CREDITS
Credits
CREDITS
- RESEARCH
- City Science, MIT Media Lab. Supervision: Kent Larson, with Luis Alonso.
- THE ANDORRA SCENARIO ENGINE (URBAN SCIENCE, 2026)
- Marcel Bartumeu Ramentol, Parfait Atchade-Adelomou, Adrian Mora-Carrero, Kent Larson, Luis Alonso-Pastor, Marta Domenech (Andorra Research + Innovation).
- h-ABM BEHAVIOR ENGINE · SOURCED
- City Science team (TechRxiv preprint: Mora, Atchade-Adelomou, Alonso-Pastor et al.). My part: the coupling, identification checks and outcome memory.
- PARTNER
- Andorra Research + Innovation.
- DATA
- Official Andorran statistics and inventories (Departament d’Estadística and the other sources cited per series in the source drawer), Observatori d’Andorra survey aggregates (no microdata), OpenStreetMap contributors where streets are drawn, and the San Francisco Assessor parcel map.
- MAPS AND DATA LICENCES
- Street geometry © OpenStreetMap contributors (ODbL). Globe land: Natural Earth 1:110m (public domain). Globe idea after Three.js Interactive Vertex Earth (Bobby Roe, MIT licence); rewritten, no code copied.
- KENDALL SQUARE WORLD · DATA
- A synthetic population of the City of Cambridge, Massachusetts, which I built in 2026; every person in it is fictional. Census geography: U.S. Census Bureau. Street and path geometry © OpenStreetMap contributors (ODbL). Transit timetables and shapes: MBTA GTFS.
- THIS SITE
- Design and code: Marcel Bartumeu Ramentol. Type: the system’s SF Pro where the device has it, otherwise Inter; Syne and Instrument Serif (both SIL OFL, self-hosted). three.js, GSAP, Lenis.
DATA SOURCES
- Resident population: official statistics (SIGDDE), 31 December, 2010 to 2024.
- GDP per capita, nominal euros, as used in the model inputs.
- Housing burden, percent of income: 2010 to 2017 back-extended, 2018 to 2024 from the household survey (ECV).
- CO2: national inventory (BTR1) for 2010 to 2021; 2022 to 2024 are fuel-based central estimates, not official.
- Electricity demand: no observed series; 2024 is the model’s own 2024 state, and later years assume 2024 use per resident.
- Visitors per year, tourists and day-trippers (2012 methodology), from the tourism series of the model. Not overnight stays.
- Natural land cover: an estimate from observed land cover, 2010 to 2024, joined to the model’s 2024 state with a 0.034 percentage-point step.
- PM2.5 in Escaldes-Engordany: European Environment Agency TEOM series, 2018 to 2024; the gravimetric years 2014 to 2017 are left out because the method changed.
- Hospital beds per 1,000 service population: observed in 2017, 2019, 2023 and 2024.
- Summer water stress (WEI): no observed series; 2024 is the model’s own 2024 state.
COLOPHON
Type: the system’s SF Pro where the device has it, otherwise Inter; Syne and Instrument Serif (both SIL OFL, self-hosted). Built with three.js, GSAP and Lenis on Vite, as a static site with no cookies and no analytics on panel use.
CHANGELOG
v1.0, built 18 October 2026: first version.
CLAIMS
This build checks 365 claims against their sources and data files; wording I drafted for myself is marked until I approve it.
