<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by NFT Chain on Medium]]></title>
        <description><![CDATA[Stories by NFT Chain on Medium]]></description>
        <link>https://medium.com/@nftchain_com?source=rss-d41e3a9c441b------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*ggqF3ZR8TCYtCALC.jpg</url>
            <title>Stories by NFT Chain on Medium</title>
            <link>https://medium.com/@nftchain_com?source=rss-d41e3a9c441b------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 00:56:20 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@nftchain_com/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[NFT Chain — Mirror Protocol]]></title>
            <link>https://medium.com/@nftchain_com/nft-chain-mirror-protocol-d581de87b622?source=rss-d41e3a9c441b------2</link>
            <guid isPermaLink="false">https://medium.com/p/d581de87b622</guid>
            <category><![CDATA[blockchain-startup]]></category>
            <category><![CDATA[nft-collectibles]]></category>
            <category><![CDATA[nft]]></category>
            <category><![CDATA[blockchain-development]]></category>
            <category><![CDATA[blockchain-technology]]></category>
            <dc:creator><![CDATA[NFT Chain]]></dc:creator>
            <pubDate>Mon, 20 Feb 2023 14:57:14 GMT</pubDate>
            <atom:updated>2023-02-20T14:57:14.523Z</atom:updated>
            <content:encoded><![CDATA[<h3>NFT Chain — Mirror Protocol</h3><figure><img alt="NFT Chain — Mirror Protocol" src="https://cdn-images-1.medium.com/max/1024/1*N30CwYXhT9Slhd-acAHbbw.jpeg" /></figure><h3>Introduction</h3><p>An NFT is usually understood to represent “ownership” over some “item” which is described by the NFT. As an example, for a typical ERC-721 NFT the “item” would be whatever is resolved by the NFTs <em>tokenURI</em> URI — e.g. an image, and the “ownership” might be the IP rights to that image.</p><p>Exactly how that “ownership” is interpreted and what it represents can of course vary widely.<br>From IP rights, to a share of revenue from artwork, on or off-chain club membership, music and financial positions, what makes NFTs unique is their flexibility to represent provable ownership to some underlying content, resource or access.</p><p>Most generally NFTs have metadata — data which describes the ownership rights and item details of the NFT. Metadata is the life blood of an NFT — it entirely describes any value that the NFT might have, and without this, or with inaccurate metadata, the NFT is likely worthless.</p><p>Whilst NFT metadata might be fixed (therefore enshrining the details of the underlying “item” forever) there are lots of use-cases where the NFTs creator will want to update the items metadata, for example to evolve a piece of art or add new features or equipement to an in-game<br>item.</p><p>We want NFTs to be able to move freely through the market, being bought and sold, lent or borrowed, whilst ensuring there is always exactly one source of truth for the the critical and potentially fluid underlying metadata of the NFT.</p><h3>NFT bridges</h3><p>There are many chains that allow the issuance and trading of NFTs. Some of these chains are more established with larger liquid NFT markets and marketplaces, whilst others provide for more innovative NFT features while potentially lacking the mature liquid markets of other chains.</p><p>Bridges allow assets to be moved between different chains, usually operating under different consensus models and with different virtual machines for smart contracts or native chain logic. Bridges for fungible assets (e.g. ERC20s or native tokens) are now common place. Since these tokens are fungible they carry no individual metadata beyond the ticker itself, and so can be easily replicated on multiple chains. A typical bridge will simply burn some quantity of the bridged token on the source chain, and mint an equivalent amount on the destination chain. How the chain verifies deposits on the source chain can vary from centralised multisigs to proof based systems that employ on-chain light clients to trustlessly verify deposits.</p><p>Bridges for NFTs are more rare. Typically they involve copying the metadata from the source chain to the destination chain which can muddy the waters around where the actual source-of-truth for the NFTs metadata is, and potentially allow the copied NFTs metadata to diverge from the original.</p><p>Additionally many bridges operate through centralised mechanisms, for example a multisig, with exploits being common as bridges act as large honey pots to hackers. Centralised controls on a bridge can lead to exploits, frozen assets, mis-assigned assets and inconsistent metadata amongst other issues.</p><h3>Mirrored NFT Protocol</h3><p>Mirrored NFTs are a novel way to address the above concerns with current bridges. They allow NFTs to have metadata routed in a single chain, whilst being liquid on multiple chains.