NFTs, Ultima Online, and player-run game economies

It’s tempting to think of games as a proving ground for technological and commercial innovation, such that any new form of human-computer interaction implemented into gaming is surely poised to permeate the aggregate of modern life. In some cases, this is true, such as with the proliferation of smartphone interaction and the explosive growth of mobile app stores, which started — and still mostly exist today — as games stores.

But it is not always true. Consider Pokemon Go, which is an enormously successful game that sits alone in the AR Gaming category. It’s true that substantive mass-market consumer trends sometimes start life resembling gimmicks in games, but it’s also true that most gimmicks in games don’t evolve into substantive mass-market consumer technologies. Establishing a successful gaming category is difficult in its own right: not every innovation that takes root in gaming will emanate to other consumer touch points. And not every new innovation introduced by a game will even find success in other games.

In my estimation, the first popular adoption of NFTs in something that could be construed as a game was in Cryptokitties, with $1.3MM being transacted inside the product within just a few days of launch in 2017. Cryptokitties at one point accounted for about 25% of traffic on the Ethereum network, although the product’s popularity has waned considerably since that time. In recent months, NFTs have seen a resurgence of interest, primarily led by art collections showcased on Twitter such as the Loot project, Bored Ape Yacht Club, etc. But while collectors of these assets , they do not constitute games — they are simply collectible art, existing independently of any more consequential or engaging media experience.

So how can NFTs be natively, deliberately, and meaningfully implemented into games to stimulate player-led economies? In this article, I propose that Ultima Online, the 24-year-old MMORPG developed by Origin Systems, provides a helpful framework to use in considering the use of NFTs to bolster a truly player-led game economy. Note that Ultima Online has changed appreciably since it was launched in 1997, and I use my own anecdotal history with the game to depict the state in which it existED in the early years after its first launch. Ultima Online is considered one of the most complex and player-driven MMORPGs to have ever been released, and while some might argue that it doesn’t serve as the best example of a player-operated economy, it’s the one I know best as a former avid player.

What are NFTs and what is play-to-earn?

While a thorough examination of NFTs is beyond the scope this article, some background is necessary in understanding their application within game economies. An NFT, or a non-fungible token, is a claim of ownership for some unique asset as represented on a blockchain. Non-fungible means unique, or not interchangeable, and so the non-fungible property of NFTs prescribes that each individual token is valued independently of others and not as a class, like Bitcoin. Most commonly, an NFT refers to a claim of ownership on a digital asset, and especially digital artwork. For a more thorough explanation of what NFTs are and how they function in the cryptocurrency ecosystem, see this article and this article.

The NFT framework has popularly been used to provide claims of ownership for digital collectibles, meaning, within the context of a collection, that a particular person owns a particular asset. As I’ll speak to later in the article, this claim on ownership is tenuous outside of the context of a particular collection: digital images can be replicated endlessly, but not only that, any visual art can be emulated to such an extent that imitations are indistinguishable from originals. This is why the construct of collections is so important within the context of NFTs: an NFT holder has a claim of ownership on a specific item within a collection, and the market for that collection exists as the totality of people that are aware of the existence of that collection. The market dynamics for NFTs are not dissimilar to those of high-end art, where a very small community of the wealthiest individuals on the planet bids against itself for items that only that specific community understands to be valuable. The book Black Edge does an extraordinary job of elucidating the motivations of this rarefied group.

The concept of NFTs is not particularly new. What’s more recent is the play-to-earn model for games, which utilizes cryptocurrencies and NFTs as in-game resources that can be sold on external marketplaces to provide real-world income from gameplay. The play-to-earn model is most recently associated with Axie Infinity, a game developed by Sky Mavis. This podcast with the COO of Sky Mavis provides some helpful guidance around the nuts and bolts of that game’s economy, although the COO views the potential market for play-to-earn games through a lens that is at least slightly more rose-colored as a result of Sky Mavis’ recent fundraise.

