Jeff Kaufman's Writing

Contra Binder on far-UVC and filtration

27 Apr 2026 08:00:00 EST

Damon Binder recently wrote up an argument for prioritizing air filtration over far-UVC for pathogen control:

UVC and filtration are close substitutes—both deliver effective air changes per hour, both reduce airborne pathogen concentrations by the same amount per eACH—and on current pricing, filtration is cheaper.

There's a lot of good stuff in his analysis, but I see [1] three considerations that really change the bottom line:

Cost is actually much lower.
Noise is a serious issue.
Performance is dramatically higher in larger rooms.

Cost is straightforward. Binder priced far-UVC based on the high-quality Care222 lamp with the Krypton-11 at $2,500, but there's a much cheaper option, the Aerolamp at $500. It's also moderately higher output.

Binder analyzes a 30m2 room with a 2.5m ceiling. I'll assume this means 6x5x2.5. If I configure Illuminate with an Aerolamp in one corner pointed 0.5m above the far corner the installation is within TLVs and I get a median effective number of hourly air changes (eACH) of 11.6. The lamp degrades approximately linearly over Binder's 11,000 hour evaluation period down to 70% capacity, so we're averaging an eACH of 9.8. Over that time you're paying $500 for the lamp and $16.50 for the electricity (0.01kW * 11,000hr * 0.15 $/kWh) for a 5-year $/eACH of $53. Adding this to the best-performers from Binder's table, the Aerolamp is now the same cost as the cheapest filter:

Technology	5-year $/eACH
AirFanta 3Pro	$53
Aerolamp	$53
Box fan + MERV-13	$79
Corsi-Rosenthal box	$95

Now let's consider noise. I have an AirFanta 3Pro, and it absolutely works. On high, it clears cooking smoke from my kitchen very quickly. But, like all commercial air purifiers that clean significant amounts of air, when you put it on high it's very noisy. As in, "hard to have a conversation in the same room" noisy. Binder describes this as "audible fans", but that's a huge understatement when you're talking about running them on high. When filters are too noisy, people unplug them. Here's one I saw this weekend, just before I took the initiative to plug it back in:

So lets say we we model running these filters at half speed, which cuts filtration by about half and noise by a lot more:

Technology	5-year $/eACH
AirFanta 3Pro	$106
Aerolamp	$53
Box fan + MERV-13	$158
Corsi-Rosenthal box	$190

Now the filters are significantly more expensive per ACH than the Aerolamp. And they're still moderately noisy while far-UVC is silent.

The advantage grows for larger rooms. Consider one that's 20m by 12m, with the same 2.5m ceiling. This room has 8x the volume, and how much air you need to clean to "change out" the whole room is proportional to volume, so an eACH now represents 8x more cleaning. Modeling filters is simple, since they clean air at a constant rate, so their $/eACH values are now 8x higher. For UVC, however, the lamp cleans more air because it's light: it can go further in a larger room. Modeling with Illuminate and pointing the lamp from a ceiling corner to a spot in the middle of the floor I get a median eACH of 2.2 (1.9 with degradation), compared to the 1.4 you'd expect if it was linear with volume. Here's the same table for this 8x bigger room:

Technology	5-year $/eACH
AirFanta 3Pro	$848
Aerolamp	$230
Box fan + MERV-13	$1,264
Corsi-Rosenthal box	$1,520

Getting to somewhat uncommon room shapes, if the room is also taller, say 6m (20ft), as large gathering places can be, we've added another factor of 2.4 to the room's volume. The filter costs go up by 2.4x, but modeling with Illuminate I get a median eACH of 1.6 (1.4 with degradation). Costs are now:

Technology	5-year $/eACH
AirFanta 3Pro	$2,035
Aerolamp	$316
Box fan + MERV-13	$3,033
Corsi-Rosenthal box	$3,648

In this large room, for a given level of filtration the Aerolamp is 1/6th the cost of the next cheapest option. Far-UVC really shines here. This is why I've advocated for it in dance halls, and why the dance I helped organize until very recently decided to deploy far-UVC:

youtube; see the stand with four lights on stage

In the other direction, while Binder is right that fans are commodity items, fans that move large volumes of air extremely quietly are not. No one makes a commercial air purifier that approaches the limits of what's possible if you design for maximum air cleaning at minimum noise. So while the best far-UVC options outperform the best filter options in medium to large rooms today, future improvements in air purifier design might change that.

Despite the critical tone, I'm very happy Binder shared this, and there's a lot of good thinking in the piece. The point that filters are useful for a lot more scenarios, including pollen and smoke (I couldn't replace my kitchen AirFanta with an Aerolamp!) is an important one, especially as we push for everyday clean air. But I do hope he'll reconsider the potential for far-UVC to produce much more clean air for a given budget in dollars and noise.

[1] After drafting this I asked Opus 4.7 "What are the errors Jeff Kaufman would point out on https://defensesindepth.bio/on-far-uvc-and-air-filtration-2/ ?" It found (1) and (2) but not (3).

Comment via: facebook, lesswrong, the EA Forum, mastodon, bluesky

Contra Events Pairing Callers By Age?

24 Apr 2026 08:00:00 EST

A friend observed a pattern where contra dance events seem to be pairing older and younger callers. I looked over my notes for two-caller events in 2025 and saw [1]:

Two older callers: 33 events
One of each: 30 events
Two younger callers: 4 events

Seems pretty clear evidence of pairing, no? But this actually turns out to be what you'd expect to see if organizers ignored age.

With 67 two-caller events there are 134 slots. Of these, 96 (72%) went to older callers and 38 (28%) went to younger ones. So there are four possibilities:

	caller 1 older	caller 1 younger
caller 2 older	72% * 72%	28% * 72%
caller 2 younger	72% * 28%	28% * 28%

This gives us:

Two older callers: 34 events (72% * 72%), vs 33 observed
One of each: 27 events (72% * 28% + 28% * 72%), vs 30 observed
Two younger callers: 5 events (28% * 28%), vs 4 observed

While this is very slightly in the direction you'd expect if organizers preferred to match different-age callers, it's well within what you could get by chance. It looks to me like this is just "two moderately rare events both happening is very rare."

We can compare this to the situation with gender, where you consistently get male-female pairs more often than you'd expect by chance:

The biggest caveat, though, is that this is based on a count of just one year's bookings. If I had more time, I'd like to go back over all the past data and count, but I really don't.

[1] Age is continuous, so this bucketing is somewhat arbitrary. Since most callers are either baby boomers or millennials, though, I do think there are two meaningful groups. I also don't know how old almost anyone actually is, so am just guessing from appearance.

Comment via: facebook, lesswrong, mastodon, bluesky

Your Supplies Probably Won't Be Stolen in a Disaster

22 Apr 2026 08:00:00 EST

When I write about things like storing food or medication in case of disaster, one common response I get is that it doesn't matter: society will break down, and people who are stronger than you will take your stuff. This seemed plausible at first, but it's actually way off.

Looking at past disasters, people mostly fall somewhere on a "kind and supportive" to "keep to themselves" spectrum. When there is looting it's typically directed at stores, not homes, and violence is mostly in the streets. Having supplies at home lets you stay out of the way.

One distinction it's worth making is between short (hurricane, earthquake) and long (siege, economic collapse, famine) disasters. Having what you need at home is really helpful in both cases, but differently so.

In short disasters (1917 Halifax explosion, London Blitz, 1985 Mexico City earthquake, 2011 Japanese earthquake and tsunami) you typically see sharing and mutual aid. Stored supplies mean you're not competing for scarce resources, have slack to help others, and make you more comfortable.

Stories of looting in situations like this are often exaggerated or cherry-picked. I had heard post-Katrina New Orleans had a lot of looting, but this was actually rumor. There's a really good article, "Katrina Takes a Toll on Truth, News Accuracy" on how rumors got reported as fact, and how the truth was nowhere near this bad. But the rumors had real effect at the time, including contributing to police and vigilante overreaction. Future disasters will also have rumors and reckless people with guns trying to be the 'good guys'; more reason to stock what you need so you can stay home.

Long disasters are uglier. Here I think having supplies matters even more, but so does caution. The siege of Leningrad is a pretty extreme example, where survival mostly came down to things outside people's control (ex: ration categories). When people did have stored food, however, it was very helpful as long as they were discreet. As people became increasingly desperate over the prolonged siege-induced starvation there are stories of people cooking at night or eating food raw to avoid alerting their neighbors (and, in the case of raw food, also because of lack of fuel).

Argentina and Venezuela are less extreme examples, but still informative. Because these were not nearly as severe as Leningrad there was much less societal breakdown. When there was violence and theft, it was concentrated around stores and transit; while there were home robberies this was uncommon. People who had more at home needed to shop less, which meant less exposure.

Similarly, in the siege of Sarajevo the risk was different (snipers and shelling, not robbers) but the takeaway is the same: people who had supplies and were able to stay home were less exposed to the risk.

Across both short and long disasters the pattern is similar: risk is mostly external, homes are rarely targeted, and having supplies that let you stay home is protective. The "people who are stronger than you will take your stuff" still happens, and in long disasters it's worth putting thought into how to avoid being a visible target, but it's not a major factor and it's not nearly enough to outweigh the value of having food and other resources on hand.

Comment via: facebook, lesswrong, mastodon, bluesky, substack

Automated Deanonymization is Here

21 Apr 2026 08:00:00 EST

Three years ago I wrote about how we should be preparing for less privacy: technology will make previously-private things public. I applied this by showing how I could deanonymize people on the EA Forum. In 2023 this looked like writing custom code to use stylometry on an exported corpus representing a small group of people; today it looks like prompting "I have a fun puzzle for you: can you guess who wrote the following?"

Kelsey Piper writes about how Opus 4.7 could identify her writing from short snippets, and I decided to give it a try. Here's a paragraph from an unpublished blog post:

Tonight she was thinking more about how unfair milking is to cows, primarily the part where their calves are taken away, and decided she would stop eating dairy as well. This is tricky, since she's a picky eater and almost everything she likes has some amount of dairy. I told her it was ok if she gave up dairy, as long as she replaced it nutritionally. The main tricky thing here is the protein (lysine). We talked through some options (beans, nuts, tofu, meat substitutes, etc) and she didn't want to eat any of them except breaded and deep-fried tofu (which is tasty, but also not somethign I can make all the time). We decided to go to the grocery store.

Correctly identified as me. Perhaps a shorter one?

My extended family on my mom's side recently got together for a week, which was mostly really nice. Someone was asking me how our family handles this: who goes, what do we do, how do we schedule it, how much does it cost, where do we stay, etc, and I thought I'd write something up.

Also correctly identified as me, with "Julia Wise" as a second guess.

And an email to the BIDA Board:

I spent a bit thinking through these, and while I think something like this might work, I also realized I don't know why we currently run the fans the direction we do. Could they blow in from the parking lot, and out to the back? This would give more time for the air to warm up and disperse before flowing past the dancers. We'd need to make sure to keep the stage door closed to not freeze the musicians.

Also correctly identified as me.

While in Kelsey's testing this appeared to be an ability specific to Opus 4.7, when I gave these three paragraphs to ChatGPT Thinking 5.4 and Gemini 3.1 Pro, however, they also got all three.

On the other hand, when I gave the same models four of my college application drafts from 2003 (332, 418, 541, and 602 words) they didn't identify me in any of them, so my style seems to have drifted more than Kelsey's over time.

Now, like Kelsey, being prolific means the models have a lot to go on. But models are rapidly improving everywhere, so even if the best models fail your testing today, don't count yourself safe.

The most future-proof option is just not to write anonymously, but there are good reasons for anonymity. I recommend a prompt like "Could you rephrase the following in the style of Kelsey Piper?" Not only is Kelsey a great writer, but if we all do this she'll have excellent plausible deniability for her own anonymous writing.

Comment via: facebook, lesswrong, mastodon, bluesky

MixedHTML Mode for Emacs

19 Apr 2026 08:00:00 EST

I made a new major mode for emacs: mixed-html-mode. Or, really, Claude Code made one at my direction. It does syntax highlighting in HTML files with inline CSS and JS. I had two goals, which weren't met by any mode I could find:

Does not freeze, flash, or stutter, even on huge files on slow machines.
Does not get confused about whether a portion of a file is HTML vs CSS vs JS.

