# I2PChat Roadmap

**Repository version:** `1.3.0` (see [`VERSION`](../VERSION)).

The milestone sections below are the **original product plan** (0.6.x → `1.0.0`). That work has largely **shipped** across releases through **1.3.x**; see [`releases/README.md`](releases/README.md) for versioned notes. **`SessionManager`** (per-peer transport lifecycle) landed in **v1.2.6**. **Text groups** shipped in **v1.3.0** (see [RELEASE_1.3.0.md](releases/RELEASE_1.3.0.md)); code under `i2pchat/groups/`, `i2pchat/storage/group_store.py`; user-facing behavior is in [MANUAL_EN.md](MANUAL_EN.md) / [MANUAL_RU.md](MANUAL_RU.md).

## Snapshot (1.2.x)

| Area | Notes |
|------|--------|
| Conversation UX | Saved peers, drafts, unread, delivery states, contacts, search |
| Trust / offline | TOFU, key-change flows, BlindBox, diagnostics, retry policies |
| Portability | PyInstaller zips, optional bundled `i2pd`, `.deb` / Homebrew / winget / AUR templates |
| Core transport | `I2PChatCore` + **`SessionManager`** (outbound policy, telemetry, reconnect) |
| Groups | Text groups (**v1.3.0**): live + pairwise BlindBox; see **PROTOCOL.md** § Text groups, **RELEASE_1.3.0.md** |

## Guiding principles

- Keep protocol and transport changes minimal unless they clearly unlock user value.
- Prefer UX improvements that expose existing capabilities before adding new complexity.
- Preserve the privacy-first character of the project.
- Treat **multi-device real-time sync**, **plugins**, and **broad automation** as higher-risk, longer-horizon work.

## Original milestones (reference)

The subsections retain the **original planning text** for context. Intended outcomes listed there match what the app does today unless a backlog item says otherwise ([`ISSUE_BACKLOG.md`](ISSUE_BACKLOG.md)).

## 0.6.5 - UX polish

Goal: make the application easier to use every day without major architectural
changes.

Planned focus areas:

- Per-contact message drafts
- Unread counters and tray/window indicators
- Simplified connection and delivery status text
- Message context actions (`Copy`, `Reply`, attachment actions where applicable)
- Notification preferences (sound, quiet mode, hide message text)
- Better attachment send feedback and clearer transfer errors

Release outcome:

- users do not lose text while switching context;
- users can see which chats need attention;
- connection and delivery behavior becomes easier to understand;
- common message interactions feel more natural.

## 0.7.0 - Contacts and conversations

Goal: evolve the app from a single-peer session client into a conversation-based
messenger.

Planned focus areas:

- Contacts sidebar / conversation list
- Local names and notes for contacts
- Last active conversation restore
- Conversation previews (last message, last activity, unread state)
- Search within the current conversation history
- Basic contact details / trust card MVP

Release outcome:

- users can move between saved contacts quickly;
- message history becomes part of normal navigation;
- contact identity is easier to manage without losing the real I2P address.

## 0.8.0 - Trust, delivery, offline clarity

Goal: make the product's strongest differentiators understandable in the UI.

Planned focus areas:

- Outgoing message queue UI
- Per-message delivery states (`sending`, `queued`, `delivered`, `failed`)
- Key change warning flow
- Clearer trust UX for pinning and lock-to-peer
- BlindBox diagnostics screen
- Retry for failed sends where behavior is safe and predictable

Release outcome:

- users can tell whether a message was sent live or queued offline;
- trust changes are surfaced explicitly instead of being hidden in low-level state;
- offline delivery failures become diagnosable.

## 0.9.0 - Portability, privacy, hardening

Goal: make the application ready for longer-term real-world use.

Planned focus areas:

- Encrypted profile export
- Profile import / restore flow
- Encrypted history export and import
- History retention controls
- Privacy mode (notification hiding, optional local lock flow)
- Drag-and-drop attachments
- Better transfer retry and media/file UX
- Protocol and transfer hardening in tests and diagnostics

Release outcome:

- users can export and restore profile/state backups through encrypted bundles;
- local privacy controls become stronger;
- reliability improves ahead of `1.0.0`.

## 1.0.0 - Stable niche release

Goal: ship a stable, privacy-focused release with a clear conversation model and
predictable delivery/trust UX.

Minimum expected state:

- convenient dialog and contact workflow;
- searchable local history;
- understandable delivery states;
- explicit trust/key-change UX;
- usable offline delivery diagnostics;
- practical backup and restore paths;
- better reliability coverage for protocol and transfers.

Release outcome:

- profile backups and history backups can be exported and restored in encrypted form;
- local history retention is configurable by count and age;
- privacy mode is available as a quick local safeguard;
- drag-and-drop attachments match existing send actions;
- pre-1.0 reliability coverage is strong enough for a stable niche release.

## Post-1.2.x / future candidates

**Shipped in v1.3.0:** text groups (multi-member chat, live + pairwise BlindBox), Saved-peers inbound model, parallel live sessions — see [RELEASE_1.3.0.md](releases/RELEASE_1.3.0.md).

Longer-horizon items (not committed on a fixed timeline):

- **real-time multi-device sync** (same identity, multiple online clients);
- **plugin or scripting** surfaces beyond local preferences;
- **broader automation** beyond built-in settings.

## Tracking

Issue-sized items: [`ISSUE_BACKLOG.md`](ISSUE_BACKLOG.md).

Per-version release notes: [`releases/README.md`](releases/README.md).