Introduction

Motivation

Digital innovation has tremendously impacted the way we do business, the way we exchange and consume information. Data has become a resource upon which many businesses have been built. It altered our society and everyday life in every possible way.

From Data to Knowledge

Data is an objective and elementary description of reality. It has no value by itself. It needs to be processed and interpreted to create valuable information and knowledge. Knowledge is usually subjective and context-dependant.

Why we need data sharing

We don’t know how to go to Mars, we don’t know how to predict cancer, we don’t know how to farm in a desert, and we don’t know what present would please our mother for her birthday.

Knowledge is the limiting factor to common good, for individuals, companies and society at large.

In our digitalized world, one of the most efficient way to unlock knowledge is through data.

But data is rarely shared. It stays in silos, isolated and protected. As such, data is under-used and under-valued. Indeed, data is a non-rival good, which means that I can share it many times without changing my ability to use it. Generating knowledge from data is theoretically infinite. That's why an incommensurable amount of value could be unlocked through data sharing.

Why is data sharing not the norm then?

Today's problem is that individuals and companies are incentivized to build gated data lakes, to keep their users captive and to restrict access to data and information, mostly because of:

• Risk (lack of trust between participants, the fear of loss of competitive advantage)
• Complexity (interoperability & governance issues)
• Costs (lack of infrastructures, R&D)
• Lack of incentives (bad business models, bad incentives, unclear benefits)

But what if we could enable trust-minimized data sharing and unlock this untapped potential?

Existing approaches

Since data is everywhere and knowledge is valuable, the issues related to data sharing and knowledge creation are far from being new concepts for individuals and organizations.

Many approaches already exist to share data between entities:

Data exchange is the most common model today, while a decentralized approach to data sharing seems to be the most promizing. In fact, it will enable data sharing without trusted third parties, without moving data from its existing place and even enable decentralized governance on data sharing rules.

Game theory applied to governance issues with tokens as tools for interest alignment is a field in which we've just scratched the surface. The future looks bright.

Technology is getting more mature by the day

Gated data lakes are the norm today because we lacked the economic incentives to fund, the technology to build, and the organizational structures to manage an open Dataverse.

It is changing now.

Through technology and governance innovation, public blockchain technology will be the key infrastructure opening the gates of trust-minimized data sharing. Thus providing opportunities to unlock new sources of valuable knowledge that will solve humanity's problems, from small to existencial ones, thanks to the next generation of Applications.

But the perfect technology doesn't exist, yet

Ethereum initiated a wave of decentralized applications thanks to smart contracts. Following Ethereum steps, many other layer-1 blockchains emerged. Some offering "more scalable versions" through bigger blocksize or higher block frequency, some offering other smart contract programming languages such as Rust or Javascript, some offering privacy-preserving technologies such as zk-SNARKs.

Then, which infrastructure should we choose to build applications based on trust-minimized data sharing?

By asking ourselves this question, we quickly realized that:

• There is no such applications designed on public blockchains, so no existing network effect to benefit from.
• General-purpose chains are not specifically designed for such complex apps: multiple layers of smart contracts generate increased risks and complexity.
• Developers are restrained by smart contracts programming languages & virtual machines of these blockchains.
• These infrastructures are not designed to enable data network effects.

To sum up, trust-minimized data sharing may need its own layer-1.

Introducing – OKP4

The previously mentioned reasons led us to create OKP4 (Open Knowledge Protocol For... everything and everyone).

What is OKP4?

OKP4 is a public PoS layer 1 blockchain built for trust-minimized data sharing.

OKP4 is an open interoperable smart contracts platform for highly customized rules, governance mechanisms and business models to align interests between participants and build useful distributed applications.

Anyone can create custom ecosystems called Data Spaces. In each Data Space, people & companies can share data, algorithms, software, storage and computation according to Data Space's rules.

The OKP4 protocol enables developers to build applications based on distributed digital resources shared by participants (datasets, algorithms, storage, computation...). These shared digital resources together constitute the Dataverse, an ever-expanding universe of resources that can interact with each other through custom ecosystems of shared rules.

