Service Details
Proof of Concept Solution
As Proof of Concept (PoC), we will develop a software tool to immediately support the Deep Funding team in the analysis of user activity on its SWAE Community Portal. The goal is to automatically detect how members of the SingularityNET community engage constructively on the portal, both in regular Deep Funding rounds as well as Governance rounds. Community members can currently create project proposals, governance proposals, comments (strengths, risks, questions), reactions (a restricted set of emojis to express sentiments) and ratings (a numerical quality indicator for proposals). So far, all these interactions have been analyzed manually in spreadsheets by the Deep Funding team, which is both a time-consuming and error-prone process that is better suited to be done by a robust data integration and analysis pipeline. The aim of an initial software tool is therefore
-
to automate the process that was performed manually in spreadsheets,
-
to introduce configurability to the performed calculations in order to enable experimentation with different evaluation approaches, and
-
to make the chosen calculation process and retrieved results transparent to both the Deep Funding team and the SingularityNET community.
On its way to become a full-fledged DAO, SingularityNET continually tries to improve its approach towards community-driven governance, which requires continuous experimentation and analysis. Two methods that are currently used in Deep Funding are 1) incentivization of constructive community involvement by distributing AGIX rewards to members for their activity on SingularityNET's Community Portal on SWAE and 2) providing greater weight to the voices of engaged community members in voting rounds on SingularityNET's Voting Portal. The software tool proposed here will support both these approaches by providing quick and reliable analytics of community activity on SingularityNET's Community Portal, which will in turn influence the voting weights applied to results from SingularityNET's Voting Portal. As a Proof of Concept (PoC), it shall demonstrate that a configurable software tool for automated analysis has immediate practical value to both the Deep Funding team and the SingularityNET community by 1) increasing the ease and speed of experimentation in community-driven governance, 2) ensuring correctness and transparency in the applied calculation processes, and 3) in future enabling more complex experiments such as “liquid reputation”, where for example receiving positive feedback from a highly-engaged community member counts more than if the same is received from a newly created account.
The goal of this Proof of Concept (PoC) is to replace the current manual calculation of contribution scores in spreadsheets with an automated process that is reliable and transparent. The project scope is therefore given by the set of functionality required to replicate the results of previous rounds and then support future rounds. The following user-centered requirements and corresponding solutions are derived from this description:
-
The Deep Funding team shall be enabled to automatically calculate contribution scores from data exported from the SWAE portal. The PoC will therefore be implemented as a web application with the following functionality: The front-end allows non-technical users to provide one or several tabular source files together with a set of configuration options that influence the downstream calculations. The back-end performs data inspection, pre-processing, calculation of scores and correctness checks. The resulting contribution scores are displayed on the front-end in the form of a table together with a description of the used configuration options. As an optional and experimental feature, the output may also contain a graph visualization similar to those that Robert Haas produced for Deep Funding round 1 and Governance round 1. It could provide an overview of all community interactions in a round and potentially highlight scores in the form of node and edge sizes. The feasibility and usefulness of such a visualization has to be tested on the actual data.
-
Results and methods of the calculation of contribution scores will be clearly communicated to community members of SingularityNET. The current practice by the Deep Funding team has been to communicate via blog posts and accompanying video presentations. To support this approach, the results from the front-end will be structured in such a way that they can be embedded directly into a blog post or hosted as a self-contained HTML site that can be referred to in an article.
-
We will work with the Deep Funding team to reproduce the current manual calculation of contribution scores and then introduce desired variations. Please note that these scores reflect the activity of community members on SingularityNET’s Community portal on SWAE, but not their AGIX holdings or their activity on SingularityNET’s Voting portal. This means that the Deep Funding team will still need to perform some manual calculations, for example to derive voting power, which depends both on contributions scores and AGIX holdings. Future extensions of the tool might also automate these tasks by incorporating further data sources.
-
We will also work with the SWAE team to design the best format for the exported data in order to support the analysis desired by SingularityNET.
