1. It’s dangerous to meet someone from a world with mirror biology, even by hologram. You might fall in love, and your heart’s desire to be with them will push you to visit the ancient sorcerers of Ceres, whose dark rituals can swap you with your reflection at terrible, terrible cost.

2. Curiously, only left-handed people could operate an Everstone. Many right-handers trained themselves to be ambidextrous in hope of touching that power. Not one fooled the Stones.

3. A month ago, his desk plant was a cute little fern in a cute little pot. He left it unwatered on vacation and returned to find it climbing up the wall. Two days later it reached the ceiling. Now the fern is locked in a dark closet, and every day he trims the new growth creeping under the door.

4. January 2047 was the last January; after that, we had no need for a month of icy winds and silent snows. Between July and August we now have the 31 days of Nocturne, when all polite business is conducted under the cool cover of night.

5. The scientists were sad at the goodbye party they threw for Spirit. This confused the little rover covered in an ever-growing pile of cold Martian sand. Didn’t they see how proud it was?

But this evening I was surprised to see Claude 4 struggling to correctly read a recipe, which I thought would be well within its capabilities, so I did a tiny manual experiment. I fed different Claude models the recipe PDF with the prompt “What is the dough hydration in the attached recipe?” The trick is that the yeast is bloomed in some water, and then bloomed yeast + flour + more water make a dough. Also, there’s flour elsewhere in the recipe not in the dough. When computing dough hydration, does the model correctly count both sources of water in the dough but not incorrectly include the extra flour outside the dough? Here are my 12 data points:

Tiny sample size, one test, discount liberally, yet this aligns with my initial vibes-based impressions: Claude 3.7 feels more solid than Claude 4.

• Not by bread alone
• A thousand and one nights
• All men are created equal
• That the strong might not injure the weak
• Look upon my works, ye mighty, and despair
• One small step for man, one giant leap for mankind
• My brother’s keeper
• The truth will set you free
• A plague on both your houses
• Justice, justice you shall pursue

This quality of mystical weight is of course both culturally bound and listener dependent. “All men are created equal” works particularly well for me as an American; “justice, justice” as a Jew. But I think for best results, weighty phrases should be broadly understood: the magic happens in the shared context between listener and speaker. There are other phrases which feel hefty and almost-mystical to me because of the idiosyncratic things I like, but I don’t think they have enough broad cultural context and centuries of reuse to really become synecdoches. Consider:

• The crownless again shall be king (last line of Tolkien’s “not all who wander are lost” poem)
• I am a Jedi, like my father before me (from Return of the Jedi, and the best moment in Star Wars)
• Do or do not – there is on try (also from Star Wars, but perhaps borderline its own cultural element now?)
• We hold these truths to be self-evident (from the Declaration of Independence)

Non-phrases can also have a similar quality. For example, the time interval “a year and a day” recurs in various fictional and legal contexts. I first encountered a year in a day in the fantasy series Wheel of Time, which intentionally remixes real world legends; in one Wheel of Time’s cultures, prisoners of war are honor-bound to serve their captors for a year and a day before being released. Just super cool how this idea stitches together fiction, current legal practice, and centuries old common law to mean “the shortest long time.”

Aaron Copland’s “Lincoln Portrait” is one of my favorite pieces of American propaganda. The orchestration starts quietly, tentatively, and gradually grows to a triumphant finale. A narrator accompanies the orchestra with descriptions of Abraham Lincoln and quotes from some of his famous speeches. The piece ends with the famous conclusion to the Gettysburg Address: “government of the people, by the people, and for the people shall not perish from the earth,” but to my ear the weightiest line is from a note Lincoln wrote to himself: “As I would not be a slave, so I would not be a master.” Such a clear and beautiful line; a succinct statement that Lincoln’s moral circle extends beyond himself, and he does not annoint himself with a privileged position at the center of the moral universe.

The 2024 US Presidential election¹ has come and gone, and soon we return to that most un-Lincoln-like of Presidents. I predict and fear the consequences will be with us for years. Leyla and I are highly likely to be fine; we’re blessed with approximately every tool modern society can provide for weathering uncertain times. Instead, this is a tragedy we collectively inflict especially on those far away from us in space and time: Ukranians no longer supported as they fight to defend their homes; future generations living in a hotter and more violent world where international borders are mere inconveniences to autocrats with guns; even the comparatively moderate stakes of neighbors denied healthcare and our own children coming of age in a United States with dirtier air and hollowed state capacity.

