An Ordinals Sin?

Wizards, Frogs And Other Degen Bitcoin Behaviour

By Ryan Shea

Key Take-aways

• “It’s early” is a common crypto phrase and it’s true. The ecosystem continues to evolve and generate new use-cases. For Bitcoin, the latest innovation is ordinals/inscriptions.
• Ordinals allows Bitcoin users to assign identifiers to satoshis and to add inscriptions which turn them into “digital artefacts”, or NFTs in more common parlance.
• When unleashed earlier in the year, they led to a surge in user interest, demand for Bitcoin block space and transaction fees, not to mention the largest Bitcoin block ever mined and the rebirth of the Bitcoin wizard.
• They also attracted the ire of the maxis, who view ordinals as blockchain-bloating spam that has no place on Bitcoin.
• Cartoon wizards, I admit, are not my thing so I have some sympathy with the Bitcoin maxis. However, many of the perceived issues associated with ordinals will have arisen anyway.
• What Bitcoin is and becomes is not set in stone. Instead, Bitcoin gets defined by what the majority of its holders wish it to be and for now that includes ordinals/inscriptions.
• Certainly, calls to sensor Bitcoin transactions containing inscriptions is a really bad idea. After all, wasn’t one of the founding principles of Bitcoin that it is supposed to be censorship resistant?

I love macro and markets, which is a good thing given I have spent my entire career, which is more years than I care to admit, sitting for 12 hours a day in front of a screen thinking, writing and investing. While tradfi markets are fun and intellectually challenging to analyze crypto is on a whole different scale – it’s like tradfi on steroids.

Unlike bonds and equities, where financial models have been developed to provide at least some guidance as to how they should be valued by investors (bonds being easier given their cash flows are reasonable well known a priori unlike equities), for crypto this is very much a work-in-progress. Part of the reason for this is the technology is still being mapped out and understood (“It’s early” in crypto lingo) and new and interesting use-cases keep popping up. Enter ordinals.

What Are Ordinals?

Ordinals were created by Casey Rodamor late last year¹. Essentially it is a numbering scheme – utilizing only 20 lines of python code - that allows Bitcoin users to assign identifiers to satoshis (sat), the atomic accounting unit for Bitcoin (each Bitcoin is worth 100 million sats).

The basis for the numbering scheme relies on the fact that sats are created via the block subsidy given to miners to incentivize them to solve Bitcoin’s hash function in order to add news blocks of transactions to the Bitcoin blockchain. Because each sat is given a unique identifier at the point of its creation, ordinals provides a way to track sats, something that was not previously possible because Bitcoin – unlike Ethereum and many other cryptocurrencies – uses the UXTO (Unspent Transaction Output) model where inputs consume existing UXTOs and outputs represent new UXTOs.

In order to track sats through this destruction-creation transaction process, ordinals uses a “first in, first out” principle. There is no reason why this ordering system is superior to other alternative methods, such as “last in, first out”, hence it is somewhat arbitrary but it is consistent with how Bitcoin wallets order transactions² and anyway whatever method is chosen doesn’t really matter as long as it is widely agreed upon and accepted. (In much the same way as it doesn’t matter what side of the road people drive on as long as everyone agrees to follow the same rule). While ordinals are themselves interesting, this innovation also provides Bitcoin with something extra. Enter inscriptions.

Inscriptions

According to ordinals.com, inscriptions “inscribe sats with arbitrary content, creating bitcoin-native digital artifacts, more commonly known as NFTs”. Yes you heard that right. Ordinals allows users to mint non-fungible tokens on the Bitcoin blockchain. Moreover, unlike NFTs minted on Ethereum, the content of the inscription is stored on-chain as opposed to relying on a decentralized storage system like IPFS. Transacting or transferring inscriptions is also easier because they do not require smart contracts. Ownership can be simply transferred via a standard Bitcoin transaction as only the holder of the wallet receiving the inscribed sat has access to the private key and hence can readily prove ownership.

Because inscriptions are stored in the taproot script-path scripts, which received the witness discount (4X less fees) inscription storage is relatively economical (at least in terms of the transaction vByte³ size although not necessarily in terms of fee density, whose values fluctuates in response to demand for Bitcoin block space – something we will touch upon later). That said, a worry many Bitcoin users unfamiliar with ordinals have expressed concern about is that inscriptions may cause the blockchain to become bloated due to the larger data footprint inscriptions have compared with regular non-inscription Bitcoin transactions.