With OKP4, builders can find development kits, governance templates and easy-to-integrate applications to design their own applications and Data Spaces. Participants can provide any off-chain digital asset they want to make it findable, available and interoperable with other digital resources of the ecosystem. This on-chain coordination unlocks revolutionary possibilities to generate applications leveraging data and services from many businesses and individuals without any trusted third party accessing these digital resources. It enables new opportunities to generate value from data without exchanging it with other parties, thus bringing trust and incentives to share it and build new, better applications.

OKP4 is designed to become the commons for builders, data providers, communities, businesses, data scientists & users to unlock tangible value creation through a new generation of web3 "real-life" use cases, way beyond Decentralized Finance.

OKP4 is built using the Cosmos SDK and is fundamentally a multichain infrastructure project. Our core value proposition relies on interchain applications leveraging existing networks inside and outside of the Cosmos ecosystem, such as Akash, Juno, Osmosis, Cheqd, Stargaze, Filecoin, Arweave, Ethereum and many more...

What is a Data Space ?

A Data Space is a digital common that makes data and services (algorithms, computing resources, etc.) available to the community, based on shared governance, for the creation of knowledge. The governance guarantees the individual and collective interests of the participants, who retain the sovereignty of their resources.

A Data Space consists of three elements:

1. Bundle of rights & rules that regulates interactions between data and services (algorithms, infrastructure and every other resources shared by participants)
2. Governance mechanisms to make these rules evolve
3. Interfaces to create, modify and interact with the Data Space

What is the Dataverse?

The Dataverse is inspired from the term Metaverse, a growing buzzword that embodies the next generation of internet experiences combining virtual reality and digital ownership. The challenge with the Metaverse is to build it on an open infrastructure, so that users are not captive of the companies and siloed environments produced by companies.

The Dataverse is an open world where users become owners. They own their data, the algorithms & resources they provide. They can even own the blockchain infrastructure and can be involved in its governance.

The Dataverse is not a datalake or a cloud where data is stored centrally. Everything can stay at the source, that's why Dataverse participants are always sovereign about what they share. The Data Spaces leveraging the Dataverse can be nested and overlapping, as one dataset can participate in many Data Spaces, and many applications can be built on top of one Data Space.

OKP4 described

To sum up, OKP4 is a blockchain for anyone to build and join Data Spaces, share data & services within these Data Spaces, and build and use applications which will leverage what has been shared.

It is important to understand these 4 key concepts:

• OKP4 Blockchain (consensus & execution layer, enabling the coordination of network participants)
• Data Spaces (rights and rules enabling data & services interactions and governance mechanisms to change these rules)
• OKP4 Dataverse comprised of Data & Services (off-chain data & services referenced via metadata and APIs; coordinated by the blockchain according to Data Space's rules)
• Applications (off-chain apps leveraging outputs from the distributed network and bringing value for the end-users)

Let's dive into each of these.

OKP4 Blockchain

OKP4 blockchain is the decentralized layer that:

• Guarantees trust through decentralization by the Tendermint Consensus & the native token $KNOW
• Indexes anything as a service (as long as it is accessible via an API) - any off-chain dataset, algorithm, software, storage or computation resource and even on-chain services from other IBC-enabled networks
• Exposes metadata of indexed data and services to make it findable, accessible and interoperable
• Defines Data Spaces rules in Prolog
• Is an open smart contracts platform powered by CosmWasm for maximum governance customization.
• Orchestrates off-chain data & services and initiates workflows which power applications
• Enforces governance rules, including access rights and value transfers
• Connects seamlessly to other blockchain networks using IBC

It is the single source of truth that is used to build autonomous and distributed applications.

Data Spaces

Data Spaces are sets of rules and governance mechanisms set in smart contracts, designed to optimize data and services sharing within a community. It is not a common database but an abstraction that defines the interaction between datasets and services, from access rights and business models. Providers share digital resources within a Data Space, according to specific conditions explicitly mentionned. They can share their data in multiple Data Spaces. Each Data Space can have its own customizable governance mechanisms, from multi-sig to a custom Data Space token. The OKP4 Blockchain can host an endless number of Data Spaces.

