Using Reputation

Koii may seem like a silly name for something so important, but it stands for Knowledgeable Open and Infinite Internet, and it's our first attempt at building a CARP.

Since the origins of the decentralized web, there has been an ongoing battle between trust and accountability, and the need to provide services and tools to billions of people, at a minimum cost. Many services have attempted to create global machines, that hope to encompass all possible scenarios in a singular data model, but this has so far been inadequate, and we've been waiting a long time for the supposed 'web3' to come to fruition.

A Compute, Attribution, and Reputation Protocol, or CARP, blends together the benefits of a 'collateralized' (a.k.a. "Proof of Stake") model with the benefits of trust that are common in the current internet. When we think of hosting applications, we usually imagine paying someone else to run our code on their device, but this model has created gate keeping and monopolies that will soon control most of the information our species has created. Typically, these exist because of the reputation of the hosting provider, and the technology that these providers are able to build to entrench their relationship with customers. With Koii, we hope to build an initial prototype of a new model, where the technology lives in a public commons, and anyone can build reputation towards providing their own hosting to the network.

Definition of CARP

Under a CARP model, a community member must have some initial way to build reputation through low-risk activities, after which they can be trusted to run computation tasks. To this end, we propose three primary objectives of any truly decentralized network.

1. Reputation Can Be Earned Without Up-Front Investment

In order for such a system to trend away from centralization, it must be possible for new entrants to join and contribute value without access to hosting resources or other personal assets. This ensures that the most honest providers can always rise to the top, without prioritizing those that would attempt to buy out the network.

2. Reputation Reduces Work Replication

The computational efficiency of a compute network defines the overall cost. If a hosting provider is not well-trusted by the community, their work must be audited by other nodes to verify it's authenticity, which wastes valuable network cycles and increases the unit cost of each individual task proportionately. As a result, if reputation systems can be employed to reduce overall audits, it is possible to implement potentially lower cost alternatives.

3. Tasks Can Prioritize High-Reputation Providers

In order for reputation to provide a significant advantage, it is necessary to ensure that hosting purchasers can give priority to honest providers. This, perhaps for the first time, necessitates a blockchain and decentralized storage architecture, where all information is public, transparent, and immutable. In a CARP, transparent money is the key to long term growth, not just a fad or a grift.

Network Growth

Under a model where honesty and reliability are the most valuable property, consistently honest participants becomes the de-facto distributors of all resources. As a result, we expect that there will arise competitive markets to provide up-stream hosting capacity to these providers, ensuring that they are able to access as many resources as necessary to satisfy demand for their services.

As a parallel, a reliable business typically receives more credit, allowing it to grow. We hope to see hosting providers use their reputation and their past earnings to receive priority treatment as re-sellers of existing hosting, as well as new hardware.

Koii is a CARP

Consumer hardware is now widespread globally, with billions of personal devices in use, spread closely around the active internet user population. These devices, under the current model, act as readers, pulling information from a central repository, but could also be employed to host much of this information as well. With the introduction of proper reputation models, new members of the community can easily join, provide hosting, and grow their reputation, before expanding their offerings with new hardware or cloud hosting capacity that they acquire through the profits of their previous activities or via promotions from external vendors.

Over time, our mission is to give each individual the opportunity to build reputation and act as a steward of the worldwide knowledge database, and in so doing, to earn their share of the value that it creates.

In Koii, we focus on building reputation first from content, and attention, because we believe this will help the network grow. As new participants join, they can then use the tokens earned from the attention economy as collateral to provide hosting, and continue to grow their share.

Koii is a work in progress, so we welcome feedback, advice, and support.

Building a CARP

Building a CARP with a Koii task is actually pretty simple. All you'll need to do is track the reputation of your node operators and other community members over time, and provide incentives for them to behave as valuable members of your community.

Koii's native Decentralized IDs are expected to launch in the coming months, so stay tuned for more updates as we build more core reputation primatives for you to use in the Task Namespace.