His father told him plainly

Jedidiah Ames was born in 1868 in Cass County, three years after the war ended. His father had served in it from the beginning to the end and talked about it for the rest of his life — not as an adventure and not as a grievance, but as a thing that had happened and could happen again.

What the son took from that was narrower than what most people take from war stories. He was not interested in the fighting. He was interested in the arrangement. The country had come apart once, along the seams between the states, and it had taken six hundred thousand men to put it back. What struck him was that almost nobody had believed it could happen — right up until it did.

The country is not held together by anybody’s good intention. It is held together by arrangements, and arrangements come apart.

So he asked the question nobody was asking. If it came apart again, and if nobody were coming, which places would still be standing? Not which had the best soldiers. Which could feed themselves.

Six years of evenings

He began in the autumn of 1897 and finished in the spring of 1903. He worked at the kitchen table after supper. His wife Almira went up alone most nights for six years and never once asked him to stop, which is a detail we would not know except that he mentions it in the only personal document he left.

What he built was a table. Forty-five states, scored on the things that would actually decide it — what a state grew, what it could refine, what ground it held, and how far it could get an army before the fuel ran out. Then every pairing the Union then admitted, worked out one at a time.

Nine hundred and ninety contests. He computed all of them by hand.

When the county paper came out to see it, he told them the work was not intended for amusement. He said a state which cannot feed itself is not a state at all, and that the question was worth settling before it was settled for us.

Asked which state should prevail, he declined to answer. He said it depended on the season and would say nothing further.

The two empty columns

There were forty-five states in 1903. Ames knew there would be more. Oklahoma was still Indian Territory; New Mexico and Arizona were still territories.

So he ruled two columns on the right-hand side of the table and left them blank, against the day somebody would need them.

He never filled them. He died in the influenza of 1918, aged fifty.

The letter

In the last week of October he wrote to Almira in pencil, on two sheets of ruled paper. It is the only thing in the collection in his own hand rather than his draughtsman’s lettering, and it is the reason this project exists in the form it does. He gave three instructions and then stopped talking about the game entirely.

Keep it in the family. Use the true figures and not the ones that make it come out even. Do not name a winner. And of the two empty columns: do not suffer anyone to fill them in badly, because better empty than wrong.

If nobody ever plays it, I am content. I did not build it to be played. I built it so that it should exist.

And then he stopped writing about the table and wrote about his wife, and about the six years of evenings, and about the fact that there was no column on it for what she had done — that he had tried for a long while to think of how to score it and could not, and that she had never known he was trying.

A hundred and one years in an attic

The farmhouse stayed in the family and so did the attic, which are two different facts that together explain the delay. Nobody hid anything. Nobody was looking.

The chest came out in 2019 — cedar, unlocked, under a window, beneath four boxes of a great-aunt’s dress patterns. Inside: the folded table, ninety-one hand-lettered cards with four of them water-damaged past reading, a rulebook of thirty-one sewn pages, two dice of turned bone, the clipping from May 1903, and the letter, folded once and worn along the fold from being read.

The letter was on top. Whoever put the chest away put it in last, and that is the only thing in the collection that tells you anything about Almira.

Seven years of finishing it

The first decision was the one he had asked for and it was easy: use the real numbers. The second took considerably longer.

The territories had arrived — Oklahoma in 1907, New Mexico and Arizona in 1912, and then two more in 1959 that Ames could not have anticipated at all. Filling his two empty columns meant adding two hundred and thirty-five contests to a table of nine hundred and ninety, and it meant deciding, on his behalf, how a state reaches an island.

That question took two of the seven years. The rulebook published now is eight pages longer than his. Every one of those pages is about the ocean.

What we found that he could not have

Ames worked his nine hundred and ninety pairings by hand and never published a summary of them. We ran all 1,225 of ours, and two things came out that we suspect would have interested him considerably.

Only a hundred and seven of the 1,225 possible matchups are between states that share a border. Everything else has to be got to. The ordinary case in this country is not a border war — it is a logistics problem.

And in more than a third of all pairings, one state can reach the other and the other cannot reach back at all. Those states cannot win. They can only take longer to lose.

That is not a design decision. That is just where everybody happens to live — and it was as true in 1903 as it is now, which is presumably why he built the thing.