THE CLAIMS REGISTER · 365 CLAIMS
| Id | Claim | Status | Source | Checked |
|---|---|---|---|---|
| ident.from | I'm from Andorra. | OWN RECORD | FIS athlete profile (nation AND, born 2005) | 2026-10-10 |
| ident.base | BOSTON · FROM ANDORRA | OWN RECORD | Own record (CV) | 2026-10-10 |
| ident.languages | Catalan, Spanish, English, French and Italian. | OWN RECORD | Own record, CV | 2026-10-10 |
| card.name | Marcel Bartumeu Ramentol | VERIFIED | ORCID record and Crossref (paper authorship) | 2026-10-10 |
| card.origin | Andorra · born 2005 | VERIFIED | FIS athlete profile (year of birth only) | 2026-10-10 |
| card.languages | Catalan, Spanish, English, French, Italian | OWN RECORD | Own record, CV | 2026-10-10 |
| card.self | The main word I would use to describe myself is technologist | OWN RECORD | My branding reflection, March 2026 (own document) | 2026-10-10 |
| card.builds | I create what doesn't exist, focusing on high-quality products. | OWN RECORD | eTower introduction slide, February 2024 (own document) | 2026-10-10 |
| card.now | Part-time Research Support Associate on the Andorra project, City Science group, MIT Media Lab (since September 2026) | OWN RECORD | Own record, CV; MIT Media Lab people page | 2026-10-10 |
| card.studies | Babson College, BSBA, concentration in Technological Entrepreneurship, May 2027. Olin College of Engineering, General Engineering certifi… | OWN RECORD | Own record, CV | 2026-10-10 |
| card.home | withheld | OWN RECORD | Withheld by choice | 2026-10-10 |
| mit.narrative | Since 2025 I have worked with the City Science group at the MIT Media Lab. | OWN WORDING | MIT Media Lab people page; offer and visiting-student letters | 2026-10-10 |
| mit.chip | MIT MEDIA LAB · CITY SCIENCE · SINCE 2025 | VERIFIED | MIT Media Lab people page | 2026-10-10 |
| mit.cv | MIT Media Lab, City Science: visiting student researcher, July 2025 to July 2026, supervised by Kent Larson with Luis Alonso. Since Septe… | OWN RECORD | Own record, CV; MIT Media Lab people page | 2026-10-10 |
| mit.profile | MIT Media Lab lists me as a researcher and student entrepreneur working with City Science, on the Andorra project. | VERIFIED | MIT Media Lab people page and Andorra project page | 2026-10-10 |
| mit.delegation | I demonstrated the model live to an Andorran government delegation at the Media Lab in May 2026. | OWN WORDING | Own record; delegation visit, May 2026 | 2026-10-10 |
| c00.line1 | Technologist from Andorra. I simulate countries and cities, and build the tables and panels people use to test decisions on them. | OWN WORDING | My branding reflection, March 2026 (the word technologist); my Media Lab application progress deck, slide 15 (the second sentence) | 2026-10-10 |
| c00.thesis | I simulate countries and cities, and build the tables and panels people use to test decisions on them. | OWN WORDING | My Media Lab application progress deck, slide 15 | 2026-10-10 |
| c00.chip.author | FIRST AUTHOR · URBAN SCIENCE · 2026 | VERIFIED | Crossref and MIT DSpace record of the paper | 2026-10-10 |
| c00.chip.founder | FOUNDER · MEVOT (2021) | OWN RECORD | Own record: Mevot brand May 2021 | 2026-10-10 |
| c00.chip.base | BASED IN BOSTON | OWN RECORD | Own record (still based in Boston) | 2026-10-10 |
| c00.chip.from | FROM ANDORRA | VERIFIED | FIS athlete profile | 2026-10-10 |
| c00.title | Marcel Bartumeu Ramentol | VERIFIED | ORCID record and Crossref (paper authorship) | 2026-10-10 |
| c00.fall.kicker | THE FALL LINE · 23.5 KM · 2,846 M → 848 M | DERIVED FROM DATA | Steepest descent computed on this site's Andorra grid (scripts/fall-line.mjs) | 2026-10-10 |
| c00.fall | The fall line is the path water takes down a mountain, and the line I looked for through every course. This is the one on my grid of Ando… | OWN WORDING | Computed on the grid; geography to check (QA 7.2 #27) | 2026-10-10 |
| c01.kicker | CITY SCIENCE · FROM 2015 TO TODAY | OWN RECORD | Own record; MIT Media Lab article on the Andorra summit, October 2017 | 2026-10-10 |
| c01.lead.2017 | I was eleven at the 2017 City Science Summit in Andorra. I had started working with the City Science group in 2015. | OWN RECORD | Own record (owner directions 3 and 4); MIT Media Lab article on the summit, October 2017 | 2026-10-10 |
