refactor: Change firehose from a drifting overlay to a fixed bottom strip and message display from a top-pinned section to a centered overlay.

Reviewed-on: #11

Mainline Renderer + ntfy Message Queue for ESP32.md

@@ -3,29 +3,29 @@
 mainline\.py does heavy work unsuitable for ESP32: 25\+ HTTPS/TLS RSS feeds, OTF font rasterization via Pillow, Google Translate API calls, and complex text layout\. Simultaneously, messages arriving on `ntfy.sh/klubhaus_terminal_mainline` need to interrupt the news ticker on the same device\.
 ## Architecture: Server \+ Thin Client
 Split the system into two halves that are designed together\.
-**Server (mainline\.py `--serve` mode, runs on any always\-on machine)**
+**Server \(mainline\.py `--serve` mode, runs on any always\-on machine\)**
 * Reuses existing feed fetching, caching, content filtering, translation, and Pillow font rendering pipeline — no duplication\.
-* Pre\-renders each headline into a 1\-bit bitmap strip (the OTF→half\-block pipeline already produces this as an intermediate step in `_render_line()`)\.
+* Pre\-renders each headline into a 1\-bit bitmap strip \(the OTF→half\-block pipeline already produces this as an intermediate step in `_render_line()`\)\.
 * Exposes a lightweight HTTP API the ESP32 polls\.
-**ESP32 thin client (Arduino sketch)**
-* Polls the mainline server for pre\-rendered headline bitmaps over plain HTTP (no TLS needed if on the same LAN)\.
-* Polls `ntfy.sh/klubhaus_terminal_mainline` directly for messages, reusing the proven `NetManager::httpGet()` \+ JSON parsing pattern from DoorbellLogic (`DoorbellLogic.cpp:155-192`)\.
-* Manages scrolling, gradient coloring, and glitch effects locally (cheap per\-frame GPU work)\.
+**ESP32 thin client \(Arduino sketch\)**
+* Polls the mainline server for pre\-rendered headline bitmaps over plain HTTP \(no TLS needed if on the same LAN\)\.
+* Polls `ntfy.sh/klubhaus_terminal_mainline` directly for messages, reusing the proven `NetManager::httpGet()` \+ JSON parsing pattern from DoorbellLogic \(`DoorbellLogic.cpp:155-192`\)\.
+* Manages scrolling, gradient coloring, and glitch effects locally \(cheap per\-frame GPU work\)\.
 * When an ntfy message arrives, the scroll is paused and the message takes over the display — same interrupt pattern as the doorbell's ALERT→DASHBOARD flow\.
-## Server API (mainline repo)
-New file: `serve.py` (or `--serve` mode in mainline\.py)\.
+## Server API \(mainline repo\)
+New file: `serve.py` \(or `--serve` mode in mainline\.py\)\.
 Endpoints:
 * `GET /api/headlines` — returns JSON array of headline metadata: `[{"id": 0, "src": "Nature", "ts": "14:30", "width": 280, "height": 16, "bitmap": "<base64 1-bit packed>"}]`\. Bitmaps are 1\-bit\-per\-pixel, row\-major, packed 8px/byte\. The ESP32 applies gradient color locally\.
 * `GET /api/config` — returns `{"count": 120, "version": "...", "mode": "news"}` so the ESP32 knows what it's getting\.
 * `GET /api/health` — `{"ok": true, "last_fetch": "...", "headline_count": 120}`
-The server renders at a configurable target width (e\.g\. 800px for Board 3, 320px for Boards 1/2) via a `--width` flag or query parameter\. Height is fixed per headline by the font size\.
-The server refreshes feeds on a timer (reusing `_SCROLL_DUR` cadence or a longer interval), re\-renders, and serves the latest set\. The ESP32 polls `/api/headlines` periodically (e\.g\. every 60s) and swaps in the new set\.
-## Render pipeline (server side)
+The server renders at a configurable target width \(e\.g\. 800px for Board 3, 320px for Boards 1/2\) via a `--width` flag or query parameter\. Height is fixed per headline by the font size\.
