Jordan-Hall / Jordan-Hall/browser

[P3][MEM-02] Opt-in adaptive patterns and declared lenses

Open
#90 2 comments 0 reactions 0 assignees View on GitHub
Dominant language
No language data
Stars
0
Forks
0
PR merge metrics
No merged PRs in 30d

Description

Programme: #1
Epic: #29

## Objective
Let the runtime learn presentation/relevance patterns and declared viewpoints without turning inferred beliefs into hidden truth filters.

## Scope
- Opt-in pattern inference from repeated explicit corrections/layout choices/workflows.
- Separate inferred pattern records from confirmed preferences.
- Sensitivity classifier/policy preventing durable sensitive inferences without deliberate consent.
- “Why did this adapt?” explanations and one-click correction/disable.
- Declared lenses/viewpoints as explicit user-selected ranking/presentation inputs.
- Stable layout adaptation through UI-03 review/lock mechanics.
- Conflict handling when a new pattern contradicts a locked/explicit preference.

## Product rules
- Personalization changes presentation/relevance, not source evidence.
- A declared lens is labeled and cannot silently remove contradictory credible evidence.
- Sensitive political/health/religious/etc. inferences must not become durable hidden preferences by default.

## Acceptance criteria
- [ ] Every adaptive change can identify the preference/pattern that caused it.
- [ ] Inferred patterns remain distinguishable from user-confirmed preferences.
- [ ] Sensitive inferred beliefs cannot silently become durable facts/preferences.
- [ ] User correction/lock overrides future adaptation.
- [ ] Declared lenses retain contradictory evidence/source access.
- [ ] Disabling adaptive patterns leaves explicit preferences and workspace functionality intact.

## Dependencies
- MEM-01
- UI-03

**First phase:** P3
**Maturity target:** P5
**Owner:** workspace-ui-data

Contributor guide

No contributing guide indexed for this repository

Research direction

No files, tests, or entry points are named. Start by reading MEM-01 and UI-03, then define how inferred patterns, confirmed preferences, sensitivity policy, explanations, lenses, and locks interact; done means all acceptance criteria are covered without hiding contradictory evidence or breaking explicit preferences.

Written by the indexing model from the issue text.

Assessment

Domain
ai, design
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Active
Clarity
Mostly clear
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.