Why Blockchain Accounts Matter 3/3: From AA to DESOC
The second part of this series was about the current problems of blockchain accounts and how contract-based accounts in the form of Universal Profiles help solve them. Now I want to outline what’s happening on the abstraction horizon and how such solutions are needed for decentralized societies to form.
Abstraction of EVM Accounts
The presented identity problems in the industry are not new. Since EOAs are a limitation for the masses, Ethereum teams themselves have also been researching concepts and standardizations for abstraction for years.
Abstraction in a brought manner is not just one single standardization. I want to give you a quick overview to get an idea: It evolved from EIP-86 to abstract signature verification and outsourcing nonce checking to EIP-725 as a new user’s pivot point, including recovery, rotating keys, and attachable storage. Mid-2018, EIP-1271 enabled signature validation on smart contracts and their functionalities. EIP-2938 later introduced a new protocol operation for abstracted transactions to pay transactions via a token. Afterward, in 2020, EIP-3074 proposed another protocol operation allowing users to delegate control of their EOA to smart contracts or even revoke rights by combining it with EIP-5003.
Solutions using EOA-controlled smart contracts are still considered state-of-the-art as protocol concepts were not adopted. Due to the architecture, user accounts cannot self-sign and are often seen as second-class citizens in a network for EOAs. The standardization of EIP-4337 has now been bringing together many individual concepts from recent years since the end of 2021. With the standardization, not only is the account itself decoupled from the signing keys, making them exchangeable but an attempt is made to integrate them natively into the network’s protocol to compensate for imbalances. The issue is fundamentally one of the entry points for new users. For existing networks, new account functionality must provide full backward compatibility. Therefore, the long-term plan is to remove EOAs from the protocol and emulate them for existing accounts with the same functionality afterward. In the long run, each account becomes an abstract account directly. Currently, contract accounts with EOAs are associated with specific circumstances. Alongside optimization, the main reason for the protocol change is that the addresses that EOAs have are no longer needed if each account already has its smart contract address- only the keys would have to rotate for security. However, the problem of the second user class does not arise if each user is given an abstracted account directly upon entry — as outlined in the LUKSO network of Universal Profiles.
For abstract accounts, according to EIP-4337, there will also be a new class of transactions, so-called user operations, which, like regular transactions, will have a separate collection pool in the network protocol. The split has the consequence that user actions do not heavily impact traditional blocks of the chain and can be worked more efficiently. Currently, each user operation still counts as a separate transaction. With completely abstracted accounts, these operations could be bundled natively using new signatures. Additional bundling would save costs and increase scalability.
The concept and functions are already in place. Since March 2023, the smart contract for validating such operations has been implemented on the Ethereum network. Since integration into the protocol will take some time due to its complexity, this bundling has only worked with relay servers outside the protocol, so-called bundlers. Prototypes like the Soul Wallet and Argent X attempt to integrate these into extensions for the first time. The development will initially spread to lower-level networks (L2) to experiment with low costs and app integration.
The functionality of ERC4337 accounts includes transaction bundling and payment with arbitrary tokens instead of the blockchain’s coins, signing multiple transactions of different keys in one go, and gaining more security and better onboarding by exchanging the keys behind an account.
The question would be how AA, as described in ERC4337 and Universal Profiles, can go along or how they differ. AA, in this regard, standardizes how the body of future blockchain accounts should be set up and what transaction flow they follow to improve security and scalability. Beyond that frame, extended identity schemes could be created on top, as the contract enables the execution of code. On this end, Universal Profiles are already extensible accounts that can be full of information, have a permission control system, can be notified on transfers to create consent, and even come with many features for assets and vaults. AA is the right direction for the blockchain space but does not compete with contract systems built today.
Current smart contract ecosystems, like Universal Profiles, can implement these methods for abstraction later on instead of still using regular EOAs. This would count as further optimization. The same applies to relay services, which can support the bundling of network actions later on. Here, both projects relieve each other by finding new standardizations on different ends.
The diagram clearly shows the path taken by account systems. Currently, EVM users start with a pure protocol and key-based account. Mainstream functionality can then be implemented with smart contracts. Further optimization and linking with transaction bundlers at the application level increase scalability. The ultimate goal is that all operations from abstracted accounts are entirely embedded in the protocol without sacrificing functionality.
Digital Societies
Through projects such as the CryptoPunks or Bored Ape Yacht Club, social concepts or groupings of people with a blockchain connection were brought out into the open for the first time. People built reputations through profile pictures and joined forces. However, all the traction has been on Web2 because there is still no mature integration of the social world into blockchain applications. DAOs can only set their governance, proposals, and tokens on the chain. Besides that, Discord is mainly used for roles and community building, Reddit for sharing posts and news, and sub-chats are formed on Telegram. Community members have different profiles, rights, and identities on each platform. Today’s concepts are pure chaos regarding participation and reputation management. Digital assets are another example: Isn’t the creator the most critical person for determining authenticity and building value? If you look on-chain, the EVM explorer states that the initiator is a randomized token address that may have a temporary domain to it. Besides cryptocurrency, the blockchain account has no identity, at least not on the ledger. Identity only comes from the association with social media on Web2. Further, based on the ERC-721 specification, most NFTs aren’t properly associated with the creators and are just minted from the zero address to indicate a fresh creation.
Abstractions and standards will pave the way for future decentralized societies and solve those problems. They give blockchain users much more user-friendliness and faces as we know them from today’s Social Media platforms. Concepts that have been around for a long time, such as organizations within decentralized networks, will receive a new boost.
In this regard, non-transferable, soul-bound tokens (SBTs) are associated with decentralized social structures. Once issued, they belong to a specific account and cannot typically be transferred or sold to a new address. Exactly these SBTs need recovery and management systems for identities. They could imitate special, inalienable certificates, achievements, proofs of presence, or interactions of a human or fictitious personality. In other words, precisely what constitutes identity in the first place.
