This captures the principles that guide design decisions, the tensions we're actively navigating, and the horizon we're pointing toward. It's about intellectual honesty: naming the trade-offs we're making and why. Every principle has a cost. We'd rather state those costs than pretend they don't exist.

Design Principles

These emerge from the core insight, from the deliberate choice to focus on surplus, and from studying why previous complementary currency systems succeeded or failed.

01
Subjective value over shared currency

Each participant maintains their own sense of balance, valued on their own terms. There is no shared unit of account, no network-wide ledger, no requirement for parties to agree on exchange value.

Valuation disputes killed many historical systems. The surplus frame means both parties are better off than their alternative, which is nothing. They don't need to agree on how much better. Value is contextual: urgent need increases it, abundance decreases it. Imposing shared valuation reintroduces the currency dynamics the protocol exists to avoid.

Trade-off: Network-wide health metrics become harder to compute, and accountability blind spots emerge. We accept this because shared valuation reintroduces the dynamics we're trying to move past.

02
Protocol over platform

SEP is designed as an open protocol that multiple implementations could use, not a proprietary platform controlled by a single operator.

Single points of control create single points of failure, and single points of capture. If the insight is valid, its benefits shouldn't accrue primarily to whoever builds the first implementation. Participants should be able to move between implementations, and the protocol should survive any single operator failing or turning hostile.

Trade-off: Protocols are slower to develop and harder to iterate than products. We accept this because long-term resilience matters more than short-term speed.

03
Business-to-business focus

SEP is designed primarily for exchanges between businesses, particularly professional services firms, not consumer-to-consumer exchange.

Historical analysis is clear: B2B systems (WIR, Sardex, ITEX) have far higher survival rates than consumer systems (LETS, Time Banks). Businesses have predictable surplus, professional accountability, higher stakes per exchange, and existing trust relationships. They also have incentive to handle tax and compliance properly.

Trade-off: A smaller initial addressable population, and some beneficial consumer use cases excluded. We accept this because network viability matters more than theoretical reach.

04
Pragmatic framing, radical intent

SEP is positioned as "a useful business tool" rather than "an alternative to capitalism." The framing is deliberately pragmatic even though the underlying insight is radical.

Systems framed as ideological alternatives attract ideological participants and repel pragmatic ones. They stay small and often die. Systems framed as useful tools attract users who want the benefits, regardless of ideology. They survive and scale. We want SEP to survive and scale, not to make a statement that nobody hears.

Trade-off: The framing may feel dishonest or diluted to people who care about the deeper implications. We accept this because impact requires adoption, and adoption requires accessibility.

05
Professional management over volunteer enthusiasm

SEP should be operated with professional rigour, not run by volunteers motivated primarily by belief in the mission.

Volunteer burnout is a primary cause of complementary currency failure. LETS and Time Banks repeatedly collapsed when key volunteers left. WIR and Sardex survived because they were run by professionals who treated it as a job. Enthusiasm doesn't scale; operational discipline does.

The governance layer around the protocol (monitoring, transparency, escalation, onboarding) is substantial. A system that claims to redirect value toward participants rather than intermediaries must be honest that operating the system itself costs money. The revenue model is unresolved and needs to be addressed without reintroducing the intermediary dynamics the protocol exists to avoid.

Trade-off: Professional operation requires sustainable economics, which creates pressure toward models that may compromise other principles. We accept this because a compromised system that exists beats a pure system that doesn't.

06
Trust through relationships, not ratings

Trust is established through verifiable identity, network position (who you've exchanged with, repeatedly), and simple satisfaction signals, not numeric ratings or reputation scores.

Ratings are easily gamed and create implicit valuation. Network position (diverse, repeated exchange relationships) is expensive to fake. It requires actual exchanges with real participants over time, making gaming costly without reintroducing currency-like metrics.

Trust is derived from relationships but may be presented as metrics. Participants need to see their trust scores to understand matching decisions. The source remains relationship-based and expensive to fake, but the experience becomes numeric. Withholding scores that determine participant opportunities is hard to defend ethically. The risk is that participants optimise for numbers rather than relationships. The mitigation is that the numbers are the relationships: you can't improve your score without genuine exchanges with real partners over time.

Trade-off: Network position takes time to accumulate, creating barriers for newcomers. We mitigate this through graduated exposure, but accept that some barrier is inherent to any trust system.

07
Human accountability in the loop

For Phase 1, human confirmation is required for all commitments. AI agents may assist with discovery, filtering, and proposal, but a human or accountable organisation must approve each exchange.

Accountability requires someone to be accountable. Until we understand how autonomous agent participation affects trust, disputes, and network dynamics, we keep humans in the decision loop. This is a pragmatic constraint for Phase 1, not a permanent philosophical commitment.

Trade-off: Human confirmation creates latency and limits the speed of exchange cycles. We accept this because accountability matters more than velocity at this stage.

08
Sustainable and efficient operation

SEP should minimise its computational and energy footprint, becoming more efficient over time rather than more resource-intensive.

A system that claims to redirect value toward participants shouldn't itself become a resource-intensive intermediary. AI inference has real energy costs. If we're building infrastructure that could scale broadly, those costs matter, both economically and environmentally. Success means the system gets lighter, not heavier.

Trade-off: Efficiency constraints may limit sophistication in some areas. We accept this because sustainable operation at scale matters more than maximum capability at small scale.