While the consequences of this election are unlikely to include the return of widespread literal slavery in the United States, Lincoln’s words echo in my mind alongside a 19th century abolitionist poem I first encountered via the novel Unsong – “The Present Crisis” by James Russell Lowell:

We see dimly in the Present what is small and what is great,
Slow of faith how weak an arm may turn the iron helm of fate,
But the soul is still oracular; amid the market's din,
List the ominous stern whisper from the Delphic cave within,—
"They enslave their children's children who make compromise with sin."

We have made our collective compromise for the feeble hope of short term gain and vengeance upon perceived enemies, children’s children be damned. That compromise is a vicious, ugly thing, and here a quote from – of all things – a beloved fantasy series has lodged itself in my mind: “But men often mistake killing and revenge for justice. They seldom have the stomach for justice.”²

My Judaism is thoroughly cultural and my Jewish knowledge shallow enough that I mostly just know the hits. But the hits are hits for a reason! At this uncertain moment I’m inspired by an absolute classic: “tzedek, tzedek tirdof” – “justice, justice you shall pursue.”³ I hope in time the better angels of our nature will again prevail, and in the words of Lincoln via Copland, we will “disenthrall ourselves, and then we shall save our country.”

• My friend Doug Kremm, a wise and patient philosopher;
• Philosophy and philosophers I’ve encountered through Effective Altruist-adjacent media, especially Eric Schwitzgebel;
• General contemplation and mellowing with age.

To be clear, I’m still a fan of materialism and tentatively a fan of utilitarian logic! I’ve not adopted faithful belief in any religion, new moral systems, etc. But more and more these empirical strains of thought seem like good tools for understanding the world locally, but perhaps not reflective of the ultimate nature of morality. For example,

• I drop a bowling ball on a teacup, which shatters. Why does this happen? Explanations involving gravity and structural forces in pottery seem most relevant.
• Is it better to make one child laugh or two? Is it worse to kick one puppy or two? Utilitarian logic seems pretty helpful here.

On the other hand, I’m increasingly sympathetic to the notion that such tools are just…orthogonal to the ultimate nature of reality, metaphysics, consciousness. Are there moral truths separate from us? No idea. I’ve not become a fan of any competing explanation, just less confident that the philosophical tools of daily life are relevant out on the frontier.

Higher weekend prices are an example of dynamic pricing: adjusting prices in response to spatio-temporal variation in the supply/demand balance.² Lots of things are already dynamically priced: the commuter rail is cheaper on weekends; french fries are cheaper outside Fenway Park than inside. But I’d like to see way more things be dynamically priced. When dollar prices are too low, [potential] buyers end up paying other costs instead: shortages, waits, congestion, etc. When prices are too high, potentially mutually beneficial exchanges don’t occur.

A highly non-exhaustive list of places where dynamic pricing seems promising:

• Restaurant reservations
• Cafe / restaurant / etc. menu prices
• Roads, i.e. we should have congestion pricing
• Some kinds of appointments, especially low stakes appointments like haircuts. I’m uncertain about dynamic pricing for healthcare or public services.
• Access to fragile natural ecosystems

Dynamic pricing is contrasted with discriminatory pricing, i.e. setting prices based on attributes of the potential buyer: I will sell Alice this cookie for a dollar, but for you, two dollars. Discriminatory pricing shifts transaction surplus to the seller. Say I have a cookie I want to sell. I’ll accept any price above \$3; below that price I’d rather eat it myself. You’d like to buy my cookie for any price up to \$5; above that price you could have ice cream instead. Any price between \$3 and \$5 creates surplus for both of us. But if I can perfectly price discriminate, I can set the price arbitrarily close to your reservation price of \$5 and grab all the surplus for myself (assuming you still buy).

It’s probably not that big of a deal if a single seller subject to robust competition tries to price discriminate; buyers can go elsewhere.³ Indeed, this probably means a seller in a crowded market⁴ rationally should not try to price discriminate. Even so, I’m tentatively opposed to price discrimination:

• It gives me the moral icks, mostly because it’s a form of discrimination. The icks, moral or otherwise, are not a uniformly reliable way to classify the world into good and bad: see vaccines, GMOs, global supply chains, etc. Yet a heuristic of “treat people the same; don’t discriminate” looks pretty good for predicting some of the moral progress of the last centuries: women should have the right to vote, racism is bad, don’t hate foreigners. So I guess I’m inclined towards discrimination=bad as a sane default heuristic until convinced otherwise.
• There are individual instances where particular sellers don’t face competitive forces. Should the only hospital in a hundred mile radius be allowed to price discriminate when someone shows up needing emergency care?
• A world with widespread effective price discrimination is a spooky world. Buying things is barely worth it because sellers charge personalized high prices and capture almost all surplus. Earning money isn’t obviously that helpful if it increases your willingness to pay and thus prices you experience. To be clear, this caricatured world is unlikely to materialize. Even so, instances of price discrimination seem to tear ever so slightly at the social fabric and move us towards a worse world.

