Your resource for web content, online publishing
and the distribution of digital products.
«  

May

  »
S M T W T F S
1
 
2
 
3
 
4
 
5
 
6
 
7
 
8
 
9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
27
 
28
 
29
 
30
 
31
 
 
 
 
 

Web3 Decentralization Debate Focuses on Wrong Questions

WEB3 advocates promise decentralization on an unprecedented scale. Excessive centralization impedes coordination and weakens freedom, democracy, and economic dynamism—decentralization should be a remedy. But the word itself is too vague to be a coherent end goal. Getting the job done requires the right decentralization, and we fear Web3 is on the wrong track so far.

Most of us focus primarily on the degree of decentralization rather than the type. The level of concern – do we want more or less decentralization, but leads Web3 advocates to mischaracterize the reality of existing centralization, and the possibility of pure decentralization. On the one hand, existing “centralized” systems are not as centralized as Web3 advocates often describe them. Traditional “banks” delegate many activities to local branches, and even central banks are often consortia. Architecturally, “centralized” are rarely so centralized in practice; they are usually spread out over different geographic areas, training large machine learning models in a distributed fashion.

On the other hand, many critics of Web3 have pointed out the extreme inefficiencies that accompany the proposed decentralized architecture, and the inevitable re-emergence of “hubs” (NFT platforms, currency exchanges, wallet providers) in Web3. Additionally, broadly targeting greater decentralization has important limitations and trade-offs. For example, technical decentralization in the narrow sense faces the conflict between censorship resistance and embedded values, which often end up in poor functionality or some centralized decision-making, as shown by content control on decentralized social networks .

Thus, in a functional system, there are (soft) limits to the degree of centralization and the feasibility of decentralization. Instead of arguing about whether the next generation of technology should be centralized or decentralized, we should ask how best to arrange the ideal decentralized model. Such debates need to articulate exactly what we want from decentralization.

We believe that the value of decentralization lies in truly empowering people to act decisively in their social context, while providing the necessary coordination mechanisms across contexts. This is in stark contrast to the current technological landscape, in which decision-making power over information, calculations, moderation, etc. is increasingly in the hands of authorities “away from” the relevant groups – for example, where platform content moderation processes attempt to Cross-community and cross-cultural, but basically fail on both counts.

Our view of decentralization is about coordination, which emphasizes problem solving through the union of “local” units that are brought together in the social context most relevant to current decision-making. This is not a new idea. Federalism in the United States, including local, state, and national governments, largely derives from this principle of subsidiarity, as does the establishment of open-source code repositories and wiki-like structures of information aggregation. The point is that these local units are composable – modular and interoperable, essentially “stacking” to a larger global scale – to enable decentralized systems to be efficiently resolved, a pattern we call composable local control.

Composable local control will distribute decisions, leveraging the core principles of markets and democracy. Those closest to the problem often have the most knowledge and the greatest interests, and by aggregating, combining, and filtering this knowledge, the best collective decisions can be made.

Auxiliary is a decentralized structure and type that enables composable local control. But Web3’s dominant trajectory is unlikely to materialize, and may even be contrary to subsidiarity. A permissionless blockchain is built as a distributed redundant ledger where storage and permissions are allocated by anonymous economic mechanisms and accessed through fungible, tradable resources such as computation and tokens. This architecture is optimized for a highly narrow set of problems and thus, by its very nature, cannot interface with the rich economic and social networks that actually need to be solved. This purely financial system has a good history of centralizing wealth, information and power, and the current Web3 ecosystem has taken these properties to the extreme. Therefore, redundant distributed ledgers are at odds with the benefits of affiliated networks and the form of decentralization we advocate.

We remain optimistic about the potential for Web3 and adjacent spaces to re-emerge as auxiliary. However, in order for this to happen, we must take steps to make Web3 a network rather than a ledger.

The “decentralization” that most typical crypto projects, like Bitcoin, currently achieve is what we call “distributed storage”: global, open access to a common, homogeneous set of data in many places , Consensus-based storage. Distributed storage depends on three factors.

1. Maximize the removal of data from the social context. (All interactions are boiled down to transactions recorded in the ledger, the external environment cannot be reflected in the technical architecture).

