Eve library

OSHUN V1 Redesign Sequencing Plan

Spec

Revision 1

On this page

Status: Accepted v1.0
Owner: OSHUN Design + Product + Engineering
Last Updated: March 26, 2026
Implements: V1-UX-055
Related:

  • docs/releases/v1/specs/surface-redesign-ranking.md
  • docs/releases/v1/specs/experience-scorecards.md
  • docs/releases/v1/scope/minimal-chrome-rules.md
  • docs/releases/v1/specs/layout-density-tiers.md
  • docs/releases/v1/scope/scan-first-content-hierarchy-rules.md
  • docs/releases/v1/specs/todos.md

1. Purpose#

This document defines the redesign sequencing plan for OSHUN V1 so the highest-impact UI overhauls ship before lower-value polish.

The principle is simple:

  • do shared launch-critical rules before route-specific redesign
  • do shared primitives before duplicating route-local UI
  • do P0 surfaces before P1 polish
  • do audits only after the thing being audited actually exists

Numeric backlog order is not sufficient on its own. The remaining UX work needs wave ordering.

2. What Is Already Complete#

The current foundation layer is already in place:

  • V1-UX-011 through V1-UX-019
  • V1-UX-045 through V1-UX-054

That means the program now has:

  • token foundations
  • domain accent behavior
  • grounded evidence, disclosure, trust, queue, persona, avatar, and review state rules
  • customer and admin visual theses
  • surface inventory, ranking, scorecards
  • minimal chrome, density, hierarchy, ergonomics, and posture rules

The redesign sequence below starts from that completed baseline.

3. Canonical Sequencing Rule#

Follow this order:

  1. launch guardrails
  2. shared primitives
  3. P0 customer and admin surface redesigns
  4. P1 and specialist polish
  5. audit, QA, and release gates

Any task that does not help one of those phases should not jump ahead of a blocked earlier phase.

4. Wave Plan#

4.1 Wave 1: Launch Guardrails And Structural Rules#

Objective:

  • finish the remaining shared rules that every redesign wave depends on

Tasks:

  • V1-UX-020 partial failure and degraded mode
  • V1-UX-056 grid, gutter, max-width behavior
  • V1-UX-057 typography measure and reading rhythm
  • V1-UX-058 imagery, illustration, texture, ambient media rules
  • V1-UX-059 card-avoidance rules
  • V1-UX-060 motion hierarchy
  • V1-UX-061 loading, skeleton, empty, offline, and error states
  • V1-UX-062 action-placement rules
  • V1-UX-063 interaction-feedback rules
  • V1-UX-034 accessibility standards for shared components
  • V1-UX-035 reduced-motion behavior
  • V1-UX-068 contrast and focus-indicator rules
  • V1-UX-073 mandatory design-review checklist
  • V1-UX-074 Playwright-backed visual QA checklist

Why this wave comes first:

  • customer and admin overhauls should not start before layout, state, motion, action-placement, and QA rules exist
  • these tasks prevent route-local redesign drift
  • they also define the acceptance bar for later implementation

4.2 Wave 2: Shared Launch-Critical Primitives#

Objective:

  • build the reusable components that unlock multiple P0 surfaces at once

Tasks:

  • V1-UX-021 domain switcher components
  • V1-UX-023 evidence card and evidence sidebar
  • V1-UX-026 assistant disclosure and memory-state components
  • V1-UX-028 admin queue row and queue filter components
  • V1-UX-029 incident severity and SLA timer components
  • V1-UX-030 review decision controls

Second-tier primitives inside the same phase, after the first group is stable:

  • V1-UX-022 continuity cards
  • V1-UX-024 claim, source, and passage list items
  • V1-UX-025 notebook and collection components
  • V1-UX-027 provenance badges
  • V1-UX-031 policy-edit and policy-diff components
  • V1-UX-032 avatar and voice pack cards
  • V1-UX-033 rights and license summary components

Why this wave comes second:

  • the P0 shell, home, evidence, assistant, queue, incident, and approval redesigns all depend on these primitives
  • building them once avoids route-local duplication and inconsistent QA burden

4.3 Wave 3: P0 Customer Surface Overhauls#

Objective:

  • redesign the highest-impact customer experience surfaces first

Primary surfaces:

  • shared shell chrome and persistent overlays
  • home dashboard
  • onboarding
  • domain route experience
  • assistant, continuity, and disclosure surfaces
  • global loading, degraded, error, and not-found states

Secondary P0 customer surface in the same wave once the above are stabilizing:

  • explore and discovery

Why this order:

  • shell, home, onboarding, domain route, and assistant posture define the core customer experience
  • global degraded/error behavior must land during the same wave, not after
  • explore is critical, but it benefits from the shell and state rules settling first