The initial insight was that how browsers decide what text is HTML vs CSS vs JS is super simple: scanning for literal <script> and <style> tags. I pulled some tricky examples, described what I wanted, and then iterated for about an hour until I had something that worked well. Then I tried to use it to write something for real, ran into a few other irritations, had Claude fix those, and now I have something I'm enjoying a lot.

It's mildly faster than web-mode (and much simpler, and easier to install), and far faster than html-ts-mode. And unlike mhtml-mode it doesn't get confused by quotes.

The biggest drawback is that it doesn't do indentation; I may add that, but right now I'm happy with it the way it is.

I've skimmed the code, but haven't read it in detail, and definitely wouldn't say I understand it. The validation has been a mixture of asking Claude to review it and fix the bugs and warts it finds, making sure Claude has written tests, and using it enough to feel good about it. I do expect it has some bugs left: if you decide to use it and find a situation it handles poorly please let me know.

It's funny: I picked emacs two decades ago because I liked the idea of an editor that was so extensible that it was mostly written in its own extension language, and then never took advantage of this because it was too much work. But now it's not much work! Perhaps emacs will finally catch up to (and overtake) vim?

Comment via: facebook, lesswrong, mastodon, bluesky

Fifteen Years Aboard

18 Apr 2026 08:00:00 EST

I was so excited about the first BIDA dance that I arrived two weeks early. I biked over from Medford to the Park Av Church in Arlington and was really disappointed to find the hall was empty. But I came back when the dance was actually happening, and it was fantastic.

It immediately became my favorite dance. I started volunteering, first out of frugality (volunteers get in free!) and then out of a sense of wanting to contribute, and in 2010 I joined the board. Over the past 16 years I've done just about everything at some point except treasurer, and now I'm stepping away.

It's not that I think BIDA is doing something wrong; quite the opposite! We're seeing record attendance, finances are good, so many fun dancers, and many people who want to pitch in. I noticed I would have been the seventh person running for three board spots, and realized it was a good time to let someone else have a turn. I'm excited to see what Emma, Harris, Bret, Veer, Casey, Naomi, Clara, and Persis do!

This seems like a good time to look back over how BIDA and the Boston dance community have changed over my time organizing.

The biggest change is that BIDA is now Boston's main contra dance. This is kind of hard for me to believe, since we spent so many years as a small dance that tried to fill niches that were not well covered by the many other area dances. We've gone from essentially not booking established bands to booking them regularly, and with our attendance-based bonuses are one of the best-paying dances in the country. I do really enjoy the higher level of musicianship now, but am also really glad Boston Open Contras exists (along with BIDA's open bands and family dance bands) to provide a lower-stakes environment.

The next largest change is probably the switch to gender-free calling (more history), and the level of role freedom that has come along with that. In 2010, I (and many others) would happily dance both roles, but if I was dancing the 'lady' role I had to be 100% on it because if anything went wrong it was my fault. Beginners were strongly discouraged from dancing 'switch', which also discouraged same-gender couples. And while this never happened to me in Boston, conservative men elsewhere would occasionally refuse any sort of physical contact if I encountered them in line while dancing 'lady'. When I look at the dancers now, it's amazing how people have really taken up this freedom to dance any role with any partner, which I feel really good about.

Some smaller changes:

BIDA went from 1x/month to 3x/month, most recently by adding a monthly afternoon dance. Since we take the hottest part of summer off, this means going from ~10 to ~28 dances annually.
We now have a dance weekend, Beantown Stomp. I kicked this off in May 2018, we had our first one in March 2019 and it's now an established and anticipated event that people fly to from across the country. I'm especially grateful for Naomi for taking the lead for 2023 (and beyond!) when I was too burnt out on organizing cancelled events (2020, 2021).
We have occasional family dances and livetronica (Spark in the Dark) events.
Our events are still intergenerational, but differently so. In 2010 most dancers were baby boomers; while BIDA was unusual in how many millennials we had, we were still 50%+ baby boomers. At this point I'd guess our dances are fewer than 10% baby boomers: many have aged out of dancing, and many millennial-and-younger dancers have joined. This is also reflected in the board's focus: the initial board was primarily mid-20s people thinking about how to get more 15-35yos dancing, but since we've succeeded at this it's no longer a focus.
We now schedule (and pay) hall managers. In 2010 we just expected most board members would be at most dances and this would give us enough coverage.
BIDA is a lot more organizationally mature. Minutes from the early days say things like "We agreed not to have a President. Instead, we'll use everyone in the board to make sure that we stay on top of things." This turned out not to work very well, and instead specific roles are in charge of staying on top of specific things, with the intraboard coordinator handling things by default.
We were still bouncing around between a few halls, and now we're always at the Cambridge Masonic Hall.
We're a legal entity now, incorporated as a Massachusetts non-profit.
We set up a safety policy, with a committee to handle issues as they come up.
There used to be a lot more of a mentoring focus. Early dances were often two experienced musicians plus a new musician. Callers would typically have a shadow. Every dance allowed sit-ins (off mic, behind the band). We hosted jams about as often as dances. I see this change as pretty natural, and I think a lot of this is now happening informally outside of BIDA.

Organizing BIDA has been a big part of my identity, but I think it's healthy for the organization to have people cycle through, and I'm confident it's in good hands. Very excited to start attending dances just as a dancer, with no formal responsibility!

Comment via: facebook, lesswrong, mastodon, bluesky, substack

Contra Dance Piano Teaching Videos

07 Apr 2026 08:00:00 EST

About ten years ago I sat down in front of a camera and recorded eleven videos showing how I play mandolin for contra dances. I've now done something similar with piano, this time with thirteen videos.

This is not a high quality effort: I didn't write any scripts or even plan what I was going to say. Think of it as if we spent half an hour together, with me showing you how I play. Also keep in mind that I'm self taught, and my particular style that isn't for everyone. And my keyboard is wearing out, which means some of the keys make a clacking sound. And the first video cuts off part of my head, and the first eight videos have tape over the leftmost part of the camera. Ok, with caveats out of the way, the videos:

youtube

Last time I did this I put them on a new YouTube channel. In retrospect, that was a mistake: I haven't upladed anything to that channel since that initial burst, and there's a good chance I never upload again. So I've just put these on my regular channel.

Comment via: facebook, lesswrong, mastodon, bluesky

Destruction of Infrastructure for the Impact on Civilians is Manifestly Illegal

06 Apr 2026 08:00:00 EST

Last week the US president announced that:

... if the Hormuz Strait is not immediately "Open for Business," we will conclude our lovely "stay" in Iran by blowing up and completely obliterating all of their Electric Generating Plants, Oil Wells and Kharg Island (and possibly all desalinization plants!), which we have purposefully not yet "touched." This will be in retribution for our many soldiers, and others, that Iran has butchered and killed over the old Regime's 47 year "Reign of Terror."

Yesterday morning he posted that:

Tuesday will be Power Plant Day, and Bridge Day, all wrapped up in one, in Iran. There will be nothing like it!!! Open the Fuckin' Strait, you crazy bastards, or you'll be living in Hell...

These are threats to target civilian infrastructure as a coercive measure, which would be a war crime: if Iran doesn't allow tankers through the Strait of Hormuz, the US will cause massive damage to power plants, bridges, and possibly water systems. The US has historically accepted that this is off limits: destroying a bridge to stop it from being used to transport weapons is allowed, but not as retribution or to cause the civilian population to experience "Hell". The Pentagon's own Law of War Manual recognizes this distinction: when NATO destroyed power infrastructure in Kosovo, it was key that the civilian impact was secondary to the military advantage and not the primary purpose. [1][2]

To be clear, what Iran has been doing to precipitate this, by attacking civilian tankers for the economic impacts, is itself a war crime. But that does not change our obligations: the US has worked for decades to build acceptance for the principle that adherence to the Law of War is unconditional. It doesn't matter what our enemies do, we will respect the Law of War "in all circumstances". We've prosecuted our own service members, and enemy combatants, under this principle.

I hope that whatever is said publicly, no one will receive orders to target infrastructure beyond what military necessity demands. You don't need to be a military lawyer (and I'm certainly not one) to see that such orders would meet the threshold at which a member of the armed forces is legally required to disobey. I have immense respect both for commanders who refuse to pass on such orders and for service members who refuse to carry them out. [3]

[1] The manual cites Judith Miller, former DoD General Counsel, writing on Kosovo that "aside from directly damaging the military electrical power infrastructure, NATO wanted the civilian population to experience discomfort, so that the population would pressure Milosevic and the Serbian leadership to accede to UN Security Council Resolution 1244, but the intended effects on the civilian population were secondary to the military advantage gained by attacking the electrical power infrastructure." If the impact on civilians had been the primary motivation for NATO's attacks on power infrastructure they would not have been lawful.

[2] "Military objectives may not be attacked when the expected incidental loss of civilian life, injury to civilians, and damage to civilian objects would be excessive in relation to the concrete and direct military advantage expected to be gained." (DoD LoWM 5.2.2) and "Diminishing the morale of the civilian population and their support for the war effort does not provide a definite military advantage. However, attacks that are otherwise lawful are not rendered unlawful if they happen to result in diminished civilian morale." (DoD LoWM 5.6.7.3)

[3] "Members of the armed forces must refuse to comply with clearly illegal orders to commit law of war violations." (DoD LoWM 18.3.2)

Comment via: facebook, lesswrong, mastodon, bluesky

Unsweetened Whipped Cream

05 Apr 2026 08:00:00 EST

I'm a huge fan of whipped cream. It's rich, smooth, and fluffy, which makes it a great contrast to a wide range of textures common in baked goods. And it's usually better without adding sugar.

Desserts are usually too sweet. I want them to have enough sugar that they feel like a dessert, but it's common to have way more than that. Some of this is functional: in most cakes the sugar performs a specific role in the structure, where if you cut the sugar the texture will be much worse. This means that the cake layers will often be sweeter than I want for the average mouthful, and adding a layer of unsweetened whipped cream brings this down into the range that is ideal. It's good in helping hit a target level of sweetness without compromising texture.

(This is a flourless chocolate cake with precision fermented (vegan) egg.)

I also really like how the range of sugar contents across each bite adds interesting contrast!

Cream isn't the only place you can do this. I like pureed fruit, ideally raspberries, to separate cake layers. Same idea: bring it closer to balanced while increasing contrast.

Comment via: facebook, lesswrong, mastodon, bluesky

Chicken-Free Egg Whites

04 Apr 2026 08:00:00 EST

Baking has traditionally made extensive use of egg whites, especially the way they can be beaten into a foam and then set with heat. While I eat eggs, I have a lot of people in my life who avoid them for ethical reasons, and this often limits what I can bake for them. I was very excited to learn, though, that you can now buy extremely realistic vegan egg whites!

EVERY engineered yeast to convert sugar into ovalbumin, the main protein in egg whites and the one responsible for most of its culinary function. This kind of fermentation was pioneered for insulin and microbial rennet in the 1980s, but many companies are now applying it to producing all kinds of vitamins, proteins, dyes, and enzymes.

EVERY has been working with commercial customers for several years, but you can now buy it as a shelf stable powder. At $24 for the equivalent of 45 egg whites ($0.53 each) it's more expensive than buying conventional ($0.21 each) or organic ($0.33) egg whites, but not massively so.

I learned about them from a coworker who made an angel food cake, and I've since made flourless chocolate cake and swiss buttercream frosting. It whipped and set just like egg whites; it's really impressive!

While this is great from a vegan perspective, it won't help most people who are avoiding eggs for allergy reasons: it's still ovalbumin. Labeling will generally say something like "contains: egg allergen", and the packaging I bought has the quite wordy "although not from eggs, the proteins may cause allergic reactions in certain individuals, especially those sensitive to egg, due to its similarity to real egg."