| c01.body.2017 | The group had been working with Andorra since 2014. The country was its living lab: sensors in school classrooms, an interactive city tab… | OWN WORDING | MIT Media Lab article, 13 October 2017 | 2026-10-10 |
| c01.quote.2026 | Nine years later I am a researcher in that group | OWN RECORD | My competition pitch, slide 6 (own document) | 2026-10-10 |
| c01.body.2026 | Today I work in that same group, on its Andorra project. I model the country I grew up in. | OWN WORDING | Own record, CV; MIT Media Lab people page | 2026-10-10 |
| c01.years | 9 YEARS | DERIVED FROM DATA | 2026 minus 2017 | 2026-10-10 |
| c01.termites.context | In 2016 the group put TerMITes sensors in classrooms of the Escola Andorrana, with 95 pupils aged 10 and 11. | VERIFIED | MIT Media Lab article on the Escola Andorrana pilot, September 2016 | 2026-10-10 |
| c01.termites.pupil | As a pupil in Andorra, I helped develop TerMITes, City Science's classroom sensors. | OWN RECORD | Own record; my current website; Andorra Research + Innovation profile | 2026-10-10 |
| c01.quote.admired | Having admired the Media Lab from a young age | OWN RECORD | My motivation letter for the Media Lab visit (own document) | 2026-10-10 |
| c01.inset.2017 | FAMILY ARCHIVE · CITY SCIENCE SUMMIT, ANDORRA · 14–15 SEP 2017 | OWN RECORD | Own photograph, already public on my site | 2026-10-10 |
| c02.kicker | BUILDING · FROM AGE TWELVE | DERIVED FROM DATA | Dates of the print files and photographs shown | 2026-10-10 |
| c02.title | Hands first | DERIVED FROM DATA | Chapter title | 2026-10-10 |
| c02.quote.create | I create what doesn't exist, focusing on high-quality products. | OWN RECORD | eTower introduction slide, February 2024 (own document) | 2026-10-10 |
| c02.body.intro | I bought my first 3D printer at twelve and started making physical things. One of them was a climbing device with electromagnets, because… | OWN RECORD | Own record; G-code file dates on my drive (file dates can be copy dates) | 2026-10-10 |
| c02.s1.caption | The prototype was a school project, printed in two halves. | OWN WORDING | Own project folder, January 2020 | 2026-10-10 |
| c02.s1.rc | At thirteen I built radio-controlled boats, cars and planes from scratch, and a device that did my laundry. | OWN RECORD | Own record; my current website | 2026-10-10 |
| c02.s1.masks | In the first COVID lockdown, at fourteen, my older brother and I made face masks, 3D-printed on my printer, and he delivered them (print … | OWN RECORD | G-code file MASK-European-Union-V2, 31 March 2020 (own file); my current website | 2026-10-10 |
| c02.s1.press | Diari d'Andorra covered the masks. The article features my brother. | VERIFIED | Diari d'Andorra: Material sanitari 3D | 2026-10-10 |
| c02.s2.unity | That year I also wrote my first game in Unity, Space Cleaner, and put it on the App Store. | OWN RECORD | Own record; my current website | 2026-10-10 |
| c02.recon | RECONSTRUCTION · APPROXIMATE 3D MODEL | DERIVED FROM DATA | Approximate 3D reconstruction made for this site from my photographs and video; not a CAD file | 2026-10-10 |
| c02.rc.title | An F1 radio-controlled car with a 3D-printed body. | OWN RECORD | Own record; my video and photograph of the car (owner direction, 12 Oct 2026) | 2026-10-10 |
| c02.rc.body | I printed the body and rear wing in green, fitted wide wheels, and put a brushless motor in it. | OWN RECORD | Own record; my video and photograph of the car (owner direction, 12 Oct 2026) | 2026-10-10 |
| c02.drone.title | A racing drone I built for FPV flying. | OWN RECORD | Own record; my photographs of the drone (owner direction, 12 Oct 2026) | 2026-10-10 |
| c02.rc.spec | 3D-PRINTED BODY · RED POWER B3650 · 3215 KV BRUSHLESS | OWN RECORD | Own record (owner direction, 12 Oct 2026); photograph of the car | 2026-10-10 |
| c02.drone.spec | 250 MM CARBON X FRAME · 5 IN PROPS · 2200 MAH 3S LIPO | OWN RECORD | Own record (owner direction, 12 Oct 2026); photographs of the drone | 2026-10-10 |
