Five

All Proceeds Go to the Artist

Somewhere in the Internet Archive’s holdings there is a capture of mycuriocards.com from 20 July 2017, and on it, under the three new cards released two days earlier, is a single line of text:

“All proceeds from the sale of cards 21-23 go directly to the artist, @robek_world”

One week later, the same line with a different name:

“All proceeds from the sale of cards 11-13 go directly to the artist, @cryptograffiti”

That is the whole business model, and it is worth pausing on how strange it was. Curio Cards was a company — it had a footer reading “© Copyright 2017 | Mothership Inc.” and a founder who spent May 2017 publicly asking for venture capital — and its product shipped with a promise that the company would keep none of the revenue from it.

Hunt said it on air before he said it anywhere else. On The Bitcoin Group #145, 9 June 2017, in the story-of-the-week segment, he gave what is still the most compact statement of the project’s purpose anyone has managed:

“My story of the week is a side project I’ve been working on. You can check it out at mycuriocards.com. These are digital trading cards on the Ethereum blockchain. This week’s cards are Crypto Graffiti cards and all proceeds from the sales of these cards go directly to the artist.”

Later in the same episode he repeated it, and the second version carries something the first does not:

“And I haven’t been promoting it much But I just wanted to mention these are CryptoGraphete cards and all the proceeds go to the artist So thank you for supporting artists with KiraOcarts.”¹

And I haven’t been promoting it much. A founder admitting on his own show that he has not been pushing the thing, in the same sentence as the pitch. It is a small moment and it gives you the temperature of 2017 exactly.

The promise was also the recruitment pitch. A tweet from 30 May 2017, 19:55 UTC, retweeting the project account:

“RT @MyCurioCards: ARTISTS WANTED! Sign up today to Create your own #CurioCards http://bit.ly/CurioArtistsWanted - all proceeds to the artists!”

And it was still the headline when the site was redesigned the following January. The FAQ on the 18 January 2018 capture states the mission in two sentences, the second of which is shorter than it needs to be:

“Curio Card’s mission is to create a platform for artists and collectors that finally benefits artists.”

“Curio is changing that.”

Whether they kept it

A promise printed on a website is evidence of an intention, not of a payment. The useful question is whether anyone was actually paid, and here the archive does something archives rarely do: it answers.

On 20 July 2017, two days after Robek’s three cards sold out, Robek was a guest on Mad Bitcoins. Partway through, the conversation turns to Hunt’s own card, which had not sold, and Robek says the generous thing. Both speakers appear inside the same unmarked passage, so both halves are given here as they stand:

“Yeah you guys should you guys should but go buy some of his cards so that way I don’t feel bad about cash now. Oh is that the truth because I know right we had ret send the Ethereum to RobEx would count and the artist has been paid the money.”²

The second half of that passage is Hunt, and what he is saying is that Rhett Creighton sent the ether from the card sales to the artist’s address, and the artist has been paid.

Robek, present on the call, does not contradict him. Hunt then adds the line that makes the whole thing sound less like policy and more like a man who has considered the alternative:

“It would look really bad if we didn’t pay and we tried to have you on the live stream anyway.”

That is as close to confirmation as a spoken archive gets: a payment described in front of the person who received it, with the mechanism named and the operator named. The project’s own documentation, written years later, makes the claim in retrospective form — “All of the Card vending machines from 2017 are sold out, where one hundred percent of the sales revenue went to the original artists” — and Wikipedia says of the launch prices that they were “$0.50–$1 each, which was paid to the artists.” Neither of those is independent of the project. The July 2017 exchange is, and it is the better evidence.

It should be said plainly what was being paid out. Robek’s set was the project’s one unambiguous commercial success, and chapter six works out what it earned. The sums are not large. The promise was kept at a scale where keeping it cost almost nothing, and the project went bankrupt anyway, and those two facts are not in tension — they are the same fact.

Seven artists, by the documentation

The number of artists is seven. It is also, in Hunt’s own telling, a matter of estimate:

“I think there’s about seven.”

He is right, but the hedge is characteristic and the confusion downstream of it is real, so this book uses the mapping published at docs.curio.cards rather than anyone’s recollection:

Three things that circulate widely are wrong, and are corrected here once.

It is Thoros of Myr, not Thoros of Mirror. Thoros of Myr is a character in Game of Thrones, which tells you something about the artist’s sense of humour and nothing about the cards.

Max Infeld is not an eighth artist. Marisol Vengas is his pseudonym. More on this below, because the error is in print in a place that matters.

Cards 1 to 10 are Phneep’s. The 2017 site’s artist list implied they were an anonymous house series; Christie’s attributes them to Phneep, as does the project’s documentation, which adds a creative-direction credit:

“Phneep’s Apples is the first art on Ethereum. His numbering sets the tone for the entire set. Guided by Mad Bitcoins, the first ten cards tell the story of the evolution of mankind into art creators.”

The story those ten cards tell, incidentally, is gone. Hunt, on the sixth-anniversary stream in May 2023:

“So they say that the first 10 cards tell a story, but it’s been lost.”