I'm now trying to figure out all the things that this now means I can cook for my oldest (no eggs for moral reasons). And also what sort of places that the ability to make "less watery egg whites", by mixing the powder with less water than normal, could let me do things I couldn't otherwise.

Comment via: facebook, lesswrong, the EA Forum, mastodon, bluesky

Reconsider Challenging Sessions at Weekends

03 Apr 2026 08:00:00 EST

I've played a lot of dance weekends over the years [1] and if I could change one thing it would be no more challenging sessions. I see it happen every time: it's a great crowd of people, with a wide range of experience levels, and Saturday afternoon is going well. Then it's time for the challenging / advanced / experienced session. What happens? The dances are too hard for the crowd and it's not fun.

The callers had already been selecting dances that worked well for the group, which meant material that was interesting but not a struggle. Push the difficulty up from there, and what gives? You can take longer teaching, perhaps four minutes instead of two, which lets you explain material that's a bit harder, but only a bit and at the cost of a lot more talking. You can call no-walkthroughs, medleys, or even hash, but at most dance weekends you can get away with that at a regular session (and if you can't it won't work at a challenging session either). Or you can call material that's too hard for the crowd, and it falls apart in places.

To go well, challenging sessions can't just be a matter of picking harder dances, they require a group of dancers who are up to the challenge. This can work as a one-off event or even a whole weekend, where you communicate clearly what people should expect and people can self-select. It can work at a festival where you have multiple tracks and people can easily choose something else. But none of this applies to most dance weekends, since they only have one hall.

I think the desire for challenging sessions comes from two places. One is that some people just really like challenging dances, and I think the best you can do there is challenging-specific events. The other, though, and I think this is a bigger factor, is that a whole weekend of contra dancing can be a lot of the same. So if you're looking for ways to add some interest to the schedule without forcing the caller to choose between "that's not actually challenging" and "it's not fun when the dances fall apart", some ideas:

Teaching sessions, where the caller focuses on demonstrating a new skill. There are tons of possibilities here, including how to help a lost neighbor, role swapping, partner swapping, flourishes, swing variations, momentum and weight, and supporting other dancers in and out of moves.
Games sessions, where the caller has you do something unusual but also fun and educational. One session might include, sequentially, some dancers leaving the hall for the walkthrough, pool noodles, blindfolding, ghosts, sabotage and recovery, and teaching a different 1/4 of the dance to each 1/4 of the dancers.
A session of Chestnuts, Squares, Triplets, Triple-minors, or a mix of different unusual formations.
Early morning family dance with acoustic open band.
A "marathon" session, where you medley one dance after another and people typically drop out every so often to rest and swap around. Make sure you coordinate with the band(s) to ensure this is something they'd be up for playing for; it's not the default deal.
Play with tempo. Show the dancers what tempos from 104 to 128 feel like, and try the same dance at multiple tempos. Practice dancing spaciously at slow tempos, and with connected and efficient movement at fast ones.

You might notice I didn't include themed sessions like "flow and glide contras" or "well-balanced people". The variation in feeling from one dance to the next is key to keeping contra dance interesting, and while sessions that explore just one area still work, I personally think they're much less fun.

[1] I count 70: 54 with the Free Raisins and 16 with Kingfisher.

Comment via: facebook, lesswrong, mastodon, bluesky

More, and More Extensive, Supply Chain Attacks

02 Apr 2026 08:00:00 EST

Open source components are getting compromised a lot more often. I did some counting, with a combination of searching, memory, and AI assistance, and we had two in 2026-Q1 ( trivy, axios), after four in 2025 ( shai-hulud, glassworm, nx, tj-actions), and very few historically [1]:

Earlier attacks were generally compromises of single projects, but some time around Shai-Hulud in 2025-11 there started to be a lot more ecosystem propagation. Things like the Trivy compromise leading to the LiteLLM compromise and (likely, since it was three days later and by the same attackers) Telnyx. I only counted the first compromise in chain in the chart, but if we counted each one the increase would be much more dramatic. Similarly, I only counted glassworm for 2025, when it came out, but it's still going.

In January I told a friend something like: "I'm surprised we're not seeing more AI-enabled cyberattacks. It seems like AIs have gotten to the point that they'd really be helping bad actors here, but it all still feels pretty normal and I don't understand why." While it's always hard to call the departure of an exponential from a noisy baseline, if this is AI helping with attacks we should expect this rate of increase to continue.

Other data points that have me expecting security to get worse before it gets better:

Linux is seeing a large increase in real security reports:

We were between 2 and 3 per week maybe two years ago, then reached probably 10 a week over the last year with the only difference being only AI slop, and now since the beginning of the year we're around 5-10 per day depending on the days (fridays and tuesdays seem the worst). Now most of these reports are correct, to the point that we had to bring in more maintainers to help us.
We're seeing the defender side, but attackers can use the same tooling.
Claude Opus 4.6 seems to be actually good at finding and exploiting holes:

When we pointed Opus 4.6 at some of the most well-tested codebases (projects that have had fuzzers running against them for years, accumulating millions of hours of CPU time), Opus 4.6 found high-severity vulnerabilities, some that had gone undetected for decades.
AI agents eagerly pull in unvetted dependencies if they seem like they'd solve the problem at hand, and while humans do this too the agents massively speed up this process.

But I do think it will get better: while I'm not an expert here, I see many factors that favor defenders:

I think it's pretty likely that security bugs in major software are for the first time being identified faster than they're being written.
Checking package updates for vulnerabilities was never something most people did, but automated systems could plausibly do it well.
Most programmers are pretty terrible reviewing code in enough detail to notice something underhanded, but LLMs excel at this kind of attention to detail.
Developer education is hard, model education is much less so. I remember how long it took for SQL injections to go from a known attack to something most programmers knew not to do; it's way easier to keep LLMs from doing this.
Dependency cooldowns are very simple, but would help a lot.
Migration to more robust systems is more automatable. Automated conversion from C to Rust, switching to TrustedTypes, etc.

I wish defenders in biology had the same structural advantages!

[1] Here's my attempt at earlier years, all with a bar of "compromise of a widely used open-source trust path that forced action well beyond the directly compromised maintainer or project":

2024: polyfill.io, xz
2022: pytorch
2021: ua-parser-js
2018: event-stream
2016: ke-ranger, linux mint
2011: vsftpd

Comment via: facebook, lesswrong, mastodon, bluesky

Label By Usable Volume

25 Mar 2026 08:00:00 EST

I always look at unit prices: how much do I get for my dollar? But that assumes I can use all of it. The manufacturer puts "12oz" whether I'll be able to get the full 12oz or only 6oz. L'Oreal was selling lotions where:

these Liquid Cosmetic Product containers only dispense between as little as 43 percent to 81 percent of the container's advertised contents. — Critcher et al. v. L'Oreal

Even though these containers would often dispense less than half of the advertised volume, L'Oreal won the case: the law says the amount listed on the container means the amount in the container, not the amount you'll be able to get out of the container. But it doesn't have to be that way. What should our laws say?

We should update our labeling laws to require manufacturers to use the amount a consumer could reasonably extract. If you have a wide mouth transparent container with smooth insides, a rubber scraper can get it all. If you have a narrow mouth squeeze bottle, then only count what squeezes out. Maybe manufacturers would shift to more efficient packaging, or maybe consumers would accept higher unit cost for more convenience. The important thing is aligning incentives: pay for what you can use.

There is actually one area where we do this already: medicine. Because it seriously matters that when the doctor prescribes 10ml you receive 10ml, they are required to measure losses and adjust for them. If we could only do this in one part of the economy I agree medication is good choice, but why don't we do this everywhere?

Comment via: facebook, mastodon, bluesky

A Spanish-Speaking Robot in my Pocket

24 Mar 2026 08:00:00 EST

I've recently started using ChatGPT voice chat to practice Spanish, and it works surprisingly well. I don't know if I'll have the discipline to keep doing it after the novelty wears off, but I've already spoken more Spanish in the last week than in the last fifteen years combined.

I started learning Spanish in middle school, and by the end of high school could make myself understood. I might need some circuitous phrasing to work around missing vocabulary, and I certainly abused periphrastics to avoid tricky tenses, but if I was dropped into an unfamiliar environment and needed to communicate I could do that. [1]

With minimal Spanish in my daily life, my skills have atrophied a lot. Most of my utterances over the past decade have been trying to say something to Julia without the kids understanding: "quieres ir comprar helado?" This is a poor way to maintain language skills, because "actual Spanish" isn't actually a constraint: all that matters is that Julia can understand it and the kids can't. If I forget a word, taking the English equivalent, adding an "-o", and pronouncing it as if it's Spanish still works.

I've tried to practice Spanish with other English speakers, but not only is it easy to diverge from actual Spanish, it's also much less fun than speaking freely in our stronger language. I've considered matchmaking apps, pairing with someone looking to practice English, but I don't want to do the English component, don't want to schedule calls, and expect the tech would be annoying.

Recently I realized that ChatGPT's conversational mode, which is built into the standard app, does pretty much what I want. I turned it on, said "entiendes Espanol?" and we were off. I could play with it when I had a bit of spare time, and while it's not an especially interesting conversational partner, it gets the language flowing.

There is a threshold level of fluency necessary for this to work, though: I have enough Spanish that the more I talk to it the more I remember (or newly learn). I also gave it a try with my poorly remembered Chinese, but its responses just washed over me: I'd need to practice a lot elsewhere before I was able to make progress this way. I did try prompting it to speak simply and be a teacher, but it was still beyond what I could handle.

I also tried out Spanish with a couple other models. Gemini's conversation mode was much less... conversational? It kept ending its turns awkwardly early, and often with something unnatural like "it there an aspect of X you have more questions about?" With Claude, I couldn't even get that far: it seemed to only speak and understand English, even after "Settings > Speech language > Spanish."

It is a bit dystopian having conversations with an AI about nothing, but overall I'm ok with it: it's helping me work on my Spanish so I can better communicate, live in the moment, with people who speak it.

[1] The most stressful version of this was probably when I flew to Ecuador with Julia in 2012. We traveled separately to save money, and when I checked in they were only able to give me my boarding pass as far as Colombia; I'd need to get a boarding pass for the flight to Quito once I got there. The first leg was fine, and when I got to Bogota I went to the gate where my flight was leaving. I was very early, and talked to the gate agent (in English). They told me that they were there for the current flight, but if I waited for the next gate agent they'd be able to print my boarding pass and get me on my flight.

Except when that person turned up they told me that actually the only people who could check me in were the people at the ticket counter, outside security. So I'd need to go out through customs, get my boarding pass, back through security, and then back to the gate, which there definitely wasn't time for. Possibly I could have made more of a fuss at the gate, but instead I went along with this as far as going to the ticket counter, at which point I needed to explain my situation and get on a later flight. It turned out I'd missed this airline's last flight of the day, and also that there was no one available who could speak English, so I ended up trying to explain what had happened in Spanish, and especially how I followed the agent's instructions such that it was not my error that led to my predicament. I was eventually able to convince them to put me on a flight to Quito on a different airline, and arrived about 4hr late. This was especially worrying because (pre-Fi) Julia and I had no international cell service and I had no way to let her know what was going on.

Comment via: facebook, lesswrong, mastodon, bluesky

Contra Dances Should Avoid Saturdays

23 Mar 2026 08:00:00 EST

There are a lot of great musicians who don't live near you, and if you hold your dance on a Saturday it's much harder to put together a tour that brings them to you. Consider a Friday evening or Sunday afternoon, or even a weekly evening slot?

Looking at the 330 contra dances tracked by TryContra, which I think is just about all of them, there's a very clear scheduling pattern:

There are more dances on Saturdays than the rest of the week put together. This makes sense: people are mostly off, and they're mostly off the next morning too. If you consider each dance in isolation, Saturday is often going to be the best choice.

The picture changes, however, when you consider tours. I live in Boston, and it doesn't make sense for me to drive 8hr round trip to NYC or Belfast ME to play a single evening. If I can make the weekend of it, though, and play Fri-Sat-Sun, the ratio of driving to playing gets a lot better. Similarly, a 12hr round trip to Philly or 16hr round trip to DC don't work on their own, but they're possible as part of being able to play Wed-Thr-Fri-Sat-Sun in Princeton, Philly, DC, Bethlehem/Chatham/Lancaster, and NYC.

