Release 0.3.0-draft.1 · Development standards
Use the system to make and judge decisions. It should allow competent designers to extend Woblar thoughtfully without reproducing a website screen on every surface.
A brief an agency can act on
Task: Create [deliverables] for [audience] to help them [specific decision/action]. Business context: [verified offer, problem and current constraints]. Core message: [one supported statement]. Evidence: [sources, permissions and claim boundaries]. Expression: [A/B/C choice and why]; preserve Logo v3, Geist regular display and the defined colour roles. Content: [required copy, legal/product facts and exact CTA destination]. Formats: [sizes, languages, applications, accessibility and supplier requirements]. Assets: [registered source files, rights and editable requirements]. Deliverables: [editable sources, review exports, production variants and provenance]. Review: Leigh / Woblar; exact release and distribution scope; staged concept, application and production checks. Success: [how the communication will be judged; do not invent commercial targets].
Provide the minimum useful package: brief, relevant chapters, rules, logo/font assets with licences, a permitted reference contact sheet, source-backed copy and a deliverable checklist. Exclude credentials, unrelated client records and internal research not intended for that recipient.
Decision scorecard
Score each dimension 0 = absent/contradicted, 1 = partly resolved, 2 = clear and evidenced. Write one reason and a correction where needed. This is a proposed review aid, not an automated brand approval.
| Dimension | Review question |
|---|---|
| Purpose | Is the audience, task and intended action clear? |
| Recognition | Does the mark/type/colour/material language feel continuous with Woblar? |
| Treatment fit | Does A, B or C suit the reader's moment? |
| Hierarchy | Can the main message be found quickly, then read comfortably? |
| Truth | Are claims, relationships and product states supported? |
| Human responsibility | Are preparation, review and action clear when relevant? |
| Inclusion | Can people read and operate it in the intended context? |
| Craft | Do crop, alignment, rhythm, detail and reproduction feel deliberate? |
| Adaptability | Does the family work across its required formats? |
| Handover | Are sources, provenance, checks and release status usable? |
Use 16/20 as a discussion threshold for moving a draft forward, with no zero in truth, inclusion or recognition. A score never overrides a hard failure. Stop-release failures: changed logo geometry; prohibited blog reference; fabricated claim; unusable essential interaction; unresolved asset rights; missing required supplier proof; or unapproved publication. Correct the affected issue and review again rather than averaging it away.
Concrete evaluation
A cinematic report with unreadable body copy may score well for recognition but fails its reading task: use B inside and preserve C on the cover. A quiet campaign with perfect spacing but no category descriptor may be elegant yet unclear: add the service context. A close-up material scene that suggests an automated action should be captioned as illustration or redesigned around a visible review point. A faithful screenshot used as a slide may look correct but fail adaptability and editability.
Record the outcome as “ready for next review”, “revise” or “blocked by [specific dependency]”. Only the owner’s explicit release decision changes approval status.
Operating rules
BRIEF-001 — Brief decisions, not adjectives
- Do: Give audience, task, evidence, message, treatment, outputs and acceptance criteria.
- When: Agency, contractor and AI assignments
- Variation: Relevant excerpts rather than the entire source archive.
- Avoid: A moodboard alone, vague premium/innovative requests or missing source rights.
- Check: Another competent maker can begin without inventing scope or claims.
EVAL-001 — Use evidence to judge the work
- Do: Apply the scorecard with reasons and correct hard failures before release.
- When: Concept and application review
- Variation: Context-specific weighting as an explicitly recorded review choice.
- Avoid: Treating a high numerical score as approval or overriding a hard failure.
- Check: Attach scores, evidence, corrections and the owner’s separate decision.
HAND-001 — Deliver only what the recipient needs
- Do: Package the right source files, rules, licences and checks for the authorised audience.
- When: External or partner handover
- Variation: Named-partner assets where access is intended.
- Avoid: Credentials, unrelated client context or unapproved public distribution.
- Check: Inspect the exact bundle and access scope before sharing.