He can still read individual cards. Card 1:

“So of course we chose the apple, which represents the fall from grace and Adam and Eve and all that kind of stuff. Plus Apple computer plus it starts with a all those things like Steve Jobs said.”

Card 10:

“And then 10 was the future card. And this is like off into the future. And you’re going off on your path on your journey.”

What has been lost is the sequence — what cards two through nine were doing between the fall from grace and the journey into the future. The images survive on IPFS. The reading does not.

There is one place in the artist mapping where this book is at genuine risk of error, and it is flagged rather than smoothed. Cards 20 and 21 are both Mad Bitcoins cards, adjacent, by two different artists: card 20, MadBitcoins, is Phneep’s, and card 21, The Wizard, is Robek’s caricature. Hunt describes card 20 as “Mad Bitt Coins head on James Bond’s body” and elsewhere says, jokingly, “I’m card number 20… Card number 20 and 21”. The documentation is clear and this book follows it, but of everything in the apparatus this is the detail most likely to have been got wrong somewhere, and a reader who knows better should say so.

How you became a Curio Cards artist

There was a Google form.

That is the entire intake process, and it produced somewhere between ten and twenty submissions, which fed a spreadsheet. The form’s URL survives in the launch-day Bitcointalk post, which is how we know it sat under /a/mother.tech/ — one of only two documentary traces of that entity anywhere.

The selection criterion was not aesthetic. It was forensic. Hunt, describing why Daniel Friedman got in:

“we found a Daniel Friedman’s entry and what I liked about it had a flicker page with lots of other artwork it was clearly hand drawn with the pen and the ink”

A Flickr page with lots of other artwork, clearly hand drawn. What he was looking for was proof that a human being had made the work and that the person submitting it was that human being — because the nightmare was publishing stolen art onto an immutable ledger. In 2017 there was no provenance tooling, no reverse-image workflow anybody used, and no way to un-deploy a contract. The test for whether you could be a Curio Cards artist was whether you had a body of work someone could look through.

The rest of the roster came through people. cryptograffiti started the whole art strand by walking up at a San Francisco meetup and saying he made art about crypto — the moment chapter two turns on. Luis Buenaventura was met at the same meetup, having flown in from the Philippines, introduced by Rick Chan. Robek came through Moon, a mutual friend who visited him in Austin. Max Infeld was known to Hunt through Hunt’s brother and through Burning Man, and submitted five aliases, of which one was accepted.

Thoros of Myr came through nobody. Hunt, on the sixth-anniversary stream, going card by card through his own project:

“we come to the eclipse card and I don’t know Thoros of Mirror. I do want to find out”

Six years after the fact, a co-founder of Curio Cards had never met one of his seven artists and wanted to. As of the latest material in this archive, he still had not.

And then there is the joke Hunt points out himself, which is the best one-line summary of the project’s position in crypto history:

“And what’s so funny is that these cards are on Ethereum and it’s clearly an Ethereum thing. But our first 10 like artist cards, wherever are all Bitcoin propaganda.”

He means cards 11 to 13 and the Bitcoin-themed Phneep cards — cryptograffiti’s card 12 is a MasterCard logo reading “MineBitcoin”, card 13 parodies a bank. Hunt’s recollection of which bank is, characteristically, not firm: “It’s a parody of one of those British banks. I forget which one. … And 13 BTC, which I think is the city bank.” The first art collection on Ethereum was substantially made by Bitcoiners, as advertising for Bitcoin, hosted on Ethereum because Ethereum was the only chain where it would work. Chapter three is about why.

Marisol Vengas

The clearest example in this history of a prestigious source getting a simple fact wrong in print concerns cards 27, 28 and 29 — Blue, Pink and Yellow, by Marisol Vengas.

The project’s documentation states the position without hedging:

“Marisol Vengas was also the pseudonym for real-life Max Infeld, the Chico-based artist whose featured work in Curio Cards was one that explored social engagement in art.”

Infeld confirms it himself, on air, twice. On Adventures in #NFTs on 3 June 2021:

“being an artist, I’ve made a ton of different artistic personalities and aliases… So I didn’t get Max art. I got Marisol Vengus art.”

And again at the Curio Plus drop party on 7 October 2021:

“the series that I made was called Marisol Vengas. And it’s also a pseudonym in Aelius”

Christie’s, in the lot essay for the October 2021 sale, says something else:

“In truth, Vengas is an anonymous artist collective comprised of various members who have never met in the physical world—essentially, a decentralized living artist.”

Both go in, because on authorship Christie’s is simply wrong — Marisol Vengas is one named man who has said so on the record — but Christie’s is describing something real about method, and the “collective” framing did not come from nowhere. The documentation describes Infeld working in cafés where, in its words, “Passing strangers would stop by and color, assemble or be in charge of the composition.” People who had never met each other did contribute to the making of those cards. They were not a decentralised collective; they were passers-by in Chico, California.

