# Sources Every dataset used, with its URL, licence, the date it was obtained and exactly what was done to it. Raw files are committed unmodified under `scripts/raw/` so anyone can re-run the scripts and get the same data files. All retrieval dates are **12 September 2026** unless stated. --- ## Population, 1971–2011 **Census of India 2011, Table A-02 — Decadal variation in population 1901–2011** Office of the Registrar General & Census Commissioner, India (ORGI), Ministry of Home Affairs. | | | |---|---| | India by State | https://censusindia.gov.in/nada/index.php/catalog/43333 | | Andhra Pradesh by District | https://censusindia.gov.in/nada/index.php/catalog/43361 | | Jammu & Kashmir by District | https://censusindia.gov.in/nada/index.php/catalog/43334 | | Licence | Government of India publication. The NADA portal records the licence as "TSB". | | Raw files | `scripts/raw/census_A2_India.xls`, `census_A2_AndhraPradesh.xlsx`, `census_A2_JammuKashmir.xlsx` | | Script | `scripts/extract_census_series.py` → `scripts/raw/census_series_a2.json` | **Why this table specifically.** Its own metadata states that "the population figures for all the previous censuses in this table are adjusted to the 2011 jurisdiction". The Registrar General has therefore already recast every earlier census onto the 2011 map, at both state and district level, which is the reconciliation this project would otherwise have had to build by hand. **What was done.** 1. State totals read from the India file, for 1971, 1981, 1991, 2001 and 2011. 2. Daman & Diu (state code 25) and Dadra & Nagar Haveli (26) summed into our single unit DH. 3. Telangana derived by summing Andhra Pradesh district codes 532–541; residual Andhra Pradesh by subtraction from the state row. 4. Ladakh derived by summing Jammu & Kashmir district codes 003 (Leh) and 004 (Kargil); residual J&K by subtraction. 5. Every column checked against the published INDIA row. All five reconcile exactly. 6. Assam 1981 and J&K/Ladakh 1991 set to null — no census was conducted. A-02 prints official interpolations for these; they are preserved in the intermediate JSON but are not served. 7. 1901–1961 refused. See `METHODOLOGY.md` §7. --- ## Population projections, 2016–2036 **Population Projections for India and States 2011–2036 — Report of the Technical Group on Population Projections** National Commission on Population, Ministry of Health & Family Welfare, published July 2020. | | | |---|---| | URL | https://nhm.gov.in/New_Updates_2018/Report_Population_Projection_2019.pdf | | Licence | Government of India publication | | Size | 11.01 MB, 268 pages | | Raw file | `scripts/raw/Report_Population_Projection_2019.pdf` | | Script | `scripts/extract_projections.py` → `scripts/raw/projections_table21.json` | **What was done.** Table 21, *Projected Total Population As on 1st March*, printed pages 259–260 (PDF pages 263–264), parsed for the "Persons" column of each of 2011, 2016, 2021, 2026, 2031 and 2036. Rows 25 and 30 — Dadra & Nagar Haveli and Daman & Diu — summed into DH. Row 05 is printed as "Uttaranchal" and row 01 as "Jammu & Kashmir (UT)"; mapped to UK and JK. Figures are printed in **thousands** and are stored in thousands in the intermediate file, multiplied to persons only in `build_units.py`, so the rounding lives in one place. Every column reconciles to the report's own published INDIA row within one thousand, which is what independent per-row rounding produces. **Secondary use.** The report's 2011 **base year** is an independent check on our five derived units, and it is the source that settled the Telangana question: it prints Telangana at 35,004 thousand and Andhra Pradesh at 49,577 thousand, and it lists Ladakh and Jammu & Kashmir as separate rows. **Caveat.** These are the 2019/2020 vintage. They predate the 2027 Census and COVID-19 mortality. --- ## Validation fixture **PRS Legislative Research — annexure to the analysis of the Delimitation Bill, 2026** | | | |---|---| | URL | https://prsindia.org/billtrack/the-delimitation-bill-2026 | | Table | "Table 1: Projected number of seats based on the 2011 Census" | | Licence | **CC BY 4.0**, as stated on the PRS page | | Stored as | `data/reference.json` | **What was done.** Transcribed as a test fixture and checked row by row against the published table, including both column totals. Used only to validate the engine, never to produce a number shown on the site. The table is on **merged** units — Andhra Pradesh includes Telangana, Jammu & Kashmir includes Ladakh, and Dadra & Nagar Haveli and Daman & Diu are listed separately — so it has 35 rows against our 36 units. The test suite validates against the merged configuration as a historical check and runs the split configuration as the live one. Its footnote is quoted here because it documents an assumption the site otherwise could not evidence: *"Andhra Pradesh includes Telangana, and Jammu & Kashmir includes Ladakh. We assume that the number of seats for Arunachal Pradesh, Goa, Manipur, and Meghalaya will not be reduced as there is an exception for states with population below 60 lakh."* **A licence wrinkle, recorded rather than resolved.