Dataverse

The Dataverse is comprised of datasets and services that are shared by sovereign network participants - they don't have to copy the digital resource somewhere else to make it available.

Datasets are bytes of information stored on a server (any data storage is possible, either decentralized like IPFS or Arweave, or centralized like AWS or Azure), indexed on the Blockchain and accessible through a dedicated gateway.

Services are algorithms, software, computation, storage, or any digital asset accessible through an API. This comprises algorithms such as data processing services, automatic description of datasets, knowledge graph query services, etc.

We designed OKP4 to enable maximum customization. Anyone can share any kind of data, algorithms, software, storage and computation into the Dataverse, you can even implement any standard or DID provider. OKP4 Dataverse enables XaaS (anything as a service) by design.

Applications

Applications are built on top of Data Spaces and their inherent rules. They use the data-products (outputs from OKP4) as inputs in applications that provides information, knowledge, initializes actions, etc. Applications are not considered as resources in the Dataverse but final consumers. However, they appear in the protocol to be allowed to access the data.

Applications can have their own governance mechanisms. For example, private companies may use outputs from a decentralized Data Space to build their own applications.

To sum up

Datasets and services provide the elementary building blocks that will interact in a distributed fashion according to the workflows initiated in the Blockchain. Each workflow initiated in the blockchain refers to one specific Data Space with its own explicit rules. Each dataset and service interacts within to one or many Data Spaces according to the access given by the data and service providers. The outputs of the distributed workflows are leveraged by Applications to provide value to the end users.

Anyone can create or customize Data Spaces and Applications using OKP4 Development Kit. It allows anyone to build and design with as much freedom as possible, as simply as possible. Developers, data scientists and communities at large can:

• Customize Data Space governance with no-code tools & templates
• Easily build applications from software to interfaces
• Use or provide digital resources in the Dataverse

Network participants

On the Blockchain layer, there are Validators, Delegators and DAO Participants.

On the Data Space layer, there are Data & Service Providers who agree to share based on Data Space's rules and Consumers who pay tokens to initiate workflows in order to create knowledge. There are also Curators who can signal promising data/services and Governors who can change these rules.

On the Application layer, there are Builders who develop and maintain applications, and also end-users who benefit from knowledge created through the whole OKP4 Ecosystem. End-users can be directly Consumers, or just a user of the Application.

Multi-layered governance

The main criticisms we can address at web2 companies and especially tech giants are their extractive nature, their willingness to keep users captive on their platforms, and their increasing power controlled by centralized governance.

But web3 also has its own sets of problems; pretty wild and self-reliant, UX is bad, allocation of capital can be messy and coordination is often suboptimal.

To build the best data sharing infrastructure, we need to enable the best of both web2 and web3 worlds through multi-layered governance:

• OKP4 Blockchain is decentralized by design, with its set of token holders that propose and vote on governance proposals - it's a trustless and censorship-resistant infrastructure.
• Data Spaces have custom governance. Data Space governance can be centralized (one or few accounts have the ability to design rules through monosig or multisig) or decentralized (through token-based voting power)

This design can be ideal for some companies who want to have control over the Data Space rules, while leveraging the decentralized infrastructure that guarantees data & services providers complete ownership over what they shared. They can remove their consents at any time and can verify that the system acted accordingly to the rules they agreed on.

• Applications also have custom governance. It can be as simple as a centralized front-end to consume workflow outputs, or more decentralized through community-run applications using only decentralized storage and computation resources.
• Datasets & services are sovereign to the data and services providers. Providers (individuals or companies) can define general permissions/requirements for their digital resources to be used in many Data Spaces, but also can hand-pick the Data Spaces (and align themselves with the Data Space's rules) in which they want to contribute.

This mixture of governance layers enables the Dataverse to combine strengths from web2 and web3 worlds: where centralized governance models can be built on top of an open decentralized infrastructure, enabling interoperability and sovereignty by design.