+The server refreshes feeds on a timer \(reusing `_SCROLL_DUR` cadence or a longer interval\), re\-renders, and serves the latest set\. The ESP32 polls `/api/headlines` periodically \(e\.g\. every 60s\) and swaps in the new set\.
+## Render pipeline \(server side\)
 The existing `_render_line()` in mainline\.py already does:
 1. `ImageFont.truetype()` → `ImageDraw.text()` → grayscale `Image`
 2. Resize to target height
-3. Threshold to 1\-bit (the `thr = 80` step)
-For the server, we stop at step 3 and pack the 1\-bit data into bytes instead of converting to half\-block Unicode\. This is the exact same pipeline, just with a different output format\. The `_big_wrap()` and `_lr_gradient()` logic stays on the server for layout; gradient *coloring* moves to the ESP32 (it's just an index lookup per pixel column)\.
+3. Threshold to 1\-bit \(the `thr = 80` step\)
+For the server, we stop at step 3 and pack the 1\-bit data into bytes instead of converting to half\-block Unicode\. This is the exact same pipeline, just with a different output format\. The `_big_wrap()` and `_lr_gradient()` logic stays on the server for layout; gradient *coloring* moves to the ESP32 \(it's just an index lookup per pixel column\)\.
 ## ESP32 client
 ### State machine
 ```warp-runnable-command
@@ -35,19 +35,19 @@ BOOT → SCROLL ⇄ MESSAGE
 * **BOOT** — WiFi connect, initial headline fetch from server\.
 * **SCROLL** — Vertical scroll through pre\-rendered headlines with local gradient \+ glitch\. Polls server for new headlines periodically\. Polls ntfy every 15s\.
 * **MESSAGE** — ntfy message arrived\. Scroll paused, message displayed\. Auto\-dismiss after timeout or touch\-dismiss\. Returns to SCROLL\.
-* **OFF** — Backlight off after inactivity (polling continues in background)\.
+* **OFF** — Backlight off after inactivity \(polling continues in background\)\.
 ### ntfy integration
 The ESP32 polls `https://ntfy.sh/klubhaus_terminal_mainline/json?since=20s&poll=1` on the same 15s interval as the doorbell polls its topics\. When a message event arrives:
 1. Parse JSON: `{"event": "message", "title": "...", "message": "..."}`
 2. Save current scroll position\.
 3. Transition to MESSAGE state\.
-4. Render message text using the display library's built\-in fonts (messages are short, no custom font needed)\.
-5. After `MESSAGE_TIMEOUT_MS` (e\.g\. 30s) or touch, restore scroll position and resume\.
-This is architecturally identical to `DoorbellLogic::onAlert()` → `dismissAlert()`, just with different content\. The ntfy polling runs independently of the server connection, so messages work even if the mainline server is offline (the device just shows the last cached headlines)\.
+4. Render message text using the display library's built\-in fonts \(messages are short, no custom font needed\)\.
+5. After `MESSAGE_TIMEOUT_MS` \(e\.g\. 30s\) or touch, restore scroll position and resume\.
+This is architecturally identical to `DoorbellLogic::onAlert()` → `dismissAlert()`, just with different content\. The ntfy polling runs independently of the server connection, so messages work even if the mainline server is offline \(the device just shows the last cached headlines\)\.
 ### Headline storage
-* Board 3 (8 MB PSRAM): store all ~120 headline bitmaps in PSRAM\. At 800px × 16px × 1 bit = 1\.6 KB each → ~192 KB total\. Trivial\.
-* Boards 1/2 (PSRAM TBD): at 320px × 16px = 640 bytes each → ~77 KB for 120 headlines\. Fits if PSRAM is present\. Without PSRAM, keep ~20 headlines in a ring buffer (~13 KB)\.