2. Aiming to achieve a generalized solution.

  • Rely on global consensus and redundant verification using alternative resource access.

Why might so many people be so obsessed with redundancy and universality? In theory, the purpose of redundancy is security against attacks. However, as we have seen, with recent supply chain challenges and the concentration of most Bitcoin mining in a small number of mining pools, market efficiency tends to concentrate activity in hyperscale centers, often very vulnerable to shocks and disruptions (e.g. local Covid lockdown policies), or located in a jurisdiction that may be vulnerable to geopolitical risk (e.g. Russia). Effective and safe redundancy requires intentionally compensating for this trend, choosing a diverse “hedging” of risk, not just the lowest cost provider. But achieving such hedging requires tracking geographic and network relationships that these purely financial systems ignore.

In stark contrast to these principles, the type of decentralization we believe is desirable, subsidiarity, focuses on:

  1. Make the data as close as possible to the created social context.
  2. 2. Link and integrate multiple solutions through coordinated alliances and interoperability mechanisms.
  3. Leverage and expand online and offline trust and institutional relationships.
  4. Perhaps the most famous one that was designed as an adjunct system from the start is the original “network”, the TCP/IP-based Internet, which was designed this way precisely for security and efficiency reasons, and can be said to be achieved unprecedented success. Recent and more freshmen examples include:

ActivityPub standard for social networks and related applications such as Mastodon.

A series of research projects from Ink and Switch, including the architecture of “local first” computing and the design of interoperability between different programs.

A range of social native identity systems such as Spritely, BrightID and BackChannel.

Federated learning and privacy-preserving machine learning more generally.

mesh network.

Data cooperatives, cooperatives and trusts.

Wikipedia and wiki-based content structure, more generally.

Community content moderation systems such as Reddit.

Community-first cloud computing, file storage and time sharing.

Unlike distributed redundancy, secondary tends to eliminate the need for trust by exploiting trust to increase efficiency rather than reduce efficiency. In the case of a community mesh network, a community initiates a decentralized wireless network with locally installed shared nodes and antennas. The design of creative economic incentives is critical to the sustainability of such networks, but these incentives are embedded in social relations, not as a substitute for social relations. Similar principles underlie recent blockchain-based alternatives, which we welcome.

The following shows the difference between accessibility and redundancy, building on some areas of potential impact on Web3.

identity and reputation

Promise: Web3 promises to liberate identity and reputation from the control of a few big tech companies, allowing “self-sovereign” identities to be used for communication, transactions and governance.

Redundancy: The basic data architecture of Web3, the anonymous ledger, is not suitable as a basis for identity or reputation. Given the ease of setting up multiple accounts on an anonymous ledger, “hybrid” attacks (or multi-identity attacks) are common, where people seek undue influence on a platform by taking control of multiple identities on that platform. Solutions that presuppose redundancy focus on common, non-contextualized unique cryptographic identifiers. Removing context leads to a reliance on “universally safe” identifiers based on clean/common features, such as biometrics, which tend to raise at least as many concerns as the centralized protocols they replace.

Auxiliary: Trust is a fundamental building block of identity, most related interactions that resort to identity mechanisms for attestation or verification are more about relationships (identity of employees, citizens, students, platform contributors) than about universal identity. Since the early days of the Internet, network-based authentication methods (often referred to as “web of trust” or “IP trust”) have envisioned verification based on strong but often informal trust relationships; more recent protocols built on this framework Examples include Spritely, BackChannel, KERI, Āhau and ACDC.

data empowerment

Promise: Web3 claims to let data creators “own” their data, and potentially monetize it, while protecting their privacy.

Redundancy: The typical vision of data ownership centers on the concept of private property for data in a “personal data store” that can be freely traded and interfaced with the market through DeFi structures. However, this structure is unlikely to facilitate data authorization in the narrower sense, for reasons including:

Most data is relational (e.g. emails between people, genetic data shared by family parts, social graph data), so the concept of private property fails. If any individual can block the transaction, then the data cannot be used; if any individual can authorize the transaction, there will be competition as each data holder tries to sell it in front of everyone else.

Most uses of data rely on aggregation, limiting the bargaining power of individuals without collective organization, just as industrial-era workers needed to bargain collectively.