</p><p>In a nutshell mirrored NFTs work by creating a *mirror* of the NFT on a different chain, with that mirrored NFT always reflecting the metadata associated with the NFT on the source chain. This is in contrast to most bridges that take a copy of metadata onto the destination chain, which can lead to mismatched metadata.</p><p>In addition to ensuring there is always consistent metadata for every NFT, the Mirrored NFT Protocol has many other benefits.</p><h4>Consistent Pricing and Simplicity</h4><p>It avoids the operators of the bridge having to directly submit transactions on the destination chain — a source of complexity and cost volatility for bridge operators and their users.</p><h4>Grouping of NFTs</h4><p>Mirrored NFTs have a one to many relationship with NFTs on the source chain. This means that a single mirrored NFT can represent ownership in a basket of NFTs on the source chain, providing flexible mechanisms for NFTs to be bought, sold and leveraged.</p><h4>Security</h4><p>For the Mirrored NFT Protocol to operate correctly, it is critical that validators on the source chain correctly observe any <em>destroy</em> transactions on the destination chain, and correctly reflect these on the source chain. Note that this part of the protocol is the only piece that requires the source chain validators to observe the destination (non-native) chain, and that no transactions are required to be submitted on the destination chain by the validators.</p><p>In particular we are interested in censorship resistance (all <em>destroy</em> transactions must be reflected on the source chain) and security (only destroy transactions that are finalised in the destination chain are reflected on the source chain).</p><p>There are quite a few ways to implement solutions to this problem, but at a high-level there are really two possible approaches — the validators act as an oracle for events on the destination chain, or the source chain implements some type of light client of the destination chain.</p><p>Given the current state of blockchain technology, the most feasible approach today is for validators to act as oracles, with appropriate incentives to behave correctly. However we anticipate that as the protocols for both the source and destination chains evolve, the more trustless approach of implementing a light client (of the destination chain on the source chain) will become more tractable.</p><p><strong>Validators as Oracles</strong></p><p>Each validator on the source chain is required to be able to observe events on the destination chain. They can do this by running a destination chain client or via a third party RPC service, such as Infura. The former is more trustless (no reliance from the validator on a third party service) but also more expensive since the validator must maintain a client for the destination chain. Provided that there is some diversity across different validators in their approach (i.e. not all validators use Infura, but that they use a mix of full or light destination chain clients and / or different RPC endpoint services) then this should be reasonably safe in practice.</p><p>This diversity can be encouraged by developing node software that makes it easy to use different destination RPC endpoints as well as embedded clients of the destination chain. Penalties can also be constructed such that if there are any reporting errors from validators, those penalties are magnified if a large number of validators make the same mistake, therefore encouraging individual validators towards more diverse approaches so as to avoid these heavy penalties.</p><p>Rather than requiring a simple majority (51%) of validators on the source chain to agree that a <em>destroy </em>event has occurred on the destination chain, we enforce a much higher limit of 95%. In other words a destroy event only un-mirrors the NFT once 95% of validators on the source chain have reported the event. This means that in order to report a fraudulent destroy event, an attacker must gain control over at least 95% of the validators on the source chain. When we say 95% of the source chain validators must report the destroy event we actually mean that validators with collectively 95% of the source chains staked tokens delegated to them, must report this event. Since reporting fraudulent destroy events would erode the value of the source chain token, the attacker must be willing to pay the validators at least enough to compensate them for a loss of 95% of the staked source chain tokens.</p><p>There are some different scenarios to analyse with this approach — in both cases we rely on governance on the source chain to resolve the dispute.