-### Gradient coloring (local)
+* Board 3 \(8 MB PSRAM\): store all ~120 headline bitmaps in PSRAM\. At 800px × 16px × 1 bit = 1\.6 KB each → ~192 KB total\. Trivial\.
+* Boards 1/2 \(PSRAM TBD\): at 320px × 16px = 640 bytes each → ~77 KB for 120 headlines\. Fits if PSRAM is present\. Without PSRAM, keep ~20 headlines in a ring buffer \(~13 KB\)\.
+### Gradient coloring \(local\)
 The 12\-step ANSI gradient in mainline\.py maps to 12 RGB565 values:
 ```warp-runnable-command
 const uint16_t GRADIENT[] = {
@@ -68,8 +68,8 @@ mainline.py          (existing, unchanged)
 serve.py             (new — HTTP server, imports mainline rendering functions)
 klubhaus-doorbell-hardware.md  (existing)
 ```
-`serve.py` imports the rendering functions from mainline\.py (after refactoring them into importable form — they're currently top\-level but not wrapped in `if __name__`)\.
-### klubhaus\-doorbell repo (or mainline repo under firmware/)
+`serve.py` imports the rendering functions from mainline\.py \(after refactoring them into importable form — they're currently top\-level but not wrapped in `if __name__`\)\.
+### klubhaus\-doorbell repo \(or mainline repo under firmware/\)
 ```warp-runnable-command
 boards/esp32-mainline/
 ├── esp32-mainline.ino    Main sketch
@@ -79,31 +79,31 @@ boards/esp32-mainline/
 ├── HeadlineStore.h/.cpp  Bitmap ring buffer in PSRAM
 └── NtfyPoller.h/.cpp     ntfy.sh polling (extracted from DoorbellLogic pattern)
 ```
-The display driver is reused from the target board (e\.g\. `DisplayDriverGFX` for Board 3)\. `MainlineLogic` replaces `DoorbellLogic` as the state machine but follows the same patterns\.
+The display driver is reused from the target board \(e\.g\. `DisplayDriverGFX` for Board 3\)\. `MainlineLogic` replaces `DoorbellLogic` as the state machine but follows the same patterns\.
 ## Branch strategy recommendation
 The work spans two repos and has clear dependency ordering\.
-### Phase 1 — Finish current branch (mainline repo)
-**Branch:** `feat/arduino` (current)
+### Phase 1 — Finish current branch \(mainline repo\)
+**Branch:** `feat/arduino` \(current\)
 **Content:** Hardware spec doc\. Already done\.
 **Action:** Merge to main when ready\.
-### Phase 2 — Server renderer (mainline repo)
-**Branch:** `feat/renderer` (branch from main after Phase 1 merges)
+### Phase 2 — Server renderer \(mainline repo\)
+**Branch:** `feat/renderer` \(branch from main after Phase 1 merges\)
 **Content:**
-* Refactor mainline\.py rendering functions to be importable (extract from `__main__` guard)
+* Refactor mainline\.py rendering functions to be importable \(extract from `__main__` guard\)
 * `serve.py` — HTTP server with `/api/headlines`, `/api/config`, `/api/health`
-* Bitmap packing utility (1\-bit row\-major)
-**Why a separate branch:** This changes mainline\.py's structure (refactoring for imports) and adds a new entry point\. It's a self\-contained, testable unit — you can verify the API with `curl` before touching any Arduino code\.
-### Phase 3 — ESP32 client (klubhaus\-doorbell repo, or mainline repo)
+* Bitmap packing utility \(1\-bit row\-major\)
+**Why a separate branch:** This changes mainline\.py's structure \(refactoring for imports\) and adds a new entry point\. It's a self\-contained, testable unit — you can verify the API with `curl` before touching any Arduino code\.
