# The Many Cities of Kochi An interactive map of monuments and historic sites across Greater Kochi — from the ancient Chera-era port of Muziris to the Kingdom of Cochin, Portuguese, Dutch and British Cochin, and Kochi's integration into Kerala — each site tagged to the era that shaped it, with citations to published historical sources. Inspired by [The Many Cities of Delhi](https://kaustubh-misra.github.io/delhi-monuments/) by Kaustubh Misra (GPLv3), which mapped Delhi's monuments through its succession of ruling kingdoms. ## Coverage Fort Kochi, Mattancherry/Jew Town, Willingdon Island, Tripunithura, Vypin, and the Kodungallur/Muziris region (~30km north, Thrissur District), spanning eight historical layers: - Chera / Muziris-adjacent Era (pre-1341) - Kingdom of Cochin (Perumpadappu Swaroopam) - Portuguese Cochin (1503–1663) - Dutch Cochin (1663–1795) - Mysorean Invasion / Tipu Sultan Era (1776–1792) - British Paramountcy / Princely State of Cochin (1795–1947) - Jewish Heritage (Paradesi & Malabar Jews) - Integration into Kerala (1947–1956 onward) The map is locked to this region (`maxBounds` in `app.js`) — panning and zooming out are clamped so the tiles shown never wander outside Greater Kochi and Kodungallur. A dark-mode toggle (top right) follows the system theme by default and switches the map between OpenFreeMap's `liberty` (light) and `dark` styles; the choice persists in `localStorage`. "Get directions" opens [osm-navigator](https://osm-navigator.siddharthshiv2798.workers.dev/), a companion turn-by-turn routing app, with the monument pre-filled as the destination. It works by building a Google-Maps-shaped URL (`.../maps/place//@,,17z`) and passing it via `?url=`, the same mechanism osm-navigator uses to resolve a pasted or shared Google Maps link — it parses the coordinates client-side and sets them as the destination search pick. The starting point isn't set by this site; osm-navigator handles that itself (its own current-location prompt or manual entry). Other UX details: a search box filters the monument list by name; marker name labels are hidden below zoom 12 to avoid overlapping near dense clusters (e.g. Kodungallur); clicking a list card flies the map to that marker; era filters persist in `localStorage` across reloads; every monument has a shareable `?monument=` URL that reopens its detail card on load; the detail card has "Copy citation" and "Copy link" buttons; list cards and markers are keyboard-operable (Tab + Enter/Space). ## How Kochi Grew (territorial history timeline) A second, complementary way to browse the site, reachable via a **Timeline / Browse all eras** toggle at the top of the sidebar (folded in there rather than as a separate full-width bar, to avoid three stacked control zones on the page): six era steps tracing how the region's political geography changed — Chera-era unified administration, fragmentation into swaroopams (Edappally holding Kochi/Vypin), the Kingdom of Cochin's formation, the Paliam family's fief at Chendamangalam, Fort Kochi's colonial-enclave/princely-state split, and the 1967 municipal merger that formed the modern Kochi Corporation. See `territories.json` for the underlying polygons and `README.md`'s citations for the historical sourcing behind each step (also documented inline as `TERRITORY_ERAS` captions in `app.js`). These are **hand-drawn illustrative approximations, not surveyed boundaries** — medieval Kerala polities didn't have any. Low-confidence eras render with a dashed outline; the 1967 merger (a real administrative record) renders solid. A **Timeline / Browse all eras** toggle switches between two modes: - **Timeline** (default): shows the selected era's territory overlay, and dims (not hides) monuments whose era tag isn't relevant to that period — relevance is a many-to-many mapping (`TERRITORY_TO_MONUMENT_ERAS`) since the two era taxonomies were never designed to align 1:1. - **Browse all eras**: today's original free multi-select checkbox browsing, territory overlay hidden, no dimming. Since the polygons are hand-drawn without any awareness of the real coastline, an early version's edges routinely cut across open water. Fixed by redrawing the basemap's own accurate `water` layer (cloned live from the current MapLibre style, so it matches both the light and dark themes automatically) on top of the territory tint — any part of a polygon that doesn't actually correspond to land gets visually erased back to water. Willingdon Island and Vypin also now use their real OSM geometry (fetched via Overpass, reconstructed from way segments