Subsidiarity: An emerging data management affiliate model combines social and legal structures (such as data cooperatives, cooperatives, and trusts) with privacy-preserving and enhancing technologies for data processing, such as federated learning and secure multi-party computation.

In these models, socially entangled interpersonal data are collected and managed by cooperative organizations (from local governments to worker cooperatives to credit unions) that are accountable to and socially connected to data subjects. These organizations can negotiate with companies and other entities to develop guidelines around the use of shared data. For example, credit unions could act as stewards of member data, exchanging specific insights only with startups building loan refinancing vehicles or with public sector agencies aiming to improve financial policy, in the process of adding value to the ecosystem and transferring benefits to members At the same time, the privacy of the underlying data is maintained. Such managers can further communicate with other credit union networks for better leverage and benefits. Similar structures could be used for different needs such as Covid-19 contact tracing or tracking carbon emissions, unlocking substantial public benefits while protecting individual and community decision-making.

organizational innovation

Promise: Web3, and in particular the Distributed Autonomous Organization (DAO) movement, promises flexible, lightweight, accountable organization, and peer-to-peer, whole-community empowerment.

Redundancy: So far, fully automated organizations have failed because of the inability to specify relevant contingencies in bug-free code. Flexibility and automation are at a fundamental tension, as permanent automated processes are very susceptible to oversights or mispredictions, and cannot adapt when faced with situations not covered in the initial code. After all, automatism is about generalization and repetition; flexibility requires the opposite. In human cognition, flexibility comes from judging specific cases and creating exceptions. As such, DAOs rely heavily on governance, but in the absence of identity infrastructure, they mostly rely on a token-by-vote structure, which is easily accessible to, for example, venture capitalists who gain control of 51% of tokens attack. DAOs, usually Web 2.0-based governance structures, can (and sometimes do) succeed.

Auxiliary: Platform Cooperativeism, Exit Community, Governance, RadicalxChange and other related movements have been developing tools such as community currencies, access tokens, innovative voting systems such as 4D voting, new democratic deliberation tools such as Pol.is and Loomio, and new ways to fund emerging democratically accountable organizations, these tools focus on community engagement and empowerment, bringing organizations together to create larger-scale collaborations rather than acquisitions or purely financial contracts. While certain functions are most effective when automated, these processes make adaptability key to how an organization operates.

WEB3 provides an important discussion on decentralization. However, now is the time to harness this energy to achieve the best of decentralization: auxiliary, not redundant, network, not ledger.

Much can be achieved with limited deviations from the current Web3 ecosystem. A non-transferable token that remains with the original issuer and is therefore non-financial and has great potential to allow social identity and community autonomy, even if limited to situations where privacy is not a major issue (e.g., a person placing things on public resumes may be included on twitter). In such an ecosystem with soul, the portability and transparency of this trust-based, non-transferable token enables not only a more personality-based network, but also innovative voting and community governance protocols.

In fact, many such protocols have flourished experimentally in less than fully secure forms, e.g., four-dimensional voting and funding, voting and collective funding mechanisms that enable more fine-grained preference expression and aggregation. Multi-signature accounts are empowering community asset management. There have been a series of attempts to build data federations etc, mostly relying on blockchain. There seems to be reason to believe that various privacy-enhancing technologies can be integrated with the existing Web3 ecosystem, expanding these technologies further, and working towards more affiliate structures.

More ambitious projects suitable for support have greater potential to address meaningful coordination challenges. These projects may face some hurdles in achieving scale.

They neither serve the power of the current profit-seeking incumbents, nor encourage speculation in a global fungible currency based on their promise of a global revolution.

Their path to implementation is more long-term and socio-technical, given that they intend to work with and leverage existing social structures, rather than try to replace them.

Currently, they are less directly tied to the energy surrounding blockchain and, therefore, benefit less directly from the resulting hype.

However, we see these challenges as a call for coordinated, multisectoral investment. While this path may be less straightforward, it also has greater potential for social change. Today’s Internet has grown out of a multisectoral collaboration between the U.S. government, academia, industrial research labs, and Internet service providers, supporting the decentralization of adjuncts on the basis of a public mission. The major coordination challenges of our time, from crisis response to global governance, require well-designed adjuncts to succeed at scale. We believe that the supporting elements outlined here could provide the basis for such a system.