Top Stories of the Week - 5/19
This week we have two great new reports. Gabe parker wrote about the state of liquidity and infrastructure competition in NFT markets. Charles yu wrote about curve’s new stablecoin, a major development for on-chain defi. Subscribe here and receive Galaxy's Weekly Top Stories, and more, directly to your inbox.
Crypto Travel Rule Advances in Europe
EU Council adopts crypto travel rule. The European Council, an important strategic body within Europe’s executive branch of government (along with the European Commission), has voted to adopt new AML rules for cryptoassets and cryptoasset service providers operating in the Eurozone. The rules amount to the “travel rule,” the financial identification rule promulgated globally by the Financial Action Task Force (FATF) that requires financial intermediaries to collect personal information for users and include that information along with money transfers to recipient organizations.
The adoption by the EU Council, which itself is comprised of members of the heads of state of EU members, follows the conclusion of Trilogue negotiations that resulted in a provisional agreement in June 2022. Trilogue negotiations occur between the European Commission, European Council, and European Parliament, and are somewhat akin to House-Senate conference negotiations in the US (in that they are a process to reconcile similar but not-identical legislative proposals). According to a press release, the adoption by the EU Council is the final legislative step for this crypto travel rule.
The rule itself would require “crypto asset service providers” to collect personal information for users of their services and, when users request transfers of crypto assets to another institution, the sending provider is required to transmit the identifying information to the receiving institution. For a detailed explanation of how this so-called travel rule applies to crypto, read this Galaxy Research report from Charles Yu. As adopted, this EU version appears to follow FATF’s proposed travel rule very closely – it would not apply to transactions that do not involve a “crypto asset service provider,” such as “person-to-person [P2P] transfers.” Read more about FATF’s definitions in this Galaxy Research report.
Jurisdictions are beginning to bring crypto into the regulatory fold around the world, though not so much in the United States. The formalization of the travel rule for crypto in Europe follows the zone’s recent advancement of MiCA, the comprehensive framework for digital assets, just several weeks ago. As we’ve repeatedly written, jurisdictions outside the United States are largely making real headway on progressive regulatory frameworks that provides clarity and space for crypto firms to operate and innovate.
For this specific item, some misconceptions were spread on Twitter that we want to correct. A major crypto “squawk” account, Watcher Guru, wrote to 1.9m followers “JUST IN: European Union passes law requiring identification for all #crypto transactions.” This is a pretty severe over-statement of what this travel rule requires. As written, and at least for now, the requirement applies only to service providers, and the circular provided by the EU Council specifically says “the regulation should not apply to person-to-person transfers of crypto assets conducted without the involvement of a crypto asset service provider.” This is a major distinction – any jurisdiction that tries to apply the travel rule to interactions between self-hosted wallets will essentially create an unenforceable prohibition on most crypto usage. Global, permissionless blockchains are – and always will be – accessible directly using free, open-source software. That’s reality policymakers need to understand. Major advances for law enforcement, consumer protection, and business integrity can be achieved by properly regulating businesses that operate in the space, but attempting to regulate the open, decentralized web will amount to an attempt at creating “Great Firewalls” and will never be successful in the long run. For now, in this application of AML rules, Europe seems to be getting it right. But there has been some movement in future AML rules for crypto that may begin to stray into the realm of network regulation (see this piece we wrote several weeks ago). That’s something to always watch for. -AT
A New Scalable Way to Issue Tokens on Bitcoin
Lightning Labs, a major builder of tools for Bitcoin’s Lightning Network, releases a new protocol to issue assets on the Lightning network. The Taproot Asset Protocol (formally TARO) enables developers to issue, send, receive, and discover assets built on Bitcoin. Over the years, there have been many proposals and even protocols to issue assets on bitcoin, from colored coins to Omni Layer to Counterparty, let alone assets on other layers like RGB, Liquid, or even Stacks. For several years, Lightning Labs has proposed and worked on a way to issue assets on the Lightning Network, and this release on testnet is a major milestone.
Taproot Assets create tokens on bitcoin through inputting arbitrary data into the taproot script. Any data that goes with the asset is carried on the asset itself, living within a Taproot output using a taptree (a merkle tree). The assets are initially issued on-chain through a standard taproot transaction. The assets functionality is enabled through a refined version of partially signed bitcoin transactions (PSBTs), which are used to trade Ordinals and BRC-20s called virtual partially signed bitcoin transactions (vPSBTs). This mechanism is a way to trade assets off-chain in a peer-to-peer environment. Although Taproot Assets are issued on bitcoin's native chain, Lightning Labs is specifically designing Taproot Assets to be compatible with the Lightning Network, Bitcoin's top layer 2 protocol. However, the final state after transacting Taproot Assets on Lightning will commit to Bitcoin's native chain to ensure finality and no double spending.
Lightning Labs will introduce Taproot Assets into Bitcoin through a set of BIPs and BLIPs (Lightning network improvement proposals). The BIPs will be informational BIPs – importantly, they do not propose any changes to the consensus layer. The Taproot Assets Protocol is now available on testnet and mainnet support will be added soon after the proposed BIPs and BLIPs are recognized.
The recent mania around BRC-20 tokens, an asset standard utilizing ordinals that we wrote about two weeks ago, shows there is clear demand for issuing assets on Bitcoin. But as we wrote, BRC-20 is a pretty awful standard in a lot of ways, including it’s lack of scalability, intentionally inefficient code design, and inability to granularly handle tokens. Quite fortuitious then, that TAP (formerly Taro) hits testnet this week.