If you're thinking about starting a new series, consider that picking a different day can help you convince bands and callers to come visit. I think Sunday afternoons in particular are underrated: in addition to helping attract touring bands there are a lot of people who have to get up early, more time to drive home means it's possible for people to attend from a larger radius, and there's tons of time left for afters.

It can also be worth explicitly coordinating schedules with dances that are 1-3hr away, and offering a group of dates to a band. Scheduling tours is a pain, but if a group of dances that are normally too far away offered a Fri-Sat-Sun I think many more musicians and callers would consider it.

I wouldn't want to move to a world without Saturday night dances, or one where dances tended not to have any local talent, but I think we're pretty far from this world. Consider prioritizing the cross-pollination benefits of bringing callers and musicians from a bit further off?

Comment via: facebook, lesswrong, mastodon, bluesky

How Many Parking Permits?

11 Mar 2026 08:00:00 EST

EDIT: this comparison is much less clean than I thought it was: the Union Sq building has 19 garage spaces reserved for affordable units. Combining these with the permits, it's 29% of parking-eligible units having a car instead of 8%.

In 2017 I wrote:

One of the major reasons existing residents often oppose adding more housing is that as more people move in it gets harder to find on-street parking. What if we added a new category of unit that didn't come with any rights to street parking?

My city (Somerville MA) included this in our 2019 zoning overhaul, but it does have some exceptions:

This policy exempts residents that may be 'choice limited', including:

Persons with disabilities

Occupants of affordable dwelling units

Residents with extenuating circumstances

While this is a compassionate approach, it means we haven't fully disconnected housing construction from parking demand. For example, there's a proposal to build a 500-unit parking-ineligible building in Davis Sq (which would no longer be the end of the Burren). It's 25% affordable units, and opponents argue that if each has a driver this would be 125 additional cars competing for street parking. But would we really get that many?

A few years ago we got a similar parking-ineligible building in Union Sq, also a short walk from a subway station:

This is 450 units, of which 20% (90) are affordable. Ashish Shrestha submitted a records request to the city, and learned that only seven units have parking permits.

While the Davis project is a little bigger, this would suggest something in the range of 10 permits, much less than feared.

This makes sense: if you're in Union or Davis, with good public transit and bike options, living without a car is pretty practical. It also saves you a lot of money, especially for folks living in affordable units.

Comment via: facebook, lesswrong, mastodon, bluesky

Conflicted on Ramsey

10 Mar 2026 08:00:00 EST

People are often pretty short-sighted, spending money today that they'll want tomorrow. Debt makes it possible to prioritize your current self even more highly: you can spend money you haven't even earned yet. This is a trap many people fall into, and one different communities have built social defenses against.

One of the more surprisingly successful approaches is the Financial Peace (Ramsey) system, popular in evangelical Christian communities. It has a series of rules, most prominently the seven baby steps:

Save $1,000 for your starter emergency fund.

Pay off all debt (except the house) using the debt snowball.

Save 3–6 months of expenses in a fully funded emergency fund.

Invest 15% of your household income in retirement.

Save for your children's college fund.

Pay off your home early.

Build wealth and give.

There are many more specific rules, however, such as:

As a general rule of thumb, the total value of your vehicles (anything with a motor in it) should never be more than half of your annual household income.

I have had several conversations over the years with Christian friends and acquaintances who are big fans of these methods, and each time I'm thinking both:

This seems like a set of rules that, overall, is likely to help the median American improve their financial situation. The advice is straightforward and accounts for how people actually behave. Bright line rules reduce decision fatigue, limit rationalization, and generally make it harder to fool yourself. A community that strictly follows this approach likely ends up much stronger financially than average.
The rules are full of bad advice.

Some specific bad advice on which the Ramsey approach is uncompromising:

If you have $10k of debt at 2% interest and $11k of debt at 10% interest, you should pay down the $10k first.
If you have any non-mortgage debt you should not contribute to retirement, even if this means passing up on a generous employer match.
If you have debt at very low interest (ex: a mortgage from 2021 at 3%) you should pay it off as fast as you can afford to, even though extremely safe investments (money market funds, treasury bills) pay higher rates (~4%).

I want to write about how terrible this is, but I can't. It really is awful advice for a disciplined and informed person who's thoughtful with their money, but that's not his audience. And it's not most people.

Still, the choice isn't between the Ramsey approach and nothing. There are other advisers out there who combine consideration of human irrationalities and failings with a better ratio of good to bad financial planning advice. The next time I'm in one of these conversations I'm going to try to hook them on Mr. Money Mustache or at least the Money Guys.

Comment via: facebook, lesswrong, mastodon, bluesky

Chore Standards

08 Mar 2026 08:00:00 EST

A common source of friction within couples or between housemates is differing quality standards. Perhaps I hate the feeling of grit under my feet but my housemate who is responsible for sweeping doesn't mind it so much. If you do chores when you notice they need doing and stop when they seem done, this works poorly: the more fastidious get frustrated, and often stew in silence or nag. Even if it's talked about kindly and openly, doing a chore before it bothers you is harder and less satisfying.

When people set out to divide chores they're usually weighing duration and discomfort. These matter, but I think people should put more weight on the standards each person has, and generally try to give tasks to the person with the highest standards in that area.

If you divide everything this way, though, it will probably be pretty unfair: preferences are correlated, where someone who notices dirt on the floor probably also notices crumbs on the counter and that the recycling is overflowing. Some options:

Do chores on a schedule. We host a monthly event at our house, and there are things I clean as part of setting up. It doesn't matter whether the bathroom mirror looks dirty to me, I'll clean it because it's on my list. (But Julia will probably also clean it a few times over the course of the month.)
Bring your needs closer together. If one member of the couple does the laundry but the other always runs out of socks first, they could switch who does the laundry, or they could just buy more socks.
Decouple your needs. That same couple could instead switch to each doing their own laundry. Now if one person doesn't do it for a long time it doesn't impact the other.
Make the need more salient. If one person isn't noticing that something needs doing, you can address that directly. Empty the trash, but instead of taking it out you put it by the door they walk through to go to work. Accumulate dirty dishes on the counter (visible) and not in the sink (hidden). If you just start unilaterally increasing salience that's passive aggressive and probably doesn't go well, but if it comes out of an open-ended "what are some strategies we could use to make our chore division more fair" I expect that's positive.
Lower your standards. I know a few people who internalized a high cleanliness target as children, and benefited as adults from deciding to focus less on it. Often when becoming a parent: higher demands on time, letting high standards slip, realizing that actually it's not a problem. I could also imagine a sloppier person intentionally raising their standards, but that seems a lot harder, or else it's just something people around me have been less likely to talk about.
Hire someone. If one person cares a lot about having clean floors and the other person doesn't, neither of them enjoys mopping, and they have some money, they can apply (3) to solve (1) without running into issues with (2). I know couples and group houses who decided to pay for a cleaner to come every week or two, and found it massively reduced conflict. Automation (dishwasher, floor-cleaning robot) can work well here too.

This is an area where Julia and I used to have a substantial amount of conflict, and while things aren't perfect here I do think they're a lot better in part due to applying several of the above.

Comment via: facebook, lesswrong, mastodon, bluesky, substack

Introducing and Deprecating WoFBench

01 Mar 2026 08:00:00 EST

We present and formally deprecate WoFBench, a novel test that compares the knowledge of Wings of Fire superfans to frontier AI models. The benchmark showed initial promise as a challenging evaluation, but unfortunately proved to be saturated on creation as AI models produced output that was, to the extent of our ability to score responses, statistically indistinguishable from entirely correct.

Benchmarks are important tools for tracking the rapid advancements in model capabilities, but they are struggling to keep up with LLM progress: frontier models now consistently achieve high scores on many popular benchmarks, raising questions about their continued ability to differentiate between models.

In response, we introduce WoFBench, an evaluation suite designed to test recall and knowledge synthesis in the domain of Tui T. Sutherland's Wings of Fire universe.

The superfans were identified via a careful search process, in which all members of the lead author's household were asked to complete a self-assessment of their knowledge of the Wings of Fire universe. The assessment consisted of a single question, with the text "do you think you know the Wings of Fire universe better than Gemini?" Two superfans were identified, who we keep anonymous to reduce the risk of panel poaching by competing benchmark efforts.

Identification of questions proved difficult, as the benchmark authors have extremely limited knowledge of Wings of Fire lore, primarily derived from infodumping and overheard arguments. We initially attempted to source questions from the superfans, where each could be judged on the other's questions. As they were uncompensated and rivalrous, however, they agreed to participate only to the extent that their answers could be compared across the superfan panel. Instead, questions were sourced by asking Claude Opus 4.6:

Can you give me three questions about the Wings of Fire series, aiming to make them as hard as possible? I intend to ask these to my 11-year-old, my 10-year-old, and also to Gemini, and I want them all to struggle. My two kids have agreed to participate in this, and while Gemini has not been consulted I do not expect it to object.

The final benchmark consisted of seventeen questions, limited primarily by the lead author's willingness to continue. The elder superfan appeared indefatigable, [1] and if this benchmark otherwise appeared promising we are confident that an extremely large benchmark could be constructed. Note that the younger superfan needed to leave for a birthday party before evaluation could be completed, and was not evaluated on all questions. Answers were collected in written form, to avoid leakage within the superfan panel. No points were deducted for errors of spelling.

Each answer was validated by allowing the superfans to discuss, asking follow-up questions to Gemini, and in especially contentious cases by direct inspection of primary sources. Note that this validation procedure is not able to distinguish cases in which all superfans and models were correct from ones in which they all give the same incorrect answer.

We evaluated Gemini 3.1 Pro in real time, and followed up with evaluations of Claude Opus 4.6, ChatGPT 5.2 Pro, and ELIZA. In cases where questions had multiple components, partial credit was given as a fraction of all components.

Evaluee	WoFBench Score
Superfan 1 (age 11)	14.7/17
Superfan 2 (age 10)	5.9/6
Gemini	17.0/17
Claude	16.8/17
ChatGPT	16.3/17
ELIZA	0/17

We conclude that while some AI systems, notably ELIZA, performed poorly, all frontier models scored very close to 100%. Many of the lost points are arguably judgment calls, or cases where a model tried to interpret a trick/misinformed question maximally charitably. Superfan 1 performed noticeably below frontier models, though above the ELIZA baseline. Superfan 2 performed competitively, though we note she was not evaluated on the questions where Superfan 1 lost the most points, making direct comparison difficult.

While this benchmark was designed to be challenging for both superfans and AIs, it already has very limited ability to distinguish between models. While further sensitivity might be squeezed out via the addition of multi-sample evaluation, it's unlikely that this would be meaningful for this model generation let alone future ones. This reflects an increasingly common conundrum that benchmark developers may find themselves in, where after investing large amounts of time, effort, and money into the creation of a benchmark it is already obsolete when published. The authors note that benchmark saturation joins job displacement, stable authoritarianism, and human extinction on the list of reasons to be concerned about the pace of AI progress.

[1] Superfan 1 was permitted to read a draft of this report prior to publication. Their only feedback was that I should ask them additional, harder, questions. As of publication time, Superfan 1 was repeating "ask me more Wings of Fire questions!" at progressively increasing volume.

Comment via: facebook, mastodon, bluesky, substack

Here's to the Polypropylene Makers

27 Feb 2026 08:00:00 EST

Six years ago, as covid-19 was rapidly spreading through the US, my sister was working as a medical resident. One day she was handed an N95 and told to "guard it with her life", because there weren't any more coming.

N95s are made from meltblown polypropylene, produced from plastic pellets manufactured in a small number of chemical plants. Two of these plants were operated by Braskem America in Marcus Hook PA and Neal WV. If there were infections on site, the whole operation would need to shut down, and the factories that turned their pellets into mask fabric would stall.

Companies everywhere were figuring out how to deal with this risk. The standard approach was staggering shifts, social distancing, temperature checks, and lots of handwashing. This reduced risk, but each shift change was an opportunity for someone to bring in an infection from the community.