| c02.drone.body | I built it on a 250-class carbon X frame: five-inch propellers, aluminium standoffs, silicone dampers, an FPV camera and a three-cell bat… | OWN RECORD | Own record; my photographs of the drone (owner direction, 12 Oct 2026) | 2026-10-10 |
| c02.s2.photo | THE FIRST BUILD · WELDED FRAME | OWN RECORD | Own photograph of the kart; no dated file, so no date is printed | 2026-10-10 |
| c02.s2.title | A 30 hp electric go-kart, built at thirteen and fourteen. | OWN RECORD | Own record (owner direction, 12 Oct 2026); photograph of the kart | 2026-10-10 |
| c02.s2.spec | ELECTRIC GO-KART · BUILT AT 13–14 · 30 HP | OWN RECORD | Own record (owner direction, 12 Oct 2026) | 2026-10-10 |
| c02.s2.body | I designed it and welded the steel frame myself, in our basement, then fitted a motor, a chain drive and printed drivetrain parts. | OWN RECORD | Own record; photograph of the kart; June 2020 print jobs | 2026-10-10 |
| c02.s2.model | The kart on screen is built from my own CAD model of it. | OWN RECORD | Own CAD file MEVOT KART.stl | 2026-10-10 |
| c02.s2.body2 | Years later I kept developing it: in 2023 and 2024 I designed and 3D-printed motor mounts, a brake disc and spacers for it. I raced karts… | OWN RECORD | Own CAD and print files, 2023-24; Automòbil Club d'Andorra events | 2026-10-10 |
| c02.s2.racing | ELECTRIC KART · 2020 | OWN RECORD | Own photograph of the kart, taken in 2020 | 2026-10-13 |
| c02.s3.body1 | I started Mevot at fifteen, building electric longboards by hand and selling them. I designed the brand, the deck graphics, the packaging… | OWN RECORD | Own Mevot archive; own record | 2026-10-10 |
| c02.s3.quote | After creating my own e-board I realized there was a market to create and sell longboards to others. | OWN RECORD | My earlier CV (own document) | 2026-10-10 |
| c02.s3.body2 | The motors, controller and battery came from a supplier's platform. The brand, the deck graphics, the box and the manual were mine. Befor… | OWN WORDING | Own Mevot archive; supplier proforma; case study of 15 June 2024 | 2026-10-10 |
| c02.s3.exit | Mevot grew to six figures in annual revenue. In 2025 I exited it through an agreement in which a manufacturing partner acquired part of M… | OWN RECORD | Own record; my current website | 2026-10-10 |
| c02.s3.tagline | Pesa poc. Arriba lluny. | OWN RECORD | Mevot MK2 product description (own document) | 2026-10-10 |
| mevot.bom.made.enclosure | Printed enclosures on the hand-built prototypes, 2021–2022; MEVOT BOX CARBON, 2024 | DERIVED FROM DATA | Own print files and prototype photographs | 2026-10-10 |
| mevot.bom.made.model | My CAD of the board (MK2) | DERIVED FROM DATA | Own CAD file | 2026-10-10 |
| mevot.bom.designed.deck | Deck shape and graphics; production decks made by the supplier | DERIVED FROM DATA | Own design files; supplier documents | 2026-10-10 |
| mevot.bom.designed.brand | Brand, die-cut box with GS1 barcode and FSC logo, manual | DERIVED FROM DATA | Own packaging files | 2026-10-10 |
| mevot.bom.designed.controller | Controller and charger industrial design (design intent) | DERIVED FROM DATA | Own design files | 2026-10-10 |
| mevot.bom.sourced.platform | Supplier platform H2B-02: 2 × 450 W motors | DERIVED FROM DATA | Supplier specification sheet | 2026-10-10 |
| mevot.bom.sourced.battery | Battery 36 V, 4,000 mAh, 144 Wh; UN 38.3 report issued to the battery supplier | DERIVED FROM DATA | Battery supplier test report | 2026-10-10 |
| c02.ev.grip | BRAND ON GRIP TAPE · MAY 2021 | OWN RECORD | Photograph of 29 May 2021 | 2026-10-10 |
| c02.ev.proto | HAND-BUILT PROTOTYPES · 2022 | OWN RECORD | Photographs of April and June 2022 | 2026-10-10 |
| c02.ev.case | BOOSTED CASE STUDY · 29 PP · JUN 2024 | OWN RECORD | My 29-page case study of 15 June 2024 (own document) | 2026-10-10 |
| c02.ev.plan | LAUNCH PLAN: 100 UNITS, NO PAID PRE-ORDERS | OWN RECORD | My launch plan, In-Stock First (own document) | 2026-10-10 |
| c02.rc.lock | DOOR LOCK · KEYPAD, SERVO, LCD · JAN 2023 | OWN RECORD | Own Arduino project files, 7 January 2023 | 2026-10-10 |
| c02.rc.safe | FINGERPRINT SAFE · APR 2024 | OWN RECORD | Babson report of 20 April 2024 (own document) | 2026-10-10 |
