Night readers,

Hi again. It’s me, Sasha. I’m covered in chocolate stains, tomato paste, and crayfish juice. My whites, most days mottled with some form of hastily-blotted-away water-soluble stain, make me feel like a walking impressionist painting, or maybe some weird kind of modernist art experiment — a canvas changing daily depending on the menu. The kitchen does strange things to a body. Imagine doing three to six hours of light to moderately difficult calisthenics in a booby-trapped medieval weapon showroom while eating small spoonfuls of Thanksgiving gravy and trying to win at jeopardy and you’ve got a close approximation. I often leave the institute in a stupor; my body, brain, and palate scrambled eggs.

Speaking of crayfish, there has perhaps never been a better example of the morphological development of Southern-American riverbank english. At some point in the 16th century, the Old French ‘crevisse’ (referring to all shrimp and crabs) transmogrified into the Middle English mispronunciation ‘crayfish’ through a shift in spelling (creveis) and pronunciation (emphasis on the -re-), and then in the 19th century split itself like a reproducing paramecium across the Appalachians. Crayfish went mostly North, while Crawfish settled in the South. Eventually, I imagine some enterprising river-dweller decided ‘fish’ was too much work for the vocal cords and replaced it with the -dad suffix (as in ‘doodad’)… which gave us the Crawdad that inhabits the lands West of the Appalachians. True to form across the ocean, the modern French word, écrevisse, remains stubbornly identical to its origin.

There is an interesting thread to pull on here regarding the etymology of language and socioeconomic evolution within groups of people. The American South, which history has been less-than-kind to, is more culturally insular than the North — particularly in rural areas. This makes sense: lack of wealth tends to correlate with less physical mobility in the world, and when people stay put culture tends to coagulate regionally, concentrating in the process. In the case of crayfish, such insulation has produced a cornucopia of wonderful appellations. Some of my favourites: crawdaddy, mudbug, crawlpappy, craydad, and waterdog. Crawjinny also has a nice twang to it (these and many other such appellations are painstakingly catalogued in the excellent and sadly pay-gated Database of American Regional English, AKA DARE, maintained by the folks over at Harvard Press).


I am just beginning to turn a corner with my cookery training; going from a completely inept beginner to a kind-of-sort-of-maybe semi-capable novice. I have a very long way to go. My technical skills are still rough, and I’m nowhere near unconscious competence. I have eternities of mincing, chopping, tasting, reducing, and basting ahead of me, but I am forming, finally, what feels like a foundation.

This might not sound like much, but it is a gigantic step. Foundations are everything. Without foundational skills, one is more or less useless, to everyone including themselves. I spent much of my first three months here feeling like a legless blindfolded matador hobbling around in a ring of angry bulls, shrieking and flapping a muleta like some kind of fabulous, brain-damaged bat. In these situations the only way through is to trust in the process. Fail, fail, and try again.

Alongside this foundation, I am getting the hang of some basics. This is the wonderful thing about foundations: they allow you to move past survival mode, begin to form opinions, and appreciate nuance. You can even begin to refine technique, because there is technique to refine, and picking yourself up after a failure is a lot easier with few wins under your belt.


On the travel front, I just returned from a brief, very wet bike trip to Belgium — which I intend to write about — to track down what is rumoured to be the world’s best beer: the Westvleteren 12. There is a mandatory amount of eye-rolling required when dealing with such grandiose claims, however I was intrigued because in this case the claims didn’t come from the brewers, and the beer seemed to have a near-mythical, axiomatic status among beer people.

Was it, in fact, the best beer in the word? Hard to say, given my incomplete knowledge of all beers and the subjective nature of taste, although I can confirm that it was exceedingly tasty. More probably in this case “best” is being used to mean “rare and really good.” The operation is small, and the brewers — Trappist monks who presumably have other duties and interests — brew just enough to support their work. The beer is not abundant, nor is it available anywhere outside the village of Westvleteren, which is really just a few buildings in a field. If you want it, you can get in line in the middle-of-nowhere-West-Flemish countryside in a pre-registered vehicle with an empty boot (or on a bike with a backpack. You will get many strange side-eyes) and buy exactly one case per day per vehicle — not a bottle more.

They do pour pints at a cafeteria next to the abbey, and word has obviously traveled. When I visited at lunch on a Sunday the place was packed like a tin of dehydrated sardines, and there was a lineup of cars around the abbey.


I will soon be launching a new newsletter! It will be called “Foood!”

As with Mexico, Fooood will be an experiment in format and content, albeit one that I am thinking about a little differently. It is a way for me to make room for other subjects in the pages of these letters, and to give myself permission to experiment with a more pure form of food writing.

But more importantly: it will be a way for you to learn with me as I burrow into this odd foreign planet called professional cooking, to share the moments of “aha!” as I have them, and to explore a strata of cookery that exists outside of most literature; certainly outside your typical food publications (there will be no recipes for “weeknight baked ziti” here). This is food with a capital F; If you’ve ever wondered how to catch, cook and eat a pigeon, this is the type of place you might find out.

