Skip to content

Public Language Guide

bijux-pollenomics should read like a calm scientific product, not like an operator notebook and not like a status theater dashboard. This page defines the vocabulary used in code-owned report surfaces and in the handbook pages that explain them.

Public language matters here because wording can inflate a weak surface without changing a single data file. The repository therefore treats naming and phrasing as part of the evidence contract, not as a cosmetic afterthought.

Use this page when your question is:

  • why does the repository use one word and reject another
  • which terms signal real review strength and which would overclaim
  • how should I describe these outputs without making them sound broader than they are

Core Rule

Public wording should help people judge evidence strength more clearly, not less clearly. If a phrase makes a weak surface sound mature, general, or final, it is the wrong phrase even when it sounds elegant.

Information Roles

  • review: public judgment about whether one bounded surface is trusted
  • validation: pass or fail structural checks on a governed surface
  • audit: systematic inspection across many files, claims, or source families
  • truth: claim-calibration surface that keeps scope and weakness explicit
  • summary: aggregate orientation surface for people or downstream tooling
  • coverage: how much of a bounded domain is currently represented
  • readiness: whether a publication surface currently clears its own bar
  • honesty: public caveat surface that keeps limits visible beside outputs
  • ledger: accumulated exclusions, conflicts, or unresolved caveats
  • matrix: repeated comparison question across several domains or surfaces
  • workflow: governed human review sequence for curation or release
  • queue: ordered recovery pressure for still-blocked work

These roles matter because different surfaces do different jobs. You should not have to guess whether a page is making a claim, exposing a caveat, or only recording a diagnostic.

Avoid These Words

  • viewer: it describes a tool posture, not an evidence responsibility
  • packets: it describes delivery format, not the question a surface answers
  • scorecard: it sounds managerial and hides what is actually being reviewed
  • polished labels that do not say what the page settles

These banned patterns are not about style preference alone. They are blocked because they often hide the real question a surface answers.

What Good Public Wording Sounds Like

  • it names the surface responsibility directly
  • it names uncertainty instead of smoothing it away
  • it distinguishes context, evidence, framing, and caveat roles
  • it helps people know what to trust and what to inspect next

Provenance Wording

  • say sample-owned, project-level, supplementary-table, or another exact provenance class when it matters
  • name uncertainty directly instead of smoothing it into confident prose
  • prefer supports, anchors, suggests, blocks, and remains unresolved over inflated verbs such as proves unless the surface genuinely does

Geographic Wording

  • distinguish Nordic, country-filtered, Europe-plus, and comparator scopes explicitly
  • do not call a comparator-heavy or region-limited output general or region-agnostic
  • name the filter boundary whenever a map or summary is scope-specific

Publication Wording

  • call a surface publishable only when the governed checks for that surface actually pass
  • keep the strongest readiness language behind the release gate
  • when evidence is partial, say partial, thin, blocked, or contextual rather than hiding that status in softer wording

A Quick Check

If a term sounds broader, cleaner, or more triumphant than the underlying evidence really is, stop and translate it into the more exact role it is trying to describe.