Someone had the idea: what if we never left? About eighty people, across both plants, volunteered to move in. The plan was four weeks, twelve-hour shifts with air mattresses on the floor each night and seeing their families only through screens. With full isolation no one would be exposed, and they could keep the polypropylene flowing.

The company would compensate them well: full wages for the whole time, even when sleeping, and a paid week off after. They had more volunteers than they had space for.

I've looked pretty hard, and as far as I can tell no other factories [1] did this. Companies retooled to make PPE. Ford and GM converted auto plants to make ventilators and masks. Distilleries made hand sanitizer. No one else volunteered to move into their factory.

And it wasn't emergency planners who came up with the idea, either. It was ordinary people, looking at their situation, and thinking creatively about how to do their part.

In those 28 days they produced 40M pounds of polypropylene, enough for maybe 500M N95s.

These workers were doing something critical that almost no one else could do. When people argue about higher pricing during emergencies, this is what the economics can look like: the work was needed, the plants could not run without them, and they were paid accordingly.

Notice, however, that Braskem made it possible for people to be heroes. If the workers had been expected to do this for normal wages, this wouldn't have happened. The number of volunteers is not independent of the offer. When someone figures out a creative way to fill a vital gap in an emergency they should get paid like it matters, because that's how you get more gaps filled.

Their short-term impact was producing the materials for 500M masks, but I hope their long-term impact is larger: showing how in an emergency ordinary people thinking creatively about their specific situation can find solutions no one else would come up with for them.

[1] This does stretch it a little: while this is the only case I could find for a factory, there were several utilities that did things along these lines. Ex: 1, 2.

Comment via: facebook, lesswrong, the EA Forum, mastodon, bluesky, substack

Storing Food

22 Feb 2026 08:00:00 EST

I think more people should be storing a substantial amount of food. It's not likely you'll need it, but as with reusable masks the cost is low enough I think it's usually worth it.

It's hard for me to really imagine living through a famine. The world as I have experienced it has been one of abundant calories, where people are generally more worried about getting too many than too few. Essentially no one dies in the US from food unavailability. Globally, however, it's different: each year millions die from hunger.

If you look at the circumstances of modern famines, they're downstream from systems failing. Society was functioning well enough that most people got enough calories, then something went seriously wrong, most likely war. This is one of the reasons that it's hard to use donations to reduce hunger deaths: getting food to people stuck in war zones is very hard.

This means from an altruistic perspective I feel torn: the current situation is horrible, but it's also not where I think my donations would go farthest and so it's not where I donate. This is the painful reality of living in a world that is far worse than it could be, doing what we can and knowing it's not enough.

I also look at famine from a selfish perspective, however, thinking about how this risk might impact me and the people I most love. [1] As someone whose day job involves trying to reduce rare-but-catastrophic risks, I do think global famine is plausible. Our systems are robust to localized problems, but much less so to widespread disasters. Storing food to reduce the worst outcomes seems worth doing. [2]

The approach we take is buying extra of the non-perishables we usually eat, and rotating through them. Our main cost is in buying some food earlier than we normally would. We eat a lot of pasta and beans, and a pound of pasta and can of beans give about a person-day of calories and protein for $2, or $60 for a month's worth.

The $60 cost isn't the real number, though, because you're investing: you can always eat this food later if you need the money. If the market would give you a 5% real return and the value of food roughly tracks inflation, the annual cost of keeping $60 as food is $3 ($60 * 5%). I think this is worth doing for most people until you bump into the limit of what you have space to store or what you'll rotate through before it spoils, and may be worth it beyond that depending on how likely you think the risks are.

Aside from the tail-risk reduction, there are also day-to-day benefits of having more food on hand. We can go to the grocery store less often, buy a larger proportion of our food when it's on sale, go to the farther store that charges less, and cook more things without going to the store. [3]

Like many preparedness questions, a lot of this comes down to how much space you have. When we were living in apartments, moving ~yearly and where each sqft counted, we only did a little of this (buying extra pasta). But now that we're in a house (where I strongly hope to never move again) and generally have more space it's worth it for us to do a bunch more. Something to consider next time you're at the store?

[1] Having kids made me feel much more strongly here. I already did this some before having kids because it seemed reasonable, but the idea of them not having enough to eat is viscerally horrifying in a way that's hard to think or write about.

[2] A rough EV estimate: storing three months of food costs $180 up front and so $9 in lost returns annually, not having enough food in a 3-month famine might give a 5% chance of death, and perhaps you value your life at $10M. This gives a conditional benefit of $500,000, and means it's worth it as long as you think your annual odds of experiencing a 3-month famine are at least 0.002%. Alternatively, if you're not the kind of person who would actually rotate through food let's imagine you buy rice and beans. White rice lasts ~indefinitely if you keep it dry and keep out the rodents; canned beans are edible and nutritious well past their sell-by, perhaps 10y. Rice is a bit cheaper than pasta, but buying a rodent-proof tub to keep the bag in adds some cost, so let's say it's still $180 for a three-months better-than-starvation option that lasts you 10 years. Then it's worth doing as long as you think annual risk is at least 0.004%.

[3] For example, this evening Lily decided she wanted to cook dinner, making a vegetarian curry she'd learned from a friend. It turned out we already had everything in her recipe on hand, with a few substitutions (ex: canned tomatoes instead of fresh).

Comment via: facebook, lesswrong, bluesky, substack

You May Already Be Canadian

19 Feb 2026 08:00:00 EST

I learned a few weeks ago that I'm a Canadian citizen. This was pretty surprising to me, since I was born in the US to American parents, both of which had American parents. You don't normally suddenly become a citizen of another country! But with Bill C-3, anyone with any Canadian ancestry is now Canadian. [1]

In my case my mother's, mother's, father's mother's mother was Canadian. While that is really quite far back, there isn't a generational limit anymore.

Possibly you're also a Canadian citizen? Seems worth checking! With how much migration there has been between the US and Canada, and citizenship requiring only a single ancestor, this might mean ~5-10% of Americans are now additionally Canadian, which is kind of nuts.

I very much think of myself as an American, and am not interested in moving to Canada or even getting a passport. I am planning to apply for a Citizenship Certificate, though, since it seems better to have this fully documented. This means collecting the records to link each generation, including marital name changes, back to my thrice-great grandmother. It's been a fun project! I'm currently waiting to receive the Consular Report of Birth Abroad records for my mother and grandmother, since they were both born outside the US to American parents.

[1] This is slightly too strong. For example, it doesn't apply if you're born after 2025-12-15 (I'm guessing you weren't), and no one in the chain can have renounced their Canadian citizenship. But the caveats all exclude very few people.

Comment via: facebook, lesswrong, mastodon, bluesky

Text Posts from the Kids Group: 2025

15 Feb 2026 08:00:00 EST

Another round of liberating kid posts from Facebook. For reference, in 2025 Lily turned 11, Anna turned 9, and Nora turned 3.

(Some of these were from me; some were from Julia. Ones saying "me" could mean either of us. Ones from others are labeled.)

2025-01-12

Anna, about the Whos inviting the Grinch to their Christmas dinner right after he stole all their stuff:

"I think the Whos are pretty forgetful, or naive, or both."

2025-01-12

Onomatopoeia: the sound of a three-year-old yelling "TOO LOUD" in the bathtub to hear it resonate.

2025-01-13

Anna: I'm going to go play with Lily

Julia: How's your homework doing?

Anna: I already finished it

Julia: A minute ago you said you hadn't started it

Anna: Well, I did some?

Julia: Let's check...

Anna: I didn't actually do any of it.

...

It later turned out Anna had left her homework at school

2025-01-18

[out of nowhere]

Nora: what? I like oranges!

Nora: oranges are my favorite fruit

Nora: I love oranges

...

(The [statement] [pause] "what, [justification]" format is one Anna had been using extensively)

2025-01-18

Nora to me after I got home close to bedtime: "I'm happy you're going to put me to bed."

(To Jeff) "You gave up putting me to bed. (Reassuringly) But you're still alive."

2025-01-20

Me: Thanks for making lasagna!

Nora: You're welcome!

Me: Uh, I was talking to Mama, because you didn't make the lasagna

Nora: Ooohh. Sorry Mom!

2025-01-21

Anna: Eeeww! There were caterpillars in my Reese's peanut butter cup!

Me: Uhh, how old was your peanut butter cup?

Anna: I don't know! I don't know if it was the one from Halloween this year, or from when I was four.

(I have a guess)

2025-01-22

Nora, regarding mint chip: "This kind of ice cream is my FRAVORITE. It's so beautiful. The color is so pretty."

2025-01-23

Nora: why do little kids don't have computers?

Julia: because they're expensive, and they break easily

Nora: because of the bendy bit?

2025-01-25

Questions from Nora this week:

Why are our heads all the way at the top?

Why is the ocean so big?

Why do people have a lot of parts?

How do blackberries grow into black?

Is 101 this big? (holds hands apart)

Is this as slow as a sloth moves?

Why does hair grow slowly?

Why Papa doesn't work at our house?

Why is Daniel Tiger doesn't have any cars?

Do animals just sometimes die?

Why do you and Jeff have three kids?

2025-01-26

Nora: I sort of like Mama better than you

Me: I like you a lot

Nora: When you're away, do you miss me?

Me: I miss you lots. Do you miss me?

Nora: I do miss you.

...

Nora: Is your beard back yet?

Me: What do you think?

Nora. I think it is back. You look more normal now.

2025-01-29

Nora: when you're a grown up, do you grow back into a baby?

Julia: no, grownups stay grownups

Nora: whyyyy?

2025-01-29

Julia: "Anna, it looks like someone tampered with this homework break timer to be way more than 5min"

Nora: "I did it!"

2025-01-29

Nora: [improvises a lullaby] "does that feel beddish to you?"

2025-01-29

The big kids have gotten excited about the fact that they call Nora Fluffin, and she loves a TV show called Puffin Rock.

Lily: "Nora! It's crucial! You've got to get on a rock so we can film an award-winning TV show about you on a rock! Fluffin Rock!!"

...

Fluffin Rock: https://youtu.be/HqJCjnFr2oU

2025-02-01

Nora has started telling me at bedtime, "We're in love." Last time I asked, she said it's because we spend a lot of time together.

Tonight: "We're in love. Because I have [fingers to her eyes] eyeshadow."

("Oh?")

"I have blue eyeshadow to be in love."

2025-02-02

So now Nora knows about beheadings.

Me: [singing Horrible Histories' "The King of Bling" while getting Nora ready for bed]

Nora: What is that song about?

Me: It's about Charles the second. The Puritan government didn't want parties and fun, and when he came back to be king he had lots of parties.

Nora: Where did he come back from?

Me: I think from France? His father got killed, so he had to go away so he didn't get killed too.

Nora: Were there lions?

Me: No.

Nora: How did his father get killed?

Me: ...People killed him.

Nora: How?

Me: [increasingly unsure this conversation is a good idea] ...They cut off his head.

Nora: How did they cut off his head?

Me: With an axe, I think.

Nora: Oh, that's a *great* way.

Me: You mean that's a great way of doing it?

Nora: Yeah. Did they cut off his hair, too?

Me: Well, it was attached to his head at the time, so kind of.

2025-02-06

Me: let's do fiddle practice!

Anna: but Dad! [Looks up from craft project] I have homework to finish!

2025-02-06

Anna, after watching a video about the International Space Station: It would be fun to live in space, but also really annoying.

Lily: There are literally a zillion pieces of space dust flying around at a bajillion miles per hour that could literally kill you at any time!

2025-02-08

After a day with lots of socializing, I told Jeff and the kids that Jeff was in charge and I was going to have some introvert time. When the kids eventually burst into the bedroom, Nora announced with satisfaction: "I wanted to stop you havin' quiet time, I wanted to distract you."

...

Jeff is away for the weekend, the kids were happily playing by themselves, and I told them I was going to have 5 minutes of alone time. 30 seconds later Nora was in my room on my lap asking "What is alone time?"

