2026-09-28 19:11:54
I worry a lot about Journalism: at its best it has a very important role in counterbalancing the big bears of Business and Government, and it's hard to see what else might fill that role in its absence. Orwell said that "journalism is printing something that someone does not want printed, everything else is public relations," and an obvious consequence of this is that powerful external parties will often want to disempower or circumvent journalism, to our collective peril. So I do believe that journalism is under basically-constant threat, and it feels bad to criticize a threatened body.
But many of the wounds on the body media are self-inflicted, and I think that the more external enemies you have, the more you should be careful not to damage yourself as well. So here is one way that I think modern journalism is hurting itself: I think it's really bad how much modern newspapers launder opinion as fact, and I think it's under-remarked because we generally read newspapers whose opinions-laundered-as-facts are agreeable to us.
Here are the top headlines from today's edition of a major newspaper:
I censored the specific names in this because I think if you see them it's too easy to evaluate on a subjective basis: if I hate X, and the headline says Allies Grumble About X, it's easy to side with it. But I think this is wrong: "Allies Grumble" is simply not a meaningfully objective news headline in almost-any situation. For almost-anyone doing almost-anything there are at least some allies grumbling about it, and the choice to write a lead story about it is extremely selective and opinionated.
I think what this newspaper actually wants to say is We Believe X Is Hindering Global Economic Growth, but that doesn't fit with their self-conception as an outlet for objective reporting, so instead they find someone else who believes the same thing and report on that belief as if impartially.
Every time I see a headline that ends in experts say I want to tear my eyes out: either the thing the experts say is objective fact, and then you don't need the attribution, or (more often) the thing your experts said is an opinion, and should be an opinion column.
On this random day, three out of three of the top headlines are opinions: aside from Allies Grumble we have Costs Are Expected To Spike in 2027. This is not a fact, it's a collection of opinions: I could just about defend it if a newspaper had one consistent polling metric it used, and reported every month on what that metric said whether it's positive or negative or neutral (or every time it moved by a certain amount, or some other objective measure). But this headline is about an arbitrary poll run by one specific company, and if you read to the bottom it turns out it's "preliminary results" that will be updated later in the year (unmentioned further up the piece).
The only honest headline of the set, to me, is What Rising Global Bond Rates Mean For Your Money. This is also opinion!, to be clear, but I think it's clearly framed that way, even if in an authoritative-opinion style. But it's not attempting to convince the reader that this thing we both believe is objective reported news.
As I understand it, there's interesting historical reasons why newspapers try to claim a mantle of objectivity; it wasn't always the case, and in some countries newspapers were historically explicitly aligned with political parties, as in some countries TV news outlets are more straightforwardly partisan while newspapers try to claim neutrality. I think it's absolutely fine to have an opinion, and I understand that as people we are kind of inclined to think of many of our opinions as near-facts. But I think it's important to maintain the distinction, and that newspapers are doing a disservice to their readers by reporting on someone's opinion to launder it into the news section. A building collapsing is news; who is to blame for shoddy building standards is opinion, no matter how many of us share that opinion.
2026-09-24 19:11:55
So I was walking down the street the other day when my friend's [X] slipped his grasp and started barreling towards the road. My friend pushed me out of the way to recover his [X] – while nothing bad happened to me, I could theoretically have been pushed into the road and in that moment he didn't seem to know or care. Was my friend's behaviour acceptable?
I would like to suggest two things:
1) the acceptableness of the behaviour is highly dependent on the value of X
2) many people in many contexts make arguments that pretend this isn't so, and come up with way-too-general justifications for what is actually a specific judgment about the values of competing things.
So in this case, if [X] is "his 5 year old child", I would personally say it's fine for him to push me out the way, even though there's some chance I would have been injured as a result. If [X] is an apple, I would personally say it's not at all acceptable. And I think most people would agree with me on each of these judgments, though that's ultimately an empirical question.
I also imagine there are various intermediate cases – a puppy, a precious ring? – where different people would have different intuitions.
But what I see quite often is people making arguments that claim to be fully general, but actually smuggle in an assessment of how reasonable the underlying prioritization is. "It's never ok to push someone potentially into harm's way!", they'll say, when really all they mean is "it's not ok to push someone into harm's way in order to save your errant apple, come on, that's silly."
I see this a ton on reddit threads like "Am I The Asshole?", which I fear I browse through something closer to rubber-necking than (what could theoretically be very helpful) learning more about different people's social intuitions.
The average AITA thread will say something like "I disinvited my parents from my wedding, am I the asshole?" And people will reply en masse that either "it is never ok to disinvite your loved ones from your wedding!" or "it is always ok to disinvite your loved ones from your wedding!," and their response is 100% based on whether they consider the Inciting Incident reasonable, and they mostly seem to show ~0 awareness that they're doing it.
