Ethereum Etheria v1.2

Information

Created: 1 November 2015
Creator: Cyrus Adkisson
Blockchain: Ethereum

Description

The Demo, the Silence and the Rediscovery

Taking the island to London

On November 1, 2015, Adkisson deployed Etheria v1.2. It was almost identical to v1.1 in its basic logic, but it became the version he would present to a wider Ethereum audience.

The ownership of v1.1 was copied across to the new contract, after which the earlier version was locked. The map retained its familiar shape; the software around it had evolved. Two weeks later, on November 13, Adkisson took Etheria to London for DEVCON 1, Ethereum's first developer conference.

A surviving video interview shows him discussing the project and its ambition to create a decentralised virtual world that could not be switched off by its creator.

It is tempting to imagine a packed hall of people immediately grasping what Etheria might mean for digital ownership. The surviving record offers a less dramatic picture. Ethereum's developer community was small, the project was difficult to use, and buying digital property had not yet become a familiar market.

Still, the demonstration gave Etheria an audience beyond the people who had discovered it through Reddit or examined its contracts. Adkisson could show the world rather than merely describe what it was supposed to do.

And the world contained something that would interest later NFT historians: actual structures recorded on the blockchain.

A cabin, a horse and a colosseum

Etheria's land was not meant to be a static map. Owners could collect blocks and arrange them into three-dimensional structures, using the rules encoded in the contract. A cabin, a horse and a colosseum were built from the hexagonal blocks. These constructions were evidence that someone had used the contract to create persistent objects within the world itself.

Today, building in a virtual world is hardly unusual. In 2015, making the world state itself live on Ethereum was a different proposition. The blockchain was not merely recording a receipt for something stored elsewhere. It held the data describing ownership and construction.

There was a price for that approach. The original building system was inefficient by modern standards. Every placement required calculations to ensure that blocks did not overlap and that the proposed structure obeyed the rules of the world. As a tile accumulated blocks, the contract had more data to inspect. Eventually, a transaction could exceed Ethereum's gas limit.

The old how-to guide warned players about the problem. According to the supplied research, farming more than about 260 blocks across thirteen sessions could make further building operations exceed a transaction limit of approximately 3.14 million gas. A player could reach a point where the tile was effectively bricked for further editing.

The world had no central server to crash, but a badly behaved transaction could still stop a building project in its tracks. Decentralisation had not abolished the practical consequences of bad engineering. It had made those consequences part of the permanent record.

Even playing the game required considerable patience. The archived instructions expected users to install an Ethereum client, paste in a lengthy interface definition and call contract functions directly. This was not a matter of opening an app and dragging a block into position. Players needed to be comfortable at the command line and willing to interact with smart contracts manually.

For a small group of Ethereum developers, that was manageable. For most people, it was a formidable barrier to entry.

A quiet world

Etheria v1.2 did not sell out. Each version of the map contained 457 ownable land tiles, but most remained unclaimed. The project eventually went quiet, and Adkisson later described it as defunct on his LinkedIn page.

The contracts continued to exist on Ethereum, preserving their ownership records and building data. That distinction is central to Etheria's story. In a conventional online game, a deserted world can disappear when its operator stops paying for servers or decides to close the service. Players may keep screenshots, but the game itself is gone.

Etheria was different. Its lack of an audience did not mean that its underlying world had ceased to exist. The blockchain continued to preserve the contracts. A tile could remain unclaimed, a structure could sit untouched, and the world could wait without requiring its creator to maintain a server.

Immutability does not guarantee that software will remain easy to use, that a world will attract players or that every asset will always be accessible through convenient tools. Etheria's own history demonstrates all three limitations. But its underlying record survived the project's years of neglect.

For a while, that survival went largely unnoticed.

The rediscovery of 2021

In March 2021, the NFT market began looking backward as well as forward. The sale of digital art for enormous sums had drawn public attention to blockchain-based ownership, and collectors started searching for older projects that might have been overlooked.

Etheria was a natural candidate. Its origins could be traced to October 2015, only weeks after Ethereum's mainnet launch. It predated ERC-721, the token standard that would later make NFT collections easier to create and trade, and it already contained a world of ownable virtual land.

The rediscovery gathered momentum through social media and contemporary coverage. Decrypt reported that a bid of 70 ETH had been made for one tile, worth about US$130,000 at the time. The figure was startling, but it also changed the project's practical circumstances. Something that had been treated as an abandoned experiment now had a market and a community of people interested in its history.

Adkisson returned to work. In March 2021, he bought the remaining unowned tiles in v0.9 and v1.0, concerned about the limitations of their original trading systems. He then spent months developing the tools needed to render the world and trade its land.

This work led to another chapter in Etheria's history: the archaeology of its own code.

The six-year-old compiler mystery

By late 2021, Adkisson wanted to verify that the surviving source code for Etheria v1.2 corresponded to the bytecode deployed on Ethereum in 2015. If researchers wanted to establish exactly what the contract did, they needed to connect the readable programming instructions to the binary code the network actually executed.