I’m planning to be more playful with the format than in the Mexico project, and unlike these rangy letters, I will be focussing on short-form content (most of the pieces I’ve written so far are 1-3 minute reads).

I’ll be sending another email out to this list when it’s ready, likely in the next week. If you’d like to sign up you can do so then (I will be maintaining an entirely separate mailing list as I prefer fully consenting readers. You will not be added automatically).

I’ve been hard at work on the first few sets of text. Usually, the formative days of writing on a new project are about as much fun as performing surgery with a rusty shovel, but words have been surprisingly forthcoming, likely because much of what I’m writing about has had five months to careen around my brain with no immediate outlet.

Unfortunately, a bunch of words don’t make a final product, no matter how easy or hard-won. As usual, I’ve spent at least as much time thinking about presentation, operational elements (newsletter publishing platform, list management, etc) and web presentation format as I have writing. Even though 95% of readers will read on their phones or in their email apps (ergo, bypassing the website entirely) there is a small subset that visits the website. This 5% is so small as to be almost insignificant, but I think of it kind of like matting, framing, and hanging a photo. Presentation matters, and insofar as I can add to the experience by enhancing the presentation, I want to.

With this in mind I’ve created a new custom template, specifically for Fooood. This gives me the ability to tweak it to better suit the content (I did something similar with the Mexico project). For example, since I intend the pieces to be shorter, I’ve reduced the column width and vertical margins to give the text a little more breathing space. These changes are all very subtle and unlikely to be noticed, but like any good design, that’s the point.

AI & Chicken Parts

Artwork in my projects can sometimes feel like an afterthought. While I put plenty of effort into the photographs that accompany my work, it is an order-of-magnitude less than the written stuff.

That’s not to say that I don’t care about images, merely that they are not usually the point. Yet, it is hard to imagine publishing without some kind of visual aid; there is a greater-than-the-sum thing about the way text and images play off one-another. Foood presented a new challenge, however, in that I just don’t have enough hours to tackle an on-demand still-life photography project every week.

In lieu of this, I was inspired to explore a medium I have become increasingly interested in over the last year: AI.

I’ve been mostly focused on Adobe Firefly, which belongs to the visual-art focused family of AI tools (Midjourney, DALL-E, Imagen, etc). You type in a prompt — “cartoon of a princess riding a cow in outer space” — and the LLM (short for Language Learning Model) will actually create novel artwork for you, training itself to do so by ingesting vast datasets of Adobe-owned and public-domain content.

In one sense, this is a new tool for an old process. Designing “prompts” for visual artists, and then working to refine their products for inclusion into larger pieces of intellectual property (books, advertisements, websites, apps) is something I’ve done for more than a decade as an art director and professional designer.

In another sense, this is entirely different, as the “artist” here is a piece of code with potentially unlimited artistic range but zero awareness and no conceptual accuracy. You get what you get, and there is no “explaining” to an AI. There is certainly no coercing or threatening it.

In this way, working with AI is mostly a “look ma, no hands” exercise. It may appear that you’re in control, but you really aren’t. No-one is. You just set some parameters, feed the model your prompt and then: hope for brilliance. More often, you get nonsense. In fact, the more specific your vision, the harder it is to get anything even remotely resembling it.

Looked at one way, this imprecision is actually a feature, and it turns the practice of “creating” images into something more akin to concepting art: your power as an author — if you can be called that — is both conceptual and curatorial: ideating, noticing details, inferring meaning, and refining vast, evanescent “maybes” into a singular primeval form… as opposed to the more technical and scrutable process of bringing a specific, contained vision to life.

Infuriatingly, brilliantly, the models often stumble over the most basic details. You may notice, for example, that the roast chicken at the top of this edition has, no wings, and some kind of pointy thigh-shaped appendage protruding from above its leg. It is exactly this wacky volatility that makes AI tools so fertile for experimental art, as the AI becomes an unwitting co-creator/conspirator in the artistic process.

To those who fear that AI will subsume and destroy the jobs of the writer, illustrator, photographer, etc, I say: “spend five minutes trying to get an AI to create anything more complex than a high-school history essay and your fears will be allayed. It is not so easy. I will concede that within the realm of purely technical content authoring this future is not far off — current LLMs seem to be able to do decent jobs of, for example, writing a 10 step tutorial for doing X thing within narrow context Y.

And yet: they get it wrong frequently enough that for now we still need a human on the other side of the screen. Someday soon, a single web-developer in India with an AI tool may replace a team of software engineers in Atlanta. Such is the way of the world. But to me, this is encouraging. It means that we become capable of reaching further and further, accomplishing what we thought, even in our own lifetimes, would be impossible. Every time a door closes another one opens, and for the curious, there will always be an open door beckoning.

I will leave you with a quote by Eric Hoffer, a migratory-worker and longshoreman who nearly lost his eye-sight, lived through the Depression, and went on to become a philosopher at UC Berkley.

“In times of change, learners inherit the earth, while the learned find themselves beautifully equipped to deal with a world that no longer exists.”

Much love, talk soon,