It’s been several months since I shared my thoughts about the world of NFT-based games and digital collectibles.

Since then, we’ve seen an explosion of interest in the category of games — as well as a huge increase in the amount of misinformation, confusion and downright cynicism.

Image by Sebastiaan Stam.

I don’t have a “dog in the fight” with games built on blockchains, other than being very enthusiastic about the potential it offers to create new types of games and experiences around open economies.

What I hope to share here is primarily for game developers who might be considering it, as well as to provide a “behind the scenes” view into anyone who is curious about the dangers and opportunities within the sector.

Some of the skepticism and cynicism is well-earned. I hope to be even-handed in this post, and cover some of that as well. But let’s start with a core belief that some seem to have:

At its core, a blockchain game is nothing more than a game that has an “open economy.” Before there were digital games, you could buy a board game and resell it afterwards. In fact, for most of videogame history you could buy a game cartridge, box or machine and resell it to anyone you want — just like any other piece of personal property. Games like Magic: The Gathering (MtG) and Pokémon were designed specifically around our real-world open economy where players would end up trading (and yes, speculating on) individual cards.

It was only with the more recent advent of online games based on virtual worlds where we started to see games where the gameplay was constructed around closed economies. For example, most MMORPGs don’t let you trade important items with other players — and most mobile games won’t even let you trade currencies. That’s because the experience (and business models) of those games is entirely based on a finely-tuned progression system.

Blockchains are replicated ledgers that enable open economies around games. They do it by providing transparent, provable and consensus-based mechanisms to allow games to exchange virtual property through decentralized markets and exchanges, and could even provide an ecosystem for interoperable items across constellations of games that choose to cooperate.

These games will be different than most current digital games.

I consider open economies to be ethically and morally neutral. Now, if you want to argue about whether capitalism and the right to trade private property in markets is unethical — then you can fight that philosophical battle if you want to. But it isn’t a battle that’s specific to game making.

Now, some people have indicated that they don’t actually like the open economy of games like Magic: The Gathering either — that they’d prefer not to have the meta constantly evolving and invalidating previous decks. Or that cards like Black Lotus should not be so scarce. It’s perfectly reasonable to debate the merits of any game design, although I’d urge you to think about what it takes to make a game sustainable over the long-run; there are plenty of CCGs other than MtG that nobody plays anymore, and it is usually because they couldn’t sustain their ecosystems longer-term. But the potential to have designs that some people dislike is not what makes an open economy into an evil one.

Although open economies are ethically neutral — you can’t say the same of some of the market participants, which brings us to…

This really sucks, but it’s true that there are just a lot of charlatans out there right now. They’re ripping people off, and spoiling it for the many people doing honest, hard work.

It resembles some of the early crowdfunded games; no doubt, if you’ve been watching game development for a while, you can think of a crowdfunded game that got some money and then left town. The difference with crypto projects is they tend to enable more anonymity along with the lure of liquidity (i.e., backers think they can exit or speculate in addition to supporting a game they’re excited about) and that’s exacerbated the problem.

Ultimately, crowdfunding for games became a lot less popular because so many games were unsuccessfully backed. In some respects, this is what you should expect to happen — game development projects fail at an alarming rate, because shipping a game (let alone building a good one) is so damn hard to do.

Selling NFTs or tokens to support a game is essentially a form of crowdfunding on the blockchain, so it adds the risk of dealing with a criminal with no intention to build something.

That said, I don’t think you should rule out backing a game you like via either NFTs or token sales. But learn what all of us have learned in crypto long ago: do some diligence, find out whether their experience matches their ambitions. Good game development teams are not afraid to put their reputations on the line.

It’s actually possible to have pseudonymous teams that are worthy of support— ordinarily, they’ve earned a reputation over a longer period of time as creators (and are able to verify their identies via public posts on platforms like Twitter). Their reputation is valuable to them as well.

And if your reason for backing a project is speculation — not the potential fun you’ll have in the play experience —realize that it is extremely risky.

Buyer beware: you’re investing in a team and their ability to execute and come together in a creative enterprise. When it’s a game, it’ll most likely fail... so expect that. But don’t let yourself be ripped off or tricked into believing that a fraud was just a failed project. If the team wants your money to build their dream, insist on a reputation and make them stake their reputation on it.

It’s surprising that this claim still circulates with respect to games, but that means that people aren’t really understanding the differences between games and the digital art markets — or how blockchain technology has changed in only the last year.