He approached Dedaub, a firm specialising in smart-contract security and analysis. The task sounded straightforward: compile the original source code with the old compiler and compare the result with the deployed contract.

It did not work.

The original code had been compiled using Solidity 0.1.6 through an early browser-based tool. Dedaub could produce code that behaved in essentially the same way, but its output differed from the original bytecode. Small differences prevented an exact match.

The team investigated the old compiler and discovered a strange problem. The JavaScript environment used to run the compiler could produce different output depending on what had been compiled earlier in the same session.

Compiling the same program was not always enough. The compiler's previous activity could affect the result.

The cause lay in the old JavaScript execution environment and its handling of memory. Solidity creator Christian Reitwiessner helped the team investigate. Dedaub then devised a way to vary the compiler's previous activity by compiling unrelated files before the Etheria source. This technique, effectively a form of fuzzing, allowed the researchers to explore a large number of possible compilation sequences.

The team ran parallel jobs and refined the results until it found a reproducible sequence that generated the deployed bytecode. The final recipe required compiling just two unrelated files first.

It was an absurdly small detail with an outsized consequence. The difference between a successful verification and a failed one depended on what the compiler had done before compiling Etheria. Six years after deployment, a piece of software that had helped create a virtual island had become an archaeological puzzle in its own right.

The result gave researchers a firmer basis for examining the contract. It also revealed something about the period in which Etheria had been built: the tools developers relied on were themselves experimental, and their quirks could survive long after the original development environment had disappeared.

Making an old world usable again

The revival required more than proving the code's identity. Etheria's original building system was too expensive to use comfortably, with gas costs that could make large structures prohibitively costly.

Adkisson developed a new building mechanism that used the tile's text-based name field to store compressed construction data. Rather than representing every building operation through the old, gas-heavy system, the new approach compressed the information needed to reconstruct a build. Etheria's own FAQ reports that the change reduced building costs by about 95 per cent.

He also created tools for rendering the world and managing tiles, making it possible to explore and build without manually calling every function from a command-line client.

In March 2022, he brought the four viable versions together in a single multiverse explorer. The combined view included v0.9, v1.0, v1.1 and v1.2, each with its own map and history. With 457 land tiles in each version, the four maps contained 1,828 ownable parcels in total, although Etheria estimates that five to ten per cent may be permanently lost.

The multiverse made the project's development visible. Each version preserved a different attempt at the same idea: how to create a digital world whose ownership records did not depend on a company maintaining them.

The first two versions could be wrapped into modern ERC-721 tokens. The later two remained constrained by the tx.origin problem, so they continued to require Etheria's own exchange. The worlds were related, but the code kept their histories distinct.

The unfinished world

Etheria's story did not end with its rediscovery. The project continues to exist, and its modern tools allow new structures to be created on the old maps. The early contracts remain available for examination, carrying both their successful experiments and their mistakes.

Even the failed drafts have acquired a second life as historical evidence. The abandoned v0.6 contract, in particular, illustrates how a world can remain technically unfinished long after its creator has moved on.

According to the X thread by @0xzxch supplied with this research, the contract's terrain and initialisation functions remained callable even though its ownership-transfer mechanism was broken. Only the first row had been initialised in 2015. In October 2026, the researcher reportedly used AI-assisted work to design a 33-by-33 elevation grid shaped like the United States, initialise the remaining tiles and assign them to an address derived from the original deployer's address. Because the destination address is not controlled by anyone and transfers remain broken, the result was effectively sealed.

The thread's transaction-level details could not be independently verified for this account, so the episode should be checked against the original on-chain records before publication. If the report is accurate, a researcher more than a decade later completed the terrain of a world abandoned in 2015 and then made its ownership permanently inaccessible.

There is a certain symmetry to it. Etheria began with the ambition to build a world that nobody could switch off. The same quality that allowed the successful versions to survive also preserved the mistakes in the failed ones.

In a conventional game, an abandoned experiment might disappear with a deleted database. On Ethereum, it can remain available for anyone to inspect, provided they know where to look and can work out what the code means.

That does not make every old contract a successful NFT project, or every digital parcel a valuable collectible. Etheria's place in the history of NFTs rests on something more specific. By October 2015, it had demonstrated that a finite world of individually ownable, transferable digital land could exist directly on Ethereum, with its state maintained by the blockchain rather than a central game server.

The project was quiet for years, but its contracts did not forget what they were built to do. In 2021, the world was rediscovered. In 2022, its four viable versions were brought together. In 2026, researchers were still investigating the abandoned experiments that preceded them.

Etheria's most interesting feature may not be its hexagonal islands or virtual buildings. It may be that the world has outlived the assumptions, tools and market conditions of the people who first made it.

The island is still there. Every so often, somebody finds another part of it.