Bored of the casino. Fascinated by the machine.

Why this site exists, what a "product lens" on crypto actually means, and the working theses I intend to test in public. A manifesto that grew up into an essay.

I. The two churches

Crypto has been written about the same two ways for fifteen years, and both of them are churches.

The first church is the casino. Its scripture is the price feed. Its sermons are charts with adjectives: "surges," "plummets," "consolidates," words that sound like analysis and carry exactly as much information as a mood ring. In this church, every protocol is a ticker, every product is a "narrative," and the only question ever asked of any technology is whether it will go up. It is the dominant church because it is the easy one. Prices update every second, and content that updates every second is content you never have to think hard about.

The second church is the cathedral of the whitepaper. Its scripture is the protocol spec. Its sermons assume you dream in Merkle trees, and its congregation measures virtue in cryptographic novelty. This church is smaller but no less devout, and it has its own blind spot, which is that it cannot tell you whether a human being would ever want the thing being specified. Elegant consensus mechanisms have shipped to precisely zero users more times than anyone in the cathedral would like to admit.

Eric Raymond wrote The Cathedral and the Bazaar in 1997 about two styles of building software. This essay borrows his method (a practitioner's field notes, organized into theses, argued from things actually shipped) for a different schism. Because the fight that matters in crypto right now is not cathedral versus bazaar. It is casino versus machine: whether these systems are primarily instruments to be traded, or primarily materials to be built with.

I am bored of the casino. I am fascinated by the machine. This site is the record of that fascination, conducted the only way I trust: by shipping products on the machine and reporting what happened.

II. The third lens

Between the price feed and the whitepaper there is a third lens, and it is almost unoccupied: crypto as a product discipline.

The product lens asks a compound question of every protocol, token, and app: what can you build with this that could not exist before, and would a human being enjoy using it? Both halves are load-bearing. Novelty without desirability is the cathedral's failure mode. Desirability without novelty is just a normal app wearing a blockchain as a costume, paying consensus-mechanism prices for a Postgres experience.

Notice what the product lens does not ask. It does not ask whether the token will go up. That is the casino's question, and it is unanswerable from first principles anyway. It does not ask whether the cryptography is maximally elegant. That is the cathedral's question, and users have never once cared. It asks the question that decides whether any of this matters in ten years: did the machine make a product possible that was impossible before, and did anyone love it?

Almost nobody covers crypto this way. This is strange, because it is the way every other technology wave was eventually understood. Nobody serious evaluates the internet by the daily price of Cisco stock. The web became legible when people stopped asking "what is TCP/IP worth" and started asking "what does it feel like to buy a book without leaving your house." Crypto is overdue for that same reframing, and the reframing is not going to come from the churches. Their incentives point the other way.

Thesis 1. Every technology is boring until someone finds the product it makes possible. Crypto's boredom is a discovery failure, not a technology failure.

III. Given enough shipped products, all hype is shallow

Raymond's most famous line was that given enough eyeballs, all bugs are shallow. The product-lens equivalent:

Thesis 2. Given enough shipped experiments, all hype is shallow.

Here is the epistemological problem with crypto discourse: nearly everyone commenting on the machine has never operated it. Reading a docs page tells you what a protocol claims. Holding its token tells you what the market feels. Only shipping a product on it tells you what it is: where the latency actually bites, which "simple integration" consumes a week, what users do in the first thirty seconds that no whitepaper anticipated, which advertised property turns out to be the boring one and which throwaway feature turns out to be the entire point.

So the method of this site is fixed: I find infrastructure, apps, and protocols doing something genuinely novel, and then I build my own product experiment around the idea to see if it survives contact with reality. The experiments are real, public, and allowed to fail. When they fail, the failure gets written up with the same enthusiasm as the successes, because a documented failure is worth ten undocumented successes. The casino deletes its bad calls. The product lens publishes its post-mortems.

This is the bazaar move, translated: release early, release often, treat your users as co-developers, and treat your own shipped code as the only argument you're entitled to make.

IV. What the machine actually offers

Strip away the jargon and blockchains offer product builders a small number of genuinely new materials. My running list has three entries, and I hold every project I review against it:

Persistence. State that outlives the session, the server, and (inconveniently for several business models) the company. A thing that keeps existing whether or not the frontend does is a different category of object than an entry in someone's database, the way a printed book is a different category of object than a page in someone else's binder.

