# Commons Collective, Example A: A Library-Led Open Education Collective

**Institutional blueprint example paper · 2026-09-19 · version 0.1**

**DRAFT / NOT EXECUTED · AI-assisted · Not peer-reviewed**

This hypothetical institution is not incorporated. No founder, host, customer, funder, payment provider, or reviewer has committed. Human roles, including Alice and Bob, are fictional; cited authors and organizations do not endorse this paper. New York is the proposed initial U.S. operating context, not an approved jurisdiction or legal form.

## Abstract

Example A specifies a library-led institution selling maintained retrieval, curation, integration, and accessibility services over already open or otherwise authorized, low-risk educational materials. It is a worked organizational design, not a general replication checklist. A prospective library, university, and open educational resources (OER) maintainer coalition would first seek an accountable existing host. Buyer and supplier chambers, affected-public representation, independent stewardship, professional operations, external assurance, and regulated payment arrangements divide authority. Public quality evidence remains open; service reliability and commissioned labor are paid. Separate ledgers and stoppable delivery prevent customer revenue from becoming ownership over knowledge. The design remains conditional on local authorization, review, funding, and demonstrated demand.

## 1. Founding choice and public purpose

The proposed locality is New York State, initially serving adult-facing public-library and university-library programs. Prospective founders are institutional library delegates, university OER staff, independent maintainers, and accessibility practitioners, not named real partners. Their hypothesized gap is recurring maintenance between initial OER publication and dependable institutional use: broken references, inaccessible formats, uncertain versions, and integration work.

The founding decision is **adapt first, federate only where useful, create only if necessary**. A convening group would compare an existing library consortium hosting this program, nonexclusive cooperation among existing providers, direct procurement, and a new body. It must document mandate gaps, costs, conflicts, dissent, and affected-public responses. No gap has yet been empirically established.

The preferred design is an adapted, hosted program with enforceable delegated controls. Federation would cover shared evidence formats and optional procurement, not centralized price setting. Creation requires demonstrated failure of hosting alternatives, constituent authorization, and funded continuity. If none qualifies, stop. Ostrom supports examining contextual accountability, not assuming one organizational recipe works everywhere. [S03]

This paper adapts the repository's [charter](../templates/collective-charter.md), [mandate](../templates/bargaining-mandate.md), [governance](../templates/governance-and-benefits.md), [business model](../templates/business-model.md), and [pilot protocol](../templates/pilot-and-evaluation.md). They remain nonbinding drafting resources.

## 2. Customers, services, and the open boundary

Customers would be authorized library and university procurement units paying from service budgets. They buy a maintained collection endpoint, versioned retrieval index, attribution and provenance records, catalog integration, accessibility remediation, staff support, and specified acceptance tests. The alternative is direct contracting or in-house maintenance, not paying for permission to read.

The initial collection comprises introductory adult-learning modules whose public-domain status, CC BY permissions, or other sufficient authorization is verified item by item. Retrieval is document discovery with source links, not unsupervised instruction or model training. Student submissions, learner records, minors' data, behavioral profiles, sensitive community material, and unresolved rights are excluded. Pilot integrations would accept catalog records, not learner activity; the design excludes learner accounts and application-level learner query/IP logs. Hosting/security metadata needs a separate minimized, disclosed, lawful processing policy; no claim is made that all network infrastructure produces no logs. Staff business-contact information receives separate privacy controls.

The **open public quality core** includes source references, license/version metadata, correction histories, accessibility checks, evaluation methods, and non-sensitive aggregate results. Newly commissioned public-benefit adaptations should be openly licensed where rights permit. Free access cannot depend on membership. UNESCO supports equitable OER access and sustainable maintenance; its open-science recommendation also recognizes justified restrictions. Neither recommendation grants missing permissions. [S73] [S72]

Paid assurance reports, integration, response commitments, and service-level agreements purchase accountable work and capacity, not exclusive quality facts, favorable rankings, or suppressed defects. Commercial arrangements cannot restrict existing public-domain or CC BY reuse; no royalties arise merely from that reuse. [S26] Crossref and Wikimedia Enterprise offer bounded service-design analogies, not market validation for this proposal. [S82] [S83]

## 3. Membership and governing bodies

An institution joins through a verified delegate; an adult maintainer joins through a short, accessible declaration of role and conflicts. No data contribution or purchase is required. Institutional members each have one buyer-chamber vote; individual supplier members each have one supplier-chamber vote. Affiliated duplicates cannot multiply institutional votes. Dual-role members declare conflicts and cannot decide their own awards.

