2025-12-27 00:45:03
Happy last days of 2025! Here's a review of the year in Blog, for those meta-interested in blogging.
By far our most popular post this year was 21 Facts About Throwing Good Parties. A lot of you probably know that writing online is very power-lawed, but maybe not quite how extreme it is: I think Party Facts already has 1000x more views than our average post, and the gap will probably keep growing over time. Enormous thanks to our editor friend A., who "commissioned" it, and needs to give us more blogpost ideas since she's clearly better at it than we are.
Another highly-reacted post from this year was How Weird Do You All Want This Blog To Be?, which generated more direct comments and emails than anything else we've written, and generally made the blog feel worth doing again. Overall it's kind of shocking how much blogging feels like shouting into a void, so thank you to everyone who turned that void into more of a soiree.
We started this blog almost-exactly four years ago in order to share the good news about Monosodium Glutamate. I had read a bunch of people online claiming that if you blog consistently for a year or two you will naturally start to develop an audience and... I don't think that's actually true? We had a bunch of unfair advantages when starting this blog, a higher-than-average connectedness in the graph of existing successful bloggers, and I would still say that mostly our audience is staggeringly small because the audience for almost-any writing like this is staggeringly small: if people could see the actual reader numbers for various prestige magazines they would cry.
Towards the end of this year I got a few big benefits from the blog that might singlehandedly justify all the time that's gone into it: I met two exceptional teachers/guides who are helping me with two major areas of existence, and made one outstanding real-life friend, for which I'm exceedingly grateful. Someone once said that posting online is a complicated, indirect search function for finding people you resonate with, and I think I'm finally feeling that. But it's probably not restricted to posting, it's more like... making stuff and sharing it with the world is a search function for finding people you'll like.
I blog largely because I have too many thoughts in my head, and left to themselves they keep repeating themselves at me, and blogging lets me clear them out and make space for new thoughts to happen. I do feel like there's a certain magic to writing, and ways that it's uniquely good at letting you think through an argument, and review and re-edit and sharpen what you actually believe. (I think that seeing something in text that you don't really believe is viscerally uncomfortable, and you therefore have to delete it, and therefore help yourself understand what isn't true).
I also think there's a physical dimension to the glory of writing, that trying to structure your thoughts on a 2-D page is inherently Good, and that physically moving segments around helps you figure out which parts are most important, and which bits are pre-requisites for which other bits, and where there's actually massive gaps that you didn't notice while thinking because of the non-linearity of internal experience.
That said... blogging has taken up a vast amount of my time, and I can't honestly say if the opportunity cost was worth it. (Well: to the extent it keeps me out of trouble, maybe that goes the other way). I often suspect that there are some people who just love writing, but other people who love idea-wrangling and then just happen to end up path-dependently doing their idea-wrangling in writing. I think if I could make myself switch from blogging to video or audio I would be better off, and if you're thinking of blogging but feel like you could channel your thoughts through games or short-form videos instead, you might realistically be better off cultivating that.
Happy end of Q1 21C to all! And wishing you an excellent Q2.
2025-12-24 21:27:28
The "endowment effect" is the claim that people overvalue things they own relative to things they don't, even if they only just got the item and don't have any reason to be attached to it (or have inside information about it).
So: participants in an experiment are rewarded with a tchotchke, and then they're asked if they'd like to trade it for a different tchotchke, and they disproportionately choose to keep their first tchotchke, even though it was randomly assigned and the people assigned tchotchke-2 generally keep that one too.
I'm sure this finding isn't entirely true, but unlike many other social psych findings it feels genuinely plausible to me.
I think there's a kind of endowment effect for our own lives, and I think I have unusually little of it: I don't generally feel that my job/beliefs/experiences etc are better just because they're mine. (My blog readership, of course, is precious and unique and I would not trade it for anything).
I'm sure you could prove in 10 seconds that I do still have a large and irrational endowment on these things, but my sense is it's far less than other people's. If you also experience this, or experience the opposite, I'd be keen to hear more.
Here's a thing that bothers me greatly. If one person insists on spending 100 hours researching a topic before having an opinion about it, while another spends 1 hour, the low-research person will publish 100x more opinions.
And just on principle I think for any given topic there'd be far more people who have researched it for an hour (or less) than people who have researched it for 100.