</p><p><em>1 . an incorrect destroy event is fraudulently reported by less than 95% of the source chain validators</em></p><p>Source chain token holders can vote to confirm that the reported destroy event (which is never processed since it didnʼt reach 95% consensus) was fraudulent (never occured on the destination chain). In such a scenario the fraudulent validators will also be slashed, with the slashing penalty proportional to percentage of fraudulent validators (so the higher the percentage of validators that fraudulently reported a non-existent destroy event, the higher the slash).</p><p><em>2. a correct destroy event is not reported by 95% of the source chain validators</em></p><p>In this scenario the destroy event will not be immediately included in the source chain, meaning that ownership of the NFT in question remains linked to the ownership of the mirrored NFT on the destination chain. Source chain token holders can vote to confirm that the destroy event did in fact occur on the destination chain, and all validators who did not correctly report it will be slashed. Again the size of the slash will be proportional to the percentage of validators that did not report it.</p><p>In both cases there is a significant incentive to correctly report destroy events for validators. Whilst slashing may not punish the validators directly (since the tokens being slashed will be those delegated to the validators, rather than necessarily owned by the validators themselves) it provides a significant incentive for token delegators to delegate to validators that they expect to act honestly.</p><p>Note that since slashing is proportional to the percentage of incorrect reports from validators, there is an incentive for validators to not all use the same approach to observe the destination chain. For example if a large percentage of validators use Infura, and Infura (either through fraud or a bug) incorrectly reflects the destination chain state, they will be penalised very heavily since a large percentage of validators will incorrectly report.</p><p>The slashed stake can either be burned, sent to the chains treasury, or distributed to those validators that did report correctly for the specific destroy event, providing a further incentive for validators to be in the correct reporting set, rather than the incorrect reporting set.</p><p><strong>Light Clients</strong></p><p>Whilst the above approach provides some crypto-economic guarantees about correct reporting, it still relies on the rational economic behaviour of an attacker.</p><p>A longer term approach would be to instead embed a destination chain light-client on the source chain. This would theoretically allow you to prove that an event has occurred on the destination chain, without the need for these economic incentives.</p><p>Whilst Ethereum is moving towards making this approach more feasible via its sync committees (a smaller randomly chosen set of validators that are required to assert the finality of blocks) so far there are still technical issues to be resolved (a great discussion on these can be seen at <a href="https://prestwich.substack.com/p/altair)">https://prestwich.substack.com/p/altair)</a>.</p><p>It is however possible to envision in the future a system by which block headers of the destination chain are reported and recorded on a Polkadot parachain (for example a common good parachain such as Statemint) allowing all other parachains to rely Merkle Proofs of destination chain transactions, which can be quickly verified on-chain within the source chain.</p><p>With this approach (a canonical representation of destination block headers on the source chain) anyone could report a <em>destroy</em> transaction by simply providing the corresponding Merkle Proof of the transaction for the relevant destination block header recorded on the source chain.</p><h3>How Does It Work?</h3><p>The Mirrored NFT Protocol operates by allowing anyone to mint a new mirrored NFT on the destination chain. Each Mirrored NFT is created with a unique identifier, and initially will not be associated with any NFT on the source chain.</p><p>Users can then take this unique identifier and link it to an existing NFT on the source chain. From that point onwards the ownership of the NFT on the source chain is derived from the ownership of the mirrored NFT on the destination chain.</p><p>To un-mirror the NFT they simply destroy the mirrored NFT on the destination chain, and receive back ownership of the original NFT on the source chain!</p><p>This approach avoids trying to copy the metadata of a bridged NFT, and instead uses simple placeholder or mirrored NFTs on the destination chain to capture only the ownership of the NFT, not its underlying content.