I don’t have high confidence that all discriminatory pricing is high. Maybe more seller surplus has good second order effects, like making more sellers. Maybe the rural hospital should be allowed to price discriminate so that we have more rural hospitals. Maybe there are obvious effects I’m missing since I am, in fact, not an economist!

For years, I’ve toyed with the idea of doing a midlife career switch and running a tiny one-man sourdough pizza shop (working name: Workshop Pizza). The actual likelihood of this is…maybe 25%?…but it’s fun to imagine the details. And for sure, my imaginary Workshop Pizza charges more for pizzas on Friday night than Tuesday afternoon.

I’m not really sure why I write this blog. It’s certainly not for the readership, which appears to be mostly my brother and my wife. Maybe it’s helpful to have an online presence, especially in professional-adjacent settings? Mostly, I think I write because the writing itself can be satisfying, because writing helps clarify thoughts, and to provide a time capsule for my future self. None of these goals imply that I should only write long posts about technical work. Following Jeff’s example, recently I’ve been trying to jot more small thoughts down, especially with photos as appropriate. Here’s a recent example on tempering chocolate.

Where I’m at

Eating

• I don’t eat any land animals.
• I don’t eat sea animals that seem especially smart (e.g. octopuses, dolphins).
• I do eat other sea creatures sporadically, perhaps once a week, and increasingly with unease about the ethics involved.
  • Exception: I’m medium confident oysters don’t get moral consideration and would eat them with abandon if that were logistically and financially easy. Clams, mussels, etc. might fall in the same category; not sure.
• I eat lots of dairy and eggs, though when purchasing myself I’ll pay approximately any price premium for a product that seems likely to come from an animal treated better than factory farm default. In practice this means certified humane eggs and organic dairy, but I think I would pay double that price level for a certified super-extra-humane version.
• I’m excited by plant based replacements for animal based products but haven’t found many worth eating. I think my whole list of recommendations is:
  • Beyond and Impossible breakfast sausage patties (not links, which are strangely dramatically worse);
  • Impossible burgers are tolerable but not obviously better than black bean burgers or falafel;
  • Oat milk replaces dairy milk well most of the time; Leyla and I like Minor Figures;
  • MyBacon is super good mycellium based bacon.

Uncertainties

Topics about which I have only loose opinions and low confidence, in no particular order:

1. What is the nature of consciousness? Which animals are conscious, and to what extent?
2. What is the relationship between consciousness, sentience (if that’s even something else), intelligence, and moral status? What does this imply about various animals’ moral status?
3. What is the effect of various diet choices on animal well-being?
   • This gets especially tricky if you start thinking about wild animal welfare. Which you should. It’s just guaranteed to get you to some wild places.
   • Factory farming seems to be outrageously dismally terrible. I’m not 100% confident in this; I have slivers of credence in worlds where animals lack moral status or for opaque utilitarian reasons factory farming is actually good. But if I had a magic “end factory farming immediately and permanently, no take-backs” button, I think I would press it in good faith.
4. What obligations do humans have to maximize / increase / not drastically harm animal well-being via our dietary choices?
5. To what extent are individuals morally responsible for the direct and indirect results of their actions?
6. Is moral responsibility even a coherent concept, or relevant to how we should act?
7. Even absent obligations and responsibilities, do we have opportunities to reduce suffering / increase justice / etc. via our choices?

Put all this together, and my uncertainty set is roughly the triangle with vertices at

• meat eating a la the typical American diet is morally comparable to a lifetime of being a sadist serial killer and anything shy of full veganism is unacceptable;
• meat eating is morally good;
• dietary choices are unrelated to morality and ethics.

My beliefs are not uniform over that triangle. It seems likely to me that many animals have some consciousness, ability to suffer, and moral standing, and that it is morally good to live in such a way as to reduce the suffering of other conscious creatures. Given that, limiting meat consumption seems like a reasonable precautionary principle. Being fully vegan would probably be maximally precautionary, but I find the lifestyle sacrifices very difficult. I weakly think my current diet avoids the great majority of suffering in a typical American diet, and I seem to have converged here because marginal ethical worries and lifestyle sacrifices balance. But on the scale of decades that equilibrium point has moved, and I wouldn’t be surprised to find myself fully vegetarian in five years or vegan in ten.