| c02.rc.game | SPACE CLEANER · FIRST UNITY GAME · APP STORE | OWN RECORD | Own record; own Unity project files | 2026-10-10 |
| c02.ev.apparel | BACKEND ENGINEERING · APPAREL BRAND | OWN RECORD | Own record; my current website | 2026-10-10 |
| c03.kicker | SKI RACING · VERMONT · BABSON · ETOWER | DERIVED FROM DATA | Chapter 03 summary of the places and years below | 2026-10-10 |
| c03.quote.competition | Competition made me who I am. Sport taught me discipline, resilience, and perseverance. | OWN RECORD | Own wording, given for the site on 17 Oct 2026; earlier form on my eTower introduction slide, February 2024 (own document) | 2026-10-10 |
| c03.body.fae | From the age of fifteen to eighteen, I competed at an elite level around the world as a member of the Andorran Ski Federation (FAE). | OWN RECORD | Own record; Federació Andorrana d'Esquí team | 2026-10-10 |
| c03.body.left | At fifteen I left Andorra and moved alone to a ski academy in Vermont, two months after starting Mevot. | OWN WORDING | Green Mountain Valley School enrolment, August 2021; Mevot from May 2021 | 2026-10-10 |
| c03.body.us | Then Babson, with engineering at Olin. I raced two seasons for Babson's men's ski team, and from 2025 to 2026 I was President of eTower, … | OWN RECORD | Babson roster; eTower board report, spring 2026; own record | 2026-10-13 |
| c03.quote.etower | Build the Tower. Build the network. Get capital and resources to the founders doing the work. | OWN RECORD | eTower semester report, spring 2026 (own document) | 2026-10-10 |
| c03.chip.gmvs | GREEN MOUNTAIN VALLEY SCHOOL · 2021–2023 | VERIFIED | Green Mountain Valley School enrolment | 2026-10-10 |
| c03.chip.babson | BABSON MEN'S SKI TEAM · TWO SEASONS | OWN RECORD | Own record; Babson athletics roster | 2026-10-13 |
| c03.chip.etower | ETOWER · FOUNDED 2001 · 21 RESIDENTS | OWN RECORD | eTower semester report, spring 2026 | 2026-10-10 |
| c03.dot.1 | WAITSFIELD, VERMONT · 2021 | VERIFIED | Green Mountain Valley School, Waitsfield, Vermont | 2026-10-10 |
| c03.dot.2 | BABSON COLLEGE, MASSACHUSETTS · 2023 | VERIFIED | Babson College matriculation, fall 2023 | 2026-10-10 |
| c03.dot.3 | MIT MEDIA LAB, CAMBRIDGE · 2025 | OWN RECORD | MIT Media Lab City Science, visiting student researcher Jul 2025 – Jul 2026; RSA since Sep 2026 | 2026-10-10 |
| model.figures_note | Figures use the corrected model run (updated September 2026), which supersedes some values in the published article; a formal correction … | MODEL · CONDITIONAL | Scenario engine, R3c corrected run, September 2026 | 2026-10-10 |
| model.footer | R3C CORRECTED RUN · FIGURES UPDATED SEPTEMBER 2026 · CONDITIONAL SCENARIOS, NOT FORECASTS | MODEL · CONDITIONAL | Scenario engine, R3c corrected run, September 2026 | 2026-10-10 |
| model.pop2024 | 87,097 | MODEL · CONDITIONAL | Official statistics: resident population, 31 December 2024 | 2026-10-10 |
| model.grid2024 | 82,456 | MODEL · CONDITIONAL | Andorra 12,474-cell grid, 2024 base | 2026-10-10 |
| model.pop2049.continuity | 116,758 | MODEL · CONDITIONAL | Scenario engine, R3c corrected run, September 2026 | 2026-10-10 |
| model.pop2049.overgrowth | 175,070 | MODEL · CONDITIONAL | Scenario engine, R3c corrected run, September 2026 | 2026-10-10 |
| model.pop2049.degrowth | 72,268 | MODEL · CONDITIONAL | Scenario engine, R3c corrected run, September 2026 | 2026-10-10 |
| model.pop2049.density | 117,810 | MODEL · CONDITIONAL | Scenario engine, R3c corrected run, September 2026 | 2026-10-10 |
| model.range | In 2049 the population ranges from 72,268 (Degrowth) to 175,070 (Overgrowth); figures updated September 2026. | MODEL · CONDITIONAL | Scenario engine, R3c corrected run, September 2026 | 2026-10-10 |
| model.drivers | Continuity grows GDP by 2.5% a year, Overgrowth by 6.0%, Degrowth shrinks it by 1.5% with an explicit migration restriction, and Density … | MODEL · CONDITIONAL | Scenario definitions in the R3c run | 2026-10-10 |
| model.grid_vs_national | The grid's 2024 base is 82,456 residents; the national series says 87,097. The terrain shows where growth lands, the national series how … | MODEL · CONDITIONAL | 12,474-cell grid and the official national series | 2026-10-10 |