Hunt’s account sits exactly between the two, which is what you would expect of a man relaying a friend’s practice at second hand:

“Max infeld was involved in setting up Marisol Vengus collective”

and, pressed on it:

“Marisol is not a person this was Max and other people as well are just basically Max doing this art as an alias”

Infeld’s own count of what he submitted is not stable either. In one breath: “I think I submitted five in total” and “how many pieces, 25, 25 or so artworks I submitted”. This book does not pick one.

The documentation makes a further claim about these cards worth recording precisely because it is the kind of claim that hardens into fact if nobody marks it: that Infeld’s submissions “are considered among the first known represented ‘generative’ artworks on Ethereum.” Considered by whom is not stated, and this book does not assert it.

Card 26

The smallest thing in the collection turned out to be the most important, and nobody knew that for four years.

Daniel Friedman’s three cards run 333, 222, 111 — Complexity, Passion, Education. Card 26, Education, is 111 copies. On chain, at 0x1cB5BF4Be53eb141B56f7E4Bb36345a353B5488c, totalSupply is 111 and decimals is 0, and has been since 2017.

Friedman is honest about how little he remembers of the arrangement. On the sixth-anniversary stream:

“the beginning is shrouded in forgetfulness for me. I don’t remember exactly how I got the web form”

One hundred and eleven is a small number for a reason that had nothing to do with strategy: by the time Friedman’s cards went up, the project had abandoned hundred-thousand-card warehouses for supplies the artist chose, and the descending sequence was elegant. The consequence is arithmetic. A complete set of Curio Cards requires one of every card, so the maximum possible number of complete sets is the supply of the scarcest card. That is card 26. One hundred and eleven.

Which means that in 2021, when a community wrapper contract was built in a hurry and a handful of cards were sent into it and locked forever, the card 26s among them did not merely reduce the supply of card 26. They reduced the number of complete sets of Curio Cards that can ever exist — on the best-sourced figure, from 111 to 106, permanently. Chapter twelve is about that, and about why the exact count is harder to pin down than it sounds.

Nobody in 2017 knew that the scarcest card was the bottleneck, because in 2017 nobody was assembling sets. There was no market, no marketplace, no floor price and almost no collectors — a point chapter nine measures. The card that would determine the ceiling on the entire collection’s completeness was, at the time, a picture of education by an artist who could not later recall how he had been recruited, selling for a dollar, mostly not selling.

The eclipse that was two days earlier

The last card closes the collection and carries its cleanest documented error, which is the project’s own.

Card 30, Eclipse, by Thoros of Myr. Contract 0x7F5B230Dc580d1e67DF6eD30dEe82684dD113D1F, token Curio30/CRO30, decimals 0, totalSupply 821. The supply is a date: 8/21, the Great American Eclipse of 21 August 2017. The project’s FAQ says so — “Card 30’s supply of 821 Cards is because of the day of the solar eclipse, 8/21, in 2017” — and Christie’s says so too, in the lot essay: “So-named for its release coinciding with an eclipse on August 21, 2017, there were also 821 (8/21) editions minted.”

The supply commemorates the eclipse. The deployment does not. The contract was created on 23 August 2017 at 07:15:26 UTC, in block 4193345 — two days after the eclipse. “Released to coincide with” is wrong, and so is the project’s own artist page, which manages to be wrong twice in one sentence: “Thoros’ Card 30 celebrates the Great American Eclipse of September 21, 2017. It was released on that day, too”. The eclipse was in August, not September, and the card went up on the 23rd.

There is an earlier, better version of card 30 that does not exist. Hunt, on the fourth anniversary:

“originally the idea actually was let’s like see if we can just time lapse photograph the eclipse and make that a card”

Dropped because it needed equipment they did not have.

And then three numbers circulate for card 30’s supply, and running the third one down is a small demonstration of what this archive can do.

821 is the minted supply. It is on chain, in the project’s FAQ, and in Christie’s catalogue. It is solid.

816 is what both curio.cards and the documentation display as “Supply”, and the curio.cards page contradicts itself inside a single paragraph — “Supply x816” sits directly above a sentence saying 821 were created. Travis Uhrig’s hedged “I believe some of them were locked” is the only explanation anyone has offered. Five cards are unaccounted for and no document explains them.

826 is an error, and its provenance is in this archive. It comes from Hunt misremembering the eclipse date on air and deriving the supply from the wrong date: “there’s only 826 of them made and that’s because it was 826 2017 which was the day of the eclipse”, said on 7 June 2021 and twice more over the following two years. An earlier recording catches him mid-wobble, on 10 May 2021: “I think it was like 827 or 821 and it was that was the month in the day.” The eclipse was 21 August. The contract holds 821. The 826 figure appears in no external source found anywhere, which means it never escaped — it lived only in his own retellings, and it can be retired here.

That is the pattern the whole book runs on, visible in miniature. A man remembers a date slightly wrong, derives a number from it, repeats the number confidently for several years, and the chain — which does not misremember — holds the answer the entire time, at an address anyone can read.