+### Phase 3 — ESP32 client \(klubhaus\-doorbell repo, or mainline repo\)
 **Branch:** `feat/mainline-client` in whichever repo hosts it
 **Content:**
 * `MainlineLogic` state machine
 * `HeadlineStore` bitmap buffer
 * `NtfyPoller` for `klubhaus_terminal_mainline`
 * Board\-specific sketch for the target board
-**Depends on:** Phase 2 (needs a running server to test against)
+**Depends on:** Phase 2 \(needs a running server to test against\)
 **Repo decision:** If you have push access to klubhaus\-doorbell, it fits naturally as a new board target alongside the existing doorbell sketches — it reuses `NetManager`, `IDisplayDriver`, and the vendored display libraries\. If not, put it under `mainline/firmware/` and vendor the shared KlubhausCore library\.
 ### Merge order
-1. `feat/arduino` → main (hardware spec)
-2. `feat/renderer` → main (server)
-3. `feat/mainline-client` → main in whichever repo (ESP32 client)
+1. `feat/arduino` → main \(hardware spec\)
+2. `feat/renderer` → main \(server\)
+3. `feat/mainline-client` → main in whichever repo \(ESP32 client\)
 Each phase is independently testable and doesn't block the other until Phase 3 needs a running server\.

README.md

@@ -2,54 +2,112 @@
 
 > *Digital consciousness stream. Matrix aesthetic · THX-1138 hue.*
 
-A full-screen terminal news ticker that renders live global headlines in large OTF-font block characters with a white-hot → deep green gradient. Headlines auto-translate into the native script of their subject region. Ambient mic input warps the glitch rate in real time. A `--poetry` mode replaces the feed with public-domain literary passages.
+A full-screen terminal news ticker that renders live global headlines in large OTF-font block characters with a white-hot → deep green gradient. Headlines auto-translate into the native script of their subject region. Ambient mic input warps the glitch rate in real time. A `--poetry` mode replaces the feed with public-domain literary passages. Live messages can be pushed to the display over [ntfy.sh](https://ntfy.sh).
 
 ---
 
 ## Run
 
 ```bash
-python3 mainline.py            # news stream
-python3 mainline.py --poetry   # literary consciousness mode
-python3 mainline.py -p         # same
+python3 mainline.py                      # news stream
+python3 mainline.py --poetry             # literary consciousness mode
+python3 mainline.py -p                   # same
+python3 mainline.py --firehose           # dense rapid-fire headline mode
+python3 mainline.py --refresh            # force re-fetch (bypass cache)
 ```
 
-First run bootstraps a local `.mainline_venv/` and installs deps (`feedparser`, `Pillow`, `sounddevice`, `numpy`). Subsequent runs start immediately.
+First run bootstraps a local `.mainline_venv/` and installs deps (`feedparser`, `Pillow`, `sounddevice`, `numpy`). Subsequent runs start immediately, loading from cache.
 
 ---
 
 ## Config
 
-At the top of `mainline.py`:
+All constants live in `engine/config.py`:
 
 | Constant | Default | What it does |
 |---|---|---|
 | `HEADLINE_LIMIT` | `1000` | Total headlines per session |
+| `FEED_TIMEOUT` | `10` | Per-feed HTTP timeout (seconds) |
 | `MIC_THRESHOLD_DB` | `50` | dB floor above which glitches spike |
-| `_FONT_PATH` | hardcoded path | Path to your OTF/TTF display font |
-| `_FONT_SZ` | `60` | Font render size (affects block density) |
-| `_RENDER_H` | `8` | Terminal rows per headline line |
+| `FONT_PATH` | hardcoded path | Path to your OTF/TTF display font |
+| `FONT_SZ` | `60` | Font render size (affects block density) |
+| `RENDER_H` | `8` | Terminal rows per headline line |
+| `SSAA` | `4` | Super-sampling factor (render at 4× then downsample) |
+| `SCROLL_DUR` | `5.625` | Seconds per headline |
+| `FRAME_DT` | `0.05` | Frame interval in seconds (20 FPS) |
+| `GRAD_SPEED` | `0.08` | Gradient sweep speed (cycles/sec, ~12s full sweep) |
+| `FIREHOSE_H` | `12` | Firehose zone height (terminal rows) |
+| `NTFY_TOPIC` | klubhaus URL | ntfy.sh JSON endpoint to poll |
+| `NTFY_POLL_INTERVAL` | `15` | Seconds between ntfy polls |
+| `MESSAGE_DISPLAY_SECS` | `30` | How long an ntfy message holds the screen |
 
-**Font:** `_FONT_PATH` is hardcoded to a local path. Update it to point to whatever display font you want — anything with strong contrast and wide letterforms works well.
