Three months ago, I relaunched this old Substack with a promise to write freely, unshackled from a past obsession. In my first post, Putting Chinese nationalism on the shelf, I explained my decision to unpublish my earlier blogs that placed too much emphasis on my ethnic identity. Now I’m looking back to see what I ended up writing about and, as Henrik Karlsson suggests, to examine the shape of these shadows.
This is what writing is like to me. I’ll work on a topic until it bores me, then I pick something else, unrelated. I’m just piling random observations. But then, in December, I skim what I’ve published during the year and realize: the essays are shadows thrown at different angles from the same object.
Race and ethnicity remained a key point of interest, but I turned my focus outward. Instead of dwelling on the Asian diaspora’s connection to their cultural homelands, I noticed far less attention given to how Foundational Black Americans—and their ostensibly domestic-focused civil rights groups—have shaped perceptions of black countries. This inspired Haiti, the Génocidaire's Republic.
Writing about black-white conflict helped me to adapt a more detached perspective, a much-needed corrective to my earlier tendency to bleed all over the page. However, I wasn’t breaking new ground. To push into fresher territory, I turned to the dynamics of conflict between black, brown, and beige, which led to Zanzibar 1964: Postcolonial Genocide Without Whites. But most of the time, I wasn’t so strategic about topic choice.
Write just to feel something
In 2024, a recurring theme that fired me up was censorship resistance. In Can tech fix crime like Ozempic did for obesity?, I took aim at how racial sensitivities choke data access for crime research. A friendlier administration in January might reduce urgent obstacles, but I remain drawn to emerging social technologies like blockchain-based Decentralized Science as lifelines for funding taboo-breaking research.
In Speech, Truth, and Survival, I shifted from the tangible excitement of tools like ShotSpotter and phenotype reports to tackle the abstract: what is free speech really for? I argued that open knowledge production is critical for rebuilding trust in institutions—like hospitals—corroded by BLM-vaxx-maxxing. My aim was to avoid another jeremiad against modernity and instead sketch a roadmap for renewal.
This year likely marked the peak of my flirtation with right-wing politics. I was moved by the Vice President-elect’s memoir, Hillbilly Elegy, which combined a story of class mobility with a message of compassion for interior hill people. Yet venturing further into the mainstream conservative sphere to review Matt Walsh’s Am I Racist? felt torturous. I struggled to find value in it beyond its role as a public opinion barometer.
The platonic ideal of the Substack salonière doesn’t just surf the waves of the Current Thing—she strives to divine the currents of the Next Big Thing. Naturally, my most-viewed (and most-hated) article was The Premature Revolt Against Right-Wing Asians. By Christmas, the MAGA Eye of Sauron had already shifted its gaze from Haitians to Asians as America’s new favorite scapegoats. I’ll have much more to say about this.
Write to feed the machine
In 2024, I wrote to learn. In Can you fact-check the price of a prediction market bet?, I explored how events markets could surpass traditional pundits in high-stakes knowledge aggregation. I concluded that their radical transparency and incentive structures make them a powerful tool for free speech: traders can cut through noise and extract sharp truths, and even profit from the backlash of innumerate haters.
In Gay men are trans women, I argued for the unity of sexual orientation and gender identity while making it a reference against confused activist terms like ‘cis gay man.’ Soon after, I read Kryptogal’s Femininity is Fake, But Masculinity is Real, which approached the topic from a different angle but arrived at many overlapping conclusions. Was it all a waste? Both pieces grasped at the same fundamental truths.
Prophets of technological stagnation often compare human (pre-2023) internet data to ‘the fossil fuel of AI’—a finite resource, destined to run out. Yet researchers are proving them wrong with synthetic data that keeps progress moving. This process mirrors how humans study: we remix the same reality into new shapes, as Kryptogal and I did. Rohit Krishnan expands on this idea in Is AI hitting a wall?:
Humans learn from seeing the same data in a lot of different ways. We read multiple textbooks, we create tests for ourselves, and we learn the material better. There are people who read a mathematics textbook and barely pass high school, and there’s Ramanujan. So you turn the data into all sorts of question and answer formats, graphs, tables, images, god forbid podcasts, combine with other sources and augment them, you can create a formidable dataset with this
Data is the ultimate bullshit detector, and is another reason why I write. The consensual non-consent election was, on some level, an egg-on-my-face moment. After predicting a narrow Kamala victory, I couldn’t conveniently misremember my position for expediency. Instead, I reflected in real-time on the pleasures of a massive preference cascade: the cope, the realignment, and the recalibration of my priors.
Write to develop better taste
But enough about systems: writing also develops taste and expression. In From Roastie to Reliable, I reflected on choosing relationship stability over fleeting thrills. I also experimented with form following function, interweaving personal struggles with structured romance advice. This experimentation parallels the way readers grow more attuned to nuance as they gain more experience, which Ozy Brennan captures:
A speculative fiction reader reads “the door dilated” and goes “oh, the doors open like pupils, and this is familiar to the narrator and the expected way for doors to work.” An inexperienced reader misses this information. A speculative fiction reader is comfortable with the normal confusion that comes from reading a book that depicts a new setting, and can distinguish it from the abnormal confusion that means she missed something. An inexperienced reader finds it unpleasant to be dropped off in a world she doesn’t immediately understand.
In the same way, I usually write for readers willing to engage with ambiguity. However, in Which Race Smells the Worst?, I broke from this pattern. Instead of letting readers ‘fill in the blanks,’ I explicitly stated that I’m not here to declare what is good or bad—I only with to describe the world as it is. Even so, the feedback reminded me that readers bring their own prejudices, and preempting this is folly.
Wrapping up the year with On the slopes, we all fam, I blended my love for winter sports with prehistory and a hope for rapprochement over the Himalayas. To everyone who read, shared, or engaged with my work this year—thank you. In the new year, I’ll continue pushing the boundaries of style, good taste, and the uncanny valley. Stay woke, be safe out there, and keep it creepy. In 2025, there’s lots more to feel.