"Web3 gaming" that doesn't suck

I just got back from ETHDenver last week, and one particular thing struck me: despite the surprising vitality of the web3 space as a whole, "web3 games" are still really, really bad. There was a little corner of the National Western Complex labeled the "Blockchain Arcade", and in it were three different kinds of installations:

• Non-web3, traditional arcade games. These were the only ones that actually had attendees playing.
• "Web3 games" that were basically low-to-medium quality traditional multiplayer games. Except some element from the 2021 crypto bull market was tacked on: tokenomics, NFTs, trading. Nobody were playing these.
• Non-interactive videos of supposed gameplay playing in a loop

Of course, blockchain games being bad is a meme at this point, and nobody seems to be surprised. But web3 failing so much in the area of gaming should be surprising. Core blockchain use cases like cryptocurrencies and DeFi do actually provide important innovations. Many of my Geph users have told me that they would not have been able to pay from their restrictive jurisdiction without using Tether or Monero, and protocols like Uniswap are obviously not legacy platforms with blockchain window-dressing the way "web3 games" are. Even somewhat dubious applications like art NFTs have proven to at least be partial successes — BAYC ape pictures still sell for $100K+! What could be so special about games?

Games are web2 but worse

The easy answer is that the anti-crypto people are right, and blockchains — at least for games — are really a Solution In Search Of A Problem™.  But that doesn't seem right — video games suffer from the exact "Web2 problems" that Web3 is supposed to solve, except much, much worse. If anything, games and not FAANG giants should be the poster child of "why Web2 is broken", and gaming should be the ideal problem for web3 to solve.

Games overall are stuck in a pre-open-source, closed model. Most games from AAA to indie build on proprietary tech stacks, use business models such as pay-per-download that require aggressive DRM, and extensively depend on the broken system of modern intellectual property rights. Compared to other tech companies, game studios only minimally benefit from or contribute back to the vast open-source ecosystem that even most "Web2" firms rely on. Open-source games are almost as bad as Web3 games — there's nothing like Linux, Chromium, or Docker in the gaming world.

Networked games suffer further from massive centralization and insecurity. Players of MMOs like PUBG or World of Warcraft lock a slice of their social interactions into a centralized platform just as unaccountable as a Big Tech social media website. Even worse, in the name of anti-cheat, multiplayer gamers endure invasive mass surveillance beyond even Web2 norms — when's the last time you had to install a rootkit on your computer to use a social media website? This is terribly unaccountable trust, of course, but even if you trust your game publisher, the attack surface of a complex C++ networked game with built-in anticheat spyware is almost too horrific to contemplate.  My guess is that your average security/privacy-conscious nerd's worst security fail is gaming on the same computer holding all their secrets.

Furthermore, the whole game industry screams incentive alignment and market failures. Game studios struggle to make ends meet, gamedevs make significantly less money than similarly experienced developers elsewhere in tech (despite gamedev being arguably a rarer and harder skill!), and gamers often complain of being "fleeced" by "greedy" game studios. The double thank you of the market is nowhere to be found. This is especially bad in networked/social games like MMOs — lacking the ad revenue of a typical Web2 social network, they are forced to extract value in incredibly inefficient and arguably unethical ways — typically by targeting the tiny minority of addicted "whales" that generate most of the income.

So in gaming, we see supercharged versions of the typical web2 problems: closed ecosystems, centralization, and bad incentives. The reason behind this is probably nontrivial — my guess is it's something to do with the combination of extremely high dev and infrastructure costs of games (especially networked games), and a very different dev/user dynamic (less technical users that can contribute back to an open-source game).

Regardless, "web3 tech" like crypto[graphy/currency] and blockchains seem designed exactly to solve these problems. But that definitely needs a vision drastically different from the current generation of "web3 games".

Games can be web3 but better

Given that games have especially bad web2 problems, web3 games can, and should be an especially big improvement.

But we need a recipe that is actually focused on solving these web2 problems, rather than shoehorning "web3" elements:

1. Envision a game that:
   • is open source
   • runs on decentralized infra and prioritizes security
   • is funded by fair and economically efficient means
   • as a result, more fun than existing games
2. Try implementing it. When you get stuck, figure out a solution using web3 tech.

This basically guarantees you'll never be stuck with a "solution in search of a problem", because you'll have your problem first before trying to shoehorn any solution.

Simple indie-style games probably don't need too much web3 magic to achieve openness, decentralization, and better incentives. A singleplayer survival game like The Long Dark already runs on decentralized infrastructure (individual PCs!); making one that is secure and open source "merely" reduces to a small public good funding problem: to come up with the cash to fund a one-time engineering, art, and security audit effort.

How you solve it is up to you — you could go the crypto-bro route of utility token presales and NFT auctions, though I personally prefer cool economics hacks like quadratic funding or dominant assurance contracts. In any case, web3 tech like smart contracts provides funding options that align profitability with a good game far more than "beg for donations" or "use DRM and make users 'buy' copies" — and it wouldn't make the game merely a traditional game with web3 tacked on.

On the other hand, complex and social networked web3 games, like MMOs, require a lot more thought. Not only do we need to solve funding, we also need to solve secure decentralized networking, anti-cheating in open-source games, decentralized dispute resolution, and a litany of other hard problems. On top of existing "web3 puzzle pieces", there's probably still quite a lot of "deep tech" development that needs to be done to build a revolutionary web3 MMO.

But I think that such development will be worth it. Web3 superpowers like aligned incentives and decentralization will make highly social MMOs dramatically more fun, profitable, and socially beneficial. The best way to illustrate this is to look at a "properly web3" MMO, how it would work, and the possible ways it could be implemented.

That would be Ostomo — a hypothetical design for an open-world, market-based, EVE Online-like MMO that:

• uses open source clients, with third-party alternatives
• runs on fully decentralized, user-operated infrastructure
• funds itself through a user-aligned market that avoids Web2 gaming pathologies like grinding, pay-to-win, DRM...

I'll try to sketch out the building blocks needed for Ostomo — a decentralized pubsub network, game mechanics designed for permissionless enforcement, a dispute-resolution market... But that will have to wait to the next blogpost 😊