+**Font:** `FONT_PATH` is hardcoded to a local path. Update it to point to whatever display font you want — anything with strong contrast and wide letterforms works well.
 
 ---
 
 ## How it works
 
-- Feeds are fetched and filtered on startup (sports and vapid content stripped)
-- Headlines are rasterized via Pillow into half-block characters (`▀▄█ `) at the configured font size
-- A left-to-right ANSI gradient colors each character: white-hot leading edge trails off to near-black
+- Feeds are fetched and filtered on startup (sports and vapid content stripped); results are cached to `.mainline_cache_news.json` / `.mainline_cache_poetry.json` for fast restarts
+- Headlines are rasterized via Pillow with 4× SSAA into half-block characters (`▀▄█ `) at the configured font size
+- A left-to-right ANSI gradient colors each character: white-hot leading edge trails off to near-black; the gradient sweeps continuously across the full scroll canvas
 - Subject-region detection runs a regex pass on each headline; matches trigger a Google Translate call and font swap to the appropriate script (CJK, Arabic, Devanagari, etc.) using macOS system fonts
 - The mic stream runs in a background thread, feeding RMS dB into the glitch probability calculation each frame
 - The viewport scrolls through a virtual canvas of pre-rendered blocks; fade zones at top and bottom dissolve characters probabilistically
+- An ntfy.sh poller runs in a background thread; incoming messages interrupt the scroll and render full-screen until dismissed or expired
+
+---
+
+## Architecture
+
+`mainline.py` is a thin entrypoint (venv bootstrap → `engine.app.main()`). All logic lives in the `engine/` package:
+
+```
+engine/
+  config.py       constants, CLI flags, glyph tables
+  sources.py      FEEDS, POETRY_SOURCES, language/script maps
+  terminal.py     ANSI codes, tw/th, type_out, boot_ln
+  filter.py       HTML stripping, content filter
+  translate.py    Google Translate wrapper + region detection
+  render.py       OTF → half-block pipeline (SSAA, gradient)
+  effects.py      noise, glitch_bar, fade, firehose
+  fetch.py        RSS/Gutenberg fetching + cache load/save
+  ntfy.py         NtfyPoller — standalone, zero internal deps
+  mic.py          MicMonitor — standalone, graceful fallback
+  scroll.py       stream() frame loop + message rendering
+  app.py          main(), boot sequence, signal handler
+```
+
+`ntfy.py` and `mic.py` have zero internal dependencies and can be imported by any other visualizer.
 
 ---
 
 ## Feeds
 
-~25 sources across four categories: **Science & Technology**, **Economics & Business**, **World & Politics**, **Culture & Ideas**. Add or swap in `FEEDS`.
+~25 sources across four categories: **Science & Technology**, **Economics & Business**, **World & Politics**, **Culture & Ideas**. Add or swap feeds in `engine/sources.py` → `FEEDS`.
 
-**Poetry mode** pulls from Project Gutenberg: Whitman, Dickinson, Thoreau, Emerson.
+**Poetry mode** pulls from Project Gutenberg: Whitman, Dickinson, Thoreau, Emerson. Sources are in `engine/sources.py` → `POETRY_SOURCES`.
+
+---
+
+## ntfy.sh Integration
+
+Mainline polls a configurable ntfy.sh topic in the background. When a message arrives, the scroll pauses and the message renders full-screen for `MESSAGE_DISPLAY_SECS` seconds, then the stream resumes.
