The Interledger Community 🌱: Joshua Tan

The Metagovernance Project — Grant Report #2

Joshua Tan — Sun, 15 Aug 2021 20:31:03 +0000

Hi GFTW community! I’m Josh, one of the founders of the Metagovernance Project. We’re building a software toolkit, Metagov, that helps online communities build their own digital governance systems. Using Metagov and its various supported services, communities can spin up democracies, governance boards, reputation systems, juries, and even full-fledged economies (featuring Web Monetization!) with the click of a few buttons.

A caveat: Metagov is a “governance layer” for interoperating and orchestrating a whole bunch of other governance apps and services. It doesn’t work without those apps. That’s why we’re starting by integrating projects like Loomio, SourceCred, and Open Collective as Metagov modules alongside in-house modules like PolicyKit, CommunityRule, and Govbase, in order to pilot this experiment.

Here, we wanted to report on the results of our work over the past 6 months and some further planned experiments with our Web Monetization plugin.

Progress on objectives

There were three mains goals outlined in the initial proposal:

To build a prototype that suits the needs of at least two of our diverse platform partners, creating the groundwork for what we hope will become a universal governance standard.
To experiment with users in the wild and better understand how to further hone the core logic of modular governance systems.
To better understand the monetization opportunities for the Metagovernance Project itself.

We’ve completed the first and third goals (see the next sections), and are still in the middle of completing the second goal (see “What’s next”).

Current implementation

The Metagov community currently uses Probabilistic Revenue Sharing to share revenue generated on the community’s Discourse server. Each time a web monetized visitor loads the page, a payment pointer is selected from a list of predefined payment pointers.The visitor pays to the chosen pointer until the page is reloaded or closed.

The revshare config is governed by Metagov via PolicyKit, a governance “driver” that ships with Metagov. We have a policy in PolicyKit that initiates a vote each time somebody requests to be added to the revshare config. That policy itself is governable by PolicyKit constitution policies. So, depending on the PolicyKit configuration, certain community members can propose changes to that policy, in order to change the way the RevShare config is governed.

Tools Used

PolicyKit
Metagov
- Discourse Plugin: discourse.poll, discourse.create-message
- RevShare Plugin: revshare.add-pointer, revshare.pick-pointer
Discourse plugin: discourse-web-monetization

Required Setup for this Use Case

A community exists in PolicyKit, and it has the Discourse Plugin enabled for Metagov.
The discourse-web-monetization plugin is installed on the Discourse server.
The plugin is enabled and configured in the Discourse admin panel.
There is a custom user field called “wallet” that was created by a Discourse admin. Users can input their payment pointer on the user profile page:
This Policy exists in PolicyKit for the community.

Voting on RevShare changes

This is what the PolicyKit Policy looks like in action:

Alice adds the payment pointer “$alice.wallet” to her profile page
The Metagov bot creates a Discourse poll for people to vote on whether or not the wallet to be added. The conditions of the vote (who is allowed to vote, what the closing conditions are, what the text says, etc) is all determined in the PolicyKit policy. This enables the community to propose changes to it. With this community, each wallet always has an equal weight. That’s also defined in the policy, so the community can control it.
The Metagov bot sends a message to Alice to notify her that the vote is underway.
The PolicyKit policy decides when the closing condition of the vote has passed. Either 24h passed, OR there was at least 1 approval from a user with Trust Level 3, OR there was at least 1 disapproval from a user with TL3, OR at least 5 people voted in total.
- The PolicyKit policy determines whether the vote PASSED or FAILED.
- If it passed, the user's payment is added to the revshare config. They will immediately start receiving revenue from visitors to the Discourse. Alice is notified.
- If it failed, the revshare config remains the same. Alice is notified.

Key activities

We recently had a technical workshop in NYC that focused on the architecture of Metagov itself and the relationship between Metagov and its integrated services (e.g. modules like PolicyKit, SourceCred, Open Collective, Agreement Engine, and Web Monetization). The goal of these discussions was to understand how we can re-design Metagov in order to optimize the experience of the developers that actually use Metagov. One question that arose in these conversations: are the actual “customers” of Metagov the communities implementing governance systems or the partner services that we integrate with? A lot of our current work has been driven by the needs of these services, e.g. PolicyKit, SourceCred, and Web Monetization all need integrators in order to function across platforms. Metagov is designed to be a backend service, and if it’s working well, the end-user should never need to know that it exists.

