By Are Ordinals the Best or Worst Thing for Bitcoin?
A long-time Bitcoin Core developer has spent years calling a large and growing category of Bitcoin transactions "fraud." A separate body of research shows that same category of transactions has paid Bitcoin's miners hundreds of BTC in fees, at exactly the moment Bitcoin's built-in economics need fee revenue to matter more, not less. Both of those things are true at once, and the fact that they can both be true is what makes Ordinals one of the more durably unresolved arguments in Bitcoin, not a debate with an obvious right answer that one side is too stubborn to see.
What actually happens when you inscribe something
The mechanism is worth understanding before taking a side. On December 14, 2022, a developer named Casey Rodarmor created the first Ordinal inscription, a pixel-art skull, assigning it to a specific satoshi, the smallest divisible unit of Bitcoin. He published the full protocol on January 21, 2023. Ordinals gives every satoshi a unique, trackable number based on the order it was mined, and inscriptions attach arbitrary data, images, text, even short video games, directly to a specific satoshi using the witness portion of a Bitcoin transaction, the part of the transaction data that SegWit (2017) separated out and Taproot (November 2021) made cheap to fill with large amounts of content. No hard fork, no rule change, no vote. The protocol works entirely by using existing Bitcoin features in a way nobody originally designed them for. A few months later, a pseudonymous developer known as Domo used the same primitive to launch BRC-20, a way of issuing fungible tokens on top of Bitcoin, which is what actually pushed inscription volume from a curiosity into a genuine flood.
The case that it's the worst thing
Luke Dashjr has been writing Bitcoin Core code since 2011 and co-founded the OCEAN mining pool. In December 2023, he stated plainly that inscriptions were "exploiting a vulnerability in Bitcoin Core to spam the blockchain," pointing to a specific technical detail: Bitcoin Core has allowed node operators to limit the size of extra, non-payment data in transactions since 2013 through a setting called `-datacarriersize`, and inscriptions get around that limit by disguising their data as program code inside a Taproot script rather than declaring it as data outright. He didn't stop at calling it a bug. Asked directly whether fixing the "vulnerability" would end Ordinals and BRC-20 tokens entirely, he answered "correct," and added, "Ordinals never existed to begin with. It's all fraud." He went on to filter inscription transactions from his own node and from OCEAN's mining pool, and later extended the same criticism to Runes, a follow-up token protocol, which he described as "only" a five-vector attack compared to what he called Ordinals' nine.
The real-world disruption behind that frustration is well documented and not exaggerated. On May 7, 2023, at block height 788,695, transaction fees paid to miners exceeded the block subsidy for the first time since December 2017, a genuine inflection point in Bitcoin's fee market. The next day, the network processed 682,281 transactions, an all-time daily record at the time, more than 372,000 of them inscriptions, and the mempool swelled past 500 megabytes with over 440,000 transactions waiting to confirm. Average fees jumped more than 560% that month to $19.20 per transaction. For a network whose stated mission includes being usable as actual peer-to-peer money, that's a real cost, not an abstract one. Anita Posch, a Bitcoiner known for her work on financial inclusion and adoption in the developing world, asked the obvious practical question in the middle of the spike: "Can anyone explain how I'm going to onboard people with these fees? Can't use on-chain, can't open channels," referring to the Lightning Network, whose channels themselves require an on-chain transaction to open.
The private key for every Bitcoin wallet on Earth is on this website, even Satoshi's. But even if you try for a million years, you'll never find a funded one.