| model.correction_effect | Correcting the model moved three 2049 endpoints by under 4%, while Density rose 19% and now edges past Continuity. | MODEL · CONDITIONAL | Scenario engine, published run against the R3c corrected run | 2026-10-10 |
| model.elasticity | Population follows GDP through a lagged labor-immigration elasticity (0.40 at one year, 0.10 at two). 500 Monte Carlo draws. Outputs are … | MODEL · CONDITIONAL | Urban Science 10(8):450 (2026), methods | 2026-10-10 |
| model.backcast | Population follows the economy through a lagged labor-immigration link. In the September 2026 update that choice back-casts population wo… | MODEL · CONDITIONAL | Scenario engine back-cast, R3c corrected run (updated September 2026) | 2026-10-10 |
| model.signature | Signature numbers | MODEL · CONDITIONAL | Scenario engine, R3c corrected run, September 2026 | 2026-10-10 |
| c04.kicker | MY WORK AT THE MIT MEDIA LAB | OWN RECORD | MIT Media Lab people page | 2026-10-10 |
| c04.card | My work at the MIT Media Lab | OWN RECORD | MIT Media Lab people page | 2026-10-10 |
| c04.card.sub | CITY SCIENCE GROUP · SINCE 2025 | VERIFIED | MIT Media Lab people page | 2026-10-10 |
| c04.title | What if you could simulate a country 25 years into the future? | OWN RECORD | Own wording (owner direction, 12 Oct 2026); the model runs 2024 to 2049 | 2026-10-10 |
| c04.body.left | Four years after I left, I started to model it. | OWN WORDING | Left Andorra August 2021; City Science from summer 2025 | 2026-10-10 |
| c04.quote.rollback | You can't roll back a country. | OWN RECORD | My vision document, July 2026, page 2 (own document) | 2026-10-10 |
| c04.quote.rehearsal | A budget, a tax reform, and a residency rule go straight onto a live population, with no rehearsal. | OWN RECORD | My vision document, July 2026, page 2 (own document) | 2026-10-10 |
| c04.body.engine | So I built a way to try them first. The Andorra Scenario Engine runs the country's economy, population, housing, tourism, energy and envi… | OWN WORDING | Urban Science 10(8):450 (2026) | 2026-10-10 |
| c04.method | Population follows the economy through a lagged labor-immigration link. | MODEL · CONDITIONAL | Urban Science 10(8):450 (2026), methods | 2026-10-10 |
| c04.quote.parish | small enough that a national-scale model can be built by a small team, and consequential enough that the model can actually inform policy | OWN RECORD | My onboarding document for an intern (own document) | 2026-10-10 |
| c04.block.degrowth | Degrowth needs an explicit migration restriction. No calibrated setting shrinks the population from economic decline alone. | OWN WORDING | Scenario definitions in the R3c run; backend summary | 2026-10-10 |
| c04.interp | Cells between 2024 and 2049 are interpolated along each scenario's national curve; the grid stores 2024 and 2049. | DERIVED FROM DATA | How this site draws the grid between its two stored years | 2026-10-10 |
| c04.record | Cell pattern from 2024; national total from official statistics. | DERIVED FROM DATA | How this site draws years up to 2024 | 2026-10-10 |
| c04.hud | {{grid}} RESIDENTS MAPPED IN THE GRID · WHERE, NOT HOW MUCH | DERIVED FROM DATA | Grid totals and the national series of this site | 2026-10-10 |
| c04.aria | {{scenario}}, {{year}}: {{pop}} residents nationally; electricity {{elec}} GWh a year; hospital beds {{beds}} per 1,000 service population. | DERIVED FROM DATA | Generated from the scenario series | 2026-10-10 |
| c05.kicker | THE PEOPLE · A SYNTHETIC POPULATION | OWN WORDING | Population run P04 | 2026-10-10 |
| c05.title | 90,000 | DERIVED FROM DATA | Population run P04, frozen September 2026 | 2026-10-10 |
| c05.quote.layer | This is the country-level layer. Underneath it are 90,000 synthetic residents. | OWN RECORD | Carousel slide 5, my own design | 2026-10-10 |
| c05.quote.question | Does making agents more human make a synthetic population behave more like the real one? | OWN RECORD | My Media Lab application progress deck, slide 14 (own document) | 2026-10-10 |