Active Tensions

Philosophy provides direction, not answers to every question. These are the places where principles conflict, where we're uncertain, or where the right path depends on context we don't yet have.

Subjective value vs accountability

Subjective value ledgers avoid valuation disputes and respect contextual differences in worth. But they create blind spots: how do we detect exploitation without shared metrics? How do we handle tax compliance when authorities require valuations? What basis exists for claims when exchanges fail?

Our position: reframe accountability as commitment fulfilment rather than balance tracking. The system tracks whether participants do what they say they'll do, not whether they give as much as they receive. That preserves generous surplus-sharing culture while providing accountability for bad faith behaviour.

Openness vs trust

Network effects require growth. Trust requires barriers to entry. Vouching and network position metrics create accountability but also create insider/outsider dynamics. Early participants accumulate advantages newcomers cannot easily match.

Our position: constrain outcomes rather than entry. Default to trust and use structural limits as the safety net instead of social gatekeeping. New participants enter through identity verification and bilateral-only exchanges, not vouching. Network-level fairness mechanisms prevent first-mover advantages from compounding indefinitely. The tension is real and irreducible. We choose to lean toward trust.

Federation vs momentum

Protocol-over-platform is a core principle, but federation is architecturally complex. "Start managed, federate later" is pragmatic but creates path dependencies. Each quarter of delay makes federation costlier and less likely.

Our position: treat federation as an escape hatch, not a roadmap item. Phase 1 deploys as a governed managed service with binding governance commitments: data portability, transparent operation, protocol conformance, participant representation, non-interference with exit. These make the escape hatch credible. The escape hatch makes the commitments enforceable.

Peer exchange vs enterprise adoption

SEP is designed for professional peers exchanging surplus. But the features that enable scale attract enterprise procurement teams who may use the system differently, extracting efficiency from smaller suppliers rather than exchanging as equals.

Our position: focus on behaviour rather than identity. The system doesn't define or exclude "enterprises" but uses structural constraints and behaviour monitoring to ensure all participants engage as peers. Any participant exchanging genuine surplus as a peer is welcome regardless of size.

Algorithm transparency vs gaming

Participants deserve to understand why they're matched or excluded. But transparency enables gaming: sophisticated actors learn to exploit visible criteria. Opacity prevents gaming but enables abuse and erodes trust.

Our position: full transparency, accepting gaming. Every participant sees their own scores, match factors, and the reasons they were or weren't selected for specific chains. If gaming the algorithm means being more trustworthy, fulfilling commitments reliably, and building genuine relationships, those are aligned incentives, not a problem.

Efficiency vs resilience

Optimal matching finds the shortest, highest-probability chains. But optimising for efficiency may route many exchanges through key nodes, creating single points of failure. A resilient network might be less efficient but more robust.

Our position: dual concentration defence. Diminishing returns in scoring (the more chains a participant is in, the less their inclusion improves a chain's score) plus a hard cap on chain participation. The design chooses resilience over pure efficiency, accepting some matching quality loss in exchange for structural robustness.

Governance weight vs participant experience

The protocol is simple: describe surplus, find matches, execute exchanges, build trust. But operating a network responsibly requires governance (accountability mechanisms, behaviour monitoring, transparency, recourse processes). Each piece is individually proportionate but collectively substantial.

The risk is that participants experience SEP as a governed system rather than a useful tool. The governance mechanisms are designed to run silently. Most participants should never encounter escalation processes or advisory body decisions. The protocol is the story; the governance is the infrastructure behind it.

Radical intent vs pragmatic survival

The insight is radical; the framing is pragmatic. This is strategic, but it creates a question: at what point does pragmatic adaptation become abandonment of the original intent? If SEP succeeds by becoming "just another B2B efficiency tool," have we succeeded or failed?

We don't have a bright line. The philosophy exists partly to maintain clarity about what we're actually trying to do, so we can recognise if we've drifted too far.

The Horizon

SEP is a first step. If it works, what might follow? We're cautious here. Speculation about future impact can become grandiose, and grandiosity is a warning sign. But understanding where this could lead helps clarify why it's worth doing.

If SEP succeeds in its narrow scope, businesses exchange surplus that would otherwise be wasted. Value that would have leaked to intermediaries or simply evaporated stays with participants. The model is proven viable in a limited context. This alone would be worthwhile. It requires nothing beyond what we're building.

If the model proves robust, the scope could expand. Surplus is a conservative starting point, but the boundary between "surplus" and "regular capacity" is somewhat arbitrary. A business that successfully exchanges surplus might choose to route more activity through the protocol. This isn't something we'd push. It would emerge from participants finding value and wanting more of it.

If the protocol becomes a standard, multiple implementations could interoperate. Different operators could serve different sectors or regions while participants exchange across boundaries. Competition between implementations could improve quality while the protocol ensures portability.

If the insight is validated, success with surplus exchange would demonstrate that AI and network technologies can meaningfully disintermediate money's matching function in at least some contexts. That might inspire adjacent experiments: other forms of non-monetary exchange, hybrid systems that reduce rather than eliminate monetary mediation, application to contexts beyond B2B professional services. We're not building these. But if SEP works, others might.

What we're not claiming

  • That SEP will achieve any of this
  • That we know how to get from here to there
  • That broader transformation is inevitable or even likely
  • That SEP is the right vehicle for anything beyond surplus exchange

The horizon is a direction, not a destination. SEP succeeds on its own terms if it enables valuable exchanges that wouldn't otherwise happen. Everything beyond that is possibility, not promise.