Try the key collider nowThe case that it's the best thing
Here's the part of Bitcoin's own design that makes the fee argument more than a talking point. The block subsidy, the new Bitcoin miners earn for finding a block, halves roughly every four years by design, and now sits at 3.125 BTC after four halvings. That number keeps shrinking toward zero over the next century-plus, which means the transaction fees miners collect have to grow to replace it or Bitcoin's long-term security budget, the total economic incentive keeping miners honest and the network expensive to attack, quietly erodes. During quiet periods, fees have historically made up as little as 1-2% of total miner revenue, which is exactly the structural problem Bitcoin's own long-term critics have pointed to for years. Inscriptions are one of the only things that has ever meaningfully moved that number. By February 2024, roughly five million inscriptions had generated a cumulative 904 BTC in fees paid directly to miners, and the May 2023 spike marked the first time fee revenue alone exceeded the subsidy in over six years. Whatever else inscriptions are, they're a real, working demonstration that Bitcoin's fee market can function as the subsidy replacement the protocol will eventually depend on entirely.
There's also a harder-edged version of the argument that doesn't rely on the security-budget case at all: nobody voted on what counts as a legitimate use of Bitcoin block space, and that's supposed to be the point. Miners have always been free to include whatever transactions they want in a block, ranked however they choose; that's the actual rule, not a loophole in it. Treating a permissionless, censorship-resistant fee market as valid only when the winning bidder is a plain payment, and illegitimate the moment it's an inscription, is a preference about what Bitcoin should be used for, dressed up as a bug report. That preference might even be correct as a matter of taste. It isn't correct as a claim about what the protocol's actual rules permit. Casey Rodarmor's own framing supports this reading: he deliberately called the project "Ordinal Theory" rather than branding it around NFTs, because he was more interested in the mathematical elegance of numbering every satoshi than in recreating Ethereum's collectibles market on Bitcoin. What followed wasn't what he set out to build, but it was built using nothing Bitcoin didn't already allow.
It didn't stay contained to Bitcoin either
Within months of Ordinals taking off, the same basic trick showed up everywhere. Ethscriptions brought an equivalent inscription mechanic to Ethereum, Doginals did the same on Dogecoin, Solana got its own inscription wave, and smaller networks like Celo, Avalanche, and Fantom collectively saw tens of millions of copycat inscriptions. That spread matters for how to read the Bitcoin-specific argument: if inscribing arbitrary data onto a blockchain were obviously a bad idea unique to how Dashjr says Bitcoin Core mishandles Taproot scripts, it's a strange coincidence that developers on entirely different codebases, with entirely different data-carrying rules, independently found ways to do roughly the same thing within the same year. The pattern looks less like one specific exploit and more like a demand that existed the moment any sufficiently flexible blockchain made it possible to satisfy.
What actually happened to the debate
Neither side won, and the pattern has repeated on a schedule rather than resolving. Fees spiked again around the Runes launch in April 2024, and that spike was the sharpest yet, dropping 74% within two days and 97% within weeks, evidence that these events behave more like weather than a permanent state of the network. Inscription activity has never returned to its 2023 peak, but it hasn't gone away either: total inscriptions passed 107 million by January 2026, and by that March, monthly inscription-market activity was still generating tens of millions of dollars in trading volume. Dashjr never got his fix into Bitcoin Core; the "vulnerability" he warned would be patched before version 27 is still there, node operators and miners simply choose for themselves whether to filter inscriptions, and most have chosen not to.
That's probably the most honest resolution available, not a technical verdict but an ongoing market one, repeated every time fees spike, every time a new inscription-based protocol launches, every time a miner decides whether to include or filter a given transaction. Bitcoin never had, and was never supposed to have, a central authority capable of declaring the question settled either way. Both the person who calls this fraud and the person who calls it free-market fee revenue are describing the same set of transactions accurately. What they disagree about is which purpose Bitcoin's scarce block space should default to serving, and that's a question about values, not one Bitcoin's code was ever going to answer for them.
For the plain technical explanation of how inscriptions actually work under the hood, see what Bitcoin Ordinals and BRC-20 tokens actually are. For more on the economic pressures shaping Bitcoin's mining industry more broadly, see how mining pool concentration affects network security, and for the environmental side of the same fee-and-hashrate story, see Bitcoin's energy debate.