The rules of play, the state dossiers and the tables are in the main article.

Audience of One is where things get made for a single person. Everything here was built to make one specific reader laugh, and that reader is never named — so anyone who finds it is free to suspect it might be them. The work is deliberately disproportionate to its purpose: forty thousand museums on one page, a 1,518-page thesis on ornamental shrimp of which one chapter exists, an announcement of a never-really-played board game with a companion article laying out its fictitious past. The concept, the development and the execution are the enjoyment, and the only measure is whether the person it was for laughed.

For more, visit omnipolar.ai/s/audience-of-one

You take one. Not a color on a board — the actual place, with the actual people, the actual food it grows, the actual fuel it makes, and the actual ground an army would have to cross to reach it.

Then you find out how you’d do.

Draw your state. Play ‘til the death.

Montana v Rhode Island ~ Rules Of Play

335KB ∙ PDF file

Download

Download

For most of you, this ends badly.

Every number in this game comes from published figures, and nothing has been adjusted to make it fair. Your state is scored on the things that would actually decide it.

Myriad considerations, including:

• Can you feed yourself if nobody sells you food? Can you keep the lights on?

• Can you turn crude oil into diesel, or do you have oil and no way to use it?

• Is there anywhere to retreat to, or is your whole population inside an hour’s drive of the coast?

• Can you physically get an army to the other side of the country, and if you got it there, could you supply it?

Most states fail two or three of those questions badly. A few fail nearly all of them. There is one state whose greatest asset is a thing no player is permitted to touch, and one whose people are the finest strategic planners in North America and who will need every bit of it.

The game’s name comes from its flagship matchup, and the two states were chosen for reasons that only become clear once you’ve played it. What can be said here is this: by the measure that would seem to matter most, Montana and Rhode Island are the same. Montana has 1,144,694 people. Rhode Island has 1,114,521. The gap is about one Bozeman’s worth of people.

But Montana is ninety-five times larger.

This game’s very real importance lives in that sentence,

We refuse to spoil the rest.

This is a time-honored game

Jedidiah Ames finished it in the spring of 1903, on a farm in Cass County, after six years of evenings. There were forty-five states then, and his table accounted for every pairing the Union admitted — nine hundred and ninety contests, computed by hand. He left two columns blank against the admission of the western territories and died in 1918 without filling them.

His papers came out of the farmhouse attic in 2019. The family has spent the seven years since adding states to a game that was finished when its author died.

On Risk, since everyone asks.

Risk is a game about armies. This is a game about whether you can feed them. In Risk, Kamchatka is a space on a board; here, Kansas is a real place with an enormous amount of wheat and no good way to defend any of it. Risk hands every player identical pieces and lets skill decide the rest. This game hands you what your state actually has, and then tells you plainly, in the third paragraph of the rulebook, that if you’ve taken Rhode Island against Iowa you have already lost and no amount of cleverness will retrieve it.

Risk settles fights with dice. This one often settles them with the question of whether your opponent has a port.

There’s a trivia element as well, on the principle that you cannot successfully invade a place you know nothing about — a principle the historical record broadly supports. Before you attack a state, its player asks you a question about it. Get it wrong and you go in weaker than you meant to. Questions about the state bird are permitted but considered poor form, on the grounds that it is the only one anybody knows.

Who plays this?

Someone comfortable losing to circumstance rather than to a rival. Someone who finds it interesting rather than maddening that most states in this country could not, under any conditions, get an army more than a few hundred miles from home. Someone willing to sit with a slow situation and watch it arrive, on the theory that watching it arrive is the honest part.

Risk rewards aggression. This rewards patience, arithmetic, and a tolerance for the unfair. It is not a game for people who like to win. It is a game for people who like to be right, which is a different and considerably lonelier pleasure.

Does it really take 3 years to play?

If you play it right: Yes.

Two neighboring states: about forty minutes. Two distant ones: an hour, give or take. Four players: two hours, and the alliances become the whole game — the rules permit any member to be expelled by majority vote of the others, for no stated reason, so four-player games are mostly about not being the third person in anything.

Ten players: eight hours — a weekend, and plan meals. All fifty states at once: sixty-two hours, which is theoretical, and would require fifty people who have all read a twenty-three-page rulebook, which is not a thing that exists.