The Wizard Of... Bitcoin

Back in February, when Udi Wertheimer launched his taproot wizard ordinals project he, in concert with Luxor Mining, created the largest ever Bitcoin block (Block 774628) by inscribing the following jpeg – see images below. The rationale for his decision to put such a large inscription on the Bitcoin blockchain was to “make Bitcoin magical again” in the hope that this PR stunt would attract more users to the Bitcoin network (according to Udi, the inspiration for the wizard jpeg was a cartoon wizard created a decade prior by mavensbot that contributed to increasing Bitcoin’s social footprint).

Bitcoin History Being Made

Source: mempool.space and taprootwizards.com

What was the size of this block? 3.96MB. Because ordinals do not require a sidechain or a fork (hard or soft) of the Bitcoin blockchain inscriptions must respect the 4MB max blockweight limit otherwise nodes on the network would reject the block as a consensus rule. (Ordinals developers two days introduced a concept called recursive inscriptions that allows inscriptions to reference the content of other inscriptions meaning that data can be stored on Bitcoin that exceeds 4MB even though each individual inscription still respects the blockweight limit).

With Bitcoin producing 52,560 blocks per year, if all blocks were at the 4MB limit the Bitcoin blockchain would expand at a maximum rate of 210GB a year, rather than the 60GB rate of increase witnessed over the past few years. However, Udi’s massive 4MB inscription is very much the exception. Since ordinals were rolled out, there have been over 11 million inscriptions of which only a minority have been images. As a result, the increase in block size observed has been around 0.7MB (1.1MB to 1.8MB – see charts).

Weekly Inscription Count by Type

Source: Dune Analytics ⁴

Bitcoin Average Block Size (MB)

Source: blockchain.com

Assuming this continues, it would imply an additional annual increase in Bitcoin’s blockchain size of 40GB give or take. This is roughly the equivalent of 80 hours of standard-definition Netflix viewing and can be stored at a cost of just over $1⁵. Not what one would consider overly onerous, suggesting fears about ordinals and inscriptions bloating Bitcoin’s block size are somewhat exaggerated. In addition, because ordinals are stored in the signature data they can also be pruned, only full nodes that wish to keep the entirety of all the blockchain data need to store them.

BRC-20 Tokens

Creating and storing magic internet wizard jpegs is, of course, not the only inscription use-case. Indeed, as the chart above shows, the most popular use of inscriptions over the past few months has been to create simple smart contracts that enable users to mint BRC-20 tokens – the name being a play on the more popular ERC-20 tokens that are minted on Ethereum. Since they were launched back in early March, some 24,677 tokens have been minted and they have a combined market cap of just under $500m (during the initial boom the market cap reached almost $1bn!). The most valued is ORDI – the first BRC-20 token issued – which has a market cap of $238m, but perhaps the most famous is the BRC-20 token is the Pepe frog meme coin because it is often confused in the popular press with the Pepe meme coin minted on Ethereum who’s popularity surge pushed its market cap up to just shy of $2bn early last month – just weeks after it was launched. Zero to $1bn in a handful of weeks for a coin that has no obvious use-case – degen indeed!

Fungibility Funk

That ordinals allow for the tracking of inscribed sats has caused some Bitcoin users to worry that it – or more accurately the atomic unit sats – are no longer fungible which is one of the characteristics normally associated with money (durability, portability etc being other characteristics).

As mentioned, ordinals are an arbitrary ordering system and, as a result, they have no on-chain footprint. If a Bitcoin user does not download a Bitcoin wallet that has “coin control” capabilities, which allows sats with inscriptions to be differentiated, then all the sats look the same ie, they are fungible. Nevertheless, what does change is that because inscriptions can be valued because of the content inscribed – in much the same way as NFTs on Ethereum are valued – their social fungibility changes.

Also, sats can be, in and of themselves, deemed valuable. Casey, for example, released a first-draft rare-ability scale that ascribes scarcity value based on whether the sat is the first in the block (uncommon), first in the block after a difficulty adjustment (rare), first sat in the block after a halving (epic), the first sat in the block after the halving and the difficulty adjustment coincide (legendary) and finally, the first sat of the genesis block (mythical).