+
+To push a message:
+
+```bash
+curl -d "Body text" -H "Title: Alert title" https://ntfy.sh/your_topic
+```
+
+Update `NTFY_TOPIC` in `engine/config.py` to point at your own topic. The `NtfyPoller` class is fully standalone and can be reused by other visualizers:
+
+```python
+from engine.ntfy import NtfyPoller
+poller = NtfyPoller("https://ntfy.sh/my_topic/json?since=20s&poll=1")
+poller.start()
+# in render loop:
+msg = poller.get_active_message()  # returns (title, body, timestamp) or None
+```
 
 ---
 
@@ -62,7 +120,6 @@ At the top of `mainline.py`:
 
 ### Graphics
 - **Matrix rain underlay** — katakana column rain rendered at low opacity beneath the scrolling blocks as a background layer
-- **Animated gradient** — shift the white-hot leading edge left/right each frame for a pulse/comet effect
 - **CRT simulation** — subtle dim scanlines every N rows, occasional brightness ripple across the full screen
 - **Sixel / iTerm2 inline images** — bypass half-blocks entirely and stream actual bitmap frames for true resolution; would require a capable terminal
 - **Parallax secondary column** — a second, dimmer, faster-scrolling stream of ambient text at reduced opacity on one side
@@ -75,6 +132,10 @@ At the top of `mainline.py`:
 - **Persona modes** — `--surveillance`, `--oracle`, `--underground` as feed presets with matching color themes and boot copy
 - **Synthesized audio** — short static bursts tied to glitch events, independent of mic input
 
+### Extensibility
+- **serve.py** — HTTP server that imports `engine.render` and `engine.fetch` directly to stream 1-bit bitmaps to an ESP32 display
+- **Rust port** — `ntfy.py` and `render.py` are the natural first targets; clear module boundaries make incremental porting viable
+
 ---
 
 *macOS only (system font paths hardcoded). Python 3.9+.*

engine/scroll.py

@@ -1,5 +1,5 @@
 """
-Scroll engine — the main frame loop with headline rendering and message display.
+Render engine — ticker content, scroll motion, message panel, and firehose overlay.
 Depends on: config, terminal, render, effects, ntfy, mic.
 """
 
@@ -16,7 +16,7 @@ from engine.effects import noise, glitch_bar, fade_line, vis_trunc, next_headlin
 
 
 def stream(items, ntfy_poller, mic_monitor):
-    """Main rendering loop. Scrolls headlines, shows ntfy messages, applies effects."""
+    """Main render loop with four layers: message, ticker, scroll motion, firehose."""
     random.shuffle(items)
     pool = list(items)
     seen = set()
@@ -28,16 +28,21 @@ def stream(items, ntfy_poller, mic_monitor):
 
     w, h = tw(), th()
     fh = config.FIREHOSE_H if config.FIREHOSE else 0
-    sh = h - fh                    # scroll zone height
+    ticker_view_h = h - fh         # reserve fixed firehose strip at bottom
     GAP = 3            # blank rows between headlines
-    scroll_interval = config.SCROLL_DUR / (sh + 15) * 2
+    scroll_step_interval = config.SCROLL_DUR / (ticker_view_h + 15) * 2
 
-    # active blocks: (content_rows, color, canvas_y, meta_idx)
+    # Taxonomy:
+    # - message: centered ntfy overlay panel
+    # - ticker: large headline text content
+    # - scroll: upward camera motion applied to ticker content