However, we also identified some major cross-service needs. For example, authoring policies that work across multiple Metagov services requires some level of identity merging / entity resolution across these services. Another requirement: monitoring, logging, and oversight of the multiple services that a community uses. Should Metagov itself be providing these functionalities, or should we integrate separate services for identity and for monitoring? And if we’re doing it ourselves, we will likely need some sort of front-facing UI so that community admins have a workflow to perform identity merging and to monitor their services. We have yet to resolve this tension.

But in the meantime, we’ve produced a draft “standard voting API” that integrates a number of different voting services, so that developers can build applications and plugins on top of a generic API rather than committing to a single voting service. (This also fulfills one of the key deliverables described in our original proposal, “an interoperable standard”.) One of our next steps is to convert the “voting on RevShare” policy described above to use this standard, so that the voting which currently occurs through Discourse Polls can happen on any other supported platform (e.g. Pol.is, Loomio, or Snapshot).

Communications and marketing

We hosted a hackathon! Over 100 hackers and developers participated in our Open Web Governance Challenge, co-hosted by NEAR, with over $25,000 given out in prizes and awards. The Metagov Slack community also continues to grow organically through our weekly seminars.

We’re planning a public release in late November, once we have time to evaluate and iterate based on our initial partner use-cases.

What community support would benefit your project?

We’ll be developing standards for identity & multi-platform communities over the course of the next quarter, and we would love to hear from the Web Monetization community about how standards could support the work that you folks are interested in.

Stay in touch through our email list (just send a message to hello@metagov.org) and let us know if you know of a community or a platform that is thinking about governance. We’re always looking for beta testers!

What’s next?

The main priorities for us in the near future are (1) to continue working on our initial partner use-cases and (2) to conduct more user interviews so we can make decisions about our architecture.

We also discussed a potential collaboration with PeerKat, another GFTW Grantee, on using Metagov for their own implementation of probabilistic revshare. However, since they are working in a smart contract, we first need to build a blockchain plugin / oracle for Metagov, which is nontrivial. Still, it's a really interesting idea and one we hope to get to at some point as we grow.

Some more concrete features on our priority list:

Add a UI component that displays the community’s revshare configuration transparently, and all historical changes to it. This UI could be in Discourse as part of the plugin, or in PolicyKit as a special RevShare view.
Support context-aware payment pointer selection. For example, if the monetizer visitor is viewing a topic authored by thelastjosh, dynamically get a payment pointer but give 50% of the revenue share config to thelastjosh, and the remaining half to the rest of the community.
- Another example: support a configuration where revenue is split equally among the topic author and the topic participants.
- Another example: support a configuration where revenue generated from topics tagged “dev” are distributed among the core dev team, and revenue generated from topics tagged “support” are distributed among the support team.
A Coil wallet integrator so that different Metagov modules and policies can directly govern wallet interactions.

A governance layer for the Internet — The Metagovernance Project

Joshua Tan — Wed, 03 Mar 2021 04:18:23 +0000

Project overview

Hi Mozfest! I’m Josh, one of the founders of the Metagovernance Project. We’re building a software toolkit, Metagov, that helps online communities build their own digital governance systems. Using Metagov, communities can spin up democracies, governance boards, reputation systems, juries, and even full-fledged economies with the click of a few buttons. Thanks to the Grant for the Web, we're building web monetization support directly into the Metagov toolset so that community websites will be able to govern a collective wallet.

Project team

Our core team:

Miriam Ashton, Metagovernance Project
Federica Carugati, Stanford University
Primavera De Filippi, Harvard & CNRS
Seth Frey, University of California, Davis
Shauna Gordon-McKeon, Galaxy Rise Consulting
Lawrence Lessig, Harvard Law School
Ricardo Saavedra
Vincent Russell, University of Colorado Boulder
Nathan Schneider, University of Colorado Boulder
Joshua Tan, University of Oxford
Michael Zargham, Block Science & University of Vienna
Amy Zhang, University of Washington

And a shout out to all the awesome people in the extended Metagov community / seminar!

What we are doing

Metagov itself is a “governance layer” for connecting, interoperating, and orchestrating a whole bunch of other governance apps and services. It doesn’t work without those apps. That’s why we’re starting by integrating projects like Loomio, SourceCred, Open Collective, and the Stanford Participatory Budgeting Platform as Metagov modules alongside in-house modules like PolicyKit, CommunityRule, and Govbase, in order to pilot this experiment.

If you're interested in learning more, feel free to drop me a line (josh@metagov.org) or drop by one of our (weekly seminars)[www.metagov.org/seminar] for a fun discussion.

What community support would benefit your project?

Do you run an online community or an independent social platform? Let us know; we are always looking for beta testers!

Link Round Up!