| c05.scope | A synthetic population at national scale: 90,000 agents for a country of about 87,000 people. | DERIVED FROM DATA | Population run P04 against the official register | 2026-10-10 |
| c05.body.method | A language model wrote 123 archetypes, for $9.55 in total. Deterministic code did the rest: 90,000 people in 40,520 households, 575,405 s… | OWN WORDING | Run metadata of population P04, frozen September 2026 (own repository) | 2026-10-10 |
| c05.trips | TRIPS · 313,527 A DAY · 99.75% ROUTED ON THE STREET NETWORK · FROZEN RUN, SEP 2026 | DERIVED FROM DATA | Run metadata of population P09, September 2026 (own repository) | 2026-10-10 |
| c05.guess | Before I show you anything: Andorra gets about 9.6 million visitors a year. What share of its residents name tourism among the country's … | OWN WORDING | Visitors per year, 2024 (tourists and day-trippers), R3c series | 2026-10-10 |
| c05.reveal | 81% of my first synthetic population did. 1% of real residents did. I wrote it up as a fail. | OWN WORDING | Round 1 findings, July 2026 (own repository) | 2026-10-10 |
| c05.reveal.incharacter | Answering in character, 21% of my first synthetic population still did, against 1% of real residents. | OWN WORDING | Round 1 findings, July 2026 (own repository) | 2026-10-10 |
| c05.quote.faithful | The population was demographically faithful and attitudinally wrong. | OWN RECORD | My independent research proposal, part II (own document) | 2026-10-10 |
| c05.round2 | The second test was pre-specified and hash-locked in my repository before a single run, an internal pre-registration. I ran it once, on 8… | OWN WORDING | Round 2 findings, 8 September 2026 (own repository) | 2026-10-10 |
| c05.rename | Then I renamed the country. | OWN WORDING | Anonymised-country control, Round 2 | 2026-10-10 |
| c05.rename.after | Results moved by less than 0.11 standard deviations. | OWN WORDING | Anonymised-country control, Round 2 findings (own repository) | 2026-10-10 |
| c05.rename.control | A twin population with Andorra's names removed reproduced the results, so the fit and the misfit travel with the generated profile, not w… | OWN WORDING | Anonymised-country control, Round 2 findings (own repository) | 2026-10-10 |
| c05.quote.recognising | The model recognising 'Andorra' therefore explains neither the housing fit nor the economic misfit. | OWN RECORD | Round 2 findings, section 4.4 (own repository) | 2026-10-10 |
| c05.quote.honest | The honest answer to the question at the top is still partly no – structure transfers and dispositions do not yet. | OWN RECORD | Speaker notes of my progress deck, slide 14 (own document) | 2026-10-10 |
| c05.credit.behavior | BEHAVIOR LAYER: h-ABM, CITY SCIENCE TEAM · SOURCED | VERIFIED | City Science h-ABM preprint (TechRxiv), Mora, Atchade-Adelomou, Alonso-Pastor et al. | 2026-10-10 |
| c05.credit.made | POPULATION, COUPLING, VALIDATION · MADE | OWN RECORD | My repository history (three coupling commits, 11-15 August 2026) | 2026-10-10 |
| c05.trips.caption | A sample of 5,239 routed synthetic trips · frozen run, Sep 2026 · not real people | DERIVED FROM DATA | public/data/agent-trips.json (sampling.sampled_trips); the card builds this line from the file | 2026-10-10 |
| c05.trips.osm | Street geometry © OpenStreetMap contributors (ODbL) | DERIVED FROM DATA | public/data/agent-trips.json and roads.json attribution | 2026-10-10 |
| c05.chip.wp | IN PREPARATION · SYNTHETIC AGENT GENERATION PIPELINE | OWN RECORD | Paper in preparation, working title (owner direction, 12 October 2026) | 2026-10-10 |
| c05.chip.telecom | IN PREPARATION · MOBILITY VALIDATION AGAINST TELECOM DATA · COMPARISON NOT YET RUN | OWN RECORD | Paper in preparation, working title (owner direction, 12 October 2026); draft telecom data request, not sent (own document) | 2026-10-10 |
| c05.chip.survey | IN PREPARATION · PUBLIC OPINION VALIDATION AGAINST SURVEYS | OWN RECORD | Paper in preparation, working title (owner direction, 12 October 2026); Round 1 and Round 2 findings (own repository) | 2026-10-10 |
| c05.quote.licenses | Every generated field carries one of four evidence licenses, so an assumption cannot be read downstream as a measurement. | OWN RECORD | Sentence from my unsubmitted method draft (own document) | 2026-10-10 |