-
We will use a project structure that maximizes modularity, so it can be easily grown over time. This includes the choice of mature and scalable libraries to withstand expectable growth of both the community and desired features.
Specification of Future Product
The ultimate goal of this project is to develop a modular solution that supports other projects within the SingularityNET ecosystem, as well as external decentralized communities that use reputation-based mechanisms for their governance processes. This will be achieved through an iterative approach, paired with extensive user research, both for the potential users of the platform and the communities producing these scores.
We develop specifications for a second phase project which detail the requirements and design considerations for a software tool to enable SingularityNET and other communities to collect contribution scores based on custom configurations. The specifications will outline ideas for extending the tool to involve additional stakeholders and generate value for a wider range of potential users. These specifications will be adaptive to allow for new use cases and the requirements of a more complete, sophisticated product including conceptual user interfaces.
We will produce a set of long term goals and a road map for achieving them. Through community engagement and user feedback, we ensure that the tool designs are continuously meeting the needs of its users and that ethical issues are well thought through and addressed.
Score Research and Community Engagement
The practice of rating users, corporations, and other entities is employed in a number of contexts. For example, Experian uses reputation scores to help businesses understand and contextualize their online reputation with scores on a 1,000 point scale. Other companies help businesses manage their reputations. Some scores, such as the reputation equation, are simple
Reputation = Trust * Credibility + Time
Others, such as the
(CSRR) score are based on robust theoretical foundations and can be quite complicated.
Credit Scores, for example Fair, Isaac, and Company (FICO) may have a proprietary construction based upon payment history, debt burden, length of credit history or "time in file", types of credit used, and recent searches for credit. Nations also have rating systems, for example China Social Credit Scores.
We will conduct a thorough exploration of relevant reputation and contribution scores, explore best practices, and determine a set of principles and a procedure for the SingularityNET community to establish a framework for contribution scores. We will determine a list of potentially important variables, winnow to a more compact list of variables that contribute to the goal we are seeking to achieve, and develop a model that uses these variables to produce a score.
We will build a process for meaningful community input into the design of contribution scores. This will ensure transparency and help to develop community endorsement of the score. We do this in an environment mindful of clearly articulated ethical considerations. This will increase the ease and speed of experimentation in community-driven governance and ensure correctness and transparency in the applied calculation processes. The community will contribute to the development of values exposition, goals refinement, and contribute user knowledge. We will determine what can be meaningfully crowdsourced and what will be better designed by this team.
Program Management
Photrek is utilizing the principles of Sociocracy in completing the proposal and execution of the project. Sociocracy, invented by a Dutch entrepreneur, operates on three principles: a) formation of governance circles, b) decisions by group consent, and c) double linking of circles with both an appointed leader and an elected representative. John Buck, an American social scientist, has been an influential consultant with the Cardano Catalyst program. His book with Sharon Villines, “We the People: Consenting to a Deeper Democracy” will be used as a guide in the program management of Community Contribution Scores.
Photrek has formed a governance circle for its company, consisting of the current owner/employee, Kenric Nelson, and three contractors working with Nelson toward transitioning to an employee/owner model; William Thistleton, Megan Hess, and Blake Anderton. Likewise, the project team (Nelson, Thistleton, Haas, and Attieh) worked with our Advisor, Jan Horlings by group consent in completing the proposal. This was particularly valuable when Haas raised concerns about the scope and budget of the Proof of Concept. We were able to modify as a team the scope and budget of this part of the project to increase the probability of success and focus on the most critical capabilities.
One of the principles of sociocracy is that the full team shares in the profits of a project. Our budget includes a 5% management reserve. The team members have provided fixed prices for each of the deliverables but if unexpected issues arise requiring additional requirements the reserves will be used for this purpose. The remaining reserve will be split between the team members (proportional to their direct cost) and Photrek (proportional to indirect costs) if it is not required to complete the project milestones.
The facilitation of effective communication within the team and with the SingularityNET will be a priority. Thus, in addition to the milestone reporting, Photrek will host regular team meetings and participate in SingularityNET community meetings.