This is clearly a win for the Lightning Network and the potential for its wider adoption. While Lightning has clearly found some product-market-fit for small payments, its use is still rather minimal given that it’s been around for 5+ years. Increased activity on Lightning will also make routing nodes more profitable, which help create positive feedback loop that improves liquidity for all users.
Broadly, with ordinals and inscriptions, renewed interest (and advancements in development) for asset issuance, growing support for an upgrade like OP_CTV or OP_VAULT to add more features to bitcoin, excitement about the possibilities of other types of L2s being built on bitcoin (such as rollups), and much more... all show that the diversity of bitcoin builder interest is at its highest level in years. All of these things are more than just payments, or secure custody, or the other foundational things people have built over the last several years. Instead, any combination of these developments can open entirely new design spaces for building on Bitcoin, and that remains exciting. -AT/GP/GG
Inactivity Leak Triggered for the First Time on Ethereum After Prolonged Finality Delays
From Thursday, May 11 to Friday, May 12, the Ethereum network experienced delays in reaching finality. Delays to finality mean assurances about the immutability of transactions on Ethereum were not guaranteed for an extended period of time. This primarily impacted exchanges and Layer-2 rollups that rely on finality assurances for the operation of their services. Most end-users and decentralized applications (dapps) were not impacted by last week’s finality delays. Transactions could still be processed, albeit with lower guarantees of immutability once included in a block.
The root cause of the issue stemmed from a high number of missed blocks. During the first finality delay on Thursday, 47 blocks were missed, while the second delay resulted in 149 blocks missing. A low number of blocks means less attestations from validators voting on valid blocks are being processed by the network. A low number of attestations and a lack of 2/3 consensus among the votes prevents Ethereum from reaching finality. If finality cannot be reached for more than 5 epochs (an epoch is a series of 32 slots or 32 opportunities for block proposals), an inactivity leak is triggered where the staked ETH of offline validators gets penalized by an increasing magnitude over time. On Friday, finality could not be reached for 9 epochs, that is one hour. On average, offline validators lost 0.00015 ETH, less than $1, during the inactivity leak.
The high number of missed blocks was caused by certain validator nodes running Prysm and Teku client software. An unusual number of old attestations were propagated on Ethereum by Lighthouse nodes last week which caused Prysm and Teku nodes to become bogged down with verifying these old attestations and replaying chain state. The nodes suffered from resource exhaustion and were unable to fulfill other requests such as the processing of new attestations and blocks in a timely manner. The Prysm and Teku client teams have since released fixes that rewire validator nodes to drop old attestations that do not contribute to the fork choice rule instead of expending work to verify them. In a post-mortem of last week’s finality delay events, the Prysm client team wrote: “Client diversity helped the chain recover with some clients still being able to propose blocks and create attestations. Lighthouse dropped the problematic attestations and stayed alive. No manual intervention or emergency release was required to resolve the immediate finality issue.”
The finality delay issues experienced by Ethereum were a non-event that did not impact the user experience in major ways or last long enough to require manual intervention from client teams to resolve. Ethereum did not experience an outage and did not stop processing transactions through finality delays. This is because part of Ethereum’s fork choice rule, LMD-GHOST, favors chain liveness over safety during times of high network stress. For the first time ever, this unique component of Ethereum’s consensus protocol that continues to add new blocks during delayed finality and eventually kicks in an inactivity penalty to restore network security was tested in production and confirmed to be working.
As minimal of an impact last week’s events had on Ethereum, even the smallest hiccup in chain operations for a blockchain of Ethereum’s size, supporting thousands of dapps and millions of dollars' worth of value, is significant and worth scrutinizing. In the Prysm client team’s post-mortem, the team details one of the lessons learned from the finality delay issues, outside of the obvious fix to handling old attestations, was the need for more realistic test networks that match the size and activity of Ethereum mainnet. Since Shanghai, there has been an influx of new validators and deposits of staked ETH. As of May 18, the total number of active validators is above 575,000 and the total amount of ETH staked is above 18.4mn. Goerli, the largest Ethereum testnet, only supports roughly 450,000 validators and 14.4mn ETH staked. Larger Ethereum testnets are needed for more accurately replicating mainnet characteristics and surfacing unexpected events like the ones triggered on mainnet last week.
It is also important to keep in mind that Ethereum due to its recent transition from a proof-of-work (PoW) to a proof-of-stake (PoS) consensus protocol is a fairly new blockchain that has yet to battle test the resiliency and robustness of its security and liveness in production. While Ethereum was launched in 2015, the LMD-GHOST and Casper FFG PoS consensus protocol, in short called Gasper, was only technically launched in 2020 with the activation of the Beacon Chain. And up until April 2023, the Beacon Chain was not fully functional, having disabled staked ETH withdrawals. Therefore, as a blockchain protocol, Ethereum is nascent. Unexpected issues in client implementations of Gasper should not be all that surprsing in the early days following Shanghai, Ethereum’s full transition to PoS. The longer Ethereum’s Gasper protocol withstands volatile market cycles, expansions and contractions in the active validator set size, and fluctuations in on-chain activity, the more it becomes hardened and proven as a reliable and secure blockchain consensus protocol.-CK