Rainmatter Foundation

How we work

Practice

Principles and techniques that may be valuable for changing either the trajectory or the velocity of change.

Border artefacts

Cooperation without agreement.

A border artefact is not an object. It is a relationship: a shared thing on the edge between a few named actors that lets them coordinate without first having to align.

What separates it from a boundary object, in Star and Griesemer’s sense, is where it sits. Its power comes from occupying neutral ground rather than any one actor’s territory. Whoever stewards that edge can only do it because they are not competing for anyone’s turf, which is also what stops the artefact turning into a platform grab.

A method rather than a description. Rainmatter Foundation works this way in projects that need a shift in trajectory or velocity and involve actors who will never share an agenda, and sets it out here so that collaboratives and consortia can name their own.

Three questions decide whether something qualifies:

  • Who is at the border. Named, distinct actors with specific unmet incentives. Never “everybody”.
  • What its purpose is. The particular misalignment it resolves, and which way things flow across the border.
  • How long it lives. Either transient, handing off to a named successor, or enduring as infrastructure.

Stated as a sentence: [artefact] sits between [named actors] so that [purpose]; it is a [form], [transient or enduring].

Two worked examples. The Open Air Quality commons sits between communities and solution actors around shared measurements of air data, and is enduring. The waste diagnostic canvas sits between innovators, funders and housing ministries to expose material recovery chokepoints, and is transient, succeeding into professionalised policy.

Seven mental models

  1. The named edge, not the public square. A border needs identifiable sides. “Open data for everyone” has no particular actors facing each other with conflicting incentives, so it has no edge.
  2. Form follows the border. The material shape is incidental. A diagnostic, a commons, a strategy, a contract, a convening coalition: whatever the edge demands.
  3. The form by life grid. Artefacts sit on two axes, data through strategy and transient through enduring, and they travel between cells. A transient diagnostic canvas typically migrates toward enduring policy infrastructure.
  4. Succession discipline. An artefact serves its purpose and then either becomes permanent or hands off to a successor and retires. The failure mode is the zombie artefact, kept alive past its usefulness because it has become an organisational deliverable.
  5. The intersection is a nursery. Where the borders of two themes overlap, new artefacts form with fresh coalitions of actors. Only an actor working across themes can see these. A single theme funder is structurally blind to them.
  6. Neutrality is load bearing. Sitting outside everyone’s territory is what lets the steward hold the edge, and holding the edge is what lets them convene. Owning the artefact would collapse that and trigger defection.
  7. It is all chlorophyll. Every form works at the same layer, converting the steward’s capitals into ownership on the ground.

The test

Six questions to put to any artefact, in any form:

  • Name the sides. If the answer is “everybody”, it is a public good, not an artefact.
  • What crosses. Name the direction: data one way and solutions back, or feedstock one way and regulatory rules the other.
  • The misalignment. The specific incentive gap or broken link being closed, between the named actors.
  • The form, for now. Diagnostic, commons, strategy, contract or convening, chosen for what the edge needs today.
  • Longevity and successor. Transient or enduring, and if transient, the successor it hands to.
  • The network the steward brings. Which actors they convene or join. If the steward would own the artefact, stop.

A commons carries extra conditions: open and interoperable standards, ownership at the edge, low cost to join with real gains at the margin, trust through convergence rather than a single authority, and a connection to feedback loops somebody can act on.

Deriving one for a theme

  1. Start with inquiry and the data work. Finish the data story first. The artefact falls out of what is found rather than being chosen in advance.
  2. Establish the border. A small set of actors on each side, and what would flow between them.
  3. Choose the form the edge needs now, while anticipating where it may go.
  4. Decide longevity, and for a transient artefact, name the successor it transfers to.
  5. Assign delivery roles to concrete actors. An unfilled role is where it will fail.
  6. Run the lifecycle: diagnose, convene, realign, institutionalise, then retire or continue, presenting evidence and advocating throughout.
  7. Watch the intersections. When one theme’s artefact touches another, as waste touches air and water, a new artefact is forming.
  8. Measure the gaps actually crossed, not the count of partners or the pile of canvases. Open, interoperable contribution is worth making a condition of support.