+    # - firehose: fixed carriage-return style strip pinned at bottom
+    # Active ticker blocks: (content_rows, color, canvas_y, meta_idx)
     active = []
-    cam = 0            # viewport top in virtual canvas coords
-    next_y = sh        # canvas-y where next block starts (off-screen bottom)
+    scroll_cam = 0     # viewport top in virtual canvas coords
+    ticker_next_y = ticker_view_h  # canvas-y where next block starts (off-screen bottom)
     noise_cache = {}
-    scroll_accum = 0.0
+    scroll_motion_accum = 0.0
 
     def _noise_at(cy):
         if cy not in noise_cache:
@@ -53,16 +58,17 @@ def stream(items, ntfy_poller, mic_monitor):
         t0 = time.monotonic()
         w, h = tw(), th()
         fh = config.FIREHOSE_H if config.FIREHOSE else 0
-        sh = h - fh
+        ticker_view_h = h - fh
 
         # ── Check for ntfy message ────────────────────────
-        msg_h = 0  # rows consumed by message zone at top
+        msg_h = 0
+        msg_overlay = []
         msg = ntfy_poller.get_active_message()
 
         buf = []
         if msg is not None:
             m_title, m_body, m_ts = msg
-            # ── Message zone: pinned to top, scroll continues below ──
+            # ── Message overlay: centered in the viewport ──
             display_text = m_body or m_title or "(empty)"
             display_text = re.sub(r"\s+", " ", display_text.upper())
             cache_key = (display_text, w)
@@ -76,10 +82,12 @@ def stream(items, ntfy_poller, mic_monitor):
             elapsed_s = int(time.monotonic() - m_ts)
             remaining = max(0, config.MESSAGE_DISPLAY_SECS - elapsed_s)
             ts_str = datetime.now().strftime("%H:%M:%S")
+            panel_h = len(msg_rows) + 2  # meta + border
+            panel_top = max(0, (h - panel_h) // 2)
             row_idx = 0
             for mr in msg_rows:
                 ln = vis_trunc(mr, w)
-                buf.append(f"\033[{row_idx+1};1H  {ln}{RST}\033[K")
+                msg_overlay.append(f"\033[{panel_top + row_idx + 1};1H  {ln}{RST}\033[K")
                 row_idx += 1
             # Meta line: title (if distinct) + source + countdown
             meta_parts = []
@@ -87,49 +95,46 @@ def stream(items, ntfy_poller, mic_monitor):
                 meta_parts.append(m_title)
             meta_parts.append(f"ntfy \u00b7 {ts_str} \u00b7 {remaining}s")
             meta = "  " + " \u00b7 ".join(meta_parts) if len(meta_parts) > 1 else "  " + meta_parts[0]
-            buf.append(f"\033[{row_idx+1};1H{MSG_META}{meta}{RST}\033[K")
+            msg_overlay.append(f"\033[{panel_top + row_idx + 1};1H{MSG_META}{meta}{RST}\033[K")
             row_idx += 1
-            # Border — constant boundary between message and scroll
+            # Border — constant boundary under message panel
             bar = "\u2500" * (w - 4)
-            buf.append(f"\033[{row_idx+1};1H  {MSG_BORDER}{bar}{RST}\033[K")
-            row_idx += 1
-            msg_h = row_idx
+            msg_overlay.append(f"\033[{panel_top + row_idx + 1};1H  {MSG_BORDER}{bar}{RST}\033[K")
 
-        # Effective scroll zone: below message, above firehose
-        scroll_h = sh - msg_h
+        # Ticker draws above the fixed firehose strip; message is a centered overlay.
