Repository navigation
docs: pistes de rétention, ADR 0080 à 0083 et maquettes - #51
Merged
Merged
Conversation
Off-navigation working page (like logo.html and cartes-visuels.html): diagnosis of what slows the game today, the measurement groundwork to ship first (J5, scheduled aggregates, J7), and four ideas, each with a demo built on Rosette and Argile words extracted from the Content repo (legendaries excluded): - the reserve as a prefix x suffix grid that remembers past attempts; - the square, a daily table quest solved by analogy; - the choice fold, two bricks offered and one kept; - the constellation, the codex drawn as a network of families. None of them is decided; each needs its own ADR. The page also flags a gap between game-design.md (rejected ideas) and the App roadmap on ingredient words whose brick is never consumed. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0151stDRaZV3ScbhwvwjrfQ7
The owner reviewed the retention working page and rejected the choice fold, the reserve shown as a prefix x suffix grid, the cross-word announcement at reveal, chronology and example-based ricochets, and special out-of-calendar fascicules. Add them to the rejected list of game-design.md and mark the special fascicules track as rejected in potentiel-evolution.md, so they only come back through an ADR. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0151stDRaZV3ScbhwvwjrfQ7
Second version of the retention page, following the owner's review: - attempt memory: small badges on reserve bricks already tried with the brick on the table, kept by the server and wiped at each release (replaces the prefix x suffix grid); - collection quests per fascicle: all prefixes, all suffixes, all words of a rarity (difficulty is never shown, ADR 0047; owner to decide); - constellation laid out on two banks, no overlap by construction, bricks never move when a word arrives; - ongoing investigations: paid hints stay visible until discovery. Demos use only released fascicles (1 to 6), legendaries excluded. The page also lists gaps verified in code, three of which reveal how many legendaries a fascicle holds (last milestone threshold, legendary word shown once its brick is found, codex numbering gaps). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0151stDRaZV3ScbhwvwjrfQ7
…situ Owner review, second round: - attempt memory: bricks already tried with the brick on the table are greyed and moved to the end of the reserve, recomputed whenever the table changes (failures, near misses and words already in the codex); the badges and the notebook variant are dropped; - constellation: shown in situ inside the app shell from the mockups, as a codex view next to Tout, Récentes and Familles, on a phone and a desktop frame side by side; the module now mounts several instances; - screenshots of the collection quests and the constellation, used by the App issues that hand both tracks over. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0151stDRaZV3ScbhwvwjrfQ7
This was referenced Oct 23, 2026
Merged
…ent-retention-lyk0n0
…them up ADR 0080 keeps bought hints readable as ongoing investigations, ADR 0081 remembers failed pairs until the next release (partly superseding the "proposals are never stored" sentence of ADR 0027), ADR 0082 adds per-fascicule collection quests by rarity, and ADR 0083 adds the two-bank Réseau view of the codex. All four are proposed. The index, ADR 0027 status, XREFS and adr.html follow; game-design.md and the data model (new `essai` table) are aligned, and the hints section now reflects ADR 0051. Mock-ups I-09 (reserve memory and investigations) and I-12 (hints à la carte) are redrawn, I-35 (Réseau view) is new, and the interface inventory lists it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_0151stDRaZV3ScbhwvwjrfQ7
Lunik
changed the title
Lunik
marked this pull request as ready for review
October 24, 2026 14:09
This was referenced Oct 24, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Résumé
Cette PR consigne les quatre pistes de rétention retenues par le propriétaire. Elle contient la page de travail et ses démos, quatre ADR (proposés) et les maquettes. Le code correspondant est dans EtymoLogique/App#92, #93, #94 et #95, et dans EtymoLogique/Content#39.
Ce que contient la PR
docs/pistes-retention.html: la page de travail hors navigation, avec un diagnostic, les fondations à livrer d’abord, une démo jouable par piste, la constellation en situation (téléphone et ordinateur), un ordre de livraison et les écarts relevés en chemin. Les démos utilisent de vrais mots de Rosette et d’Argile, sans légendaire.XREFS,adr.htmlrégénéré ;game-design.md: mémoire des essais, vue Réseau, quêtes de collection, et la section « Indices » réécrite selon l’ADR 0051 ;essai, « Ce qui n’est jamais stocké » mis à jour.interfaces.htmlet àAGENTS.md.docs/assets/pistes-retention/.Écarts relevés, non corrigés ici
game-design.mdécarte « un mot découvert qui deviendrait une brique sans compteur d’exemplaires », alors que la roadmap de l’App en fait une brique qui ne se consomme jamais. Aucun ADR ne tranche.design-system.html(§ 3.5) montre encore une « piste » de sens sur une carte inconnue, que l’ADR 0051 a retirée.Vérifications
build_adr: 77 ADR écrits.check_identity --base origin/master: 0 erreur, 0 avertissement sur les lignes modifiées.rule-verification:master.🤖 Generated with Claude Code
https://claude.ai/code/session_0151stDRaZV3ScbhwvwjrfQ7