# Addendum: Parked Features & Architectural Notes

## Source

Derived from brainstorming session: `_bmad-output/brainstorming/brainstorm-wordle-clone-2026-07-07/`

## Architectural Centerpiece

The computer solver engine is the load-bearing system. It pre-computes attempt counts for every word in the bank using multiple algorithms. That data drives: word difficulty ranking (v1), word tier assignment (post-v1), adaptive difficulty (post-v1), percentile ranking (post-v1), streak quality weighting (post-v1), solvability validation (v1), and the computer-guess subgame (post-v1). Build the solver once, unlock features incrementally.

## Parked Features (Post-V1)

These are prioritized by friend feedback after v1 ships. Listed in rough dependency order on the solver engine.

### Tier 1 — Builds directly on solver data

| Feature | Description | Depends on |
|---|---|---|
| Adaptive difficulty | Track per-word global averages; words self-sort into difficulty tiers; player MMR drifts with performance | Solver baseline, per-word stats |
| Percentile ranking | "Top 15%" instead of "#14,327" on results screen | Per-word global stats |
| Streak quality weighting | Streaks on harder words count more than streaks on easy words | Word difficulty tiers |
| Daily word + infinite after-play | One shared daily word, then optional endless play | Word bank, game loop |

### Tier 2 — Extends the experience

| Feature | Description | Depends on |
|---|---|---|
| Standardized hints | Same hint given to every player on attempt 5 (weak) and 6 (stronger) | Per-word hint definitions |
| Post-solve learning layer | Definition, etymology, example sentence, picture after solving | External dictionary/image APIs |
| Computer-guess subgame | Player picks word, computer solves — teaches strategy | Solver engine |

### Tier 3 — New game modes

| Feature | Description | Depends on |
|---|---|---|
| Real-time 2-player | Both players guess same word simultaneously | Multiplayer infrastructure |
| Variable word length | 3-letter speed rounds, 7-letter marathon | Word bank expansion |
| Position-blind hard mode | Yellow/gray only, no green position hints | Game mode framework |

## Permanently Rejected

- **User-submitted words** — can't trust community adequacy/moderation
- **Time-penalty mechanics** — doesn't add value
- **Mid-game hints (before attempt 5)** — breaks statistical comparability
- **No-color-feedback mode** — too few attempts for 5-letter deduction without position hints

## Prioritization Process

Post-v1: put the game in front of friends → ask which feature they want first → build that → repeat.