+        ticker_h = ticker_view_h - msg_h
 
-        # ── Scroll: headline rendering (always runs) ──────
-        # Advance scroll on schedule
-        scroll_accum += config.FRAME_DT
-        while scroll_accum >= scroll_interval:
-            scroll_accum -= scroll_interval
-            cam += 1
+        # ── Ticker content + scroll motion (always runs) ──
+        scroll_motion_accum += config.FRAME_DT
+        while scroll_motion_accum >= scroll_step_interval:
+            scroll_motion_accum -= scroll_step_interval
+            scroll_cam += 1
 
             # Enqueue new headlines when room at the bottom
-            while next_y < cam + sh + 10 and queued < config.HEADLINE_LIMIT:
+            while ticker_next_y < scroll_cam + ticker_view_h + 10 and queued < config.HEADLINE_LIMIT:
                 t, src, ts = next_headline(pool, items, seen)
-                content, hc, midx = make_block(t, src, ts, w)
-                active.append((content, hc, next_y, midx))
-                next_y += len(content) + GAP
+                ticker_content, hc, midx = make_block(t, src, ts, w)
+                active.append((ticker_content, hc, ticker_next_y, midx))
+                ticker_next_y += len(ticker_content) + GAP
                 queued += 1
 
             # Prune off-screen blocks and stale noise
             active = [(c, hc, by, mi) for c, hc, by, mi in active
-                      if by + len(c) > cam]
+                      if by + len(c) > scroll_cam]
             for k in list(noise_cache):
-                if k < cam:
+                if k < scroll_cam:
                     del noise_cache[k]
 
-        # Draw scroll zone (below message zone, above firehose)
-        top_zone = max(1, int(scroll_h * 0.25))
-        bot_zone = max(1, int(scroll_h * 0.10))
+        # Draw ticker zone (above fixed firehose strip)
+        top_zone = max(1, int(ticker_h * 0.25))
+        bot_zone = max(1, int(ticker_h * 0.10))
         grad_offset = (time.monotonic() * config.GRAD_SPEED) % 1.0
-        scroll_buf_start = len(buf)  # track where scroll rows start in buf
-        for r in range(scroll_h):
-            scr_row = msg_h + r + 1  # 1-indexed ANSI screen row
-            cy = cam + r
+        ticker_buf_start = len(buf)  # track where ticker rows start in buf
+        for r in range(ticker_h):
+            scr_row = r + 1  # 1-indexed ANSI screen row
+            cy = scroll_cam + r
             top_f = min(1.0, r / top_zone) if top_zone > 0 else 1.0
-            bot_f = min(1.0, (scroll_h - 1 - r) / bot_zone) if bot_zone > 0 else 1.0
+            bot_f = min(1.0, (ticker_h - 1 - r) / bot_zone) if bot_zone > 0 else 1.0
             row_fade = min(top_f, bot_f)
             drawn = False
             for content, hc, by, midx in active:
@@ -160,22 +165,24 @@ def stream(items, ntfy_poller, mic_monitor):
                 else:
                     buf.append(f"\033[{scr_row};1H\033[K")
 
-        # Draw firehose zone
-        if config.FIREHOSE and fh > 0:
-            for fr in range(fh):
-                fline = firehose_line(items, w)
-                buf.append(f"\033[{sh + fr + 1};1H{fline}\033[K")
-
-        # Glitch — base rate + mic-reactive spikes (scroll zone only)
+        # Glitch — base rate + mic-reactive spikes (ticker zone only)
         mic_excess = mic_monitor.excess
         glitch_prob = 0.32 + min(0.9, mic_excess * 0.16)
         n_hits = 4 + int(mic_excess / 2)
-        scroll_buf_len = len(buf) - scroll_buf_start
-        if random.random() < glitch_prob and scroll_buf_len > 0:
-            for _ in range(min(n_hits, scroll_buf_len)):
-                gi = random.randint(0, scroll_buf_len - 1)
-                scr_row = msg_h + gi + 1
-                buf[scroll_buf_start + gi] = f"\033[{scr_row};1H{glitch_bar(w)}"
+        ticker_buf_len = len(buf) - ticker_buf_start
+        if random.random() < glitch_prob and ticker_buf_len > 0:
+            for _ in range(min(n_hits, ticker_buf_len)):
+                gi = random.randint(0, ticker_buf_len - 1)
+                scr_row = gi + 1
+                buf[ticker_buf_start + gi] = f"\033[{scr_row};1H{glitch_bar(w)}"
+
+        if config.FIREHOSE and fh > 0:
+            for fr in range(fh):
+                scr_row = h - fh + fr + 1
+                fline = firehose_line(items, w)
+                buf.append(f"\033[{scr_row};1H{fline}\033[K")
+        if msg_overlay:
+            buf.extend(msg_overlay)
 
         sys.stdout.buffer.write("".join(buf).encode())
         sys.stdout.flush()