Checks Transparency • Complete summary of the project's development

Transparency time:

Working on the next iteration of Checks and itching to share some of the ideas/revisions/thinking.

Went back and forth on doing this — but ultimately, the process is as much a part of the art as the art itself, so will be sharing more in the thread below.

This began life as a single piece of artwork, captioned:

“This artwork may or may not be notable”

Priced at $8 per edition, a nod to the new meta of “purchasing” status. No prizes for guessing the very notable “legacy” artist that inspired the visual.

An edition limited by time only (a nod to the idea that the only requirement to get a check, was $8). It ran 24 for hours on @ourZORA, and 16,031 identical pieces were minted.

Candidly, as is always the case — you have no idea how the world will react to something you made, so that number could have also been very different.

To have the world “validate” an idea that you have thought deeply about (but always second guess) is a profound experience.

This piece from March 2021 “NFTs, explained.” contains the genesis of this idea: Since then I have been exploring the overlap between social contracts, NFTs, digital provenance, what art is, how it evolves, and who decides all of the above.

Obviously all of this has been massively controversial, which is exactly why it’s fertile ground for making art.

Bit of context setting out of the way, back to the Checks idea. One of the other big recurring themes in my work @visualizevalue has been positive-sum games vs. status games, inspired heavily by the writings of @naval


It’s my belief that on a long time horizon, permissionless markets are a positive sum game, but in the short term they can feel incredibly zero sum.

Which led me to my first idea (that didn’t make the cut) for an extension of Checks, an on-chain status game:

While I think this would make a great artistic statement, it leaves the possibility open for someone to burn the art they purchased and get nothing in return.

It’s the original piece, sliced in to 80 1/1s — distributed to the top 80 burners of the edition.

Not quite right.

So back to the creative process, but this time starting with some numbers:

There are 2,561 owners of the original Checks piece at time of writing. 16,031 in circulation. ~6 per person.

Each artwork contains 80 checks.

In my agency art director days, I always used to start with wikipedia if I was trying to find a creative angle on something that was rooted in truth.

Full transparency, the 80 checks in the original piece were not intentional so this is a fully reverse engineered rationale:

The semi-perfect number above gave me an idea that I am still working on, and will continue to document in this thread shortly.

Thanks for reading this far, and thanks for collecting my work. I appreciate it more than I can articulate.

A concept that gives us some constraints to work within.

Also worth noting how paradoxically perfect the word “semiperfect” is for this. A mathematical oxymoron.

If you have been around NFTs for a while, you’ll recall Damian Hirst’s “The Currency” — which worked like this:

Burn the NFT, get the physical, or keep the NFT, and the physical gets burned.

Of the original 10,000, there are 1,846 NFTs left.

The supply stays the “same” but the market reveals its preference for each medium in the process.

(Numbers above are apparently not correct, collection is split across a couple of chains/not displaying perfectly on OS, anyone who knows actual numbers please enlighten me)

A lengthier detour than planned, correction below:

The majority of the market opted for the physical version.

This is not what we are going to do, but an important precedent to recognize.

Next steps for Checks — checking technical feasibility on a few things, and double checking some math.

Will be back to share soon, thanks for reading.

Some meta-transparency on this thread to leave you with.

To extend the premise of the original piece — the question we are trying to answer is: are checks more desirable if they are harder to get?

We can find out by incentivizing a reduction in supply.

But we want to do it without destroying anything in the process.

Some new wireframe sketches informed by our semiperfect concept, and true to the invisible grid of the original piece:

80, 40, 20, 10, 5, 4, 1, 0


Recruited the legendary @jalil_eth & @traf to build some proper tech for the next round. Below courtesy of the giga talented @traf (all this animation is done in-browser)

Recruited the legendary @jalil_eth & @traf to build some proper tech for the next round.

Below courtesy of the giga talented @traf (all this animation is done in-browser)

— Jack Butcher (@jackbutcher) January 8, 2023

Current working mechanic (not final):

Burn your edition for an on-chain original.

Burn 2 originals for the next smallest edition original.

80 > 40 > 20 > 10 > 5 > 4 > 1

Stop wherever you like.

Hard to explain in text so here’s a visual (again, a work in progress):

What this does: rewards every collector, creates new, original, on-chain work, and every interaction either maintains or reduces supply.

An idea for everyone following along in real-time, minting some of the concept sketches as a free claim for holders.