Membership initially requires no capital unless the eventual lawful form requires it. Any required amount, liability, and repayment restrictions must be disclosed. A reviewed solidarity arrangement should subsidize required capital for people unable to pay, without donor voting control. If lawful inclusion cannot be achieved, choose another structure.

| Body | Composition and mandate | Accountability |
| --- | --- | --- |
| Buyer chamber | Authorized institutional delegates; sets service priorities and procurement envelopes | Equal institutional votes; purchasing volume buys no extra votes |
| Supplier chamber | Maintainers and delivery practitioners; proposes feasible standards and working conditions | No rival bids; recusal from own awards and complaints |
| Affected-public council | Adult nonmember users, disability advocates, and underserved-community representatives | Open recruitment and documented selection; funded access and participation; can trigger review |
| Independent stewards | Three qualified people, without a buyer/operator majority | Protect rights and allocate commons funds; publish reasons; removal only for documented cause |
| Professional operator | Contracted service lead, technical/accessibility staff, and finance lead | Bounded delegation, milestone reporting, replaceable contract |
| Independent assurance and appeal | External reviewer plus a separate appeal adjudicator | Protected budget and publication rights; neither reviews their own work |

Chambers elect their delegates for renewable one-year pilot terms; majority turnout establishes quorum and ordinary decisions require a majority of unconflicted votes. Each chamber can recall its delegates. The public council nominates stewards; both chambers confirm appointments after published conflict checks. Protected-rule changes require two-thirds approval in each chamber and steward concurrence; no vote overrides third-party rights.

Council recommendations receive a reasoned response within thirty days. Relational governance motivates representation of affected noncontributors; trust scholarship informs bounded delegation without making this program a legally constituted trust. [S25] [S07]

## 4. Six functions and an authority matrix

Six necessary functions are scoped member procurement, contributor agency, independent commons stewardship, professional operation, independent assurance, and regulated payments. The chambers and mandate registry cover the first two; the bodies above cover the next three; an appropriately regulated external payment partner covers settlement where required.

These functions do **not** automatically require three incorporated entities. The proposed host may house several with enforceable separation. Counsel must determine whether liability, independence, custody, or regulation instead requires separate entities. Neither the program nor its commons fund has assumed charitable, tax-exempt, fiduciary, or financial-services status.

| Decision | Who proposes or executes | Who authorizes or checks |
| --- | --- | --- |
| Service specification and shared budget | Operator with both chambers' input | Buyer chamber; individual institutions retain purchasing authorization |
| Negotiate and sign a purchase | Operator negotiates within recorded scope | Principal approves; only an expressly authorized signatory binds it |
| Accept materials or commission work | Rights officer checks item-level evidence | Authorized rights holder controls necessary new permissions; stewards can suspend disputed use |
| Award supplier work | Conflict-screened procurement panel | Unconflicted delegated approver; supplier chamber cannot award its own work |
| Allocate restricted commons funds | Independent stewards | Funding conditions and external finance review; no customer editorial veto |
| Settle contributor entitlement | Finance prepares verified instruction | Separate approver and payment partner; reconciler checks settlement |
| Suspend activity or replace operator | Stewards or independent reviewer may suspend | Buyer chamber appoints replacement with steward concurrence and independent review |

Awards, appointment reasons, aggregate accounts, and redacted minutes are public; mandates, live bids, and private evidence are restricted. Competition review precedes joint negotiation: no shared price floors, compulsory exclusivity, market allocation, or automatic boycott. EU solo-self-employed guidance is not U.S. authorization or general licensing immunity. [S13]

## 5. Contracts, Alice and Bob, and enforceable exit

The contract package would contain a hosting agreement, operator delegation, procurement schedule, supplier-work agreement, voluntary commons covenant, payment arrangement, and assurance/appeal terms. Each specifies responsible parties, deliverables, permitted processing, acceptance, fees, liability review, termination, and surviving duties. Every mandate records scope, budget authority, expiry, and exclusions. New permissions require a separate verified schedule; no blanket copyright or personal-data assignment accompanies membership.