For example, if the disinvitation is because the parents are gay, that is judged Bad, and if the disinvitation is because the parents are homophobes, that is judged Good. But instead of saying "I think it's bad/good you disinvited them because I disagree/agree with your substantive assessment," they pretend that a universal principle about disinvitations is being followed. (I say pretend but I have no real sense of how much people realise they're doing this, whether it's conscious or unconscious or somewhere in between).
After (and only after) writing this blogpost, I realised I have recently made this exact same mistake myself: I was talking to a friend who was deciding whether to date someone, and unsure whether his reason for being hesitant to date this someone was reasonable or not.
The thing I said to him was "of course it's reasonable, you can choose not to date someone for any reason at all!", but this is not exactly what I actually believe. The more accurate statement would have been "I believe that the reason you just gave seems like a reasonable reason," yet the thing I responded with was unnecessarily more general than that. If the reason he'd given had been that he didn't like their hairstyle, I would not have said the same thing.[^1]
It's hard to suppress the feeling that the difficult part of everything is having good judgment, knowing when to apply A and when to apply B, and that this can't be boiled down into universal rules. Two thousand years later, it's hard to argue with Hippocrates:
Life is short,
and craft long,
opportunity fleeting,
experimentation perilous,
and judgment difficult.
[^1]: On some level it is true regardless, you can not-date someone for any reason at all, and sometimes your seemingly-insignificant reason for hesitance is really a mask for a deeper underlying reason, etc. If you are unsure whether to date someone and you want a second opinion, email me.
2026-09-24 00:11:01
There's a thing that happens sometimes where you catch a glimpse of yourself and feel like you look terrible, and you say to someone "I'm having a bad hair day", and they say "no you look completely normal."
And the options are basically
1) they're politely lying to you
2) the difference is just in your head, you look great as always
3) you always look about this bad, but usually you don't realise it
The other day I was hanging out with friend of the blog M. and feeling exceptionally mopey. When I said sorry I'm being so mopey she said I seemed completely normal.
I believe we should call this a Bad Hair Day Of The Soul.
2026-09-22 23:11:27
Look, I'm sorry if this is Not Even Wrong but:
1) the AIs are apparently ingesting All The Books, many of which have been rarely read by humans
2) surely some of those books are very good
3) the AIs are good at pattern-matching
Is there any way to get an AI to tell me about books that are extremely good but entirely neglected by humans? The ideal case here would be someone's PhD thesis that has been read by 2 humans, but would be transformative for me intellectually if I knew that I should read it.
But I don't personally know how to pose this question to an AI in a way that gets truly neglected but excellent books, rather than (say) books that have featured on some human's list of Neglected Books, which is a different thing. Do you?
2026-09-22 00:11:41
Some time ago — never mind how long precisely – I announced a new weekly polling thing, where I would ask you questions about What You Are Like and also to predict how other people would answer the question. I wanted to get at a cluster of questions that I don't know how to describe but which are ultimately related to feelings and experiences that I wonder about people having, but which I don't usually talk with other people about.
I originally sent out a random form and collated the results by hand, and then I decided it would be better to make a proper web app, and that's why you haven't heard from me in a month. But lo, it is ready now! So please join me at the new and improved PWATL:
Thank you to the 93 people who answered the previous question:
Which of the following feels more true to you: I hold imaginary conversations (or monologues) with specific people I know in my head
The results to which were:
Based on an early commenter's observation, I added this bonus question for the Imaginary Conversation Havers:
are those conversations mostly positive, mostly negative, or mixed?
Based on another early commenter's observation, I also added a bonus question for those who DON'T have imaginary conversations: do you have an inner monologue at all? I only got 4 responses to this, but 3/4 of those 4 people said they don't have an inner monologue.
I will copy out and post your comments below this page on the website; going forwards, you'll be able to comment and interact live on the new web app, check it out!
2026-09-21 19:11:24
Gratuities are always and everywhere framed as a % of the customer's payment: you add 20% on top of your meal/ride/etc, and that goes to the provider.
It strikes me that there's another possible framing: namely, how much did you increase the provider's take-home?
Let's take Uber as an example, since each driver only has 1 customer at any given time and there is only one worker involved in each transaction. I've been told by drivers that they receive 40% of what I pay the company: as a result, if I tip 20% then they're effectively receiving 50% more for the ride. (This is before you even calculate their profit; since their costs are unaffected, a tip might well be increasing their profit by 100% or 200% or more).
In restaurants and coffeeshops, the tip is less directly tied to the person's main income – the waiter isn't getting a specific % of the cost of my meal – but (at least in the US) the general principle is roughly the same, where tips add 20% to the customer's costs but double or triple the waiter's salaried pay.
I found myself wondering what would happen if you framed it this way to customers. Personally I feel much more generous (and happy!) thinking I tipped someone 50% of their take than 20% of my spend, even though it's the exact same dollar amount.
Perhaps this means it would be bad to frame it this way, since people might tip less and still feel more generous. Or perhaps it can't be framed this way since it would confuse too many people. Still, nobody can stop you from doing it in your own brain.