So unless there's some gating on whose opinions get published, or some reason to assume that informed opinions would get more traction than uninformed ones, most of the opinions you read will be from people who know very little about the topic.
And note that this is fractal: even if you decide to only listen to (say) Harvard Trained Historians, there will still be far more published content by the Harvard Trained Historians With A Low Bar For Having An Opinion than from the ones who insist on researching a lot before opining.
This seems very bad.
Of all the convoluted pointless bureaucratic loopholes I jump through, one of the least important but most poignant is borrowing ebooks from the library. There's a digital file somewhere that has 0 marginal cost of reproduction, and me and a bunch of other people queue up to have access to it, and when it's my turn to borrow I have three weeks to read it (which usually expire before I get to it), while other people are needlessly excluded from reading at the same time, and all for... what? As I said, there are other more-important pretzels we tie ourselves in for legal fiction reasons, but this one is just so vivid as a pointless game we play to pretend that something is what it isn't.
2025-12-22 21:30:58
I have been asked before if ATVBT has any official editorial positions, and I believe the only one is "aliens are real and extremely nearby".
This position was widely mocked until extremely recently, is currently on the cusp of social acceptability, and (I think) will soon be extremely mainstream, with everyone pretending that they were always open to it and never mocked it (while of course shunning the weirdos who were saying it twenty years ago).
Luckily, if you start investigating aliens today, you can be at just the right time to be part of the "early majority" – your friends will remember that you were talking about it just before it became popular, but after the New York Times published testimony from US Navy pilots that they'd seen objects doing things that no known human technology can do.
A good place to start for alien exposure is the recently-released movie Age of Disclosure. It tries to convey that there's a bipartisan understanding at the top of the US government that
1) aliens exist, in a non-trivial near-earth sense,
2) there's been some shady maneuvering for a while now to prevent democratic oversight of the government's alien knowledge, and
3) the dam is breaking and it’s all coming out soon, one way or another.
I have some quibbles with the movie. There is some really unnecessary equivocation between the extremely uncontroversial position "humans are not the only intelligent life in the universe," which ~most people I know believe in the abstract, and the not-yet-popular position "aliens exist and are in contact with the earth."
In an effort to assert bipartisanity, the trailer for the film gives equal billing to headline interviewees Marco Rubio (Republican) and Kirsten Gillibrand (Democrat), but in the movie itself Gillibrand basically just says that alien life probably exists somewhere and that the intelligence services and defense contractors should be accountable to democratically elected representatives, both of which I think are uncontroversial but not groundbreaking.
I think this is a useful pattern to remember when looking at alien stuff: it will simultaneously be true that most of the claimed evidence for aliens is shoddy and/or fake, and also that the best evidence is sufficient to be convincing. (This is actually a helpful thing to remember for many other arguments, but that’s another story).
The good news is, I think the movie is convincing after you chop out the fluff. Basically:
Overall, for me, the movie backs up the claim that either near-earth aliens are real, or the US government wants you to think that near-earth aliens are real, and either way it should be a massive story. That's really the number one question for me to people who think it's all a hoax or misdirection: ok sure, but isn't that kind of huge deal too? I have to stress this is from very senior people in both political parties.
One interesting thing to do if you’re alien-curious is just to go around asking your friends if they’ve seen UFOs. One of the first things that pushed me towards belief in aliens was sitting at a random hangout with 6 people and one of them brought up aliens, turned out 4 out of 6 had seen a UFO (alas I was one of the remaining two). Everyone seemed surprised to learn about each other’s UFO stories, and all the stories were more detailed and meaningful than I would have expected.
2025-12-19 21:21:40
I’m James, a friend and regular contributor to ATVBT. This is my weeklyish list of links, curiosities, and musings.
In case this is the last Good Tokens of the year, have a great holiday season and end of 2025. See you in 2026!
Christmas is just 7 days away. If you’re here, you already know about Uri’s hit new game Person Do Thing, but out of love for the game and it’s creator, I have to say one final time, this is a perfect stocking stuffer and a great way to spend time with your friends and family without a screen. Highly recommended!
Human Invariant: “In simple terms, good work gets noticed by everybody who matters.”. On some level, I think this is a useful reduction. At the same time, I also think that HI underates the dance between creators and audiences. Most great things are to some degree co-created.
Also, Human Invariant interviews a YouTube screenwriter.