Scratch that, not interesting. Right click save only. A few closeups of @jalil_eth 's color tests for on-chain generation of new originals using only the 80 colors from the first piece: 80s (left) 20s (right)

Math Q:

How many possible permutations of the 80 grid are there?

Assuming 80 colors w/ duplicates allowed, 80 checks, 80 positions.


one hundred seventy-six novemquadragintillion six hundred eighty-four octoquadragintillion seven hundred six septenquadragintillion four hundred seventy-seven sexquadragintillion eight hundred thirty-eight quinquadragintillion four hundred thirty quattuorquadragintillion

We will only be generating a fraction of a fraction of a fraction of a percentage of possible permutations, capped at 16,031.

A closer look at yesterday’s burn mechanic diagram:

2x editioned 80s → 2x on-chain 80s
2 x on-chain 80s → 1x on-chain 40

They are very meditative. And 100% on-chain. Live test:

The above video is on 2 second rotation, and the live link is at 3 seconds.

Think we’re liking 3.

One issue with the perpetual motion to represent each piece is losing the low-probability in the outputs like the below.

Great in isolation but gets homogenous at scale.

Might make sense to make motion downloadable for display in a frame vs. running with it as the token.

At time of writing there are 2,964 holders (will try to track change from here on): Avg: 5.4 editions per collector Top: 247 editions (1.5% of supply) 20 collectors w/ over 100 editions 56 collectors w/ over 50 editions And we hit 9th on opensea today :saluting_face:

A visual recap.

Massively tempting to buy and burn 27 editions for a cleaner start, but we work with what we were given…

Work in progress burn data visualizer here to track macro:


Working on front end stuff today.

Data and burn UI.

Updates to follow.

One thing I have always known but had trouble articulating: markets are art.

In further developing this piece, we are producing a living piece of art that collectively represents our taste, behavior & psychology.

Thanks for being a part of it.

“Token as canvas” enables this connection and interactivity, and the underlying infrastructure is as much a part of the work as the visual component.

Without it, the piece has an entirely different (and limited) meaning.

We want to design something that simply reflects the behavior of everyone who interacts with it, forever.

Once the rules are defined and deployed immutably, the infinite art show begins.

We’re building some components in framer to make our recap visuals above into live modules:

And here they are in a browser:

Quick bounty: 3x Checks to someone who can help me set up a simple twitter sales bot. :pray:


In the spirit of network effects, follow the bot


and drop your ens under this tweet, will airdrop 5 editions at random at 8 EST.

Minor interruption but we should be back now. :pray:

Some transparency on listings, looks like 2 editions (12163 and 12165) have been stuck all day, h/t


. Pinged


so hopefully resolved soon. :pray:

You can pinpoint when they got listed below:

Secured a domain for the front end:

Meta commentary, but popped in my head when I saw this so will share to keep it transparent: 10/10 meme placement, entangled with the meaning of the project itself.:saluting_face:

v0 of site now working w/ real market data. Some writing, leaderboard and burn UI design in the works. Likely sharable tomorrow.

Big problem with that stuck listing, but incredible day, thank you all. 89 ETH of volume 2,133 sales 2,739 holders Tomorrow we continue :saluting_face:

Since OS is bugged and throttling volume, removed all zero-royalty trading restrictions, and dropped optional royalty to 2.5%. :saluting_face:

Contentious topic so more detail:

Royalties are essentially unenforceable based on the way I deployed the edition, so the option I had was to block zero-fee markets, and set an optional royalty on OS and buyers opt-in to paying. (ty)

Opened it up since OS listings are bugging.

Even more candidly: Given the our dwindling attention spans, throttling the volume and subsequently the attention on the project is a bigger risk than not earning royalties from the tiny percentage of people who do opt in (legends all the same).:pray:

For proper context, here’s what the “creator earnings” dashboard looks like on a collection with optional royalties.

This slice is pretty indicative of opt-in percentage, and it goes down as the price goes up (as number gets bigger, less people want to pay)

I am v sympathetic to the idea of capturing upside from your work into the future — and royalties are an important way to do that.

But there’s also an argument to be made for reducing friction/maximizing attention/reach:

These 1/1s just kicked off (have been sitting for months)

gm, weather check

One “semiperfect” set.

Would take 128 editions to produce.

Currently doable by 5 collectors.

Excludes the final check.

Checking in:

We just finalized the logic for the token IDs for the on-chain originals post-burn.