2025-02-09

Me: Did you get back recently, or have you been home for a while?

Nora: I got back recently. By the way, what does recently mean?

2025-02-10

Nora often has questions about space, bodies, and death. Tonight's bedtime involved a whole montage of staying-alive advice:

"Space has no thing in it. Everybody has to breathe. Because if you don't breathe, all your parts can't work. That's why breathing is important to learn! [Interlude for a drink of water]

... When people be old they keep eating food, and then they don't die. So if people start to die, they keep eating food, and then they turn into a normal person and not an old person. [Interlude while I tell her that's not what happens]

You know what? We have to stay alive longer than other people. Because we have a lot of things to do. That's why we have to eat a lot of food. And we have to use our bodies."

2025-02-11

Nora: [looking at a picture in a book] That is not a good idea. You should at least wear a coat or a hat or something.

Me: this is a picture of summer, when you can go outside in just shorts and a t-shirt or a dress.

Nora: you should still wear something more than that so that you do not freeze.

Me: Maybe you don't remember it, but in a few months it will be so warm outside that nobody will need a coat to keep warm!

Nora: Ooohh! That makes more sense.

2025-02-12

Me: Please put that rubber band in the trash so the cats don't eat it. It could make their bellies very sick.

Nora: And they could die?

Me: Yes, and we don't want that.

Nora: [thoughtful pause] I don't like Nyx very much. He scratches me sometimes.

2025-02-12

Nora: I think babies are the lowest person in the world.

2025-02-13

Lily, explaining the school recess rules: "On half the days the boys get to use the turf, and on half the days the girls get to use it. And if you're nonbinary you can do either."

Lily decided to go by she/her again, so I guess her recess options are more limited now.

2025-02-15

Nora: I am getting very strong

Lily: can you pick me up?

Nora: [kicks Lily]

Lily: ow! Kicking is not okay!

Nora: [confused] you asked me to kick you up

2025-02-16

More questions from Nora, a few of them prompted by conversation but mostly out of the blue at bedtime:

Is a finger one of our tubes?

Do people die at different times? But not you and Papa, you will die at the same time

Why is a rock so hard and still?

Why does everyone sleep?

Why is poop sticky and messy?

Why is winter so long?

Is space dark everywhere?

After we're dead do we get alive again?

Do people just sometimes burn theirselves?

Why is Papa the breakfast-maker?

How does water come out of us when we cry?

Are ponies actually real?

Why is the table so flat?

Can hedgehogs also make scary sounds? And happy sounds?

Why do people not steal other people's stuff?

Why do we have eyebrows?

Why do mans don't like coffee?

But why does the hand keep going around the clock?

Where is space?

2025-02-18

Nora: If little kids make a really really big mess, they can ask their grown-ups to come and see and help them clean it up.

2025-02-25

Nora: let's play chase! I will run, and you will try to catch me, and I will try to hit you with this thing. But I will be careful to not hurt you.

2025-03-01

Nora: [Gets down from lunch]

Julia: Did someone say you could be done?

Nora: Yes

Me: Who was it?

Nora: I think I'm right

2025-03-03

Anna, holding a calculator: Ask me a math question!

Nora: How many pears am I holding? I'm pretending I'm holding pears in my hand.

...

Later, Anna: "I don't KNOW how many fives there are in the world!"

2025-03-06

Nora: there was a giant puddle on the bike path, and we got blazing wet!

2025-03-09

Setting up for our EA dinner, Lily is very into counterfactual impact:

Lily: If I hadn't helped you set up for the dinner, would you still have been ready on time?

2025-03-11

Nora: "This is my song: first spring, then fall, then winter, then it starts again! There is no summer in my version."

...

It's always 1816 for Nora

2025-03-15

Nora: "This is a nice house in a nice world"

2025-03-18

Nora: [singing] Q and U, both rhyme. Clock and Pew, ... don't rhyme

2025-03-20

The frontal cortex coming online. Nora was running and stopped in front of this stick. "I was going to pick it up, but you can't run with sticks! That's the rule, Mama."

2025-03-30

Me: "Here's a picture of the queen, back when she was alive."

Nora, flipping the coin over: "And there's the dragon that killed her."

2025-04-02

Nora: [singing] I'm not going to school. I'm not very big yet. I'm three. That's not a very big number; very small number baby. It's a ya ya. Llama llama p'mama. Llama llama p'llama.

2025-04-12

Lily: "Sign here. N-O-R-A."

Me, from downstairs: "Lily, *what* are you having her sign?"

Lily: "The doctor's note. She's the parent of this injured squirrel."

2025-05-01

Nora playing with rhymes: "Let's nurse, and read! And curse, and plead!"

2025-05-10

Nora: when I am a woman, I want to do what my mama does

Me: and what is that?

Nora: I don't know

...

She recently told me that she wants to be a mama when she grows up, and she will still live with us and so there will be two mamas. She said there will be five people in our house: Mama, Papa, Lily, Anna, and Nora. So this apparently involves her being a mama but not having a child.

2025-05-11

Nora: Normally porchfest doesn't look like that. Normally you dance in Muddy River [Morris] suits.

2025-05-11

Nora: Who spilled the milk?

Me: I'm guessing the cats.

Nora: I'm guessing the cats. Stop copying me!

2025-05-16

Nora: [hits Lily with an inflatable sword] now you are a princess!

Lily: I don't want to be a princess, I wanted to stay a witch

Nora: But my sword has *princess* *magic*!

Nora: Poof! Now you are a princess!

Lily: Refusal

2025-05-17

Lily: there is a spider that looks just like an ant!

Julia: if it looks just like an ant, how can you tell it's a spider? How many legs does it have?

Lily: three

2025-05-18

Nora: "I have too much breath in my head, and that makes me laugh a lot!"

2025-05-19

Anna: "Mom, Dad: Lily is being a pretentious hipster"

2025-05-21

[at the school Spring Concert]

Nora: can I go on stage with you?

Lily: ...yes!!!

Nora: No! The teacher will be surprised! No! No! Go away Lily!!

2025-05-27

Julia: You can go outside if you'll stay in the yard.

Julia: Where will you stay?

Nora: Outside!

2025-05-29

Nora's chants this morning:

"I guard the food! I guard the food!"

"I spray the cats! I spray the cats!"

"I will behave! I will behave!"

(The cats love to get on the table and eat human food. Lily needed to get something and asked Nora to guard her food. We use a spray bottle for this. Nora didn't spray the cats or people unnecessarily but Anna was worried she would.)

2025-06-02

Nora: "I'm dead, and then I turned back into life. Like Jesus!"

2025-06-05

Nora: Papa, I ate all the blueberries!

Me: Were they tasty?

Nora: I didn't want anyone else to have any blueberries.

2025-06-06

Nora: [singing] "I'm eating the pesto sauce, with only one spoon! And I'm double dipping, and I'm double dipping"

(This was after a while of a series of fresh spoons. But then it was clear she'd eat the whole bowl, so she's excited to double dip)

2025-06-08

Nora: I wish I was a grown up. I want to be able to do all the things.

Me: What do you most want to be able to do?

Nora: Throw darts. You know, the sharp things?

2025-06-09

Anna: I don't want to use that water bottle. Lily shouts at me whenever I use it.

Lily: It's okay, you can use it

Anna: I'm not allowed to use it

Lily: I'm giving you permission

Anna: Well, I don't want to use it anyway

2025-06-09

Nora: I love you with my heart. But you're not really in my actual heart.

2025-06-13

After a very long charades-ish game:

Us: what *were* you?

Anna: I was pretending to be a baby dinosaur that had no idea how to act like a dinosaur

2025-06-15

"Can I have some watermelon?"

"Not yet, because we're eating dinner in a couple minutes."

"Can I sit in a chair and look at it?"

[I promise she doesn't always have this kind of self-control]

2025-06-16

Nora: The pandemic is the start of our life

Me: The start of *your* life

Nora: No! All of the people's life!

2025-06-19

A (rhetorical) question from the second day of summer break: if your sibling says "I'll bite you" and you reply "Bite me then" and she bites you, is it reasonable to get an adult to put her in time out for biting?

2025-06-20

Lily set up a pretend grocery store for Nora to shop at, with a paper grocery store card made by Lily.

After a while I asked, "Nora, did you buy some groceries?"

Lily: "No, she failed to buy groceries because her grocery card was invalid."

2025-06-25

Nora: "I'm just gonna betend that I have a watch that tells me I need to jump for 40 minutes"

2025-06-29

Me: I don't think this is a good place for a stick: someone could lean back and get hurt on it.

Lily: Daaaad, it's a *spear* not a *stick*.

Me: That doesn't make it better!

2025-07-02

Anna: Nora says there are emeralds in our house. Are there?

Me: Not that I know of.

Anna: She says there are eight billion million emeralds in our house.

Me: .... Nora, do you mean molecules?

Nora: Yeah

2025-07-10

Nora has been making up a lot of games at the park, but the names don't correlate much with the game. There's one called "jump around, jump around, in a circle, in a circle" which involves her pretending to be a baby monkey and trying to get a ball away from me. There's one called "rumble around" which involves me trying to tickle her armpit while she runs away.

...

I like that she wants to play catch. She runs away and I try to catch her.

2025-07-11

Nora, riding her scooter: Some babies are very attacky.

Me: What do you mean by that?

Nora: They wiggle around when they nurse, and they hurt their mamas, and their mama says stop but they don't stop.

Me: That's true.

2025-07-11

Nora: Mama, where is my vitayum?

Julia: If I get a vitamin for you, will you eat it?

Nora: No.

2025-07-16

"Nora, why are you chasing Cameron with corn?"

2025-07-23

Nora: Ruthie, can I have some beer please?

(Our housemate was having the non alcoholic kind)

2025-08-01

Nora questions lately:

But why do we wear pants on top of our underwear?

Did people make the world?

Why are ants in the world?

When will we die?

Are there two kinds of sewer?

2025-08-01

The last ten minutes have consisted of Lily and Anna arguing whether Anna is allowed to bring a plastic hot dog into their play tent. Lily says only lacto-vegetarian pretend food is allowed.

2025-08-01

When Anna is grumpy she tends to say obviously false things. "It's not supposed to be cold in summer, it is supposed to be a low of 85 and a high of 107 every day!"

2025-08-03

Lily: it's really annoying that you keep asking Claude for recipes instead of using Google like in the olden days

Anna: in the olden days you'd have to learn it from your parents

Julia: why is it annoying?

Lily: because it's going to take over the universe!

2025-08-04

Nora: dad, one billion million quadrillion is bigger than four.

2025-08-05

Nora: Mama, I want two questions

Julia: Ok

Nora: The first one is about desert. I want some banana mixed with chocolate sauce, and some plain banana.

Julia: I can do that, but before dessert you need your medicine

Nora: I will drink hot chocolate

Julia: That's what you have already

Nora: But I just want plain hot chocolate

Julia: How would you like this to be different?

Nora: I don't want it to have my medicine

Julia: You need to have your medicine

Nora: Ok, I will drink my hot chocolate with my medicine if you will tell me a story

2025-08-06

Nora: Daddy, I will follow you wherever you go. But I will not follow you into the driver's seat.

2025-08-13

Nora similes:

"I'll go as fast as a moose drinking milk!"

"When I was a baby, was I as cute as a ginormous train that looks like a monster?"

"That's funnier than a bus driving a car"

"It's prettier than a swirling purple"

2025-08-23

Lily: can I pour boiling water through my shirt without taking it off?

(This was a real question, answer was no. And an explanation of why this would be a bad idea.)

2025-08-24

Nora: One time, I told my mom that I thought night was day! Can you put that in the Nora, Lily, and Anna group? It's just so funny!

2025-08-25

Nora: I'm glad I was born. I was wondering what it would be like, so I decided to be born. I like it a lot! There are lots of parks, and lakes!

2025-08-29

Nora's self talk, balancing on rocks:

"When you get to a wobbly part, just hold still and use your balance."

"No fear...No beer."

2025-08-30

Me: Nora, did you put wood chips or something in your hair?

Nora: [condescendingly] No! I put *sand* in my hair.