** The PRS page states CC BY 4.0. PRS boilerplate elsewhere says content may be reproduced "for non-commercial purposes". Those are not the same permission — CC BY 4.0 permits commercial use. Both strings are recorded here rather than picking the convenient one. This site is non-commercial research either way. --- ## The 2014 territorial transfer **The Andhra Pradesh Reorganisation (Amendment) Act, 2014 (No. 19 of 2014)** | | | |---|---| | URL | https://prsindia.org/files/bills_acts/acts_parliament/2014/the-andhra-pradesh-reorganisation-(amendment)-act,-2014.pdf | | Licence | Government of India, Gazette of India Extraordinary | | Raw file | `scripts/raw/AP_Reorganisation_Amendment_Act_2014.pdf` | **District Census Handbook 2011, Khammam (district code 2810)** — Village Directory inset tables. | | | |---|---| | URL | https://censusindia.gov.in/nada/index.php/catalog/142 | | Raw file | `scripts/raw/khammam_dchb/Appendix_VD_2810.xls` | | Script | `scripts/verify_bhadrachalam.py` | **What was done.** Section 2 of the Amendment Act gives the exact schedule of mandals and revenue villages transferred to Andhra Pradesh. Those were summed from the village directory to establish that the territory accounts for exactly 190,304 people, which is the gap between the two circulating Telangana 2011 figures. See `METHODOLOGY.md` §8. --- ## Current seat allocation **First Schedule, Representation of the People Act, 1950**, as amended by the Andhra Pradesh Reorganisation Act 2014 and the Jammu & Kashmir Reorganisation Act 2019. Cross-checked against the state-wise allocation published by the Ministry of External Affairs and against PRS's "current seats" column, which agree. Andhra Pradesh 25 and Telangana 17, together the 42 the undivided state held. Jammu & Kashmir 5 and Ladakh 1, together the 6 the undivided state held. Total 543. --- ## Boundary geometry **udit-001/india-maps-data**, `topojson/india.json`, states layer. | | | |---|---| | URL | https://github.com/udit-001/india-maps-data | | Licence | **None declared.** See the caveat below. | | Raw files | `scripts/raw/udit-001_india.topo.json`, `udit-001_india_districts.geojson` | | Script | `scripts/build_boundaries.mjs` → `data/boundaries.topo.json` | | Output | 45.3 KB uncompressed, 16.9 KB gzipped, 36 units, 3,873 points | **What was done.** The states layer extracted from a topology shared with a districts layer; 1,571 district-only arcs dropped, which is the entire size saving. **No simplification applied** — the sweep in the script shows every level from 0.25 to 16 buys 147 bytes in total, 0.3%, because the source is already generalised for a national choropleth. Coordinates left as longitude and latitude; projection happens at runtime. **Compliance.** Verified by point-in-polygon test against fifteen landmarks on the **final** file. See `docs/boundary-check.svg` and §9 of `METHODOLOGY.md`. **Provenance caveat, stated plainly.** The repository states its data is "curated from publicly available sources on the internet", declares no licence, and makes no Survey of India claim. Our confidence rests entirely on our own geometry test. Two things corroborate official derivation — the attributes are Census 2011 state codes, and Gilgit-Baltistan is assigned to *Ladakh*, which is the specific and counter-intuitive arrangement of the Survey of India map published in November 2019 — but neither is a citation. **Before any wider publication this should be replaced with a citable government-hosted equivalent, or the repository owner asked to declare a licence.** ### Rejected candidates, recorded so the next person does not repeat the work | Source | Verdict | |---|---| | DataMeet / `geohacker/india`, `state/india_state.geojson` | **Fails 8/15.** GADM-derived (`ID_0`/`NAME_1`/`VARNAME_1` schema), northern extent stops at 35.501°N. Gilgit-Baltistan, Muzaffarabad, Mirpur, Aksai Chin and Shaksgam all outside the India polygon. Also pre-2014 vintage: "Orissa", "Uttaranchal", no Telangana, no Ladakh. | | `datta07/INDIAN-SHAPEFILES`, `INDIA_STATES.geojson` | **Passes 15/15**, 37 features with Dadra & Nagar Haveli and Daman & Diu still separate. Kept in `scripts/raw/` as a corroborating second opinion. Not used because the chosen source already matches our 36 units and ships TopoJSON. Same undeclared-provenance problem. | | Survey of India Online Maps Portal | Authoritative, but no clear free download path for state boundary vector data was found. | --- ## Legislative status | Bill | Status | Source | |---|---|---| | Constitution (131st Amendment) Bill, 2026 | Introduced 16 April 2026, **negatived** 17 April 2026. 298 for, 230 against, 528 present and voting; two-thirds threshold 352. | https://prsindia.org/billtrack/the-constitution-131st-amendment-bill-2026 | | Delimitation Bill, 2026 | Introduced 16 April 2026, **infructuous** 17 April 2026 | https://prsindia.org/billtrack/the-delimitation-bill-2026 | | Union Territories Laws (Amendment) Bill, 2026 | Introduced 16 April 2026, fell the same day | as above | Checked 12–13 September 2026: no revised bill has been introduced. This is the most perishable fact on the site and the page carries a visible "status as at" date for that reason. --- ## Development-time dependencies None of these are loaded by the site. The site itself has exactly one runtime dependency, d3 v7 from cdnjs, used for the geographic projection and path generation. | Tool | Used for | |---|---| | Node 22.7+ | test suite, boundary and snapshot scripts | | Python 3 | data extraction scripts | | `pypdf` | reading the projections report | | `xlrd`, `openpyxl` | reading the census workbooks | ## Licences summary - Code in this repository: **MIT**, see `LICENSE`. - PRS Legislative Research projection table: **CC BY 4.0**, attributed. - Census of India and Registrar General material: Government of India. - Boundary geometry: **no licence declared upstream** — see the caveat above.