First, there are two main types of blockchain algorithms: proof-of-work (PoW) and proof-of-stake (PoS). PoW needs to solve massive cryptographic puzzles to reach consensus, and it’s what’s used by Bitcoin and the Ethereum mainnet. It requires substantial energy to complete these operations.

Blockchain games are not being built on Bitcoin or the Ethereum mainnet. Games are being built on proof-of-stake blockchains like Solana, Avalanche, Hedera, Polygon and Immutable X. Proof-of-stake does not take any appreciably more energy than any other cloud-based service, such as the backend technologies powering whatever you’re reading this on right now. Games are building on these blockchains for pragmatic reasons that include decreased energy use, as well as the high throughput and faster finality of microtransactions.

Indeed, if one were to look at the total energy utilization for a popular game: then the aggregate energy consumed by the GPUs on the front-ends by your players will end up being millions of times greater than the minuscule energy consumption of the proof-of-stake backend.

If you want to critique proof-of-work algorithms like Bitcoin, or other non-game applications of NFTs (like digital art collectibles, that do tend to mostly run on Ethereum mainnet) then that’s fine — just realize it has nothing to do with the games being built today.

A question to consider: if this is one of the aspects that has worried you, and all of this news about games using proof-of-stake is new to you… is it possible that there are other aspects of blockchain games that you dismissed or recirculated prematurely?

Let’s get something straight: game development is a craft. That means it depends on constant refinement, learning and iteration.

Almost nothing looks good in the earliest phase of a new market, a new business model, or a new genre.

Even games that are extremely popular today looked really bad in some of their earliest prototypes.

Right now we’ve seen a lot of junk. Just like there’s a lot of junk on PC games and mobile. There’s nothing wrong with junk. Everyone needs to learn. We’ll see improvement over time. The presence of junk at this present moment is not an indictment of blockchain; in fact, I’m aware of several games under development with AAA production value… Of course, those take the longest to bring to market.

“You don’t really own anything with an NFT, you just have a token in a ledger and it points at a JPG in a random server somewhere.”

First, the “JPG in a random server” applies to things like digital art collectibles. It isn’t particularly relevant to what happens with a game. It’s also a bit unfair to what it means in the art markets, but that’s a discussion for another article.

Here’s what a blockchain’s purpose is for a game: it retains the state and accounting history of the game’s economy (and this is typically done with a snipped of JSON, which is a good way of serializing this sort of data).

Individual games will decide how to interpret the data in the blockchain. That includes deciding to block utilization of items they don’t like. That means that yes, many game companies will likely get drawn into things like blocking item use that has passed through the wallets of bad actors.

Do you “own” it? We’re getting into a bit of a philosophical argument here: but if “own” means you have the agency to decide whether to dispose of an asset for another asset or currency with a third party — then yes, you “own” it. You own it as much as you own a Magic: the Gathering card that you’ve chosen to sell to another person, even if you don’t own the rights to the art on the card itself. The cardboard it is printed on isn’t much different than the token on the blockchain.

If you show up at an MtG tournament with a Black Lotus will you be allowed to use it in every format? No, because they make up rules for what you can and can’t do with your cards. If you steal a Black Lotus and get found out, will you be banned for life from playing? Most likely (in addition to other legal charges). Games built on blockchain can do the same thing. Again, all that blockchains enable are games with open economies.

It is possible for an NFT to point to a URL on a server, and if that server goes away, then the content of the NFT is gone. The ownership of what that NFT represents remains intact.

There are alternatives to centralized server dependences. For example, blockchain-based storage systems like Arweave provide for permanent storage that doesn’t depend on whether a server stays up.

The deeper issue with games is that the server is more complex: it’s a game server, after all — not simply a file store. And that does bring with it a lot more complexity. Since we haven’t seen the entire lifecycle of a blockchain-based game become popular and then shutdown, we can’t know for sure how this will play out in all cases. But there are a few ways it could go that are not entirely negative:

In fact, the latter case suggests an area of innovation: the digital-collectibles-first game company which focuses on creating a game economy, and open sources game implementation to the community.

People often point out how games that tried semi-open economies (for example, Diablo 3’s failed auction house) and failed. They usually don’t seem to remember that games with open economies like Magic: the Gathering exist.

Here’s the thing: games with open economies need to have different game designs. Most of them won’t be successful by bolting an open economy (blockchain) on top of a closed-economy MMORPG. Although there are a couple exceptions, most MMORPGs that tried to go from subscription-based to free-to-play failed for similar reasons — the game design is too dependent on the business model.