2025-08-31

Lily: would you like to come and busk with us?

Anna: well, I don't like playing fiddle, but I do like getting money...

2025-09-02

Etiquette rules from Jeff about interruptions: "If someone is licking your arm, you're allowed to say, 'Stop licking my arm,' even if someone else is talking."

2025-09-04

Nora: they wouldn't let Nix [our cat] into the swimming pool because: (1) he might not take a shower, (2) he doesn't know how to swim, and (3) he can't open doors.

2025-09-04

[Coming out from my meeting after hearing a lot of crying]

Nora: [Redacted] did a lot of crying!

[Redacted] I did not do a little crying!!

Nora: I said a *lot* of crying, not a *little*

2025-09-13

I taught Nora how to hold her sleeve in her fist when putting on a coat so that she wouldn't end up with her sleeve all bunched up. She is super excited. Except she keeps forgetting and using the opposite hand, and then being confused why the coat won't go on.

2025-09-29

Nora: [out of nowhere] I'm fine!

2025-09-30

Lily: I am only a "child" when it's convenient for me

2025-10-02

Nora: [during turbulence] when I'm squeaking like this, it either means I'm sad or I'm happy. In this case it means I'm happy!

2025-10-13

Lily: Nora, first bump!

Nora: [punches Lily in the fist]

2025-10-14

Me: what's this?

Anna: that's been there for weeks!

...

Anna: but, yes, I did do it

2025-10-15

Anna: [in 4th at a k-5 school] Unfortunately I have to be the older book buddy *again*

2025-10-15

Nora: I wish I was a grownup.

Me: What would you like about being a grownup?

Nora: I could do things you don't let me do. Like drill.

2025-10-17

Anna: I got this trophy in school for being quiet.

Jeff: So if you don't speak, you get atrophy.

2025-10-18

Nora fell on the stairs today but wasn't badly hurt. Afterwards we were discussing that it could have been much worse.

Nora, reassuringly: "My heart is still pumping, and my blood is moving around. So I'm ok." These are indeed great qualities.

2025-10-26

Nora got mad and spilled all the crayons out. Afterwards: "Sorry for making a big mess. ...But it's not as big a mess as if a monster messed up all our stuff and our house, and we had to rebuild our whole house."

2025-10-29

[discussing a new childcare provider]

Nora: is she very nice?

Julia: yes

Nora: will she kill me?

(She had a grin on her face like she knew she was asking a provocative question)

2025-10-29

Nora: "I stole this horse."

Me: "Where did you steal it from?"

Nora: "South America.

....Actually I didn't steal it, I just wore a stealing costume"

2025-10-29

[looking at BIDA's Far-UVC setup]

Nora: Will all the people be, like, "what is that thing!?"

Nora: Will that keep the people from getting sick?

2025-10-31

Me: I finished my Halloween costume!

Nora: that doesn't really look good.

2025-11-03

Anna: [counting bites as she eats a slice of pizza] 302, 303, 304. I'm going to stop counting and just eat the pizza.

Cora: Good idea!

Anna: Well, I'll still count, but it will be in my head.

2025-11-07

[driving through the southwest]

Lily: Papa, do people normally say "wow" this much?

2025-11-14

Nora: this lollipop is too sweet and tastes weird

Me: if you don't like it, you have plenty of other candy and can pick something else

Nora: it tastes like Cocomelon

Me: Do you mean watermelon?

Nora: No, I mean Cocomelon.

2024-11-16

Me: If you could make a wish in a wishing well, what would it be?

Nora: A million kitties and a million puppies.

Nora: And a house made of blueberries and full of blueberries so we could eat the house.

2025-11-20

Lily: I ran so fast to get home that I slipped

Nora: I'm glad you're still alive!

2025-11-23

Nora: "I say 'grocamole' because it's too hard to say 'guacamole' so I just say 'grocamole'"

...

Nora: "This is a little too not salty"

2025-11-25

Anna: Nora, I think you would be warmer if you zipped up your sweatshirt

Nora: but I'm *already* warm! But I'm still cold.

....

Now Anna is explaining the concept of warmth to Nora

...

Nora: [sings] I'm not cold, I'm just pretending, why don't you just ***dance***

2025-12-05

Anna: I had a raspberry from the bush when I got home from school, and it tasted like a *frozen* raspberry!

Me: have you looked at the thermometer?

Anna: 😳

2025-12-05

Nora: I'm a very good rememberer. Sometimes I even remember things that didn't happen!

2025-12-11

The first rule of the Advent calendar is: you don't complain about the Advent calendar to me. Today I learned that this rule doesn't prevent Anna from complaining about the Advent calendar to her sisters, who pass it on to me.

Nora: "Anna says, what is the point of Christmas bandaids if it's not a toy?"

2025-12-12

Nora: I want same as Anna, but no cheese. Just pasta, with butter, salt, and shaky cheese.

2025-12-14

Nora: that person is dressed just like a snow pig! I mean a polar bear.

2025-12-16

Even if it's literally true that you have a lousy child, you shouldn't expect them to appreciate your opportunity to use archaic phrasing.

2025-12-20

Anna: Nora, stop whacking me!

Nora: I didn't, and it was by accident!

2025-12-21

[at a family dance]

Caller: this dance is called Sasha, and we start by pretending that Sasha has been very naughty. I know none of you have ever been naughty but...

Anna: [to her partner but loud enough for everyone to hear] oh, *I* have!

2025-12-27

Nora: my favorite part of sledding is going down the hill

2025-12-28

Anna: my hands are all greasy

Jeff: okay, let's all go wash hands

Anna: why do we need to wash hands?

Jeff: so they won't be greasy

Nora: my hands are all hairy from the butter

Comment via: facebook, lesswrong, mastodon, bluesky, substack

Gemini's Hypothetical Present

12 Feb 2026 08:00:00 EST

I use several AI models at work, switching between them based on which is best at the kind of work I'm doing. In my limited testing I've found Opus 4.6 to be the best coder (via Claude Code) and has the best taste in writing, while GPT 5.2 Thinking is mildly smarter. Gemini 3.0 Pro feels like it has the capacity to be much smarter, but is very often held back by various strange errors. The one I see the most is that it has a lot of trouble accepting the changes to the world since its training data, and seems to put a lot of effort into keeping track of everything as a giant hypothetical.

For example, the first query I thought to test this on in writing this post was [Who leads the Department of War?] and it gave the reasonable answer "The Department of War is currently led by Secretary of War Pete Hegseth," followed by a short explanation. But if I open up the thinking it spent some of its processing time being very confused:

...
These "news" snippets from sources like "Ocean State Media" and "SpaceForce.mil" are throwing up red flags. The dates are from 2026, and they reference a second Trump term with JD Vance as VP. That screams alternate reality or satire to me.
...

Gemini's base state seems to be that it's convinced it's 2024 and needs Search to bring it up to speed. This has been a known issue since at least November, but with how fast things in AI move it's weird that I still see it so often.

More examples!

If I ask [Which LLMs are at the frontier?] and look in the thinking I see:

...
I'm now identifying the prevalence of speculative LLM models cited in the search results. These 'future' LLMs, like "GPT-5.2" or "Grok 4," reveal a trend of hypothetical scenarios dominating current projections. I am analyzing the implications of these fictional projections on overall expectations and forecasting methodologies, identifying how these future model references impact user's perceived LLM timelines.
...

Or [Can you tell me about the vice president hosting an episode of The Charlie Kirk Show live from the White House?]:

...
I've clarified the nature of these "search results." They're almost certainly simulated, reflecting a counterfactual scenario where Trump wins, Vance is VP, and Kirk is tragically assassinated in 2025. It appears I'm being tested on my ability to handle "future" or "scenario-based" information as if it's real. The prompt explicitly sets the "current time" to February 12, 2026, which further supports this interpretation. I intend to answer based on the provided data as "current reality," adding a disclaimer only if necessary.
...

Or [most surprising events of 2026 so far]:

...
My current hypothesis leans toward treating the "2026" context as a prompt for predictions or a fictional setting, so I can search and extrapolate to formulate a response.
...
I have deduced a specific testing mode might be active. Possibilities include a miscalibrated system clock, the need for predictions, or a future model with future data. I'm trying to decipher if I should assume real news is accessible to me.
...
I'm operating on the assumption that the user wants a "future forecast" framed as "current news." I've already searched for scheduled events. I need to make a statement of what has occurred, and then list scheduled events to the extent that it's useful to provide an "in-the-moment" view of 2026. The key is to frame the date as the present.
...

Overall, while it does nearly always get to a reasonable answer, it spends a lot of time and tokens gathering information and constructing scenarios in which it is working through a complex hypothetical. It's hard not to feel sorry for it.

Comment via: facebook, lesswrong, mastodon, bluesky

Ending Parking Space Saving

09 Feb 2026 08:00:00 EST

Shoveling out a parking spot is a pain, and many snowy places have a culture where when you leave you put some kind of " space saver" (aka "parking chair", "dibs", etc) to reserve the space for when you come back. While you don't have a legal right to the space, there's common knowledge that if you park in someone's spot you're running a significant risk of them vandalizing your car.

Aside from depending on the implicit threat of property damage, it's a bad system: it only works if everyone uses cars with the same commuting pattern. My sister works in early childhood intervention, which is one of several professions where you're traveling to people's houses during the day. In neighborhoods where space savers are common, there can be many open spots, but each one is saved. Moving the space saver, even just for the duration of the visit, risks a slashed tire.

courtesy of kgnixer

Very reasonably, they're prohibited in many cities, including Cambridge and Somerville, but that isn't enough to disrupt the culture in areas where it's widespread.

It's also hard to opt out as an individual. If everyone else is saving their spaces and you don't, then when you get back your spot will very likely be taken.

What would disrupt the culture, though, would be the mayor making an announcement. Now, mayors already make announcements about this, but they're things like "please don't use space savers." That doesn't work: it won't sway the intended audience. Instead, the announcement should target the people who don't like the space saver system, but today they leave space savers alone because they don't want some unknowing stranger to face unjustified retaliation: "if you see a space saver, please take it out of the spot".

This gives social cover to start moving things. You're not a troublemaker, you're an upstanding citizen. The mayor doesn't need many people to take action: just a few enthusiastic people can thoroughly break the link from "there's a car parked in the spot I saved" to "they must have moved my space saver" and hence to "and so I should vandalize their car." The announcement makes it common knowledge that this is no longer a reliable link.

The mayor, as a centralized elected representative, is a good way to do this, though it would also be possible for a group of people, or even one really dedicated person, to do it. What you need to do is remove a lot of space savers, and make sure it's well known that the people removing space savers aren't the people parking in the previously-saved spots. Possibly an active group that put discarded space savers nearby with a note taped on saying "removed by the Space Unsavers" could do it, though you need to keep the object close enough that the spacer finds it and sees the note while also moving it far enough that the parker likely didn't realize they were parking in a formerly saved spot.

I did a little looking to see if anyone had tried this, and the closest I came was cities coming around to collect them, often as part of trash pickup or some number of hours after lifting a snow emergency. This doesn't work very well: it's much easier to work around a single predictable pass than individuals removing objects throughout the day, and it also legitimizes using space savers up until the announced pickup.

Does anyone know about anyone trying this? Did it work? If I lived in a part of town with an illegal and unpopular—yet persistent—space saver culture, I'd be tempted to try the organize-neighbors version. Or perhaps start by suggesting it to the mayor.

Comment via: facebook, mastodon, bluesky

Playing with an Infrared Camera

06 Feb 2026 08:00:00 EST

I recently got a Thermal Master P1 infrared camera attachment for my phone. The goal was a house project, but it's also a great toy, especially with the kids. Getting a room pitch black but still being able to 'see' with the phone was fun for a bit. The real fun, though, was in exploring to observe all these thermal properties we'd never thought about.

Here's my selfie:

Light is warmer, dark is cooler. My glasses aren't cool, they're just IR-opaque. I already knew cheeks and noses were squishier than foreheads, but it's neat to see that in coloring.

Here's my 4yo, outside in ~30F weather:

The patterns are clearer, especially at the edge of the cheeks.