On Bill Snyder’s career at Kansas State.. This level of obsession makes me wonder if I have What It Takes.
I feel I don’t spend enough time thinking about Esmerelda and how ambitious of a project it is.
The story of kelp pots (seed starter pots made from seaweed), a fun story in which I played a small part.
New to me: The Gettier Problem
The Lost Generation. This one is controversial because it deals with race and DEI, but if you can distance yourself from that a little bit, it’s really informative. It made me believe more in the Elite Overproduction Hypothesis.
The story of the fight over Romansch. Particularly enjoyable for me because the Engadin is among my favorite places in the world.
The average boomer will get paid out significantly more in medicare and social security than they paid in taxes — Russ Greene. Soon we’ll need a Boomer Corner.
This was the first year where no Pearl Harbor survivors were able to attend the commemoration ceremony.
Americans drive 3.3 trillion miles each year — Rohit
The EU makes more from fines on US tech companies than it does from taxes on all EU tech companies — David Fant
Lebron James has played against 35% of players in NBA history — CBS Sports. To be fair, he has played in ~28% of the league’s seasons. I think this makes him the Queen Elizabeth of athletes.
I feel like the Grinch saying this, but we’ve got to cut down on the number of special clothing days (e.g., pajama day) that are happening in schools or daycares. All it does is create stress for me as a parent and I don’t get the sense that my kids actually enjoy these. Who is this for?
Some shameless self promotion: The latest episode of —dangerously-skip-permissions: How Penny Schiffer works. Penny is another technical product manager turned AI software developer who I think has really mastered creating with AI.
Dexter, Claude Code for researching stocks.
GenAI created ads outperform human created ads by 19%… unless they are disclosed as created by AI, in which case the performance goes down by 32% — Eric Seufert
The coming AI wildfire.
Dev Browser, a better tool for agents to view local webpages. I can’t wait to play with this one.
2025-12-17 19:55:16
By far the best poem to share with people who don't like poetry is Failing and Flying, by Jack Gilbert. I have never seen any other poem connect so much with so many people on first reading. Honestly the highest-value move right now is to stop reading this post and go read that poem.
I once heard Seamus Heaney at a recital and he read each poem twice – as in, he'd finish a poem and then immediately read it again. He said he was copying this off some other famous poet, presumably William Butler Yeats. It was weird but also truly a good experience – the first read situates you, big picture, and in the second read you hear the details.
When I was in high school I would memorize poems to the point where I now have probably 20 poems lodged in my brain for life. I no longer know how I did it, I have tried again since and largely failed to remember anything. Maybe the brain-plasticity stuff is real, or maybe I've just gotten lazy, my guess is the second.
I do think poem-memorization is an immensely valuable practice, it embeds a particular music into the language of your thoughts.
Probably unsurprisingly, highly rhyme-y poems are easier to memorize; perhaps less-obviously, having a clear progression and narrative in the poem also helps a lot. So e.g. Robert Frost's sonnets go down easy, and Milton's On His Blindness (extra fun if you do the voices), and some Carol Ann Duffy: all have strong rhymes and clear storylines.
But some of the most beloved poems, like Wild Geese, are hard to memorize because many of the lines could plausibly be flipped, which adds to the cognitive burden, and this is also a problem for some lovely structured rhyming poems, like Missing Dates.
I guess my point is that not every poem needs to be memorized, and if you're struggling with one poem maybe try another.
I am broadly of the opinion that most aesthetic experiences should be judged by their peaks not their averages, and that very few artists can consistently produce outstanding work.
E.g. I used to be disappointed when I saw a new book by a "favourite author", and inevitably discovered that I didn't like it half as much as the book I fell in love with them for; now I accept that the meaning of "favorite author" is just "person who has ever written one great book," not "person who consistently writes great books."
The same goes for restaurants: a great restaurant is a restaurant that has at least one exceptional dish, and I should not expect their other dishes to be exceptional, and some of their other dishes will be actively bad.
My point is, this all goes even more so for poetry. I do not enjoy most poems in most poetry books; even an exceptional poetry collection might only have a small handful of poems I love. But it's those poems that the book exists for, and possibly that we exist for.
(Unsurprisingly, my favorite single poetry collection is Jack Gilbert's The Great Fires. But I still don't like most of the poems in it, and some of the poems I love only in parts).