Such sats may be viewed as valuable to blockchain nerds but, in my view, there are many other inscribed sats that will be deemed valuable to a broader audience. For example, ordinal numbers are not just represented by what is in effect a reverse sat count. To reduce the sat notation Casey added a base 26 translation of the ordinal number, which converts the sat number to a sat name based on the letters in the English alphabet. As a result, each sat also has a unique name. Due to the declining length of ordinal numbers, the set of sat names also declines (the final base 26 encoding being the letter a). This means there are a whole range of interesting words associated with ordinals. For instance, it is possible to own a sat with your name. The sat that has my name as its sat name will be mined in 2063 and hopefully I will be around to own it – see chart.

My Name Sat

Source: https://ordinals.com/

Given the importance humans attach to identity and language more generally it is quite easily to see how demand for named sats could increase steadily over time.

There is another reason why certain sats may become popular and valued. Given sats can be traced back in time via ordinals, some Bitcoin users have been attempting to find and own the sat that has the earliest mint date. So far, the earliest sat ever found was one that was sent to Hal Finney from Satoshi Nakamoto from the ninth Bitcoin block mined. The sat was found by @null_ish and has subsequently been inscribed with the number nine – see below.

Earliest Sat Found So Far

Source: Twitter (the details of the sat can be found here: https://www.ord.io/9235048 )

Clearly, it is early days for ordinals and inscriptions, and no one can be certain as to how this will play out. My best guess would be that ordinals will around for a long time for the simple reason that people like to collect art (cartoon wizards are not my thing personally, but art, like beauty, is in the eye of the beholder), own things that are in some way connected to their identity, and/or perceived as scarce.

Bitcoin Maxi Backlash

Needless to say, Bitcoin maxis are livid that their protocol has been co-opted by jpeg wizards, green frog inspired meme coins and turned into nonfungible collectibles. They see it as a corruption of Bitcoin and a hindrance to its ability to serve as a money. One of the most prominent critics is Luke Dashjr, a Bitcoin core developers known to voice his strongly-held opinions. Last month he posted on the Bitcoin-dev forum calling ordinals and inscriptions spam that should be removed from the network – see image. He is far from alone in his opposition.

Source: https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2023-May/021631.html

Are the maxis right in thinking that ordinals are damaging for Bitcoin’s and its monetary ambitions?

In my view, no.

Law Of Large Numbers

Bank notes, which in almost all circumstances⁶ are considered fungible, each have a unique identifier - their serial number - and there are many people around the world who collect bank notes whose serial numbers are somehow considered special⁷. These collectible bank notes are bought and sold for values that far exceed their face value, but because their supply relative to the overall supply of bank notes is so small, it in no way diminishes the fungibility of bank notes.

The same is likely to hold true for ordinals. As noted, each Bitcoin contains 100 million sats. When Bitcoin’s supply finally dries up sometime around 2140, there will be 2.1 quadrillion sats (less a few) in existence. To put that number into context it is the equivalent to the number of stars estimated to be in the milky way⁸ times 21,000!Alternatively, if every one of the 8 billion people on earth decided to own ordinals they could each own 262,5000 without hitting a capacity constraint. To my mind it is simply unimaginable that inscriptions will become so popular that Bitcoin’s fungibility will ever be remotely compromised. The larger issue comes from the potential impact on transaction fees.

Transaction Fees: A Double Edged Sword

Since ordinals were unleashed demand for Bitcoin block space has increased and this has directly contributed to pushing up overall Bitcoin transaction fees. In early May the cost of a Bitcoin transaction jumped to around $30, although more recently it has since fallen back to around $4 .

Higher fees clearly makes Bitcoin less suitable to making small-value transactions, which is a negative, but they are also positive because transaction fees get paid to Bitcoin miners and it’s their combined hash power that protects the blockchain (that's why miner revenue is also labelled the security budget). As a result of the demand surge attributable to ordinals, fees increased last month to 15% of overall miner revenue, a substantial increase relative to prior months. It meant Bitcoin miners were less reliant on the other income source available to them, namely the issuance of new Bitcoins in the form of the block subsidy.

Bitcoin Transaction Fee (% of Total Miner Revenue)

Source: The Block

The reason why this is important is because the asymptotic decline in the block subsidy is required to ensure Bitcoin supply is ultimately capped at 21 million. As a result, it was always going to be the case that Bitcoin’s fee market would have to develop, irrespective of ordinals, to ensure the ongoing security of the blockchain.

Furthermore, this fee market would have to come about in the not-too-distant future because although the block subsidy will not be totally removed for another 120 years, beyond 2032 (three halving cycles away), the block subsidy declines to less than a single Bitcoin. As I noted in a previous note, back-of-the-envelope calculations suggest that unless Bitcoin ‘s price rallies above $120,000 by then (a scenario I don’t rule out by the way as I am bullish on crypto prices over the medium-term for reasons I have discussed in many prior research notes) fees in the order of $22 will be required before too much longer.