Playing every possible matchup once, at one per day, takes three years and five months. Nobody has done this.

We do not recommend beginning.

Or do we.

What states would you like to see battle?

If you’re interested in a simulated battle between two states, click this button to send us a message with the names of the two states in the subject line.

Audience of One is where things get made for a single person. Everything here was built to make one specific reader laugh, and that reader is never named — so anyone who finds it is free to suspect it might be them. The work is deliberately disproportionate to its purpose: forty thousand museums on one page, a 1,518-page thesis on ornamental shrimp of which one chapter exists. The concept, the development and the execution are the enjoyment, and the only measure is whether the person it was for laughed.

For more, visit omnipolar.ai/s/audience-of-one

Having never heard of such a shrimp, I felt compelled to start a ridiculous side-bar exchange with the one person who really gets my sense of humor (also in attendance at the wedding). For the sake of anonymity, I shall refer to this person as they/them/friend.

Quickly, a one-off Claude prompt fed a text that, well, turned into a full-blown research thesis with the following stated objective:

To establish that ornamental shrimp breeding is the most consequential domestication event in the history of life on Earth, and the only one its participants have been awake for.

Not wanting to put in any real work of my own — I will only go so far for a joke — I went to Benny, the Claude chat instance who serves as my trusty Research Analyst.

The Prompt

My original prompt to Benny:

The Abstract

After a little back-and-forth dialing in the degree of absurdity, we produced this abstract for what Benny specced to be a 1,518-page thesis officially titled “THE COMPRESSED NEOLITHIC: Ornamental Caridina Breeding as the Terminal Event in the History of Selective Agency.”

Abstract ~ The Compressed Neolithic (aka Thesis about Ornamental Shrimp)

56.5KB ∙ PDF file

Download

Download

Pick a Chapter for Review

And since, why stop there, I gave my friend the option of reviewing whichever chapter they wanted from the following:

1. The Problem of the Invisible Domestication: On the Impossibility of Noticing Fifteen Thousand Years, With Reference to the Dog, Who Did Not Notice Either
   Antecedents in the Domestication Record: Wheat, Cattle, the Horse, the Silkworm, and Four Species That Declined

2. The Compression: Ninety Days and the Collapse of the Neolithic Interval, Wherein an Entire Agricultural Revolution Is Observed to Occur in a Spare Bedroom in Under Six Years

3. Grading as Epistemology: S, SS, SSS, and the Fourth Grade That Two Men in Osaka Refuse to Recognise

4. The Price of Information: Market Structure in the Ornamental Trade, With Attention to the Four-Hundred-Dollar Animal That Weighs Less Than a Postage Stamp

5. Manufactured Scarcity in a Species That Reproduces Exponentially: The Breeder's Dilemma, Being the Obligation to Sell Before One's Own Success Destroys the Price

6. Reversion, or the Second Law in a Two-Gallon Container: On the Impossibility of Rest, and What Is Found Upon Returning From an Eleven-Day Holiday
   The Rehearsal: What Is Owed to the Specimen That Graded Poorly

7. The Propagation Thesis: A Business Plan for the Cessation of Human Conflict, Including Unit Economics, Milestones to Year Seven, and an Exit Strategy
   Objections, Anticipated and Received: Including the Objection That This Is Not a Thesis, Answered at Length

8. The Inversion: On the Original Surviving Only in the Form of Its Improvements

• Appendix A — Complete Grading Nomenclature, with Contested Cases: Including Eleven Patterns Named by Marketers and Subsequently Accepted by Everyone

• Appendix B — The Culling Register, 2019–2026: 41,206 Entries, Graded, Dated, and Dispositioned, Without Commentary

• Appendix C — Daily Water Parameter Logs: Two Thousand Five Hundred and Fifty-Five Consecutive Mornings

• Appendix D — Survey Instrument and Schedule of Non-Responses: The Instrument, and the Two Hundred and Nine People Who Did Not Return It

• Appendix E — Correspondence with Breeders: Being Letters Sent, and the Four Replies

• Appendix F — Financial Model, Propagation Thesis: Seven-Year Projection to Global Harmony, in Constant Dollars

• Appendix G — Colour Reproduction Failures, a Portfolio: Forty Pages Demonstrating That the Animal Cannot Be Photographed

• Appendix H — Glossary: Terms of Art, Including Nine the Hobby Uses Constantly and Cannot Define

• Appendix I — Index of Named Specimens: Every Shrimp Given a Name, and the Date It Stopped

• Appendix J — Figures: Sixty-Four Plates, of Which Figure 4 Is Gerald

They wisely chose Chapter 7.