It is no secret that “low number good” has been a pervasive idea in collectibles and art way before we had these tools for digital provenance…

Not to mention numbers that mean something to individuals — culturally significant numbers, birthdays, or flexing single digits.

Another aspect to your journey through Checks will be taking agency over the numbers associated with your originals.

I’ll have a UI sketch soon.

Congrats to both! :saluting_face: :heavy_check_mark:

Sent for this just after mint, no physical plans but thought I’d share.


Back to token IDs:

By no means a final design, but shows the mechanics — each on-chain burned pair passes on the ID of the lower original.

Our decision here came down to two choices:

1/ The new collection is ordered by each original, sequentially (80 checks at the high end, 1 at the low). Nixed this.

2/ What you see above, which gives collectors control over the token IDs they continue to create.

Wrapping for the day. Tomorrow: motion (think we found a great fix), contract tests and making some of the above UI real. :pray: :heavy_check_mark:

gm :ballot_box_with_check:

Color testing underway.

Organizing our original 80 colors into a gradient to ensure any motion runs buttery smooth.

Left here represents an original, right is in color order.

Will become clearer with a real example shortly.

Stage 1 verification complete, collection pending. :pray: :ballot_box_with_check:

Playing w/ what can be communicated with color using our v low-density grid (for meme purposes only).

Feel free to remix the svg if you have an idea.



Last one then back to work

Link to a playground, enjoy. h/t


Checks on checks.

Found a proper fix for motion.

Unique, static originals will display on listing pages (maintaining rarity and aesthetic preference), motion will occur only on click, eventually looping perfectly back to the beginning.

All on-chain.

Motion starts at 6/7 secs in clip below:

More subtle in our bigger compositions, here’s a test in browser — scrub the footage if you want to see how much it’s actually changing.

Please excuse any twitter video compression, color and sharpness much crispier first hand.

We should have a v1 of the contract done tomorrow, then most of the complexity lives in getting the front end right. Thanks for reading this far :saluting_face:

I will attempt a tl;dr version of everything above in a separate thread tomorrow also, we are deep in show more replies territory down here.

Interesting suggestion raised via dm re: token IDs:

ID should be selectable from the two burned tokens vs. always defaulting to the lowest.

Need to think through trade offs.

gm update: Giving full agency to the collector to choose which ID to carry through each burn. Thanks


for the thoughtful input on this.

A very happy accident:

Choosing this composition last week before twitter decided to axe the square crop.

Meme velocity hopefully less compromised as a result.

Working on some animated explainers to explain burn logic that should circumvent the need to read this entire thread to understand how this thing is going to work.

Pictured 64 → 1:

Editions → Originals

To get philosophical for a second, this process usually happens in reverse for traditional art:

The original is created and then editioned.

We are doing the opposite.


Figure we should share all the ugly details here, for some reason godaddy suspended my account and reversed the domain purchase.

We make lemonade though:

New burn UI sketch w/ collector ID choice:

@jalil_eth has been in the lab cooking up some beautiful new details that I am excited to share in the am. Back to some color stuff and how evolution will work. We may crack open a twitter space if there is interest. :saluting_face: :ballot_box_with_check:

Pretty sure we passed $1m in secondary volume today but opensea stats are glitching. :heavy_check_mark:

gm all,

We have been deliberating the language in the contract to describe the burn function. (obviously unchangeable after deployment so has to be perfect)

Landed on “composite”

definition: combine (two or more images) to make a single picture, especially electronically.

We thought it important to not imply better/worse with the language here, simply describe what is happening.

The contract is as much a part of the art as the visuals.

OK time for another L: Another


bug resulted in someone buying the wrong token thinking it was the last in the collection (for ATH): 16,031 were minted, and one was burned (#10845) Last token ID is #16031 but only 16,030 exist. cont…

We were using the opensea api to pull data into the frontend, so have tweeted out that 16,027 number a couple of times.

Extended an offer to buyer to buy back at price paid.

Hypothetical visual to demonstrate token ID selection.

“What’s in a number?”

Some more UI module sketches.

gn weather check Space tomorrow at 3pm EST, rsvp a few tweets up :pray::white_check_mark:

gm please help me torch this account:




What’s the difference between an image and an NFT?

Refresh your Checks metadata on opensea, and you will find out.

Ideas > images.

Perhaps that was a bit too subtle, let’s try again.

Editions will change, but they will all stay the same.

Shared w/ permission.

