blair has not called for Starmer to go. He has done something more irritating. He has said that Labour cannot solve its problem by changing the wrapping paper.

What a Turn-up!

As I was getting dressed this morning, Hayley pointed out that one leg of my shorts had somehow acquired an accidental turn-up.

One side neat and cuffed like a 1950s Italian film star strolling around Portofino with a cigarette and inherited confidence. The other hanging normally like a man about to go to B&Q for weedkiller and wood screws.

That made me think.

What exactly is the point of turn-ups?

The official explanation is always practicality. Mud. Rain. Protecting the trouser hem. That's how they supposedly started. Men in the nineteenth century folding their trousers up to avoid puddles and horse filth. Entirely sensible. Britain used to excel at sensible things. Drainage. Steam engines. Naval logistics. Then, as always happens, society got hold of the idea and converted it into fashion.

A practical emergency fold became a permanent sewn-in feature. A temporary adjustment transformed into a signal of refinement. The same species that invented the adjustable spanner somehow ended up paying extra for trousers deliberately designed to look as though they are perpetually avoiding a puddle outside Swindon station.

And the odd thing is that nobody ever questions it.

You can walk into a tailor and ask for turn-ups with complete seriousness. Measurements are taken. Cloth discussed. Solemn nodding occurs. Somewhere deep in Savile Row there are men speaking quietly about "a one-and-three-quarter-inch cuff" as though discussing naval gunnery tables.

Yet fashion has spent the last century behaving as though the existence of turn-ups is a matter of civilisation itself.

They drifted in and out of favour decade by decade. Wide turn-ups in the 1930s. Wartime austerity in the 1940s, when Britain suddenly decided extra cloth at the bottom of trousers was practically aiding Hitler. Narrower, sharper styles later on. Then flared trousers without turn-ups. Then power suits. Then designer minimalism. Then suddenly fashion rediscovered cuffs again because somebody in Milan spotted an old photograph of Cary Grant looking pleased with himself.

During rationing the government even frowned upon turn-ups because they wasted cloth. Which is marvellous when you think about it. Somewhere in Whitehall, civil servants were effectively conducting strategic calculations about trouser hems while German bombers crossed the Channel. The nation that built the Empire ended up auditing cuffs in the national interest.

And, absurdly enough, it probably mattered. Wartime Britain counted everything. Steel. Coal. Rubber. Fabric. Housewives saved bacon fat for the war effort while men were quietly expected not to swan about using unnecessary wool around their ankles. One imagines a Ministry leaflet urging citizens to defeat fascism by surrendering two inches of unnecessary trouser.

But the logic completely collapses the moment you notice turn-ups on shorts.

Shorts.

There is no puddle-protection issue with shorts. If floodwater has reached the hem of your shorts, your concerns have moved beyond tailoring and into survival strategy. At that point you need a lifeboat, not elegant drape.

Which means turn-ups long ago ceased to be practical and became entirely psychological. They exist because clothing designers fear a plain edge. Left unattended, a simple hem apparently causes existential panic within the fashion industry. Somebody somewhere sees an ordinary trouser leg and thinks, "No. It needs... extra trouser."

The fashion world does this constantly. Buttons that do nothing. Zips leading nowhere. Fake pockets stitched shut. Shoes designed for walking which visibly prevent walking. Men's fashion has its own outbreaks of madness too. Tiny suit jackets that only fit if the wearer survives entirely on almonds and despair. Trousers cropped halfway up the shin so grown men resemble Edwardian newspaper boys waiting outside a pie shop.

And now, apparently, my shorts had joined the movement.

The disturbing thing is that the accidental turn-up genuinely did make them look slightly smarter. One tiny fold of cloth and suddenly the shorts looked less "retired man trying to remember where he left the hose connector" and more "casual Mediterranean leisurewear". This is worrying. It means I may be only one laundry accident away from style.

I briefly considered deliberately turning the other leg up to match. Then I caught sight of myself in the mirror and realised I was only about three styling decisions away from owning a linen fedora and discussing olives.