**One illustrative flow:** Alice is an adult OER maintainer; Bob is an authorized library buyer. Alice's existing CC BY module remains freely reusable. Bob requests an accessible, maintained version integrated into his library catalog. His institution gives the operator a time-limited negotiation mandate; Alice separately authorizes negotiation of remediation and update labor. Neither mandate alone authorizes signature.

The rights officer verifies the module and embedded third-party material. Unresolved components are excluded. Alice approves the work terms; Bob's authorized institution signs the service order. Acceptance requires the agreed accessibility checks and functioning source-linked retrieval. The invoice separates operator services from any agency amount contractually owed to Alice. A payment partner settles verified entitlements; finance cannot book Alice's pass-through as operator revenue. Public corrections and licensed outputs remain accessible.

If Alice withdraws representation, registry receipt establishes the effective UTC timestamp and immediately blocks new offers, renewals, and commitments under that mandate. The operator acknowledges within two business days and notifies counterparties. Existing work follows its explicit termination and accrued-payment schedule. CC BY permissions persist under their terms; withdrawal is not license revocation. No trained-model erasure is promised. [S26]

Anyone affected can complain by accessible web form, telephone, or post without joining. Urgent harms are triaged within one business day; ordinary complaints acknowledged within five, with a reasoned decision within thirty. A separate adjudicator hears appeals. Protected funding covers access assistance and remedy. Exit includes authorized-record export, migration support, settlement, and retention/deletion schedules. GDPR is a comparative reference, not presumed New York governing law; applicability and any consent withdrawal require specific review. [S10]

## 6. Separate accounts and conditional economics

The following **annual USD planning assumptions** reproduce the canonical [service-operator worksheet](../examples/service-operator-economics.json); they are not market quotes, forecasts, commitments, or a six-month pilot budget. This service model is separate from the consumer-dues model.

| Ledger | Receipts and uses | Closing position |
| --- | --- | --- |
| Operator | $660,000 service receipts; $300,000 direct costs; $240,000 continuing operations, including $24,000 CCF fee; $60,000 commons transfer | $60,000 unrestricted cash |
| Contributor agency | $100,000 pass-through receipts; $80,000 settled | $20,000 cash owed to contributors; **not revenue** |
| Restricted commons fund | $120,000 external grant plus $60,000 transfer; $170,000 spending | $10,000 restricted cash |
| Consolidated | $880,000 external receipts; $790,000 external outflows; internal $60,000 transfer eliminated | $90,000 total cash, only $60,000 unrestricted |

The CCF fee is a hypothetical shared-services assumption, not payment to an existing federation or an executed obligation. A three-month unrestricted reserve target is $150,000 against $600,000 annual operator cash costs excluding pass-through, leaving a **$90,000 reserve gap**. Annual operating balance does not establish launch liquidity. Taxes, launch costs, debt service, depreciation, deferred revenue, grant conditions, and payment timing remain outside this worksheet.

Operator, contributor, and restricted-fund accounts or safeguarded subaccounts require distinct purpose controls and monthly independent reconciliation. Procurement commitments, complaints/remedy funding, and exit reserves receive separate tracking. Bookkeeping alone provides no insolvency protection. Founders cannot informally hold client money or borrow restricted balances.

Stewards prioritize publicly useful accessibility and maintenance awards using published criteria, milestones, and conflict checks. The operator earns disclosed service consideration; suppliers earn contracted labor payments; the public receives maintained open resources. A qualified finance reviewer must approve principal-versus-agent treatment, lawful custody, monthly liquidity, reserve funding, and funded wind-down before launch.

## 7. Delivery and stopping gates

**Preflight:** Record all approvals as **not-reviewed** until competent reviewers document evidence, conditions, and expiry. Required reviews include New York/U.S. legal form and contracting powers, rights, competition, privacy, finance/payments/tax, accessibility, governance, and ethics. No real service launch precedes approval, verified buyer authority, signed bounded orders, or funded incident, appeal, and exit provision.

Preregister a small collection and an identical service basket against credible in-house or direct-provider alternatives. Count labor, member time, legal work, assurance, integration, accessibility, payment costs, transition, and exit provisioning. Publish grants separately from both full resource cost and customer price. Lack of a credible comparator means savings remain untested.