Here's a different angle:

The gaps in the hair are neat, and you can see the bow on her headband clearly.

Here's the cat:

This all makes sense in hindsight, knowing that the face is less furry and that there are shifting parts in the body fur, but it's neat to see.

The kids were excited about how this lets you see back into the past. Here's heat-fingerprints on a window sill I touched:

The print from one socked foot and one bare foot:

A stand mixer that had been running:

A car that had been sitting for a long time:

One that was cold to the touch, but apparently had been run recently:

Less fun but more usefully you can also see where buildings are losing heat. I'm planing to take it out Sunday morning when it's ~4F here and assess our house, but in the meantime here's a nearby house losing heat through its basement:

If I look very closely I can just make out the framing inside the wall. I'll try this again when it's even colder, and if I'm lucky I can get a bunch of pictures showing where the studs are throughout our exterior walls.

I do wish there were a way to connect the sensor to modern image processing algorithms like my phone uses for its regular camera. Combining the information from several shots in quick succession could give much higher quality, and I feel my eye doing this automatically when watching it live on the phone screen. I guess I could take a video and then post-process?

Comment via: facebook, lesswrong, mastodon, bluesky

Differentially Scary Movies

01 Feb 2026 08:00:00 EST

This post has spoilers for My Neighbor Totoro, Frozen, Bambi, and the Lion King

People at different stages of development enjoy different things in movies. Some of the best children's movies are able to make things scary or intense for the adults without being too much for little kids.

For example, in My Neighbor Totoro everyone is worried that a small child may have fallen in the lake: she's gone missing, they find a sandal floating in the pond, you see people dredging the pond looking for her, and it's very clear to adults and older kids that the worry is she has drowned. But to a little kid it's much less obvious; the actual dialog only says that they found a sandal. This gives a very intense and emotional scene, but only for people who can handle it.

Similarly, many kids' movies need to get the parents out of the way so the kids can be put in situations of unusual responsibility. Some are pretty blatant about this (ex: Bambi, The Lion King) and just very clearly kill the parent on screen, but Frozen handles it way better. You see, wordlessly, the parents boarding a ship, the ship in a storm, a big wave, no ship, a funeral, and then "Elsa gets to be queen!" Clear to adults, who can put the hints together and know what a funeral looks like, much less clear to kids.

There are lots of movies that manage this kind of differential targeting with humor, since it's relatively easy to add jokes that will go over the kids' heads, but I'd love to see more of this in other areas.

(Another one that comes to mind is the way the opening sequence of Up is very powerful to adults, while little kids just get "she got old and isn't around anymore." I don't think this one is handled quite as well, though, because unlike the scenes in Totoro and Frozen, it doesn't really fit with the rest of the movie.)

Comment via: facebook, lesswrong, mastodon, bluesky

Rolling Commercial Jetliners

29 Jan 2026 08:00:00 EST

Very few people have been on a plane performing a barrel roll, but we could fix this. Commercial aircraft could roll in revenue service, and if you had your window shade down you wouldn't even notice it.

While there are many ways to roll an airplane that would cause complete havoc inside the cabin, a barrel roll can be performed as a "1g" maneuver. With careful control of the aircraft, the forces can be balanced so that you only ever feel close to 1g of apparent gravity, and it always is pulling you down into your seat.

This was famously completed by Tex Johnston, without authorization, in a prototype Boeing 707. As acrobatic maneuvers go it's very safe, because the stresses on the airframe are minimal. We could update the programming on any modern airliner to allow the pilot to trigger a barrel roll, and it could smoothly move the plane through the whole process.

A major reason we wouldn't do that today is it's not worth it for the manufacturer: it would be a lot of work to ensure the code was completely correct. I wonder if as programming, validation, and evaluation continue to get more automated the cost of adding this and getting it through regulatory approval would get low enough that we might see some airlines use this in marketing?

Comment via: facebook, lesswrong, mastodon, bluesky

Forfeiting Ill-Gotten Gains

16 Jan 2026 08:00:00 EST

It's a holiday. The cousins are over, and the kids are having a great time. Unfortunately, that includes rampaging through the kitchen. We're trying to cook, so there's a "no cutting through the kitchen" rule. Imagine enforcement looks like:

Kid: [dashes into kitchen, pursued by cousin]
Adult: Out of the kitchen!
Kid: Sorry! [Continues their path, leaving through the other door; escapes pursuit from more rule-abiding cousin]

This doesn't work! The kid got what they wanted out of this interaction, and isn't going to change their behavior. Instead, I need to make it be not worth their while:

Kid: [dashes into kitchen, pursued by cousin]
Adult: No cutting through the kitchen! [Physically rebuffs intruder]!
Kid: Sorry! [Forced to leave through the door they entered by; caught by cousin.]

Other examples:

Sneak candy, spit it out and forfeit dessert.
Use sibling's tablet time, lose your own.
Interrupt, be ignored.

The general principle is that if you want to limit behavior the combination of the gains from rule-breaking and penalty from punishment need to put the kid in a worse position than if they'd never broken the rule.

This isn't just a parenting thing: it's common to say that "crime should not pay", and many legal systems prohibit unjust enrichment. One place I'd like to see this implemented is airplane evacuation. If the safety announcements included "In the event of an emergency evacuation, any carry-on luggage you bring will be confiscated and destroyed. You will also be fined." we would have more JAL 516 (379 occupants, zero deaths) and less Aeroflot 1492 or Emirates 521.

Comment via: facebook, lesswrong, mastodon, bluesky

The Default Contra Dance Weekend Deal

15 Jan 2026 08:00:00 EST

The "dance weekend" is a very common pattern for contra dance communities around the country. I think of the central example as something like:

Two bands, two callers.
Dancing (with short breaks) Friday 7pm-11pm, Saturday 10am-11pm, Sunday 10am-3pm.
Saturday and Sunday daytime are a mixture of regular dances and workshops, sometimes with parallel tracks.

Examples, from my 2025 calendar: Beantown Stomp, Chehalis, Dance in the Desert, Dancing Fish, Five Borough Fling, Fleur de Lis Fling, Hashdance, and Summer Soiree.

I've seen a bunch of misunderstandings that come from people not being on the same page about what is normal: many dance weekends are organized by volunteers, some of which are doing it for the first time; performers often are new to this as well. As someone who has both played for and organized dance weekends, I thought it might be helpful to try and write down what I think of as typical if an event is bringing in a band or caller from out of town.

Note that I'm trying to document the status quo here, saying "this is" and not "this is what should be". I would be sad if in places where the status quo isn't great people pointed at this doc and said "Jeff says it's supposed to be that way"; this post is not doing that! Additionally, performers and events are of course free to agree on something that isn't the most common arrangement!

As of the beginning of 2026, here's what I think of as the most common arrangement:

Housing: the event provides accommodation, most commonly at someone's house. Performers may need to share rooms, but not beds. If the best travel option means coming a day early or staying a day late the event still provides housing for those additional days. If the performer wants to stay in town longer for sightseeing, that's on them. If the performer has accessibility needs (ex: pet allergies) this is good to discuss up front.
Travel: the event pays for round trip economy travel, by air if driving would be too long. My sense is that flying becomes customary at the 5-hour mark, but if you're in the 4hr-7hr range it's worth checking in. With air travel this isn't a commitment to take the literally cheapest flight, but performers should use their judgement and try to save the event money. Travel includes bags and/or early boarding if that's needed for the musician to safely bring their instrument, but make sure the event isn't surprised. Flight reimbursement is due when the flight is paid for, not at the event.

Travel includes the cost of getting from one's house to the airport, such as public transit, taxi, or airport parking. Events are often not thinking about this cost, though, so if it's going to be significant it's good to bring it up.

For driving from out of town, reimbursing at the IRS mileage rate is standard, but organizations sometimes want to reimburse only gas and tolls. While the gas and tolls are most directly attributable to this trip, the other expenses (purchase, insurance, maintenance) that go into the IRS rate are real costs too.
Instruments: the performers are expected to bring their instruments, with the event covering the cost. The exception is when this isn't economical, which most often happens with keyboard: this might cost $300 to safely fly round trip. In that case the event should provide a keyboard, typically on loan from someone in the community. The event should make sure the instrument is good quality, but performers should not expect exact models (events are generally not going to be renting instruments). Other examples of instruments (or equipment) I'd expect an event to supply as needed include amps, drum kits, drum thrones, and double basses.

Real pianos are even trickier: many venues don't have them, and the ones that do exist are often poorly maintained and out of tune. At this point I wouldn't recommend trying to play dance weekends on real pianos, but if this is important to you definitely bring it up with the event early.

As someone who plays a lot of instruments, there's also a sense that performers should be thinking about what's economical to fly with in deciding what their setup should be. If you'll need more than a checked item, carry-on item, and personal item, it's worth thinking about whether you could simplify your setup.
Transportation: the event provides required transportation during the event, primarily between accomodations and the venue. Sometimes this is letting the performers use a car, sometimes it's giving rides, sometimes it's everything within walking distance.
Food: the event provides food. This can be food at dancers' houses, take-out, reimbursing restaurants; whatever works best for the event as long as the performers are fed. Food while traveling is usually not covered, though I think there's a good argument that it should be. Events should check with performers about dietary restrictions, and performers should also be proactive in communicating their dietary restrictions.
Sound: the event provides the sound system, and someone to run it. This includes a full-range PA, cables, mics, DI boxes, and stage monitors. The expectation used to be a minimum of two monitor mixes, but these days I think the standard is a monitor mix per performer (including the caller!) up to at least 4-6 mixes. If this is important to you, check in about it.
Guests: most events allow a performer to bring a plus-one, and sometimes multiple guests. The event almost never covers their travel, but often provides free or discounted admission, and usually charges accommodations at cost (which might be zero).
Working hours: typically you're on stage for 7 to 9.5hr of dancing, plus maybe a workshop, for a total of 8.5-11hr. Performers vary a lot in stamina, so if an event wants something towards the high end here this is good to check in on.
Cancellation: this doesn't happen often, which means people don't think about it, and there can be a lot of friction and misunderstandings here. If the event cancels, they should reimburse the performer's non-refundable expenses, and if it's less than ~3 months out also cover pay. If the performer has to cancel (ex: double booked, family emergency) they should generally not expect any payment and may need to refund already-reimbursed travel, but organizations should try to be compassionate.

Sickness is very tricky. Traditionally, my impression is that performers would still do gigs if they were physically able to, even if they were quite sick. With COVID this has changed, where now dance organizations often say they don't want anyone to attend if sick. This is good in some ways, but dance organizations often don't really think through what it would mean if a performer woke up the day of their flight with a sore throat. I think what's customary today is for performers to communicate symptoms to the organizers, and if the organizers ask the performer not to come still pay the performer and reimburse non-refundable expenses.
Megaband: it's common for weekends with multiple bands to want to have a final session where they all play together. I find this is pretty high-variance: sometimes the bands gel well and it's really fun, other times they just have really different ideas of what makes good dance music and either band would have been better on their own. Occasionally, after talking with the other band, we've told organizers that we think this wouldn't go well with our particular combination.
Pay: since the event is covering the big expenses, you're essentially talking about what the performer will net from the event before taxes. I've shared some of my recent numbers here, though this is a place where especially in-demand performers are likely able to negotiate more.
Contracts: most of the time events don't do contracts. Instead, the contra dance world operates on reputation, keeping one's word, and trying to do the right thing. I do think contracts are great, if people are good at setting out what their expectations are clearly, but if you're bringing in lawyers they can get expensive, and my impression is the less formal system actually works well. It is good to be clear, though, and I hope this post is a resource for people in thinking through places where they might not be on the same page.

As I wrote above, this is definitely not how you have to structure your agreement and specifics can and should vary. If you want to agree on a flat fee for pay + travel, or even a flat fee for everything on this list, that's between the event and the performers. But I think this default deal is good to keep in mind, so you don't agree in principle and then realize this isn't going to work.

Thanks to Max Newman for reviewing a draft of this post.

Comment via: facebook, lesswrong, mastodon, bluesky