What will these new games look like? We don’t entirely know yet. Some of the earliest successes seem to be drawing a lot from the collectible card games like MtG, or from open world games that had land-based economies (such as Second Life). I expect people to experiment and there will be a lot of surprises. Game development is a craft.

The usual argument is that we have database technology that is faster, more scalable and more performant than blockchains. While that’s absolutely true, that is not the problem that blockchains solve.

The problem it addresses is giving ecosystems of people and applications the ability to exchange value with each other without requiring central gatekeepers.

Prior to blockchain, the main solution for creating open economies required a centralized service to provide APIs for all the participants to use. Even if you trusted those services to run reliably and forever, you’d get a hub-and-spoke model where everyone could build off the same economy. This exists in Second Life and their Tilia payments platform, and to a limited extent on Steam today.

“But didn’t various scrip systems, like the one George Pullman created, just do this in the past and blow up on everyone?”

No. That’s a centralized value-exchange system, not a decentralized one. It depended on the trustworthiness and control of a central authority.

Centralized systems are hub-and-spoke in structure. That means every game built around one hooks into a single service, where you’re constantly routing all transactions back through them. These networks are inherently more constrained:

As far as games are concerned, blockchain allows creators to leverage decentralized marketplaces, exchanges and other application code they don’t need to build themselves — and without needing to ask publishers for the privilege to participate. Whenever you need to rely on a centralized service of some kind (like publishing and selling your game or its in-game items), you give them the ability to extract a high rent for the service. You’ll also be subject to their whims about what is and isn’t allowed. The goal of blockchain is to enable permissionless development, where small teams can dream big — relying on a growing ecosystem they don’t need to build on their own.

In fact, there’s entirely new ways that games will be able to extend their ecosystems through modding and even open-sourcing their client software. We already know people like to mod user interfaces, levels, or add features. (and if you don’t know that, please take a look at Minecraft and Terraria and about a zillion other games you can find on CurseForge). Blockchain expands modding to include modding that interacts with the economy (custom auction houses, charting, etc.) as well as even outsourcing gameplay to the community itself (as LootProject is attempting).

Gatekeeping results in onerous rent-taking, which decreases innovation and creativity.

Blockchain is less efficient in terms of performance than centralized databases, because they need to reach consensus with all the validators. That’s what gives blockchain its power. Fortunately, some of the newer proof-of-stake blockchains have high performance so that the inherent inefficiency of this consensus does not get in the way of making a good game.

“But doesn’t a decentralized economy just mean a Wild West where anything goes?”

No. Each game can still choose how it wishes to interpret digital assets from the blockchain; authenticity can be checked by inspecting the provenance of items and ensuring that they originated from certain wallets. Items that were the subject of a fraudulent sale can be frozen. Items can be instantiated with their own game properties, and modified to fit the evolving game system (just as the rules affecting the Black Lotus in MtG changed over time).

In more open ecosystems that have interoperability between worlds, nothing forces a world to display or utilize an item in a way that clashes with the intention of the new world.

This is a reasonable concern. The critique is usually leveled against OpenSea: essentially, “meet the new boss, same as the old boss.” — that they’re creating a new central ecosystem people will depend upon for buying/selling their digital assets. And they’re also starting to introduce rules that allow them to control the sale process, or even freeze assets.

Online auction platforms did become (mostly) a winner-take-all market if one uses Ebay as an example.

Platforms like OpenSea attract a lot of eyeballs. Their value is not just the utility they provide, but the size of the audience they harness. Since the goal of a seller is to get the most bids, it does naturally favor the largest platform. In that sense, there’s not much difference form Web2.

It should also be noted that there’s going to continue to be a role for tastemakers and curators. In essence, OpenSea is currently the largest curator. But there are also curators like SuperRare who are focused on specific kinds of content, and we’ll be likely to see the same happen within games.

However, the underlying structure of NFTs is such that there’s far less friction in the ability to move an asset from one selling platform to another. We’ve already seen that new platforms like LooksRare can stand-up rapid alternatives to a platform like OpenSea. And I believe we’ll see the emergence of fully on-chain auction systems as well, which could potentially aggregate auctions across the entire ecosystem — enabling maximal auction bids without requiring everyone enter into the same auction house.

It’s a little early to declare any total victories in the marketplace ecosystem. I suspect it will be far more dynamic and competitive than the early successes indicate, but only time will show us how this unfolds.