With this abbreviated background deployed, I’ll next present a depiction of the player-run economy of Ultima Online, and I’ll explain how I believe it serves as a worthy framework for considering the use of NFTs in game economies oriented around player interactions.

The player-run economy of Ultima Online

Ultima Online is a MMORPG (massively multiplayer online role-playing game) that was originally released by Origin Systems, a subsidiary of EA, in 1997. Despite its age, the game retains a loyal player base, and EA last released an expansion pack for the game in 2015. Ultima Online is generally considered the first popular, scaled MMORPG — the creator of the Ultima franchise, Richard Garriot, coined the term — although various MUDs and smaller-scale MMORPGs predate the game. The game has changed materially since it was first released, but the focus of this article is the original design of the game, which is captured in this paper, titled The In-game Economics of Ultima Online and published in 1999.

What was perhaps most innovative about Ultima Online when it was released: as an open world, it featured a truly player-run economy. Resources (gold, wood, iron ore, fish, meat from animals, etc.) existed in the open world that could be harvested by players and improved in various ways to fuel their exploits and adventures. Some raw materials and natural resources found throughout the world held intrinsic value and could be sold to NPCs or other players, but most didn’t: they needed to be improved through some sort of crafting process in order to be fashioned into something valuable. Gold is the primary currency of Ultima Online’s economy, although open bartering between players allows for any item to be priced in and sold for any other item.

A good example of a raw natural resource in the game is iron ore: iron ore is mined by players, and in its base state it is heavy, difficult to transport, and either worthless or of very little value. But iron ore can be smelted into iron ingots, which are very valuable because they can be used as inputs in a blacksmithing process to create a multitude of items, such as weapons and armor. And those products of the blacksmithing process are even more valuable than the raw iron ingots, especially when crafted to a high level of quality by a grandmaster blacksmith. The “economic flow” of the game, as outlined in the aforelinked paper, is presented below. The value chain of resource collection to item crafting is captured in points 4-6.

Point seven captures the crux of the player-run economy: vendors. Vendors exist in two flavors: NPC shopkeepers and player-owned vendors. NPC shopkeepers occupy the shops that populate towns, and they buy and sell specific categories of goods (eg. NPC shopkeepers in a Mage Shop sell items related to the practice of magic: reagents, spells, magic books, etc.). Players can sell resources and finished goods to NPC shopkeepers, but the NPC-centric game economy is mostly divorced from the player-run economy and thus NPC shopkeepers pay players for goods and resources on a different and much lower scale. This “dual-track” economy dynamic is illustrated in the aforelinked paper with this diagram:

Players mostly sell crafted goods and other harvested resources to other players: either via in-person bartering or through player-owned vendors, which are NPCs that players can “hire” to sell goods on their behalf as attachments to homes that they own, paying a fee on the basis of the value of the goods held in inventory. Player-owned vendors allow players to sell anything, and they are permanently open, meaning the player doesn’t need to be online in order for their goods to be sold through their vendor. Vendors are the centerpiece of Ultima Online’s player-run economy because they facilitate the sale of not just undistinguished wares and resources (such as player-crafted weapons or harvested natural resources) but also rare items such as high-value weapons with special, magic properties.

Player-owned houses allow for the placement of vendors, but houses also provide many more benefits to their owners:

  • Storage. Every player in Ultima Online has access to a bank account, which is universally available at the bank in every town and cannot be accessed or interfered with by other players. But bank accounts provide very limited storage, whereas houses can be filled with chests and other forms of repositories that can be used to store all manner of items collected over time;
  • Guild assemblage. In order to form a guild, which is the core social structure within Ultima Online, a guildstone must be purchased and placed at a house owned by the player. But more importantly: houses allow guilds to gather, store wares, and spar. Many of the larger homes placed in Ultima Online are owned collectively by guilds and not individuals and serve as their headquarters in the game. Guilds are the centerpiece of social activity in Ultima Online;
  • Skill training. Training skills in Ultima Online eventually became an exercise in operating “macros,” or pre-recorded sets of keystrokes that could be run by a third-party program so as to automate much of the tedious, repetitive activity needed to progress a specific player skill. “Macroing,” or the unnatended use of a macro program to direct the player’s avatar to perform some activity, was banned in Ultima Online, but that ban can only be enforced when another player witnesses the activity and reports it. The conduct of players outside of a house is hidden from them, and houses are placed outside of cities in areas where foot traffic is light. So home ownership provides a safe place for macroing to be done.

It’s worth noting here that the social and gameplay environment in Ultima Online, at least in the first few years after its launch, was ruthlessly, brutally savage. Players were protected in towns from any sort of criminal behavior by guards who would kill offenders at the moment of their offense. But the territory outside of cities was utterly lawless, with murderers, bandits, and scheming hustlers lurking everywhere. Being murdered by another player in Ultima Online carries consequences: all of the player’s gear can be looted, and the deceased player takes the visage of a ghost and is left to wander their surroundings until encountering a Healer NPC or a player kind enough to resurrect them with magic. Anything looted from the player’s corpse is gone for good, including house keys. If a house key is stolen or looted from a player, then the beneficiary of that key can simply stroll into the house in question and pick it clean.

Very few player assets in Ultima Online are permanent: items can be stolen and looted, but many also have limited durability and break after enough use. Houses are no exception: houses can decay if they are not used by their owners, and abandoned houses will disintegrate after a long enough period of neglect. When this happens, the structure of the house disappears but its contents are scattered across the parcel of land it once occupied, available for any spectator to pilfer. Once a house decays completely and its erstwhile plot is cleared of debris, a new house can be placed on that parcel. For this reason, groups of opportunists often gather around houses that are “IDOC,” or “in danger of collapse,” meaning they will collapse imminently unless used by their owner. These people hope to not only scrounge for valuable items but also to place a new house in the newly-vacant lot.

Houses are placed on vacant lots by executing deeds, which can be purchased from NPC vendors. House types differ in size and storage capacity; small, single-story houses offer much less storage and space than multi-story homes, towers, keeps, and castles. The footprints of house types (for classic Ultima Online) range from 7×7 units for a small house to 31×31 units for a castle. Ultimately, most Ultima Online servers (called “shards”) became populated to the extent that placing new houses via deeds was impossible: a home could only be acquired by waiting for an existing house to decay, or, more commonly, by purchasing an existing home from another player.

The player-run economy of Ultima Online eventually migrated away from the game client and onto eBay, where in-game items, in-game currency, and entire accounts are sold for fiat currency. But housing became the most valuable asset in the game, whether priced in gold coins or in US dollars — all other in-game assets could be harvested given enough time and commitment, but housing was truly scarce relative to the size of the Ultima Online player base. And, after some point in the game’s lifecycle, the only reliable and immediate avenue for acquiring in-game housing was eBay. Note that the game didn’t feature a server-wide marketplace mechanism, and placed housing couldn’t be sold by vendors. Either of these product features might have alleviated the need to use eBay to sell in-game items.

It’s interesting to consider housing in Ultima Online, which connects to fiat monetary value quite directly, as the economic foundation of the player-run economy. The supply of houses isn’t limited: NPC vendors can theoretically sell an infinite number of house deeds. What is limited in supply in Ultima Online is land: the parcels on which houses can be placed. A placed house might sell for 10 or 100 times the price of the deed for that house, such is the scarcity of available plots of land for certain types of structures. And land in Ultima Online has unique properties:

  • Non-tradeable. Land in Ultima Online cannot be “owned”: anyone can place a house on an empty plot of land. Placing a house confers de facto ownership of the plot of land underneath it because only one house can be placed on a plot of land at a time. But no player can make claim to, or sell, a virgin plot of land — an empty plot can be claimed by anyone quick enough to place a house on it;
  • Comprised of atomic units that increase in value with contiguous scale. As noted earlier, the smallest footprint of a house that can be placed in the game is 7×7. Contiguous lots create opportunity for plottage value, meaning a large plot of land has more value than some collection of non-contiguous plots of equivalent total area because a larger house can be placed there;
  • Value partially derived from commercial usage. Housing has inherent value in Ultima Online because of the use cases identified earlier (storage, guild assemblage, skill training). But housing can also generate income through vendor placement: houses near highly-trafficked roads are worth more than houses in remote regions of forest for that reason. The wealthiest in-game merchants own houses in commercially convenient locations, often just outside of city boundaries (and thus guard protection), where entire shopping districts emerged.

The entirety of the player-run economy in Ultima Online is predicated on the ownership of houses. Houses provided the space required for players to store items, operate and convene their guilds, to train skills, and to generate income from vendor sales. And this in-game economy connects to the fiat economy primarily through housing: because selling housing inside of the game is complicated and fraught with risk, players primarily traded in housing by selling entire accounts on eBay, which provided some level of transaction security. The value of housing is derived from the limited supply of land on which houses can be built but also the intrinsic utility of housing, which provides many of the requisite conditions for successfully engaging with the game’s social structures. Namely: guild participation.

NFTs and successful player-run economies

NFTs are helpful instruments for tracking and revealing ownership in highly curated, command-controlled environments. When an artist or organization declares that a collection of NFTs is limited in supply and pronounces that collection as authentic and original, then that authority and brand identity is what confers value on the associated assets (versus ownership of mere copies, or “right-click usurpation”). Note that, while the blockchain-enabled record of ownership captured in an NFT confers trust in the system of account, it’s the credibility and authority of the original issuer that imbues that trust with any value. This is actually a highly-centralized means of operating: without the authority of the issuer as to the authenticity of the item being traded — eg. the artist that created the limited-edition JPEGs declaring that those JPEGs belong to the collection — then the ability to endlessly reproduce those items, or create forged or inauthentic versions of them, renders any claim on them impossible to authenticate, despite whatever entries in the blockchain may exist as a record of ownership.

In-game economies are well-suited to allow for decentralized trading of NFTs while still providing the imprimatur of authenticity in a command-controlled, centralized environment free from outside influence or intervention. A market for NFTs can exist outside of a game, just as the player-run economy exists independently from the NPC-run economy in Ultima Online, but the game provides boundaries and limitations for what in-game assets are tokenized, and how they are used. This control by design prevents “right-click usurpation,” constrains the surface area over which NFTs are minted and how they are used, and ultimately serves as a function of authenticity. Games are closed environments: if an NFT for an in-game asset confirms ownership, then the purchaser knows the asset is legitimate and authentic, because the game publisher controls which assets exist in the game.

The promise that NFTs present for games, then, is to facilitate the transaction of in-game assets such that:

  1. The marketplace isn’t controlled or operated by the game developer, as an in-game marketplace is susceptible to manipulation of trading prices, volumes, and availability;
  2. Transparency is enabled across the marketplace for in-game assets. The playerbase can engage in real-time price discovery and potentially uncover who owns what;
  3. Fractional ownership of in-game assets is made possible;
  4. Players can sell their in-game assets at will without the game developer serving as a gatekeeper for those transactions.

This promise, however, necessitates a number of game design imperatives. The first and most important of these is that the value of in-game assets not be determined by an inflow of new players to the game. When this happens — when prices are a mere function of new player adoption — then the game’s economy has broken, and engaging in transactions becomes a speculative exercise. The game’s design, and especially the designation of in-game assets for tokenization, must obviate this outcome.