As I also argued in the aforementioned research note, the sort of transaction fees that fall out of these calculations are not dissimilar to those associated with large denomination bank wires transfer or cross-border payments. To my mind, these are the sort of transactions that Bitcoin should be used for, so I have some sympathy with the Bitcoin maxis in this regard. However, to the extent that ordinals help develop a fee market for Bitcoin, and as long as they do not crowd out Bitcoin’s potential to be used for such purely financial transactions, they are really just a sideshow. (Yes the Bitcoin blockchain got a bit congested for a while in the first flurry of excitement over ordinals, but more recently the mempool size measured in vBytes has been moving lower – see chart.)

Mempool by vBytes (sat/vByte)

Source: mempool.space

Where higher transaction fees might still cause some issues is in relation to something long recognized in the world of Ethereum - MEV, or mining/maximum extractable value. Historically, because of its perceived limited functionality, Bitcoin has never really been seen as a source of MEV. However, with the advent of ordinals, MEV opportunities are generated. There is even a risk that miners are incentivized to remine blocks with high value transactions associated with rare ordinals to increase revenues causing a blockchain reorg which clearly is far from optimal. In addition, because identifying and extracting MEV opportunities is a complex business this could provide larger, more sophisticated, miners the edge resulting in more centralization. That said, one solution to this problem that has been applied on Ethereum is Flashbots, an open-source software that quantifies MEV and allows greater numbers of miners (or validators post Merge) to participate in MEV extraction. To my mind, I see no reason why it is beyond the ability of Bitcoiners to deploy a similar approach on the Bitcoin blockchain as a way to mitigate the centralization risk from ordinals. (NB: This centralization risk occurs because of higher transaction fees whether they are the result of ordinals or not).

Nuclear Option

For the reasons outlined, I do not think that ordinals present much of a problem for Bitcoin. I don’t see a future where cartoon wizard jpegs, or rare sats, trump financial transactions as the best – or most popular - Bitcoin use-case. Rather I think the evolution of Bitcoin to include other use-cases, such as collectibles, is a positive in that it will bring Bitcoin to the attention of larger numbers of people. This is valuable because as I have pointed out in previous research notes for anything to be valuable it must be in demand by a network of people. This principle applies to companies and their products, social media platforms and crucially money.

Nevertheless, Bitcoin maxis are clearly are unhappy with ordinals and inscriptions and it has been suggested the Bitcoin code is tweaked – potentially even hard forked - to allow inscription transactions to be censored. Such a radical outcome is, in my view, neither necessary nor likely but is certainly risky.

What Bitcoin is and becomes is not set in stone. Instead, Bitcoin gets defined by what the majority of its holders wish it to be. We know this from the simple fact that over the years there have been many hard forks of the Bitcoin blockchain (Bitcoin Cash, BitcoinSV etc etc) and none have challenged the supremacy of Bitcoin. The crowd gets to decide what Bitcoin is, and if the crowd is content to have ordinals and inscriptions then so be it. Any move to try stop and filter out inscriptions via hard/soft forks would risk throwing the baby out with the bath water and in the long run may do more harm than good to Bitcoin’s future prospects. After all, wasn’t one of the founding principles of Bitcoin that it is supposed to be censorship resistant?

Until next time.

Ryan Shea, Crypto economist

1As Casey readily acknowledges the idea of applying identifiers to satoshis was first outlined back in 2012, but as often occurs in science the idea fell by the wayside only to be rediscovered later on - see: https://docs.ordinals.com/introduction.html

2For a good explanation with a toy example – see: https://kf106.medium.com/rodamors-bitcoin-ordinals-explained-e0cb775c68dd

3See: https://river.com/learn/terms/v/vbyte

4See: https://dune.com/dataalways/ordinals

5Amazon sells 2TB external hard drives for around $60.

6During the 21012 Eurozone crisis, which fuelled fears that the single currency could split, some individuals took to checking the serial numbers of EUR bank notes to see which EU member printed the bank note. For rather obvious reasons German Euro bank notes were perceived as being more desirable that those printed by the likes of Greece. However, this is very much the exception that proves the rule.

7See: https://www.theweek.co.uk/87798/rare-10-notes-which-ones-are-worth-thousands

8See: https://solarsystem.nasa.gov/solar-system/beyond/overview/