with shapely, simplified, baked into `territories.json`) instead of a generated shape, for the two entities where a real one exists — most other named places here (Fort Kochi, Mattancherry, Ernakulam, Edappally, Palluruthy, Vyttila) only exist as single point nodes in OSM, not boundary polygons. For eras where **one polity holds the entire visible peninsula** (Chera unified administration, Kingdom of Cochin formed, the Paliam era's broad Cochin extent), the polygon is deliberately drawn oversized — big enough to fully contain the whole map area — rather than hand-fitted to the coast. Combined with the water-mask, this makes the tint reveal the *exact* real landmass shape with zero invented edges, rather than an approximation. Eras that split the same peninsula into pieces with no real boundary data (the colonial Fort Kochi enclave vs. the rest of Cochin State; the 1967 merger's constituent municipalities/panchayats) still use hand-drawn approximate fills — flagged with a dashed outline — since no real sub-boundary exists to substitute. Each territory is labelled directly on the map with its **period-appropriate name** — e.g. "Mahodayapuram" (not the modern "Kodungallur") for the Chera-era capital, and "Cochin (English)" for the colonial Fort Kochi enclave, taken directly from a real early-20th-century "Kingdom of Cochin" map ([Wikimedia Commons](https://upload.wikimedia.org/wikipedia/commons/b/b3/Map_of_the_Kingdom_of_Cochin.jpg)) which also confirms the enclave-within-princely-state structure modelled here. ## Running locally ``` python3 -m http.server 8000 ``` Then open `http://localhost:8000`. (Opening `index.html` directly via `file://` will not work — the browser blocks `fetch()` of `monuments.json` under the `file://` scheme.) ## Validating the data ``` python3 scripts/validate_monuments.py ``` Checks schema completeness, coordinate sanity for the Greater Kochi/Kodungallur bounding box, duplicate IDs, presence of a citation clause in every description, and per-era coverage. Pass `--write` to refresh the `generatedAt` timestamp after an edit. ## Historical sources Every monument's description ends with its own citations. The works drawn on across the dataset include: - K.P. Padmanabha Menon, *History of Kerala* (1924–1937) - A. Sreedhara Menon, *A Survey of Kerala History* (1967); Kerala District Gazetteers (Ernakulam & Thrissur) - C. Achyutha Menon, *The Cochin State Manual* (1911, Government of Cochin) - William Logan, *Malabar Manual* (1887) - Pius Malekandathil, *Portuguese Cochin and the Maritime Trade of India, 1500–1663* (2001) - Robert Bristow, *Cochin Saga* (1959) - Roland E. Miller, *Mappila Muslims of Kerala: A Study in Islamic Trends* (1976) - J.B. Segal, *A History of the Jews of Cochin* (1993) - Nathan Katz, *Who Are the Jews of India?* (2000) - Ruby Daniel & Barbara C. Johnson, *Ruby of Cochin: An Indian Jewish Woman Remembers* (1995) - P.J. Cherian (ed.), *Pattanam Excavations: Interim Reports*, Kerala Council for Historical Research (2007–2015) - Roberta Tomber, *Indo-Roman Trade: From Pots to Pepper* (2008) - Pliny the Elder, *Natural History* (c. 77 CE); *Periplus Maris Erythraei* (1st century CE), trans. Lionel Casson (1989) This list is a starting bibliography, not exhaustive — contributions of better or additional sources for any entry are welcome. ## Data notes Coordinates were originally placed from general geographic knowledge, then cross-checked against OpenStreetMap Nominatim and Wikipedia/Wikidata coordinates and corrected where those sources gave a materially different, named-POI location — this caught several meaningful errors (e.g. Paradesi Synagogue previously sat ~190m off, over water; Vypin Lighthouse and Kottapuram Fort were off by 3km+). Sites without a confidently-matched named source (e.g. Fort Immanuel, largely demolished) are still best-effort estimates. Further corrections are welcome via pull request. The `mysorean` era currently has no monument tagged to it directly — the 1776 Mysorean invasion is referenced within the Kingdom of Cochin and Hill Palace entries instead of a fabricated standalone site, since no discrete surviving structure could be confidently identified. ## Tech Static HTML/CSS/JS, no build step. Map rendering by [MapLibre GL JS](https://maplibre.org), tiles from [OpenFreeMap](https://openfreemap.org), map data © [OpenStreetMap contributors](https://www.openstreetmap.org/copyright). No API key required. ## License GNU General Public License v3.0 — see [LICENSE](LICENSE). Following the license of the project this one is modeled on.