Project website
Project whitepaper
Govbase
PolicyKit
CommunityRule
Our most recent progress report to GftW.

Highlight other projects

Metagov is built on top of a whole host of awesome open-source projects; in some sense, we built Metagov because we wanted to make all these projects more accessible and more interoperable. We listed a bunch of projects above already, but here are some more that we love: Civic Tech Field Guide, Constitute, Aragon, DAOstack, Kleros.

The Metagovernance Project — Grant Report #1

Joshua Tan — Wed, 03 Mar 2021 03:46:19 +0000

The Metagovernance Project — Grant Report #1

Hi GFTW community! I’m Josh, one of the founders of the Metagovernance Project. We’re building a software toolkit, Metagov, that helps online communities build their own digital governance systems. Using Metagov, communities can spin up democracies, governance boards, reputation systems, juries, and even full-fledged economies (featuring Web Monetization!) with the click of a few buttons.

To clarify: Metagov itself is a “governance layer” for connecting, interoperating, and orchestrating a whole bunch of other governance apps and services. It doesn’t work without those apps. That’s why we’re starting by integrating projects like Loomio, SourceCred, Open Collective, and the Stanford Participatory Budgeting Platform as Metagov modules alongside in-house modules like PolicyKit, CommunityRule, and Govbase, in order to pilot this experiment.

Project Update

In January, we brought on board Miriam Ashton as our lead research engineer. Miriam has a background in standards-based development at Spotify and Commure (e.g. working with healthcare standards like FHIR) and she’s been instrumental in helping us develop the Metagov standard for online governance. She hit the ground running and helped us commit to a bunch of initial architecture decisions around the relationship between the Metagov Core (an API service) and the Metagov Driver—currently Amy Zhang’s PolicyKit, but possibly folding in other drivers such as Shauna Gordon-McKeon’s Concord—which directly enforces governance actions on a social platform. Now we’re in the middle of the first phase of prototyping: we’ve hooked up governance apps like Loomio, SoureCred, and Web Monetization, developed a Discourse integrator for our driver, built a stable of use-cases, launched a reference implementation of Metagov on Discourse, and are testing a bunch of governance policies involving the external libraries we’ve integrated. In other news, we also just got a fresh infusion of funds from Stanford's Digital Civil Society Lab!

image - current policykit.metagov.org screenshot

Web Monetization Support

The prototype will include a module for implementing probabilistic revenue sharing, with weights based on an external reputation system such as SourceCred. Discourse will fetch this rev share so that the Discourse instance is monetized according to the external configuration, and changes to that configuration can be governed. We have spec’d out this module and will build it in March.

Progress on objectives

There were three mains goals outlined in the initial proposal:

To build a prototype that suits the needs of at least two of our diverse platform partners, creating the groundwork for what we hope will become a universal governance standard.
To experiment with users in the wild and better understand how to further hone the core logic of modular governance systems.
To better understand the monetization opportunities for the Metagovernance Project itself.

We’ve made good progress on the first goal, and are in the middle of implementing the prototype. But there’s a lot more to do before we can deploy Metagov with live users on our "launch partner", DADA.art. In the meantime, we’ve built a series of architecture diagrams with DADA that explore the different kinds of institutional structures and governance systems that we might implement using Metagov and its associated tools.

Key activities

In the proposal, we identified three key deliverables:
An open standard that allows platforms using Metagov to be interoperable and to share common governance modules
A software framework with which platform developers can easily implement a Metagov instance for their users
Test deployments with existing blockchain communities, Discord, and MMOs

Each of these deliverables are really just different ways of describing the core deliverable / objective: a prototype of a software framework. As mentioned above, for now we’re focusing on the test deployments of the software framework. That doesn’t mean we’ve lost sight of the two important requirements referenced in the first and second deliverables: cross-platform interoperability and ease-of-use. We’re actually just this week building our first “cross-platform” policies, namely governance policies that take data and emit actions across multiple social platforms (in this case, forum posts in Loomio and forum posts in Discourse).

Communications and marketing

We recently hosted an AMA with Jazear Brooks of Sifchain, one of our platform partners. And look for an upcoming governance hackathon in late April that we are organizing with another one of our platform partners (announcement to go out soon!). The Metagov Slack community also continues to grow organically through our weekly seminars. Otherwise, we haven’t spent much time on marketing in the first half of the project.

What’s next?

There’s a lot still to do! In March, we’ll be building a Coil wallet integrator so that different Metagov modules can govern wallet interactions. Our partner integrations are slated for May; in April, we plan on introducing the prototype to Metagov’s own community of platform operators and governance researchers.