Every government now governs mainly by announcing things, and the announcement has become the policy rather than a description of one.

My handwriting has become genuinely illegible and I only found out because I had to fill in a form in front of somebody.

I have stopped believing any product review I read and I do not have a replacement, which is a genuinely new situation.

I started leaving my phone in another room and I feel slightly deranged about how much better it is. That cannot be right. It is a phone.

Cricket commentary has got unbearable. Nobody describes anything any more, they just wait for a stat to read out.

unpopular opinion but most "food deserts" are actually time deserts. the shop is there, people are working two jobs.

A thing about friendship after about thirty that I do not think anybody warns you about. It stops happening by default. School, university and your first job all put you in a room with the same people repeatedly for years, and every friendship you have was quietly manufactured by that. Then it stops, and the machinery goes with it, and nobody tells you because the people it has already happened to have stopped noticing. What replaces it is admin. You have to want it enough to put a date in a calendar six weeks out, and then to still want it on the night, which after a full week is a genuinely different question. The friendships that survive are not the deepest ones. They are the ones where somebody was willing to do the admin. I find that slightly bleak and also I have made my peace with it. The alternative was assuming it was about me, and it took about four years to work out that it was structural. Everyone I have said this to since has said some version of yes, and then said they thought it was just them.

Every organisation says it wants honest feedback and has exactly one person who gives it, and that person is always halfway out the door.

Right, the thing nobody says about "learn to code" advice: It was decent advice in about 2012 and terrible advice by about 2020, and the people giving it never updated. It got handed to redundant workers as though the shortage was permanent and the barrier was purely knowledge. Both of those turned out to be wrong within a decade. What actually happened is that the entry level filled up while the senior end stayed short, which is the worst possible shape for someone retraining at forty. You can learn the language in a year. You cannot manufacture eight years of having seen things go wrong, and that is what the remaining jobs are actually asking for. I do not think the advice was cynical. I think it was given by people who got in early and genuinely could not see which parts of their own path were the ladder and which were the timing.

I have changed my mind about anonymity online, and I want to set out why, because I argued the other side for years. The case I used to make was straightforward. Anonymity protects the vulnerable — whistleblowers, people in hostile countries, anyone whose opinion could cost them a job. Real-name policies were introduced by companies with a commercial interest in identity, and they were sold with a safety argument that the evidence never really supported. All of that is still true and I have not stopped believing any of it. What changed is that I stopped treating anonymity as one thing. There is anonymity from the state, anonymity from your employer, anonymity from other users, and anonymity from the platform itself. These are four different properties. A system can give you the first three and none of the fourth, and for almost everyone that combination is what they actually want. The mistake I was making was arguing about a slider when the real question was who each end of it protects you from. Once I separated them, most of the disagreements I had been having turned out to be people defending different columns of the same table.

Most of what psychology published before about 2015 does not replicate, and the field has been much more graceful about admitting it than any other would have been.

A thing I have noticed after fifteen years of interviewing people: almost every interview process is measuring how comfortable someone is being evaluated, and then reporting that as a measure of how good they are at the job. The tell is that the same candidate performs completely differently depending on the room. Give someone a whiteboard and three strangers and you learn how they handle being watched. Give them a real problem and an afternoon and you learn something about how they think. These correlate weakly, and we have built an entire industry on assuming they correlate strongly. What makes it hard to fix is that the noisy signal is cheap and the good signal is expensive. An hour of whiteboarding costs an hour. A realistic work sample costs a day of the candidate's time and half a day of yours, and multiplied across a funnel that is a real budget line that somebody has to defend. So we keep the cheap test and then add stages to compensate for its noise, which is how a process ends up with six rounds that are each individually indefensible. I do not have a clean answer. I have stopped pretending the current thing measures what we say it measures.

The most useful thing I was ever taught about science is that a result with no error bars is a story.

The friends I have kept are not the ones I had most in common with, they are the ones who happened to live nearby at the right time.

My mother has started every phone call for about eleven years by asking whether I have eaten, and I used to find it maddening. It is not really a question. She does not want to know what I ate. If I say yes she moves on immediately and if I say no she is briefly upset, so it is not information she is gathering, it is a check that I am alright, in a form that does not require either of us to say anything embarrassing. I have come round to thinking most of what people find irritating about their parents is this — a real thing being expressed in a code that got set decades ago and never updated. The content is fine. The format is from 1994. What changed my mind was catching myself asking my nephew whether he is keeping busy, which is the same question wearing a different coat, and realising I had inherited the whole apparatus without noticing.

Getting a GP appointment has become a competitive sport and the people who are worst at sport are the ones who need it most.

An unpopular position I hold about hospitals: the food is bad for a reason that has nothing to do with catering, and fixing it would cost more than anyone will admit. Hospital food has to be safe for people with immune systems that barely work, edible by someone who cannot chew properly, compatible with about fourteen medical diets, transportable across a large building, and hold at temperature for the twenty minutes it takes to get to the far ward. Given those constraints, what arrives is roughly what has to arrive. Every campaign to improve it treats it as a failure of will, as though somebody in charge simply does not care whether patients enjoy lunch. Then a celebrity chef gets involved, produces something excellent in one ward for six weeks, and the programme quietly ends when it meets the other three hundred wards and the actual budget. I am not saying it could not be better. I am saying that "why is hospital food bad" has a real answer, and nobody campaigning about it ever engages with the answer.

Weddings have got completely out of hand and everybody involved seems to be having a worse time than they would at a decent party.

Lower league football is better to attend and worse to watch, and the people who only do one of those never believe the other.

Owning things is overrated, says the man with a garage he cannot park in.

What I actually miss is software that was finished. You bought it, it did the thing, and nobody moved the buttons on a Thursday.

Term limits, everywhere, no exceptions. Nobody should be able to build a forty-year career out of a job that was meant to be temporary.

Films are too long. Not artistically, logistically — I have children and a bus to catch.

Being late is a decision that your time is worth more than mine. There is no version of it that isn't that, however pleasantly it gets phrased.

Peer review does more harm than good at this point. Two anonymous strangers, no accountability, eighteen months, and a result nobody will ever try to reproduce.