# IARPG-OPS-2 2.0.11-wip Release Notes

## Status

**Work in progress.** This release improves first-click clarity and adds optional guided walkthroughs to the three densest browser-local tools. It does not claim that representative visitors understand the site or complete tasks successfully: no human participants were recruited or observed in this round.

## Release focus

The 2.0.10 release established a clear definition of IARPG and a canonical `/start` guide. The 2.0.11 round addresses the remaining ambiguity between product identity and first action:

- **Game** becomes **Game Concept** so visitors are less likely to infer that a live persistent game client is available.
- **Create** becomes **Create Missions** so the object of the action is explicit.
- The first game-oriented destination now opens `/how-to-play`, which explains the proposed player loop and its distinction from the publication toolchain.
- Create Missions, Operation Package, and Simulate now provide optional five-step walkthroughs.

## Guided walkthroughs

The shared walkthrough layer is manually launched, dismissible, restartable, and non-blocking. It uses native dialogs and points to existing controls without taking ownership of tool state.

Each walkthrough:

- never auto-opens or gates direct expert access;
- never loads an example, imports a file, fills a field, starts a simulation, saves, exports, or changes work data;
- performs no `fetch`, `XMLHttpRequest`, `WebSocket`, or `sendBeacon` call;
- keeps optional completion metadata only in browser-local storage;
- remains usable when local storage is blocked;
- restores focus to its launcher after dismissal;
- supports Escape, explicit Exit Tour, previous/next controls, and backdrop dismissal;
- is constrained at narrow viewports and high zoom;
- respects reduced motion, reduced transparency, forced colors, light/dark themes, print, and offline use;
- does not add another sticky surface.

The centered `.menu-strip` remains the only sticky element.

## Onboarding evidence boundary

A new deterministic onboarding task-routing validator checks the rendered publication for:

- the plain-language IARPG definition;
- current-versus-future product truth;
- first-time and audience routes;
- the player-loop versus publication-toolchain distinction;
- direct mission, creation, simulation, verification, merge, publication, international-methodology, and research routes;
- walkthrough markup and progressive-enhancement safeguards;
- absence of walkthrough network APIs.

The generated evidence and the new public `/onboarding-evidence` page explicitly record:

- `human_participants: 0`;
- `moderated_human_study_status: not-run`;
- a truth boundary that excludes comprehension, confidence, preference, and participant task success.

The Answer Center now includes a direct answer explaining that no real first-time visitor study has been completed.

## Discovery, AEO, and GEO continuity

The new evidence route and implementation report are included in search, sitemap, feed, answer routing, well-known discovery, documentation manifests, checksums, UAI memory, and public UAI mirrors. Canonical HTML remains the primary answer and citation surface; machine records remain supplementary routing aids.

## Memory and documentation

- `.uai/` remains active first-load memory.
- Durable bodies remain under `docs/long-term-memory/`.
- All canonical reports remain under `docs/long-term-memory/reports/`.
- Added `docs/long-term-memory/reports/onboarding-comprehension-and-guided-walkthrough-report.md`.
- Added a corresponding pointer-ledger record and deep links from current hot-memory records.
- Current inventory is 189 canonical reports, 198 authoritative durable pointer records, 255 preserved source files, and 95 active UAI records.

## Validation results and boundary

The final local completion suite records:

| Area | Result |
|---|---:|
| Deterministic onboarding task routing | **15/15 passed** |
| Site identity and first-visit contract | **24/24 passed** |
| Managed Chromium fixture groups | **15/15 passed** |
| Production-edge aggregate | **59/59 passed** |
| Isolated Apache host groups | **22/22 passed** |
| HTTPS sitemap responses and metadata | **338/338 passed** |
| Structural accessibility | **89 routes, 0 findings** |
| Local delivery budgets | **10/10 passed** |
| PHP lint | **90/90 passed** |
| JavaScript syntax | **6/6 passed** |
| Python compilation | **10/10 passed** |
| JSON parsing | **310/310 passed before final validation artifacts** |
| XML parsing | **3/3 passed** |
| Local JSON Schema records | **64/64 passed** |
| IARPG-CJSON-1 vectors | **20/20 passed** |
| Memory links and anchors | **2,518 links and 2,136 anchors** |
| Active UAI records and mirrors | **95/95 byte-identical** |
| Canonical reports / durable pointers | **189 / 198** |
| Preserved research sources | **255/255 verified** |
| Documentation checksums | **728/728 verified** |

The browser fixture suite now includes all three walkthroughs, local-only completion, zero work-data mutation, zero network calls, narrow-viewport containment, reduced motion, blocked storage, required viewports, high zoom, theme behavior, media fallbacks, round-trip fixtures, and offline navigation. Controlled local TLS negotiated TLS 1.3 with certificate verification. These are direct local results, not participant, public-edge, native Firefox, native screen-reader, or field-performance evidence.

The release remains WIP because the following evidence is still unavailable:

- moderated comprehension, first-click, and task-completion results from unfamiliar participants;
- unrestricted navigated Chromium against an externally reachable host;
- native Firefox runtime behavior;
- native assistive-technology speech output;
- externally reachable production-matched HTTPS/CDN or reverse-proxy behavior;
- field Core Web Vitals and representative real-device results;
- host-owned Google, Bing, IndexNow, production crawler-log, and external generative-citation evidence.

## Best next action

Run the real moderated study defined in the public onboarding evidence and the 2.0.12 prompt. Use observed results to decide whether the labels and tours help, need revision, or should be reduced. Do not infer participant success from automated checks.

## Next prompt

See `NEXT-PROMPT-IARPG-OPS-2-2.0.12-wip.md`.