So I flattened it back down again and went downstairs for tea. Britain was spared.

Categories

Human beings are obsessed with boxes.

We like things labelled, sorted and filed away neatly because ambiguity is exhausting and most of us have enough trouble remembering where we left the car keys.

Reality, unfortunately, keeps producing continuums.

Light was one of the great scientific irritations. Physicists spent centuries demanding to know whether it was a wave or a particle, as though the universe had a legal obligation to tick one box on the form. Waves were waves. Particles were particles. Nice solid Victorian categories. Then quantum mechanics arrived and light effectively replied, "Depends what you're doing with it."

That was not the answer anyone wanted.

Sometimes light behaves like a wave. It interferes with itself and produces rainbows. Then it abruptly starts behaving like a stream of particles smashing electrons out of metal like microscopic shotgun pellets. The categories themselves turned out to be incomplete. Reality had quietly wandered off while the scientists were still labelling the drawers.

Even the rainbow is cheating. We talk about colours as though they are distinct things. Red. Orange. Yellow. Green. But there is no actual line where red stops and orange begins. Nature does not install borders. Humans do. We stare at an uninterrupted spectrum and start naming regions because otherwise we'd spend all afternoon in B&Q debating whether the bathroom should be painted "Sunset Coral" or "Tuscan Apricot".

And once you notice this tendency, you start seeing it everywhere.

Politics now functions almost entirely through categorical collapse. Are you left or right? Patriot or traitor? Woke or fascist? Modern political tribes cannot tolerate spectral positions because nuance performs horribly online. Somebody saying, "Well, this issue contains competing pressures and trade-offs" will be flattened instantly by a man with a Union Jack avatar screaming, "ANSWER THE QUESTION."

Take the endless debate about whether Trump is fascist. The internet demands a box. Yes or no. But reality is more awkward than that. Trump does not fit neatly into classical fascism as a coherent ideology, yet he clearly rummages through parts of the authoritarian toolbox whenever useful. Both sides are convinced their box is the correct one. Reality, meanwhile, sits somewhere awkwardly between the shelves.

And then there is my GT6.

I have umpteen boxes of parts for it. Electrical. Interior. Engine. Suspension. Trim. That at least is the theory. In practice, opening any given box resembles an archaeological dig conducted by somebody with attention deficit disorder.

The electrical box contains relays, certainly, but also two bonnet catches, what may be part of a door mechanism, three unidentified brackets, and a bolt that looks as though it came off agricultural machinery during the Attlee government.

Now, according to me, these things do not belong in the electrical box. They are categorically not electrical. But clearly the person who originally filled the boxes operated under a different philosophical framework. Perhaps the criterion was merely, "small metal things I found near the wiring loom."

And it never ends. Every few months I rearrange the shelves in the garage in pursuit of some glorious final system of categorisation that will supposedly bring order to the universe. Sanders here. Drills there. Paint equipment on that shelf. Electrical testing kit over there.

This lasts about three weeks.

Then some object appears that is not quite a drill, not quite a sander, and not entirely clear in purpose. It may polish. It may grind. It may remove rust. It may, for all I know, prepare cappuccino. Suddenly the entire classification system starts wobbling because reality has once again produced a thing that sits awkwardly between the boxes.

And that is the point. Categories are not laws of nature. They are human convenience systems. Another person creates different boxes entirely. One man's "electrical components" is another man's "bits that were on the same shelf at the time."

The internet has made all this dramatically worse because algorithms reward certainty. Nuance spreads across social media with all the speed and grace of a wardrobe falling downstairs. Certainty, meanwhile, races around the world in under a minute carrying a flamethrower.

So we keep trying to compress continuums into categories because categories are easier to weaponise, easier to administrate and easier to store mentally.

Reality, meanwhile, continues behaving suspiciously like my GT6 garage shelves. Full of awkward objects that refuse to stay obediently in the box somebody assigned them to years ago and which, if thrown away in frustration, will almost certainly turn out to be absolutely essential six months later when fitting the driver's door seal.