Said better than I could’ve put it:

:dove: :raised_hands:

Checks is an infinite canvas for expression that is designed to challenge the concept of ownership and authorship in the age of the internet.

“Whose image is this?”

If no one is verified, everyone is verified.
If no one is notable, everyone is notable.

“Check the chain.”

gn weather check

Overwhelmed by the creativity in the notifications tab.

Back end coming along, testing underway.

Shared this statement earlier but adding to the gigathread for provenance purposes:

gn weather check unique owners creeping up :pray:

Will share a full run in the am. Round of applause for

@jalil_eth, incredible work.

Before we do that, some color updates: 🧬

Hex values of colors are not final, some updates and explanations coming on that today. Embedding some more memeing. :white_check_mark:

Each original will search all the way back through its family tree in order to display itself. Provenance within provenance. :white_check_mark:

Please let @jalil_eth know that this interesting. :saluting_face:

“The RGB color model is additive in the sense that the three light beams are added together, and their light spectra add, wavelength for wavelength, to make the final color’s spectrum.”

64/16,031 → 1/250 (possible)

Color still in progress.

We toyed today with the idea of introducing 255 colors (an ode to RGB) — which would mean it would be possible to generate 250 unique single checks (quite unlikely in itself)

What that would do is break the chain of inheritance.

So we are leaning toward natural selection…

With this system in place, it will be interesting to see how our 80 colors reach the penultimate stage.

If you’re playing for the black check, does it even matter?

This also plays into the overall logic and philosophy of the piece, the further you play — the less “unique” the result (only in the visual sense of course).

The first original contains the most randomness, the last original contains the least.:white_check_mark:

The distribution of single check colors in a mathematically perfect world (if all 250 single checks are reached) is 3.125 of each.

I would bet it won’t play out that way, though.

If you take a snapshot of collectors right now, only 43 have the resources for a single color check.

We haven’t fully cracked the technical feasibility of this but figured I would share anyway.

Trying to answer the question: can we assign varying bands of randomness to initial original color migration?

With each originals ability to pass on color traits, introducing a mechanic like this would increase collectors ability to control the visual outputs of their pieces, and introduce degrees of rarity during the initial migration.

Huge nod to @ArtOnBlockchain’s Chromie Squiggle here — the slicing up of the color spectrum to create distinct feeling pieces, but build a beautifully cohesive collection overall.

gn weather check :pray::white_check_mark:

Looks like the above is possible.

Back in figma to mock it up.

Have been agonizing over color the last few days.

We were missing a little VV DNA.

More updates on color bands, gradients, inheritance logic getting dialed in.

This system does much more justice to the nuance in color, really punches out subtle differences and highlights the details in beautiful code @jalil_eth is writing.

Semiperfect trait distribution.

More on this tomorrow.

80 = 100%
40 = 50%
20 = 25%
10 = 12.5%
5 = 6.25%
4 = 5%
1 = 1.25%


Couldn’t wait, some previews of gradient and pattern logic.

And motion on an individual 80 check.


Lots of time spent handpicking colors and sequencing, pretty much locked in. Feeling v good about it, back to burn interfaces and illustrating traits/inheritance logic.:pray::white_check_mark:

Quick look at one of the frontend modules for a trait browser (very much work in progress)

Some concept notes:

Yes, “check” is a play on a number of cultural phenomenons we’re living through, and how we refer to the ubiquitous symbol that connotes status…

But we are also trying to communicate the act of “checking” something into the work here.


All of these memes have been useful to get us from zero to one (in a very technical community), but lack broader memetic velocity:

Not everyone sees the benefits in decentralization, but it’s not difficult to understand why something should be checked.

Yes this is an art project with (hopefully) a very beautiful visual and technical output, but it is also an invitation to understand a set of tools for building consensus from the bottom up.

With the above in mind, we will also be building tools for displaying provenance beyond who created the work and who currently owns it.

Another layer to defining a “perfect” set.

Art history check:

Updated burn sketch.

Trait strength here not fully baked just playing with UI for now.

Rambling update on provenance incoming:

“What’s the incentive to burn?”

Will be as transparent as my ability to articulate allows on our thinking here.

Any mechanic we introduce that incentivizes “speed to burn” will inevitably get gamed (bad)…

Our objective this entire time is to deliver an experience that a collector has agency over.

I think we solve the above with provenance data that already exists.