The second of these imperatives is that the assets available to be tokenized contribute to core gameplay and are not mere collectibles. Collectible NFTs are a wholly distinct category from NFTs that are components of a game’s economy. Collectibles and cosmetic items are helpful tools as currency sinks to remove hoarded money from an in-game economy as a means of militating against inflation, but they don’t otherwise contribute to gameplay and only superficially support an in-game economy.

And the third of these imperatives is that the in-game assets tokenized as NFTs be derived from some combination of in-game resources or items that are freely accessible and can be harvested by players without participating in the NFT trade. In other words: there is no hard cap on the number of NFTs that have been tokenized, even if there is a hard cap on the number of NFTs that can be traded at any given point in time.

These game design imperatives can be instituted through the deliberate construction of an NFT-capable game economy through these design choices:

  • NFT non-permanence. As with houses in Ultima Online, the in-game assets that are tokenized as NFTs might decay as a result of non-use such that players that stop engaging with those assets or abandon the game altogether might forfeit them. NFTs might also simply disappear with the death of a character, or be stolen or looted by another player. NFT decay might mean that the in-game representation of the NFT simply evaporate, with a connection severed between the in-game asset and the NFT, potentially allowing for another NFT to be minted in its place (increasing the total number of NFTs minted but leaving the number of NFTs effectively available for trade constant);
  • Derived scarcity. Again, as with houses in Ultima Online, the NFTs themselves cannot be scarce or limited if the economy is designed to be fortified against a dependency on a constant influx of new players. To achieve this, NFTs can be minted on the basis of access to a limited resource in the game that is not tradeable or ownable but that nonetheless serves as a prerequisite for manifesting the in-game asset to be created and tokenized. In the case of houses in Ultima Online, this limited resource is land: players can’t own the plot of land on which they place a house, but only one house can be placed on a given plot at a time, and relinquishing ownership of a house (eg. through non-use decay) frees the plot to be used by someone else;
  • Intrinsic utility of NFTs. The in-game assets made available as NFTs should hold utility in the game and present the owner with value beyond the speculative potential of the NFT. In the case of houses in Ultima Online, this value is captured through the storage, guild assemblage, and skill training benefits bestowed through house ownership. If the NFT’s only value to the player is potential market appreciation and not of some in-game use case, then the NFT doesn’t contribute to the in-game economy.

These game design choices ensure that the player-run economy remains yoked to the mechanics of the game and doesn’t exist as a wholly independent marketplace that serves only to allow players (and non-players) to speculate on assets. The in-game environment should impact NFT demand: while a well-functioning NFT market might have influence over in-game behaviors, the game’s design and economy ultimately dictate the value of NFTs with unidirectional authority. A genuine integration of NFTs into a game economy provides an avenue to players to monetize their time spent in the game but it also creates an independent, neutral marketplace for trading in-game assets.

The inescapable reality of supply and demand

Note that a functional integration of NFTs into a game’s economy renders those assets more, not less, centralized. If in-game economy decisions can change the intrinsic value of in-game assets, and it is that value that is captured with NFTs, then the market for NFTs is completely centralized within the operational domain of the game studio. But this is inherently true for all NFTs, since the non-fungible property means that the issuer defines the differences between the assets and can potentially render those differences more or less meaningful in some other jurisdiction.

“Scarcity” is sometimes proposed as the prevailing valuation input for NFTs, but scarcity is a fabricated construct: things can only be “scarce” on a relative basis, and even then, only demand influences prices. Francium, for instance, is the rarest naturally occurring element on Earth, but because it is so unstable, no use for it has emerged, and no market for it exists. And because it can’t be used to do anything, it wields a theoretical price of $1BN per gram, based only on very thin trading for trace amounts. Of course, even if a full gram of Francium was available to purchase, no one would want to buy it, especially for $1BN. Any price point, given the lack of demand for the substance, is an exercise in extrapolation and demand dilation. The same happens frequently in the market for NFTs by abusing the concept of “scarcity” and ignoring the inescapable realities of supply and demand.