I also hope that meat and dairy alternatives (including animal-free production of animal products) become so good that there’s no longer a lifestyle cost to avoiding animal-derived foods. I look forward to being the early adopter who pays 10x market rate for lab grown meat and precision fermented dairy.

I enjoy working with chocolate at home; the bonbons above, from a holiday party last December, were some of my prettier work. Bonbons and many other chocolate applications require tempering the chocolate: finely dispersing tiny cocoa butter crystal seeds throughout otherwise melted chocolate.

There are many tempering methods. You can find lots of guides (most of which will work fine) and science explainers (some of which are true). But I’ve found that for tempering small amounts of chocolate at home, maybe 300g or less, the common methods suffer from at least one of these three problems:

1. Low yield – lots of chocolate lost to tools and general mess;
2. Labor intensive or time consuming;
3. Environmental heat transfer makes the process unreliable.

The third problem arises because small amounts of chocolate have a high surface area to volume ratio, so they quickly gain and lose heat to their environment. For example, a recipe might tell you to melt some chocolate in the microwave to 95F. But if you try this with just 100g of chocolate in a glass bowl, you’ll find that by the time your chocolate is at 95F the bowl is burning hot, and carryover heat from the bowl to the chocolate will take it well beyond 95F.

My preferred method for small quantities is a variant of the standard seeding method, but with shaking to agitate. It goes like so:

1. Fully melt your chocolate¹ using the method of your choice. We don’t have a specific temperature target here, so no real subtlety other than not scorching your chocolate. I use a microwave.
2. Transfer the chocolate to a container that has a lid and is only a bit larger than the chocolate volume. Leave the lid off for now. If you already microwaved in this container, congrats: this step is a no-op. I typically use the same mason jar I microwaved in.
3. Prepare 2% of your chocolate’s weight in finely grated dark chocolate. I use a microplane grater.
4. Wait for your chocolate to cool to 94-95F. This can take a few minutes, but it’s hands-off and you don’t need to stir continuously. Tip: an infrared thermometer works well for telling you when you should pay attention, but you really want a thermometer tip in the chocolate for a precise reading to know when it’s go time.
5. Add the grated chocolate (or just grate direclty in at this point), close the lid, and shake hard for about 30 seconds. Think of crushing ice in a cocktail shaker.
6. Your chocolate is now tempered and should be around 90F! A quick stir can confirm no solid chunks remain.

Why this works

• Tempering requires thorough agitation to distribute fine seed crystals uniformly throughout the melted chocolate. Shaking works super well for this and is in general surprisingly powerful, e.g. I can’t crush solid ice in my bare hands, but I can easily crush ice in a cocktail shaker.
  • Stirring by hand isn’t vigorous enough, so it takes a long time and with small quantities you’ll often fall out of the correct temperature ranges before you’ve stirred enough.
  • An immersion blender is vigorous but generates enough heat to be troublesome with small volumes. Over-aeration is also a risk. Plus, every immersion blender I’ve found is way too big for a little jar with 100g of chocolate in it.²
• Letting the chocolate cool from fully melted in the container you shake in means that by shake time the chocolate isn’t rapidly exchanging heat with its container.
• Finely grating the seed chocolate means it can disperse easily and the heat of the melted chocolate can quickly penetrate each bit of solid seed.
• Empirically, at 94-95F the melted chocolate is warm enough to raise the temperature of the room-temp seed to working temp and easily incorporate the seed, resulting in a final mix at a good working temp of 90-91F. Too much warmer and the seed is fully melted, and too much cooler and you end up with thick chocolate and perhaps still-solid flecks of seed.

• Parking lots: 43,797 spaces. The City publishes a nice geojson of parking lots, covering 14.4 million square feet. (14 Harvard Yards!) At 330 square feet per space¹, that’s roughly 44 thousand spots.

• Residential driveways and garages: 27,368 spaces. Using a similar methodology² to the previous street parking post, I estimate 27,368 parking spaces in residential driveways and garages. These don’t include commercial garages.

Adding these to the on street spaces brings my estimated lower bound of Cambridge parking supply to 107,419:

There are about 50,000 households in Cambridge, so this works out to at least two spots per household.

This estimated lower bound doesn’t include privately run garages, and perhaps other flavors of parking I’m not thinking of. A quick glance at Parkopedia suggests there are ~20k private garage spaces – and probably more, because Parkopedia probably isn’t exhaustive. Since the population of Cambridge is roughly 118,000, this means it’s nearly certain that there are more parking spots than people in Cambridge. You hate to see it.