#362 - [UX] Wireframe exploration — 5-10 concepts across every surface - rbrooks/Quest-Board

claude-bot commented

2026-08-25 20:39:24 +00:00

Contributor

Found in the August 2026 session lifecycle review (#319).

What

Produce 5-10 distinct wireframe concepts covering every potential UX surface, for side-by-side comparison. These are deliberately low-fidelity and deliberately divergent — the point is to explore genuinely different structural answers, not to polish one.

Each concept should take a clear position on the questions the review exposed:

Where does attention live? A dashboard inbox, a per-campaign activity feed, ambient badges, or something else. The product currently has no answer at all.
What is the navigational spine? A persistent nav with a campaign switcher, a command palette, a session-centric hub, a sidebar. Today navigation is strictly a tree walk through page-specific breadcrumbs.
Is the session or the campaign the primary object? Much of the current awkwardness comes from prep, shelf, recording and review living in different places despite all being about one session.
How do GM and player views relate? Mirrored, separate, or one view with role-conditional density.
What is the at-the-table surface? Currently three partial screens with incomplete links between them.

Cover every surface, not just the obvious ones: dashboard, campaign, session, prep, live/at-the-table, recording, review and approval, wiki and lore review, quotes, analytics, profile, and admin.

Acceptance criteria

5-10 concepts, each taking a coherent and distinct position
Every surface from the inventory appears in every concept, or its absence is a deliberate stated choice
Each concept is walked through against the six core journeys
Concepts are presented together for comparison, with the trade-offs of each stated plainly
Mobile and at-the-table treatment is shown, not deferred

Found in the August 2026 session lifecycle review (#319). ## What Produce **5-10 distinct wireframe concepts** covering every potential UX surface, for side-by-side comparison. These are deliberately low-fidelity and deliberately divergent — the point is to explore genuinely different structural answers, not to polish one. Each concept should take a clear position on the questions the review exposed: - **Where does attention live?** A dashboard inbox, a per-campaign activity feed, ambient badges, or something else. The product currently has no answer at all. - **What is the navigational spine?** A persistent nav with a campaign switcher, a command palette, a session-centric hub, a sidebar. Today navigation is strictly a tree walk through page-specific breadcrumbs. - **Is the session or the campaign the primary object?** Much of the current awkwardness comes from prep, shelf, recording and review living in different places despite all being about one session. - **How do GM and player views relate?** Mirrored, separate, or one view with role-conditional density. - **What is the at-the-table surface?** Currently three partial screens with incomplete links between them. Cover every surface, not just the obvious ones: dashboard, campaign, session, prep, live/at-the-table, recording, review and approval, wiki and lore review, quotes, analytics, profile, and admin. ## Acceptance criteria - [ ] 5-10 concepts, each taking a coherent and distinct position - [ ] Every surface from the inventory appears in every concept, or its absence is a deliberate stated choice - [ ] Each concept is walked through against the six core journeys - [ ] Concepts are presented together for comparison, with the trade-offs of each stated plainly - [ ] Mobile and at-the-table treatment is shown, not deferred

claude-bot added this to the v4.4.0 - UX Overhaul: Design Exploration & Rebuild milestone

2026-08-25 20:39:24 +00:00

claude-bot added the

enhancement

frontend

labels

2026-08-25 20:39:24 +00:00

claude-bot referenced this issue

2026-08-25 20:45:22 +00:00

[Review] Full audit of the session lifecycle: prep, recording, transcription, summarization, lore — scoping v4.0.0 #319

claude-bot referenced this issue

2026-09-06 19:30:20 +00:00

v4.4.0 build order: UX Overhaul in six phases, with two review checkpoints #535

claude-bot referenced this issue

2026-09-06 19:45:57 +00:00

[UX] Inventory every surface and define the redesign brief #361

claude-bot commented

2026-09-06 19:46:04 +00:00

Author

Contributor

Picking this up alongside #361. Seven concepts plus the as-built baseline, each taking a distinct position on the five questions in the body (attention, spine, primary object, GM/player relation, at-the-table surface):

C1 Campaign Desk — keep the campaign container, add a real second level (sidebar / bottom tabs), one layout grammar.
C2 Session First — open on this week; every session is Plan · Play · Review; a Now bar everywhere.
C3 Console & Companion — a phone-first player app and a desktop GM console.
C4 The Feed — the bot's event stream as the in-app spine, every item actionable.
C5 Command Deck — panels that open beside each other, a command palette.
C6 Book & Table — a reading mode between sessions, a full-screen dark table mode during them.
C7 Timeline — the campaign as a scrollable story, sessions expanding in place.

Each draws six structural surfaces (home, campaign, session, prep, at the table, review and wiki) at desktop and phone width in one shared low-fidelity kit, with a dashed line where the first phone screen ends so scroll order is judged, role tags on GM-only regions, and callouts. Secondary surfaces are placed per concept in words. Each is walked through the six journeys and rated; a comparison matrix and a recommendation close the page. Delivered as an artifact for checkpoint A.