Compared to VCs covered in previous parts, SBTs are about things regularly shared publicly or on the Internet. They need the blockchain data layer as they are a crypto-native approach and imagined as assets locked in on an identity account within social constructs like DAOs. On the other hand, VCs are ideal for private claims and sensitive data that users want to use in digital worlds, which do not need to be docked to the blockchain layer per se.
As they are on-chain assets, it must be ensured that there is a valid token and social structure. If bound goods had to be reissued every time a key was lost or updated, this would result in immense additional expense. The complexity of recovery or carryover for locked assets is also why current issuers, POAP being the most prominent representative, opt for transferable certificates, although the sale is not desired.
There should also be a consensus between the issuing and receiving parties. Otherwise, unintentionally credible identities would be at risk of spam without a way to remove it from their accounts. Selling entire accounts is also dangerous: objects would linger on the exact identity address even though the owner changes. SBTs only unfold their power in trustworthy social networks and should be linked to specific requirements. If SBTs were transferred, people would likely recognize it and outlaw the seller’s reputation if he “sold its soul.” In the long term, ways must be found to restrict bots from forming social circles and fake relationships.
For decentralized societies, however, there is a chicken-and-egg problem when using EOAs: Regular wallets need SBTs to become valid identities. Yet, these cannot be issued if one quickly loses access to them. On the other hand, EOAs need authentic social group networks for identity recovery, which can be implemented exclusively with SBTs. An EOA is just the sum of its particles: Since they cannot carry any information, the social space is only about what the address has acquired. What is a soul-bound token without a connected soul? Relatively meaningless.
Account abstraction, especially the Universal Profiles ecosystem, solves the chicken-and-egg problem. It provides a static address as an identity, delivers natively upgradable security, and links the account to public user data that can generate reputation: It enables data to be stored and shared more securely from the outset. Thanks to the configurable rights interface, variable recovery methods are now conceivable. Here, the emphasis is no longer on what a user accumulates but on who they are and how they present themselves to the outside world. Universal Profiles solve the identity problem and make up the ideal soul framework.
Concepts for community-based backups are possible results. In this way, “lost souls” could be helped back on their feet without requiring action from the user. Vaults or rights could be cleverly combined to embed backups in social structures. SBTs can strengthen the dynamics of a DAO and blur the boundaries between Web2 and Web3.
Outlook on Blockchain Socials
Currently, many prototypes for protocols and social media applications are arousing interest and showing what integration into blockchain technology can look like. Pioneers are the LENS Protocol with many apps based on a decentralized graph and Farcaster. Both have great properties: self-sovereign profiles, individual feed algorithms, open-source code, censorship-resistant base layer, and no intermediaries. However, the limitations of the widespread EOAs become apparent in all services — since user information has to be outsourced to tokens or external registers, and recovery is either not included or super limited. Every application would also have niche profile information instead of a generalized account.
The emerging nexus of abstract profiles and intertwined digital goods is fundamental to advancing Internet technology and will steadily insert itself into society over the next decade. However, demonstrated economies will only spread within separate or subordinate networks due to the higher computational power of using smart contracts as the foundation of identities and the scaling problem of reasonably decentralized networks. Nevertheless, this generally benefits the integration of new user groups. Upon entering the network, new functions can thus be established without an existing user group reducing the range of possible functionality. Demonstrated blockchain standards can only achieve lasting and appropriate comfort if the entire ecosystem relies on them.
Blockchain accounts, social environments, and claims will represent a groundbreaking step forward regarding data protection and regulation. A Web3 data economy is needed to overcome current submissive ways of prancing around digital worlds. To this end, the foundation for public profiles, DIDs, and abstraction is already in place. However, while the possibilities sound promising, public accounts and ledgers should be used cautiously and anonymized where appropriate. They are not a panacea. SBTs have great potential to balance in-group dynamics, yet there is a risk that they could also be used to marginalize or misuse social groups for attacks. An excess of SBTs could make users glassy and put them at risk. Blockchain-based systems used for social media are public by default, and so is the data within smart contracts. Unlike Web2, however, these created bindings cannot be undone in the decentralized space, even if the data behind them is deleted or protected by zero-knowledge proofs.
If managed improperly, multiple previously anonymous or pseudonymous accounts could quickly be linked in undesirable ways. Conversely, too many private SBTs and connections could lead to hidden communication channels. The dichotomy in such decentralized social circles does not rule out manipulative bubbles, which can hardly be contained due to the lack of central regulation. Therefore, it should be considered carefully whether SBTs are necessary and safe as on-chain tokens or whether signed claims such as VCs should be used instead. The same cautions apply to the allocation of rights. SBTs should not give direct access to rights in a DAO. Roles are also mapped within the organization’s smart contract, and administrators would have to manage two places- that might interfere with each other. What if someone leaves the DAO and still has rights written into the old SBT, or if roles are adjusted? It would be cumbersome to keep track. They are better suited for public memories or achievements- that might even state that you had certain roles at a certain time. Yet, signing in with an account should be the only key. The SBTs information can then separately be tracked to qualify for access and rewards.
Especially in the political context, the prevailing decentralization of accounts and new data-sharing schemes leads to generally more democracy on the Internet and in society. Judging by all the emergence of decentralized organizations, profiles, and related SBTs, despite their dangers, are precisely the right step with which more money-oriented mindsets of the blockchain industry can be transformed into a more social space. In conclusion, decentralized networks and societies not only give people back the power over their data and interactions but also put the focus on what matters: Individuals and their genuine, unique relationships- as I already mentioned once before on Medium.
Thanks for still reading human-generated text. Have a great day; I hope you learned new things ;)