There’s definitely backlash against NFTs.

This is classic innovator’s dilemma stuff: adding a new business model on top of an existing busieness has a high risk of alienating existing customers. This is what Ubisoft recently encountered, and what Star Wars Battlefront encountered when they tried to add free-to-play elements on top of a premium game.

Part of it is a branding problem — NFTs have, unfortunately, acquired a negative reputation due to misunderstandings related to their ecological impact on games, and due to the reality of the bad actors (criminals) who are taking advantage of the naive.

While there’s a vocal group of players who don’t want to see NFTs added to their games, there also seem to be a lot of players who are receptive to the idea:

Now, there are some criticisms of the above research: that the selection criteria for the survey biased towards a high number of “interested” players. That’s because the selection criteria for the survey required that people either own or previously owned an NFT, or had heard of them. One can assume that there’s still a huge number of people who have no idea what an NFT is (or are otherwise unsure enough that they said they didn’t know). So perhaps there’s a lot fewer than 53% in the population of gamers who are intrigued by the idea. Nevertheless, this doesn’t eliminate the evidence support a sizable segment of players who are receptive to it; even if it were just 1%, one could build a whole industry around it.

What I expect we’ll see is that — just as the games that use NFTs will be new types of game designs — they’ll also appeal to different audiences than the average AAA game player (just as mobile f2p opened up entirely new market segments in the past).

The core idea here is that we will have the ability to buy an NFT representing an item in a game that has some sort of utility in another game.

Sometimes the confusion is that people are claiming this means all NFTs will offer interoperability.

Some will, some won’t. And in the cases where they do, they’ll likely be limited to certain worlds and avatar systems. Some of those will be a single world, some of those will be constellations of cooperating worlds, and sometimes it will be large networks united by common avatar systems.

Interoperability does have a lot of challenges. When we talk about interoperability, there are actually several issues:

Blockchain really only addresses the “persistence” problem. It’s rather good at it in situations that envision multiple parties to interact without requiring a central rent-taking authority (cooperating worlds, interoperable marketplaces, galleries, economic applications, etc.)

The other layers either require people to cooperate (agree on how things can be presented, what they mean or behave) — or there could be default/undefined behaviors, meanings and presentation.

Sometimes this is pointed out more as a business criticism: i.e., that interoperability with NFTs is a solution in search of a problem. Players won’t care. Or that it will be such a terrible implementation that people won’t like it.

We already have interoperability within walled-garden ecosystems. For example, Second Life — despite exaggerated reports of its demise — continues to have a thriving economy in which creators build content for each other and take it from place to place. It happens at massive scale in Minecraft. It happens in Roblox (250M+ MAU as of the end of 2021). It’s happening now in VRChat. And for a long time, we’ve had interoperability at the modding level within games like Warcraft, Neverwinter Nights and a zillion other examples.

So, we already know that players are interested in interoperability within walled gardens.

Will we see interoperability happen between wildly different worlds with different aesthetics, or between different worlds with varying game rules? Maybe not, but that just ignores the interest in bringing your avatar from world to world as is already happening in more social environments — and between constellations of cooperating worlds.

All that blockchains do is enable an open economy. I can’t think of a reason why people wouldn’t want interoperability in an open economy if they want it in a closed economy.

Sometimes, advocates of blockchain say that NFTs will have a value beyond the life of a game.

Often, this life is expressed at two levels:

We know that players already have a desire to resell their gaming property — whether permissible (open market games like MtG) or illicit (third party markets that exploit World of Warcraft). Whether games like Axie Infinity are able to keep this flywheel spinning will be a matter of whether they keep the long-term enjoyment of the game high enough to continue to attract players. There’s probably room for a few games like this, just as there’s room for a few collectible card games that have endured over time (and as a digital product, you’d think there’d be even more room due to the higher availability, accessibility and abundance of options).

Once a game no longer exists, the utility value of the item goes away and there’s only the affinity value.

That means you’d expect an item to fall in value once it no longer has use as a playable object… But it doesn’t mean it loses all its value.

Speaking for myself, there are a few games I’ve played in the past where I wish I had retained some sort of digital collectible from the play experience. I don’t know if that’s something enough other people care about.

I don’t know what the long-term value of these collectibles will be. Nobody does. But I do know that collecting things is not new, and lots of people want to do it.

Perhaps there will be cases of interoperability where games cooperate and there’s more utility than simply the collecting value. Again, that’s an unknown — time will tell how important that will be to the ecosystem.