4.4 Wave 4: P0 Admin Cockpit Overhauls#

Objective:

  • raise the operator cockpit to launch-grade after the shared operator primitives exist

Primary surfaces:

  • explicit admin shell boundary and operator chrome
  • Studio root cockpit directory
  • identity, access, and policy cluster
  • review, audit, notification, and retention cluster
  • reliability, incident, and launch-readiness cluster

Why this is after Wave 2:

  • these surfaces depend directly on queue rows, incident severity, review controls, policy diffing, and shared feedback/state rules
  • admin quality should be built on reusable operator primitives, not bespoke cockpit fragments

4.5 Wave 5: P1 Surfaces, Tone, And Specialist Ergonomics#

Objective:

  • complete the second layer of quality once the launch-critical surfaces are structurally correct

Tasks:

  • V1-UX-036 Tara-led content tone standards
  • V1-UX-037 Veritas grounded tone standards
  • V1-UX-038 Nisaba scholarly tone standards
  • V1-UX-039 AI disclosure copy guidance
  • V1-UX-040 error and fallback copy guidance
  • V1-UX-064 one-handed mobile ergonomics rules
  • V1-UX-065 dense-data ergonomics for admin tables and compare views
  • V1-UX-066 form-entry ergonomics
  • V1-UX-067 copy-compression rules

Surface polish focus:

  • search
  • activity
  • library
  • profile
  • public domain marketing
  • download/install
  • integration and composition admin cluster
  • design-system and navigation-governance specialist workspaces

Why this wave is later:

  • these tasks matter, but they should refine a stable core rather than define it
  • tone and specialist ergonomics are most valuable after the primary surface model exists

4.6 Wave 6: Audits, QA, And Launch Signoff#

Objective:

  • prove the redesign actually meets the bar

Tasks:

  • V1-UX-041 two-taps-to-value audit
  • V1-UX-042 one-handed reachability audit
  • V1-UX-043 calm visual rhythm audit
  • V1-UX-044 keyboard and focus-path audit
  • V1-UX-069 dashboard-card sprawl audit
  • V1-UX-070 long-form readability audit
  • V1-UX-071 viewport-fit audit
  • V1-UX-072 progressive disclosure audit

Persistent release controls from earlier wave outputs:

  • V1-UX-073 design-review checklist
  • V1-UX-074 Playwright-backed visual QA checklist

Why this is last:

  • audits should validate implemented redesigns, not placeholders
  • these tasks become the launch gate once the major surfaces are redesigned

5. Explicit Priority Rules#

5.1 Do Not Jump Ahead To Polish#

These should not move ahead of launch-critical surface work:

  • specialist imagery polish
  • secondary profile or library refinement
  • late specialist admin family polish
  • audit-only work before the redesigned surfaces exist

5.2 Do Not Build Primitives Before Their Rule Set#

Component work should not start before its rule layer is available.

Examples:

  • queue-row primitives should follow action-placement, interaction-feedback, and degraded-state rules
  • domain switchers should follow grid, chrome, and hierarchy rules
  • evidence and notebook primitives should follow reading-rhythm and density guidance

5.3 Do Not Audit The Old Product And Call It Done#

Audit tasks are not substitutes for redesign implementation.

The required sequence is:

  1. define the rule
  2. build the primitive
  3. redesign the surface
  4. audit and verify it

6. Dependency View#

Depends on Enables
V1-UX-020, 056-063, 034-035, 068, 073-074 all major customer and admin redesign waves
V1-UX-021, 023, 026, 028-030 shell, evidence, assistant, queue, incident, and approval redesigns
completed V1-UX-045-054 foundation docs all remaining UX implementation and audit work
P0 customer/admin redesign implementation meaningful completion of audit tasks 041-044, 069-072

From the current backlog state, the recommended next-order execution is:

  1. V1-UX-020
  2. V1-UX-056 through V1-UX-063
  3. V1-UX-034, V1-UX-035, V1-UX-068, V1-UX-073, V1-UX-074
  4. V1-UX-021, V1-UX-023, V1-UX-026, V1-UX-028, V1-UX-029, V1-UX-030
  5. P0 customer surface redesign implementation
  6. P0 admin surface redesign implementation
  7. remaining reusable components and P1 surface refinement
  8. audit and launch signoff tasks

8. Exit Standard#

The sequence is being followed correctly only when:

  • shared rules appear before shared primitives
  • shared primitives appear before route-local duplication
  • P0 shells and core surfaces land before P1 polish
  • audits are used as signoff, not as delay camouflage

If the team is polishing a lower-value surface while a P0 shell, queue, assistant, onboarding, or degraded-state dependency remains unresolved, the sequence has broken.