Public open commons

Published is not the same as usable.

A critical asset in changing the velocity of change is creating shared commons. Collaboratives for people, technology for information, or tooling for livelihoods. Making them available to others as infrastructure unlocks velocity.

Especially in a publicly funded sector, we need to look for avenues where more open and freely available common goods or infrastructure are available to people.

What a commons can be

The word covers more than software. Four kinds come up repeatedly, and a piece of work is often more than one of them at once.

  • Software. Free and open source software is the oldest and best understood version of this: Linux, PostgreSQL and QGIS are infrastructure that thousands of organisations build on without any of them owning it. In India, FOSS United works on the community and the practice around it rather than on any one project.
  • Data. OpenStreetMap and the India Biodiversity Portal are commons because the contributing is distributed and the using is unrestricted. Air quality is the clearest case in our own work: measurements are worth more pooled than held, and pooled they can be acted on by anyone.
  • Method. A methodology can be a commons. A survey protocol, a measurement standard, a restoration technique or a governance model, published so that the next group does not have to rediscover it, is infrastructure in every sense except the physical.
  • Tooling. Instruments, hardware designs and field applications that lower the cost of doing the work. A low-cost sensor design or an open data collection app decides who can afford to gather evidence at all.

What makes it a commons rather than a release

The conditions are the same ones the commons form of a border artefact carries above: open and interoperable standards, ownership at the edge, low cost to join with real gains at the margin, trust through convergence rather than a single authority, and a connection to feedback loops somebody can act on.

The test worth applying is whether a group with no relationship to whoever built it can pick it up, use it, and change it. If using it requires permission, it is a product with a generous licence.

What PDGI names here

PDGI (opens in a new tab), a framework Rainmatter Foundation works on as part of its coalition, gathers eleven strands of thinking that grew up separately and rarely speak to each other. Five of them are this section’s argument in other words:

  • FOSS
  • Digital commons
  • Open data
  • Decentralisation and federation
  • Platform cooperativism

Its case is that these are worth a shared grammar rather than a replacement, so that what each has already learned can be used with the others. The remaining six are about what a person keeps hold of once the technology is built, and are below.

Humans in the loop

Access is not agency.

With technology permeating every sphere of life, and AI increasingly making decisions on our behalf, building technology which keeps humans in the loop becomes increasingly important.

It is also important to ensure not just access, but that the use of technology becomes more accessible and democratically decided rather than imposed.

People’s Digital Goods and Infrastructure

PDGI (opens in a new tab), named in the commons section above, is where we are doing that thinking, as part of the coalition rather than as an observer of it. It sets out to be an evidence-based, forward-looking, normative framework of principles, standards, rights and obligations, and practices covering the entire spectrum of technology, from conception, design and creation through to deployment and governance.

The order of the words is the argument. People come before digital: P before D.

The problem it starts from

Digital infrastructure now intermediates a great deal of human interaction while being opaquely governed and operated by large, powerful private and public entities. That is a power imbalance, and AI widens it. As more of what decides an outcome moves inside systems nobody outside can inspect, the risk is that most people end up spectators, both as users of technology and as citizens.

What it names here

Six of PDGI’s eleven strands are about what a person keeps hold of once the technology is built, rather than about how it is shared. The other five sit with the commons above.

  • Rights and justice
  • Transparency and accountability
  • Humane technology
  • Ethical technology
  • Digital privacy
  • Algorithmic fairness

One of its principles shows what that means in practice, and is worth stating on its own: no critical digital service on which a person’s life and wellbeing depends should exist without easily accessible non-digital alternatives.

The framework is released under CC BY-SA 4.0, and is open to anyone who wants to contribute to it.