</p><h4>Detailed Mechanism</h4><p><strong>Source Blockchain</strong></p><ul><li>users can create, mint, transfer NFTs</li><li>NFT creators can update NFT metadata as needed</li><li>each NFT on the source chain has a unique <em>sourceId</em> (which can not be modified)</li><li>the owner of a particular NFT can submit a transaction on the source chain to link their NFT (identified via its <em>sourceId</em> ) to an NFT on the destination chain (identified by its <em>destinationId</em> )</li><li>the source chain stores a <em>NFTMirror</em> table with columns <em>sourceId</em> and <em>destinationId</em> which is initially empty</li></ul><p><strong>Destination blockchain</strong></p><ul><li>users can mint new mirror NFTs which are assigned a unique <em>destinationId</em> on creation (which can not be modified)</li><li>the owner of a mirror NFT (on the destination blockchain) can submit a transaction on the destination blockchain to destroy their mirror NFT and transfer ownership back to a specified address on the source blockchain</li></ul><p><strong>Mirroring NFTs</strong></p><ul><li>Alice owns an NFT created on the source blockchain with a <em>sourceId</em> of <em>X</em></li><li>Alice mints a mirror NFT on the destination blockchain which is assigned a <em>destinationId</em> of <em>Y</em></li><li>Alice submits a transaction on the source blockchain to link <em>X</em> to <em>Y</em></li><li>Source blockchain validators check:<br>- does Alice own the NFT with a <em>sourceId</em> of <em>X</em> on the source blockchain</li><li>If the above checks pass, then the source blockchain validators update state as follows:<br>- set the owner of the NFT with unique sourceId of <em>X</em> to be some unreachable account to indicate that the NFT is locked and has no defined owner on the source blockchain<br>- add an entry to the <em>NFTMirror</em> table (with columns [<em>sourceId</em>, <em>destinationId</em>] ) for [<em>X</em>, <em>Y</em>]</li><li>At this point the ownership of the NFT with <em>sourceId</em> of <em>X</em> is effectively recorded via the owner of the mirror NFT on the destination blockchain.</li></ul><p><strong>Unmirroring NFTs</strong></p><ul><li>Bob owns the mirror NFT on the destination blockchain with a unique <em>destinationId</em> of <em>Z</em></li><li>Bob submits a transaction on the destination blockchain to destroy the mirror NFT with <em>desintationId</em> of <em>Z</em> and transfer ownership to address K on the source blockchain</li><li>On the destination chain the ownership of the mirror NFT with <em>destinationId</em> of <em>Z</em> is set to some unreachable account of the NFT is destroyed.</li><li>On the destination chain an event recording this action is emitted. Source blockchain validators check:<br>- observe that an event as above was emitted<br>- check that the <em>NFTMirror</em> table includes at least one row where <em>destinationId</em> is <em>Z</em> remove any matching rows in the <em>NFTMirror</em> table that corresponds to a <em>destinationId</em> of <em>Z</em><br>- change the owner of the corresponding NFTs with <em>sourceId</em> from the removed rows above to the address <em>K</em></li></ul><p><strong>NFT Baskets</strong></p><p>A mirror NFT on the destination blockchain can represent multiple NFTs on the source blockchain.</p><p>For example suppose we have a mirror NFT with a unique <em>destinationId</em> of <em>X</em> and two NFTs on the source blockchain with unique <em>sourceId</em>’s of <em>Y</em> and <em>Z</em> then the <em>NFTMirror</em> table count include the rows:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/358/1*fX9NJ6erDTNVQRbviVS7XA.png" /></figure><p>This implies that ownership of the mirror NFT with a <em>destinationId</em> of <em>X</em> on the destination blockchain represents ownership of both the NFTs with <em>sourceId</em>’s <em>Y</em> and <em>Z</em> on the source blockchain.</p><p>Anyone can associate a source blockchain NFT with a mirror NFT, even if they don’t own the mirror NFT, but that there is no advantage to doing so since the mirrored NFTs can be destroyed and ownership decomposed on the source blockchain.</p><p><strong>Metadata</strong></p><p>NFT metadata (i.e. the data that indicates to what the ownership of the NFT pertains) is recorded only on the source blockchain.</p><p>This means that there is no requirement for mirror NFTs to record a separate independent copy of this metadata, and removes the risk that these become inconsistent.</p><p>To determine the metadata of a mirror NFT you:</p><ul><li>determine the <em>destinationId</em> of the mirror NFT</li><li>retrieve the corresponding rows of the <em>NFTMirror</em> table that reference this specific <em>destinationId</em></li><li>for each row, the metadata can be dereferenced via the corresponding <em>sourceId</em></li><li>since a single mirror NFT can represent multiple NFTs on the source blockchain, metadata for multiple NFTs can be composed as appropriate</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d581de87b622" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>