Shared state. Everyone reads the same world. Your asset's condition is not a number in my database that I could edit, lose, or ransom back to you. It is a fact we both agree on without trusting each other. Multiplayer-by-default is a property no web2 stack offers without a trusted referee.

Provenance. The machine remembers. Who created, who held, who acted, who stayed through the winter. History becomes a first-class material you can build mechanics, status, and meaning out of, rather than a marketing screenshot.

Thesis 3. If your product uses none of these three properties, it does not need a blockchain, and pretending otherwise is how you pay consensus prices for a Postgres experience.

The list is deliberately short. Note what's absent: "decentralization" as an end in itself, "trustlessness" as a slogan, "community" as a euphemism for exit liquidity. Those are church words. Persistence, shared state, and provenance are product words. You can design a feature against them, test the feature, and watch a user care or not care.

V. Three experiments, three theses

Field notes are only credible if there is a field. Three experiments are live, each one a deliberate probe of a different thesis.

APE Works, the payments probe. The dominant crypto onboarding flow has the emotional texture of filing taxes at customs. APE Works asks whether buying a token can feel like buying a coffee: tap, confirm, done, using MoonPay's PayBox rails inside an economy of governed treasuries and missions. The thesis under test: friction, not volatility, is crypto's real adoption ceiling. Every step removed from the first transaction is worth more than any feature added after it.

TTTRMNL, the transparency probe. Ten thousand tokens, every one visible, in a Bloomberg-grade terminal for a protocol where NFTs burn into ERC20 launches. The thesis under test: radical legibility is a feature, not a leak. Traditional finance treats information asymmetry as the product. Onchain systems invert this. Everything is already public, so the product question becomes interface: what happens when supply itself is the UI, when any user can see what only insiders used to see? Shared state, weaponized for the small guy.

OEGP Garden, the retention probe. 9,888 dynamic plant NFTs on Ethereum, nine growth stages, tended daily through an app that looks like a botanical conservatory instead of a trading floor. The thesis under test: care mechanics beat reward mechanics. Points-on-a-swap is not gamification, it's a coupon. What brought people back daily was streaks and growing things: persistence and provenance turned into an emotional loop. (This one got its own essay: Gardens, not slot machines.)

Three probes, three materials from the list in section IV, three answers being collected in public. That is the entire editorial strategy.

VI. Necessary heresies

Positions held with confidence, documented so that being wrong later is maximally embarrassing:

Heresy 1: The speculation era was the scaffolding, not the building. Casinos bootstrap liquidity and attention; they do not create durable products. Every technology wave has its tulip phase, and every tulip phase gets mistaken for the point by the people standing inside it. The scaffolding is coming down. What remains standing is what this site is about.

Heresy 2: "Community" is not a product feature. A Discord with 40,000 members and nothing to do is a queue, not a community. Products create communities as exhaust, from shared activity, shared status, shared history (provenance again). Any project that lists community before product has the causality backwards, and the machine keeps receipts on how that ends.

Heresy 3: The interface is the strategy. OEGP looks like a greenhouse. TTTRMNL looks like Bloomberg. Neither is decoration. An interface tells users how to behave: watch the number, or come back tomorrow and see what's grown. In a space where every product ships the same dark dashboard with neon candles, choosing a different visual language is choosing a different user relationship. Design is positioning wearing work clothes.

Heresy 4: Boring chains make interesting products. The stack is finally good enough to be ignorable, and infrastructure becoming boring is the single most bullish product signal there is. Nobody built Amazon while TCP/IP was still exciting. The sooner the base layer is as thrilling as plumbing, the sooner the product era actually starts.

VII. How to read this site

The format follows the argument. Articles when an idea needs room. Short videos when a point fits in ninety seconds. Longform video when a walkthrough beats prose. Every experiment gets a project page with its stack and status, and posts carry reactions. Tap one; that's onchain interactivity, minus the chain, plus the honesty.

The tone stays sarcastic because the material deserves both jokes and rigor. The sarcasm is aimed at the churches, never at the machine. Underneath the jokes, one dead-serious claim: some of this genuinely rewires how products get built, and the rewiring is happening now.

Raymond ended his essay convinced the bazaar would win because it converted more eyeballs into more fixed bugs. The product-lens bet is the same shape: the builders who treat crypto as material will, product by shipped product, quietly out-compete the churches that treat it as scripture or as slot machine. Not because they're smarter, but because shipping compounds and sermons don't.

Welcome to BORED ONCHAIN. The machine is running. Let's build something on it and see what it's actually made of.