A year ago I told you about Loot, a collection of non-fungible tokens that had inspired an energetic community to form around it. A series of brief, text-based descriptions of fantasy genre items like swords and amulets, Loot captured the imaginations of builders and speculators who wondered if the underlying NFTs might someday serve as the basis for graphic novels, movies, video games and more — an open-source art project that could eventually become the foundation of a crypto-flavored Marvel Cinematic Universe.

Loot was also the first NFT project that I found personally compelling, at least as a subject for journalism. Its creator, Dom Hofmann, was well known to me as a co-founder of the short-form video pioneer Vine and the idiosyncratic social network Peach. It was an art project more than a startup — Hofmann made 7,777 randomized “bags” of loot available for free to anyone who would pay the transaction fees necessary to mint them on Ethereum. And the project blew up overnight: five days after launch, Coindesk reported, Loot bags had generated sales of $46 million, and had a market capitalization of $180 million.

Of course, in those days lots of NFTs were selling for eye-popping sums. What made Loot stand out was the way a community of builders stepped forward almost immediately to begin building out an ecosystem: creating art for the items contained in the bags; forming guilds for people who owned the same “rare” items; and writing smart contracts to let people trade the items held within their NFTs.

The dream was that all this would coalesce into something more than the sum of its parts. But 2022 has been catastrophic for the larger NFT market, wiping out the majority of the notional value for even the most richly valued collections. And so as we near the anniversary of Loot’s launch on August 27th, I got curious: what had happened to Loot in the year since it went viral?

“I think it’s doing well, in the sense that the things that are being built on top of it right now are really interesting, or really ambitious, and in some cases both,” Hofmann told me over the phone Thursday. “A lot of them are closer to the types of things that I thought we might see in the early days of the project.”

Hofmann had spent the previous hour and a half on the phone with a developer planning to launch a Loot-related project, something he said he does fairly frequently. His full-time job is running Sup, a 12-person game studio working on various projects that he mostly declined to discuss on the record. But Hofmann continues to lurk in Loot forums, piping up when he feels like he has something to offer, and messages frequently with Loot developers.

That development continues despite the fact that the speculative mania around Loot has cooled. The project’s market cap has fallen to just under $6 million, and the price of the cheapest Loot bag available for sale has fallen from about $20,000 when I wrote about it to about $1,438 today.

So what are developers building? One project all the people I spoke with pointed me to is HyperLoot, which its creators call a “visual building block on top of Loot.” Essentially, Loot owners log in to its website with their crypto wallets and HyperLoot generates a picture of whatever is in their bag on the body of a digital adventurer. That image might could then serve as your profile picture, the basis for a novel, a character in a game, and so on. Like the original Loot, HyperLoot creations are released under the CC0 “no rights reserved” license, effectively placing them in the public domain.

HyperLoot is building a software development kit that will allow people to build using these creations more easily, co-founder Thanakron Tandavas told me over direct messages. The company is also building its own video game, CCO Wars — essentially a Super Smash Bros.-style fighting game, populated by NFT characters in the public domain. Think Bored Apes vs. CryptoKitties, coming eventually to a console near you.

“I feel that the same success could happen to Loot like other successful open-source projects that embrace permission-less innovations, [like] Linux or Android,” Tandavas told me.

The artist Shahruz, who uses one name, told me about a similar public-domain project he’s working on called 32 Swords. It’s “an elimination reality series” in which players must have both Loot and another NFT known as a Manny. The game is human chess, with each team given 12 hours to discuss and vote on their next move. As players are eliminated, their Loot bags are given away to the other participants or spectators.

Shahruz, too, was drawn to the idea of creating art from the public domain.

“I think there’s a lot of unique kinds of fun to be had working with existing projects and communities this way, instead of only creating new properties all the time,” he told me.

So what is there to learn from Loot a year out?

One, Hofmann told me, the early wave of hype around the project wasn’t as helpful as you might have guessed. Because the project was public domain, the early days were essentially a free-for-all of people making cheap clones and other associated projects that distracted from the more serious developers. It took a while for the hype to subside and the better projects to rise to the top.

Two, early enthusiasts likely under-estimated just how challenging it is for public-domain projects to coalesce. At the height of Loot’s hype cycle, plenty of folks wrote blog posts and tweetstorms about how Loot could form the basis of a video game. But it’s a really long way from writing a list of magical items to shipping Elden Ring, and the entropic forces that can cause multi-year delays for even the best-funded video games are magnified exponentially for games built by part-timers and volunteers.

Even in 2021, it was obvious that turning Loot into a video game would be incredibly difficult. But some of the intermediate steps have proven difficult, too.

One of the aspects of NFTs that Hofmann finds appealing is “composability,” the idea that blockchain applications are modular and can be joined together like Legos. In the weeks after Loot launched, other developers “composed” various secondary elements meant to work with it in this way: a compendium of beasts in the Loot universe, a list of dungeons, some visual elements, and so on.

But the result was a kind of Frankenstein’s monster that was less than the sum of its parts. You can write a new smart contract that attempts to join the pieces together, but as soon as someone wants to add a new piece to the stack you might have to start over.

“It’s really, really, really difficult to make a bunch of pieces of something that are supposed to be related to one another, relate to one another — and also be open to future pieces that need to connect,” Hofmann said. “Being future-proof is sort of difficult.”

Three, while much of the focus on NFTs over the past few months has been on crashing prices, Hofmann said that one of the biggest reasons web3 projects aren’t making much progress lately is how expensive Ethereum is to use. Transaction fees routinely run $10 or more to take a single action, making projects inaccessible to anyone other than rich nerds.

“Because of that, I think there is a whole slew of use cases — I would even say the majority of use cases — that are not within the reach,” Hofmann said. “Under normal conditions, you’re talking about closer to $10 and maybe even $100 to perform a transaction. … The price is prohibitive.”

All that said, Hofmann remains deeply interested in the technology. Making films used to be expensive, too, he said — recalling the time when, as a young teenager, he bought a book about filmmaking and learned he was expected to scrape together $50,000 for an entry-level film camera. Today every smartphone comes with a fully capable camera; that barrier to entry has disappeared.

And he still likes the basic properties of NFTs: the fact that they represent a piece of data that lies outside an application; the fact that the data has a verified owner; the fact that the data can be transferred to another verified owner.

“Those are pretty big inversions of the way most of the internet works right now,” he said. “Removing the value part from it altogether, just thinking about what it means to be the owner of the data and being able to prove that you’re the owner of the data. What that enables, just as a technology primitive, it’s super interesting. That’s not something that exists right now.”

One project Hofmann is working on, Corruptions, attempts to build on those properties. It’s a role-playing game that takes place entirely on the blockchain — it doesn’t even have a website — with holders of the relevant NFTs playing by interacting with Discord bots. Among the game’s more outlandish features is the CPC, or “community-playable character,” whose actions are voted upon the community.

I don’t know if Corruptions — or Loot — will ever reach an audience of more than a few thousand people. (Hofmann estimates Loot has about 200 developers currently involved in some capacity.)

But I’m struck by the degree to which developers like him continue working to make something out of the NFT and its strange properties. No one I spoke with today suggested that developers have gotten particularly close to cracking the code — just that it all continues to feel interesting enough to keep working on.

A year from now, Hofmann said, there would likely be Loot-related fiction in the world, and perhaps the first Loot-related video games. And as other CC0 projects join the public domain, they will likely continue to collide with Loot in surprising new ways.

“It just moves so fast,” Hofmann said of the NFT landscape. “And it changes so much.”