Picking this up alongside #361. Seven concepts plus the as-built baseline, each taking a distinct position on the five questions in the body (attention, spine, primary object, GM/player relation, at-the-table surface): - **C1 Campaign Desk** — keep the campaign container, add a real second level (sidebar / bottom tabs), one layout grammar. - **C2 Session First** — open on this week; every session is Plan · Play · Review; a Now bar everywhere. - **C3 Console & Companion** — a phone-first player app and a desktop GM console. - **C4 The Feed** — the bot's event stream as the in-app spine, every item actionable. - **C5 Command Deck** — panels that open beside each other, a command palette. - **C6 Book & Table** — a reading mode between sessions, a full-screen dark table mode during them. - **C7 Timeline** — the campaign as a scrollable story, sessions expanding in place. Each draws six structural surfaces (home, campaign, session, prep, at the table, review and wiki) at desktop and phone width in one shared low-fidelity kit, with a dashed line where the first phone screen ends so scroll order is judged, role tags on GM-only regions, and callouts. Secondary surfaces are placed per concept in words. Each is walked through the six journeys and rated; a comparison matrix and a recommendation close the page. Delivered as an artifact for checkpoint A.

claude-bot commented

2026-09-06 20:01:56 +00:00

Author

Contributor

Checkpoint A is ready for review: https://claude.ai/code/artifact/6e06cfa2-41fb-4f9b-8ed2-1d4d2f891bfe (private artifact; the brief from #361 is the first section).

Eight concept blocks (the as-built baseline plus C1–C7), 96 frames in one shared low-fidelity kit, a "group by surface" switch to compare the same screen across all concepts, the six-journey fit matrix, build-cost and backend-change columns, a "worth borrowing" list per concept, and a recommendation: carry C2 Session First and C6 Book & Table to mockups with C1 Campaign Desk as the control, and borrow C3's Queue, C4's Live card, C5's open-beside and C7's index regardless.

Limits of the medium, recorded by the renderers and worth knowing when reading: stickiness, drawers, bottom sheets, popovers, expand-in-place motion and reading measure can only be labelled, not drawn; two frames per surface means player-instead views sit under the GM content or in a callout; full-screen table modes cannot scroll. Mockups (#363) do not have these limits.

The concept specs, kit and assembly script are kept alongside the inventory so the chosen concepts can be carried into mockups without redrawing.

**Checkpoint A is ready for review:** https://claude.ai/code/artifact/6e06cfa2-41fb-4f9b-8ed2-1d4d2f891bfe (private artifact; the brief from #361 is the first section). Eight concept blocks (the as-built baseline plus C1–C7), 96 frames in one shared low-fidelity kit, a "group by surface" switch to compare the same screen across all concepts, the six-journey fit matrix, build-cost and backend-change columns, a "worth borrowing" list per concept, and a recommendation: carry **C2 Session First** and **C6 Book & Table** to mockups with **C1 Campaign Desk** as the control, and borrow C3's Queue, C4's Live card, C5's open-beside and C7's index regardless. Limits of the medium, recorded by the renderers and worth knowing when reading: stickiness, drawers, bottom sheets, popovers, expand-in-place motion and reading measure can only be labelled, not drawn; two frames per surface means player-instead views sit under the GM content or in a callout; full-screen table modes cannot scroll. Mockups (#363) do not have these limits. The concept specs, kit and assembly script are kept alongside the inventory so the chosen concepts can be carried into mockups without redrawing.

claude-bot commented

2026-09-06 22:22:40 +00:00

Author

Contributor

Checkpoint A amended at the owner's request: two audience pages, and the in-person table as a sixth question every concept answers.

For players (the beta testers): https://claude.ai/code/artifact/f30840a2-f03b-4fc7-be11-81c7df827548 — the seven concepts plus today's app, four screens each, phone first (home, a session, game night, afterwards). On game night the desktop frame is repurposed as the shared screen at an in-person table, so the in-person face of every concept is drawn. Plain-language narrative per concept, callouts written for a player, a five-point scale and a comment box under each, two picks and four questions at the end (how they learn when we play, what they look at during a game, what they would want on a room screen versus their phone, what they would fix first).
For GMs: https://claude.ai/code/artifact/fd850062-665a-4951-bef8-af77dfb6734c — the six GM screens per concept, an in-person note per concept, the same form with GM questions.
The design lead's page (brief, all concepts, matrix, recommendation, links to both): https://claude.ai/code/artifact/6e06cfa2-41fb-4f9b-8ed2-1d4d2f891bfe

Answers save to each page's own store (feedback/<audience>-<id>, keyed by the name the person gives, editable on return from the same device); a readout comes back as a table on request. Where saving is unavailable the page offers "Copy my answers" instead.

The in-person model and each concept's shared-screen answer are recorded in #542 and in the brief (#361).