If you think about art on a long enough timeframe, time matters…

When was the inception of this particular check’s oldest ancestor?

Was it the first day the burn mechanism was available? Or the 100th?

The value of those datapoints is not for us to decide, just to make them visible so you can.

Full transparency, we considered “generation” trait that would be mapped with relative percentages to our semiperfect system:

But I am almost certain this would be a magnet for bad action.

We are very happy with the other mechanics we’ve designed and will put a huge focus on displaying provenance that doesn’t punish collectors by the resources they have available to game a tranche-based system.

gn weather check

should be checking some big items off the list this week

  • contracts going into testing
  • live front end being built

thank you all for following the process

One of the traits that has taken the longest time to dial in: color banding.

We designed this on top of our semiperfect system (1, 4, 5, 10, 20, 40, 80)

Imagine the grid below reconfigured in infinite combinations to influence order.

Some test outputs in next tweet.

These all represent single pieces (each would also have its own unique motion sequence on click)

This set is “unordered” — meaning there is no gradient filter applied to order the sequence of color, that’s another trait.

One of the hardest macro challenges to solve with art at this scale is the fine line between uniqueness and cohesion. We are close though :white_check_mark:


127 editions to originals:

Featuring a few of my (personal) favorites in the next couple tweets:

Too many good ones to share.

Should also be noted that this is only one of the 7 possible grids.

16,029 possible unique onchain 80 check originals via a single edition burn.

Other (highest possible) edition numbers below:

gm weather check, welcome new collectors

A thought/question we have been kicking around this morning:

Should we make it possible for collectors to retain an ID and “consume” another piece of artwork at the same check level?

Solves for collector agency, but makes the system a little more complex.

If people chose to participate in this mechanic, it would affect the overall supply in a way that the previous system didn’t. The “weight” of the overall collection could change, and would potentially prevent certain volume based outcomes. (h/t to merge by @muratpak )

For illustrative purposes:

I have ID 100 and I burn for an 80 check piece, but I see another 80 check piece I like the look of more.

What we’re thinking through is the ability to keep the ID but replace the visual, without burning down to the next level.

Again, solves for ultimate collector agency — but does have a potentially strong effect on the overall outcome of the piece.

My gut is this would be taken advantage of very rarely, so I lean toward implementing — but curious for your thoughts.

One macro goal we have with checks is to force the consideration of one of many nft megamemes:

“in it for the art”

I think this mechanic would help do that.


Contracts ready for testing tomorrow.


UI build pushing forward.

Contracts in code review.

We had a big UI breakthrough today.

“The contract is the canvas.”

Introducing the constraint in the art itself as a system for the interface.

With the intention of turning any and every moment of the experience into art:


we interrupt this broadcast to say a heartfelt thank you to our sponsors

(everyone who doesn’t uncheck creator royalties when relisting on opensea)

Another module.

The goal is to give collectors the ability to slide through different views of the art without leaving the canvas.

The art is both the individual check compositions, and the entire ecosystem. (more on this soon, likely in a v3 update post burn)

Testing… ty @ourZORA for these tools and you all for all the goerli eth contributions :pray:

Compositing and ID inheritance :white_check_mark:

Color inheritance :white_check_mark:

All for tonight Thanks for checking :saluting_face:

Some slight visual tweaks to do from these tests, then straight into UI build w/ refactored modules based on the canvas grid.

Welcome to all new collectors, thank you to all

Burned all the way down to a black check on Goerli :white_check_mark:

new butter from @traf check terminal:

A few snapshots of the collection showing varying degrees of burn on testnet:

massive shout out and thank you to @backseats_eth for putting fresh eyes on the contract a few more passes and tests and we should be 🔒

working on some pre-burn UI (very work in progress)

important that is part of the experience/flow

“check yourself”

Which brings us back to a point made maybe 50 tweets up in the thread.

The check is the both the image and the idea.

UI check

check balance

small win, collection name updated frrom “checks v4” on blur

big win, live frontend work underway

gm, shared “checkonomics” last night

image below depicts the logic behind the evolution of the piece

largely well received, some comments on the layouts of the smaller originals (10, 5, 4) not hitting as well as the others, so we jumped back in the lab…

poll incoming

v1, v2

drop a comment w/ your preference (can’t run a poll on images):pray:

poll for the above

also, to be clear - the results will be considered but we will ultimately decide :saluting_face:

thanks all, dialing in some more, poll in next tweet

v3, v4