Governance: who decides, and how they answer for it

In plain words

Every survey, ballot, and consultation has a hidden power: someone chose the topics and wrote the questions. Change the list or the wording, and you change what people appear to want.

So the important questions about My Vote Forward are not only technical. They are about people and rules:

• Who picks the subjects people rank?
• Who checks the wording of each question for fairness?
• Who pays for the work, and could money influence it?
• Who runs the systems, and what are they allowed to see or change?
• What can you do if you think a decision was wrong?

This page sets out the framework's answers, explains where they come from, and says plainly which answers are still unsettled.

Defined terms

• Agenda authority. The power to decide which subjects, proposals, and questions are presented. See proposals.
• Question review. An independent check of question wording, order, endpoints, and options before use. See question design.
• Conflict of interest. A relationship or interest that could reasonably be seen to influence a decision, such as funding from a party with a stake in the outcome.
• Operator. The person or organization that runs the websites and data systems.
• Privileged action. An action only an administrator can take, such as publishing a page, changing a theme, or approving a release.
• Appeal. A request to reconsider a decision, decided by someone other than the original decision maker.
• Formal authority. The legal power to make a binding decision. A participation framework has none unless law grants it. See jurisdictions.

See also the glossary.

Evidence

Electoral practice: impartial administration, stable rules, appeal

The Venice Commission's Code of Good Practice in Electoral Matters states that "an impartial body must be in charge of applying electoral law" ¹. It says "the fundamental elements of electoral law, in particular the electoral system proper, membership of electoral commissions and the drawing of constituency boundaries, should not be open to amendment less than one year before an election" ¹. It calls for "an effective system of appeal" in which "final appeal to a court must be possible" and the procedure is "simple and devoid of formalism," and it says observation should cover "respect by the authorities of their duty of neutrality" ¹. It also lists "funding of parties and candidates" among the areas where state authorities "must observe their duty of neutrality" ¹.

The UN Human Rights Committee, in General Comment 25, says "an independent electoral authority should be established to supervise the electoral process," calls for "independent scrutiny of the voting and counting process and access to judicial review," and observes that limits on campaign spending may be justified to prevent "disproportionate expenditure on behalf of any candidate or party" ². It also says "the results of genuine elections should be respected and implemented" ².

These standards govern official elections, not an educational framework. They show the governance problems that any process asking people what they want eventually faces: neutrality, timing of rule changes, money, and appeal.

Statistical practice: professional independence and disclosure

The UN Fundamental Principles of Official Statistics say statistical agencies "need to decide according to strictly professional considerations" on methods and procedures, and are "entitled to comment on erroneous interpretation and misuse of statistics" ³. The American Association for Public Opinion Research requires published research to "name the sponsor of the research and the party(ies) who conducted it" and to disclose "the exact wording and presentation of any measurement tool from which results are reported" ⁴.

Wording and order shape answers

Pew Research Center reports that "even small wording differences can substantially affect the answers people provide," that "earlier questions can unintentionally provide context for the questions that follow," and that questions asking about "more than one concept" should be avoided ⁵. This is the empirical basis for independent question review. A question writer cannot reliably detect their own framing effects.

How existing participation programs handle agenda and review

Existing programs show different allocations of agenda power. In Reykjavík's My Neighborhood program, "residents can impact their immediate surroundings by submitting ideas," which are then "reviewed and ideas are selected for the ballot," after which "people vote for the best ideas," and chosen ideas become projects the following year ⁶. The city's page does not say who conducts the review. vTaiwan describes itself as "an experiment that prototypes an open consultation process," created by the civic technology community g0v with government participation, and says formal discussions begin "when needs are articulated"; in one case it reports that more than 4,000 participants crowd-sourced a meeting agenda ⁷. More cases, with strengths and failures, are in cases and participation between elections.

In Switzerland's e-voting trials, the Federal Chancellery commissions "independent examinations" of the system and its operation, and tracks identified gaps in a catalogue of measures ⁸. That is one model of operator accountability: an outside examiner, a published list of findings, and monitored remediation.

Competing interpretations

Expert curation versus open agenda setting. Curated subject lists are coherent and comparable over time but concentrate power in the curators. Open submission spreads power but can be dominated by organized or well-resourced groups. Reykjavík and vTaiwan combine both in different ways ^{6 7}. The framework's ten starting subjects are a demo choice, not a claim that ten is right. See priorities.

Funding independence versus sustainability. Some argue that only public or philanthropic funding is compatible with neutrality. Others argue that disclosure and firewalls matter more than the source. The framework rules out the most direct conflicts (data sale, paid influence, pay-to-rank, sponsorship by candidates) and requires disclosure; it does not claim disclosure alone removes influence.

Binding versus advisory. Some participation advocates want results to bind institutions. Others hold that representatives must retain judgment and protect minority rights. General Comment 25's call to respect results applies to genuine elections ², not to advisory consultation. See representation.

Application to this framework

These are working rules. They are labeled provisional and recorded in the project's decision register.

1. Agenda authority. Subject lists and question sets are versioned. Each version records who proposed it, who reviewed it, the date, and the reason for changes. The demo's subject list is a demonstration, not a universal agenda.
2. Question review. Questions are reviewed before use against the question design guidance: one concept per question, labeled endpoints, an uncertainty option, and no loaded terms. Review notes are kept. The goal is independent review; where none is available, the page says so rather than claiming it.
3. Stable rules near a decision. Following the logic of the Venice Commission's one-year guideline ¹, question wording and method for a given release should not change once collection has begun, except to correct an error, which is disclosed.
4. Conflicts and funding. No data sale, paid influence, pay-to-rank, candidate sponsorship, or campaign targeting. Funding sources and material conflicts are disclosed on the about page when they exist. The ".org" domain is not a claim of nonprofit or charitable status.
5. Operator accountability. Administrators have the least privilege their role requires, use multi-factor authentication, and every privileged action (publish, rollback, theme or hero change, release approval) is logged as metadata without political answer content. Routine administrators cannot browse individual political responses. See security.
6. Corrections and appeals. Anyone can report an error or challenge a decision through the contact page, which is a structured submission, not a public forum. Corrections are published in the changelog. A decision should be reconsidered by someone other than the person who made it; whether an outside party can fill that role is an open question.
7. No official authority. The framework cannot make any institution act on a result. Official elections remain with election authorities. See jurisdictions.

Unresolved questions

• Who should hold final agenda authority over subject lists if the framework is used by a real community: the operator, an independent panel, participants, or a public body?
• How can organized or well-funded groups be kept from dominating open submissions without excluding legitimate organizing? See collective priorities.
• What independent body could review questions and hear appeals, and who would pay for it without creating a new conflict?
• How should governance change if a public authority ever adopts parts of the framework, since its legal duties would then apply?