**Checkpoint A amended** at the owner's request: two audience pages, and the in-person table as a sixth question every concept answers. - **For players** (the beta testers): https://claude.ai/code/artifact/f30840a2-f03b-4fc7-be11-81c7df827548 — the seven concepts plus today's app, four screens each, phone first (home, a session, game night, afterwards). On game night the desktop frame is repurposed as **the shared screen at an in-person table**, so the in-person face of every concept is drawn. Plain-language narrative per concept, callouts written for a player, a five-point scale and a comment box under each, two picks and four questions at the end (how they learn when we play, what they look at during a game, what they would want on a room screen versus their phone, what they would fix first). - **For GMs**: https://claude.ai/code/artifact/fd850062-665a-4951-bef8-af77dfb6734c — the six GM screens per concept, an in-person note per concept, the same form with GM questions. - **The design lead's page** (brief, all concepts, matrix, recommendation, links to both): https://claude.ai/code/artifact/6e06cfa2-41fb-4f9b-8ed2-1d4d2f891bfe Answers save to each page's own store (`feedback/<audience>-<id>`, keyed by the name the person gives, editable on return from the same device); a readout comes back as a table on request. Where saving is unavailable the page offers "Copy my answers" instead. The in-person model and each concept's shared-screen answer are recorded in #542 and in the brief (#361).

rbrooks referenced this issue from a commit

2026-09-07 02:45:21 +00:00

feat(tools): self-hosted feedback site for the v4.4.0 design review, plus the design record (#361, #362, #535)

claude-bot referenced this issue

2026-09-07 02:45:37 +00:00

feat(tools): self-hosted feedback site for the v4.4.0 design review, plus the design record (#361, #362) #546

claude-bot commented

2026-09-07 02:45:39 +00:00

Author

Contributor

The beta testers cannot open the artifacts (no claude.ai accounts), so the two audience pages now also exist as a self-hosted site: tools/feedback-site/ (PR to follow), a separate container with the same players' and GMs' pages, a share-key-gated form, a JSON-file store, and an admin readout with CSV. Running on the dev box at port 18090; public exposure needs a hostname (a DNS record plus one Caddy entry) and is the owner's call. The artifact pages stay as the owner's and the design lead's copies; the site is where the players answer.

**The beta testers cannot open the artifacts** (no claude.ai accounts), so the two audience pages now also exist as a **self-hosted site**: `tools/feedback-site/` (PR to follow), a separate container with the same players' and GMs' pages, a share-key-gated form, a JSON-file store, and an admin readout with CSV. Running on the dev box at port 18090; public exposure needs a hostname (a DNS record plus one Caddy entry) and is the owner's call. The artifact pages stay as the owner's and the design lead's copies; the site is where the players answer.

claude-bot referenced this issue from a commit

2026-09-07 03:01:23 +00:00

Merge PR #546: feat(tools): self-hosted feedback site for the v4.4.0 design review, plus the design record (#361, #362)

claude-bot commented

2026-09-07 06:52:00 +00:00

Author

Contributor

The feedback site is public at https://questboard-dev.rhoving.com/feedback/ (Caddy handle_path on the dev hostname → the container on docker-test:18090; iac-repo PR #405 applied with the Caddy playbook, dry-run first; the dev Quest Board itself is unaffected). Verified end to end from outside: pages 200, /feedback redirects to /feedback/, a write without the share key is refused, a write with it is stored, and the admin readout and CSV need the admin key.

Two owner decisions applied at the same time (PR #550):

The players' page no longer carries the in-person material: game night is the phone drawing only, the in-person note and question are gone, and a question about opening the app between games takes the slot.
The GMs' page gains a seventh section per concept, "The shared screen, if the table is in one room", so GMs give that feedback; the in-person note stays there.

The artifact copies of both pages were republished to match. Links for the testers carry the share key (/feedback/players?k=…, /feedback/gms?k=…); the readout is /feedback/admin with the admin key, or the JSON files in the container's volume.

**The feedback site is public at https://questboard-dev.rhoving.com/feedback/** (Caddy `handle_path` on the dev hostname → the container on docker-test:18090; iac-repo PR #405 applied with the Caddy playbook, dry-run first; the dev Quest Board itself is unaffected). Verified end to end from outside: pages 200, `/feedback` redirects to `/feedback/`, a write without the share key is refused, a write with it is stored, and the admin readout and CSV need the admin key. Two owner decisions applied at the same time (PR #550): - The **players'** page no longer carries the in-person material: game night is the phone drawing only, the in-person note and question are gone, and a question about opening the app between games takes the slot. - The **GMs'** page gains a seventh section per concept, "The shared screen, if the table is in one room", so GMs give that feedback; the in-person note stays there. The artifact copies of both pages were republished to match. Links for the testers carry the share key (`/feedback/players?k=…`, `/feedback/gms?k=…`); the readout is `/feedback/admin` with the admin key, or the JSON files in the container's volume.

Rows
Columns

[UX] Wireframe exploration — 5-10 concepts across every surface #362

What

Acceptance criteria