Days 1–30 establish the rights manifest and synthetic withdrawal/payment drills. Days 31–90 deliver the endpoint and initial accessibility work. Days 91–180 test updates, portability, remediation, and renewal from actual institutional budgets. Evaluation uses adult staff acceptance exercises and aggregate service records, not learner tracking; no learning-outcome claim follows.

**Day 90 gate:** Independent assurance requires verified authority for every applicable activity, reconciled accounts with no unexplained shortfall, completed acceptance tests, a successful withdrawal drill, and an accessible funded appeal route. At least 90% of participating adult representatives must explain scope, refusal, withdrawal, and surviving open permissions; missing responses count as not demonstrated. Failed comprehension suspends the relevant mandates pending redesign.

**Day 180 gate:** Continue only with no unresolved serious rights harm, maintained free access, independent quality acceptance, credible renewals, recurring-cost coverage, and funded reserves and exit. Test the preregistered target of at least 10% lower quality-adjusted total cost without diminished access or quality. Missing that target or lacking a credible comparator blocks scaling and any savings claim; continuation requires a newly reviewed, explicitly subsidized public-benefit case, not retroactively changed success criteria.

Discovery of learner records, minors' data, disputed authority, inaccessible essential delivery, or contributor-fund misuse triggers immediate suspension of affected activity at either gate. Unresolved systemic failures stop the pilot. Publish negative findings and a redacted decision record.

## 8. Counterfactuals and limitations

Direct purchasing may outperform this collective: governance could cost more than it saves. A grant may sustain public maintenance without proving commercial demand. Losing an anchor customer or receiving invoices late can exhaust unrestricted cash while contributor and restricted balances remain unavailable. Removing a grant requires reducing grant-supported scope, not raiding contributor funds.

Supplier concentration could allow a contractor to withhold integrations; require documented interfaces, export tests, and replacement support. Buyer dominance could favor marketable English-language modules over accessible public-interest work; independent allocation and affected-public review must expose that trade-off. Public quality claims can be wrong: publish correction evidence, not an unqualified certification seal.

This is a normative, untested design, not legal advice, a feasibility study, or evidence of willingness to pay. Initial exclusion of learner data reduces risk but does not resolve all privacy, procurement, employment, accessibility, or copyright duties. Local reviewers must validate enforceability and representation. Institutional success means accountable service and public benefit, or an orderly decision not to proceed, rather than incorporation itself.

## Bibliography

These ten references reproduce identifiers and bibliographic records from the existing research source registers. They support bounded concepts and comparisons, not approval of this institution.

- **[S03]** Elinor Ostrom. 2009. *Beyond Markets and States: Polycentric Governance of Complex Economic Systems*. https://www.nobelprize.org/prizes/economic-sciences/2009/ostrom/lecture/
- **[S07]** Sylvie Delacroix and Neil D. Lawrence. 2019. *Bottom-up data Trusts: disturbing the 'one size fits all' approach to data governance*. https://www.repository.cam.ac.uk/handle/1810/301307
- **[S10]** European Union. 2016. *Regulation (EU) 2016/679 (General Data Protection Regulation)*. https://eur-lex.europa.eu/eli/reg/2016/679/oj/eng
- **[S13]** European Commission. 2022. *Guidelines on the application of Union competition law to collective agreements regarding the working conditions of solo self-employed persons*. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52022XC0930%2802%29
- **[S25]** Salome Viljoen. 2021. *A Relational Theory of Data Governance*. https://yalelawjournal.org/feature/a-relational-theory-of-data-governance
- **[S26]** Creative Commons. 2013; version 4.0. *Attribution 4.0 International: Legal Code*. https://creativecommons.org/licenses/by/4.0/legalcode.en
- **[S72]** UNESCO. 2021. *Recommendation on Open Science*. https://www.unesco.org/en/legal-affairs/recommendation-open-science
- **[S73]** UNESCO. 2019. *Recommendation on Open Educational Resources (OER)*. https://www.unesco.org/en/legal-affairs/recommendation-open-educational-resources-oer
- **[S82]** Crossref. Undated; accessed 2026. *Fees*. https://www.crossref.org/fees/
- **[S83]** Wikimedia Enterprise. Undated; accessed 2026. *Wikimedia Enterprise/Principles*. https://meta.wikimedia.org/wiki/Wikimedia_Enterprise/Principles

**Reuse:** Original draft text is offered under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). External works retain their own rights and license terms. Reuse implies no author or organizational endorsement.