I hate the standard Poetry Reading Voice, it feels so fake to me. I wonder if there was one poet once who everyone else is imitating, someone whose natural reading voice had that slow insistent lilt to it, and they were super charismatic and incredible to listen to. And then everyone else tried to copy that voice but now they're facsimiling facsimiles, and it's like kids dressing up in their parent's clothes, it's bathetic.
Someone pointed out once that we think of poetry as niche and unpopular, but the most popular commercial music of the 1990s to 2010s was spoken word poetry (i.e. rap). I don't know, I think that's pretty great.
2025-12-15 21:05:15
Mostly, when an armchair guy (gender neutral) has views on what a heavily-optimized company Ought To Do, the armchair guy is obviously going to be wrong: the company wants to make money AND knows their market in detail, while the armchair guy is just spouting off based on ~1 data point and ~0 skin in the game; if the thing he wants them to do made sense, they would presumably already be doing it.
My tiny sliver of justification for Why This Time Might Be Different is that – based 100% on stereotype and 0% on facts – companies like Uber are managed by yuppies in tier-1 cities who would never dream of taking an intercity US bus, and therefore might be missing a genuine market opportunity. This is not a good justification, so what I'm about to say is still presumably wrong, but I'm going to say it anyway.
I once took a Greyhound bus between two medium-sized American cities about 200 miles apart. Booking the $30 ticket was a pain in the ass to start with, the Greyhound website rejected every payment method repeatedly. I later needed to change my ticket by one day and Greyhound's change fee was so exorbitant that I basically lost this $30 and had to spend another $50 on a second ticket.
When the day came and I got to the bus station, I tried to board the waiting bus that was headed to my destination city, but was told that I did not in fact have a ticket for this particular bus, but for a different bus heading to the same place scheduled 15 minutes later. (These were either literally or approximately the only two busses headed to my city that day, so why they were 15 minutes apart is beyond me).
Unfortunately, that other bus was delayed indefinitely. Since the delay was not my fault, I wondered, would they honor my ticket on the already-existing, nearly-empty bus going to the same destination at almost exactly the same time? No, but I could buy another ticket for this other bus. So I ponied up another $50 to save myself from waiting an indefinite number of hours to possibly-or-not get the transport I'd originally paid for, twice.
Just for fun, I looked up the cost of an Uber between these cities and found it was $120 for a 2.5 hour ride. In my particular case, at least, ubering the entire way could have saved me 2+ hours, taken me door-to-door in comfort, allowed me to leave home at any time of my choosing, and ultimately cost about the same as I spent on busses.
Admittedly this was partly because I bought three tickets, one of which was literally last-minute; on the other hand, I was travelling alone and if I'd been with two other people I think the Uber would have come out cheaper regardless.
This is, frankly, bananas. I don't understand how the economics can possibly work this way, and I think Uber could very plausibly offer an intercity transit service competitively with the bus companies.
You might think this is an unfair comparison: Greyhound has to pay for stations with staff, etc, while Uber doesn't. But I think this is actually an argument in Uber's favour: massive, central bus stations were once necessary to co-ordinate passengers, in a time before ubiquitous cellphones and internet. Nowadays, one of the motivating purposes of the central bus station is gone, and it's plausible that on certain routes we should change our mode of transport accordingly.
A more relevant criticism, I think, is that price parity would only be true on short-ish routes; I bet Greyhound is far cheaper for a 12-hour ride. The problem for Greyhound is that if Uber cannibalized the most cost-effective part of the route, they're screwed on the longer ones. This is what kinda happened for the mail, I think, where private carriers ate the profitable parts and left the public carriers with the expensive bits. Greyhound is a private company, though it wouldn't at all surprise me to learn they have implicit or explicit government subsidies for "taking care" of long-distance transportation in the US.
Still, my experience suggests that they're not always doing the best job of this. I think the long-distance bus-riders of the United States deserve better, though (to circle back to the start) I suspect their under-representation among the decision makers on US transport makes it unlikely they'll get it.
p.s. a harder-to-value benefit of taking the bus was that it got me into a conversation I would never have heard or dreamed of otherwise, most of which I can't reasonably post on this blog, but which included an exchange that I still think of fondly:
"The thing is, everyone has free will."
"Yeah, but some people shouldn't."