The Chapter That’ll Never Get Read

So Benny and I set out to write Chapter 7, knowing it would never be read.

Chapter 7: The Propagation Thesis: A Business Plan for the Cessation of Human Conflict

950KB ∙ PDF file

Download

Download

The joke, as far as I was concerned, was in the creation.

If you’d like to see another chapter, or even the entire thesis, please let us know.

Use code-word “Dumb.”

Epilogue: Questions for an Ornamental Shrimp Breeder

To be clear, this ‘project’ was always about humoring ourselves; it was never about mocking the person with this passion.

That said, in case we ended up speaking with this person at the wedding, we thought it wise to have legit questions that we could raise sincerely, simply to learn more.

1. Are your ornamental shrimp Neocaridina or Caridina? — I ask because my understanding is that determines everything else, since one breeds in tap water and sells for a few dollars and the other needs precise water chemistry and can run to hundreds per animal.

2. Which color line are you running, and do you keep the colors in separate tanks? — Is it true that every Neocaridina color is the same species, so reds and blues in one tank produce brown offspring within a few generations? I’ve read that’s the most common way people lose a colony without realizing it.

3. Are you running RO with a remineralizer, or straight tap? — I ask because my understanding is that RO means stripping the water to nothing and rebuilding it to a target, which sounds like a weekly ritual and a real commitment.

4. Do you cull? — My understanding is that’s what actually separates keeping shrimp from holding a line, and that nobody in the hobby brings it up unprompted because it’s the least pleasant part.

5. Have you tried the local aquarium club auctions? — I’ve read that’s the standard way breeders build a reputation, because buyers see the animals in person and word travels regionally in a way it doesn’t online.

Alas, I never had the opportunity to speak with the designer shrimp breeder.

But feel free to use these if you ever do run into them.

Audience of One is where things get made for a single person. Everything here was built to make one specific reader laugh, and that reader is never named — so anyone who finds it is free to suspect it might be them. The work is deliberately disproportionate to its purpose: forty thousand museums on one page, a 1,518-page thesis on ornamental shrimp of which one chapter exists, an announcement of a never-really-played board game with a companion article laying out its fictious past. The concept, the development and the execution are the enjoyment, and the only measure is whether the person it was for laughed.

For more, visit omnipolar.ai/s/audience-of-one

We at The Museum of Museums firmly believe that museums, for all the honoring they do of other museums, deserve to be honored themselves.

Museums are the unsung heroes of our history and culture.

The list of museums attached below — 39,980 museums across the country — has been said to be the most comprehensive directory of our country’s actively operational museums ever assembled. It’s definitely the only list with made-up descriptions for museums in cases where we could not surface a real description.

Fabricated descriptions are marked with an asterisk. Everything else is true.

The Museum Of Museums Directory (july 31 2026)

9.72MB ∙ PDF file

Download

Download

Also, after vigorous debate, though mostly because our lawyer insisted, we did not invent descriptions for museums concerned with sensitive topics such as war, genocide, slavery, tribal history, ethnicities, or religion. Too bad; we had material.

If you know of an active museum that’s not listed here, or if you spot any incorrect information, let us know and we’ll update it within one hour.

And remember: Now museum, now you don’t.

—Chad Barker
Museum Curator
The Museum of Museums

p.s. Exciting news! We’ll soon open the physical location of The Museum of Museums. The project is in stealth mode and security is a factor, so we cannot yet reveal the location or timing, But, rest assured, it’ll be great. Subscribe to this Substack to receive updates.

Audience of One is where things get made for a single person. Everything here was built to make one specific reader laugh, and that reader is never named — so anyone who finds it is free to suspect it might be them. The work is deliberately disproportionate to its purpose: forty thousand museums on one page, a 1,518-page thesis on ornamental shrimp of which one chapter exists. The concept, the development and the execution are the enjoyment, and the only measure is whether the person it was for laughed.

For more, visit omnipolar.ai/s/audience-of-one