| c05.block.constructs | Only four constructs in the whole schema can be checked against real surveys at all. The synthetic unemployment rate is 9.63%; the offici… | OWN WORDING | Validation schema and unemployment check, Round 2 (own repository) | 2026-10-10 |
| c06.kicker | ROUND 2 · WHERE I'M WRONG | OWN WORDING | Round 2 findings, 8 September 2026 | 2026-10-10 |
| c06.quote.negative | A negative result that is correctly measured is a contribution. A positive result that is a measurement artifact is a liability. | OWN RECORD | My onboarding document for an intern (own document) | 2026-10-10 |
| c06.body.racing | In racing, a missed gate is on the record and everyone sees it. I keep my research the same way. | OWN WORDING | Round 2 findings and Round 1 write-up | 2026-10-10 |
| c06.slalom | ROUND 2 · 24 ITEMS · 5 PASSED · 12 MISSED (6 NOT EQUIVALENT, 6 OUTSIDE POLICY TOLERANCE) · 7 NOT SCORED (1 INCONCLUSIVE, 4 DESCRIPTIVE ON… | DERIVED FROM DATA | Round 2 scorecard, 8 September 2026 (own repository) | 2026-10-10 |
| c06.verdicts | Round 2 by verdict class | DERIVED FROM DATA | Round 2 scorecard, 8 September 2026 (own repository) | 2026-10-10 |
| c06.quote.frozen | No verdict rule, prompt, model, matching or population parameter was changed after seeing any result. | OWN RECORD | Round 2 findings (own repository) | 2026-10-10 |
| c06.replay.1 | JUL 2026 · MY OWN ENGINE | OWN WORDING | Engine audit, July 2026 (own repository) | 2026-10-10 |
| c06.replay.2 | JUL 2026 · ROUND 1 | OWN WORDING | Round 1 findings, July 2026 (own repository) | 2026-10-10 |
| c06.replay.3 | JUN–JUL 2026 · PEER REVIEW | OWN WORDING | Review correspondence and the published abstract | 2026-10-10 |
| c06.replay.4 | 8 SEP 2026 · ROUND 2 | OWN WORDING | Round 2 scorecard, 8 September 2026 (own repository) | 2026-10-10 |
| c07.kicker | ELSEWHERE · SAME PIPELINE, OTHER PLACES | OWN WORDING | Pipeline repositories | 2026-10-10 |
| c07.headline.default | The pipeline is built to move. | OWN WORDING | Pipeline repositories | 2026-10-10 |
| c07.headline.d4 | I ran the same pipeline on the Media Lab's own neighborhood. | OWN WORDING | Kendall Square world README, frozen commit e94f175 (private repository) | 2026-10-13 |
| c07.quote.pack | Adding another place requires a new pack and compatible assets, not an engine fork. | OWN RECORD | README of my coupling repository (private) | 2026-10-10 |
| c07.kendall.body | Pointed at Kendall Square, the same code built 118,403 residents in 564 populated cells of Cambridge, and 141,368 inbound commuters, with… | OWN WORDING | Kendall Square world README and held-out benchmark v2, frozen commit e94f175 (private repository); public/data/kendall-cells.json facts and totals | 2026-10-13 |
| c07.kendall.synthetic | Every one of them is synthetic. The map shows how many live in each cell, never a person. | OWN RECORD | public/data/kendall-cells.json (resident counts per H3 cell only) | 2026-10-13 |
| c07.trips.label | TRIPS · TIER 3.1 DEMO EXPORT · EXPERIMENTAL, NOT ADOPTED | DERIVED FROM DATA | Kendall Square world docs/TIER3.1_DEMO.md ('Status: experimental, not adopted'); public/data/kendall-trips.json label | 2026-10-13 |
| c07.kendall.credits | MAP DATA · OPENSTREETMAP (ODBL) · MBTA GTFS · U.S. CENSUS | DERIVED FROM DATA | Attribution strings of public/data/kendall-cells.json, kendall-trips.json and kendall-transit.json | 2026-10-13 |
| c07.quote.intercepts | Fitting intercepts moves the miss, it does not remove it. | OWN RECORD | HOW_WE_BUILT_IT, section 8 (own repository) | 2026-10-10 |
| c07.chip.walk | WALKED WORK LEGS · 9.6% VS 28.2% (CENSUS) | DERIVED FROM DATA | Kendall Square world README (private repository) | 2026-10-13 |
| c07.chip.tier | TIER 3.1 · NOT ADOPTED (ONE METRIC WORSENED) | DERIVED FROM DATA | Kendall Square world README (private repository) | 2026-10-13 |
| c07.concepcion | CONCEPCIÓN, CHILE · SCOPED · NOT GENERATED | OWN RECORD | Concepción world dry run; paid step not run (own repository) | 2026-10-10 |
| c07.concepcion.body | A third place is in progress. Its population has not been generated. | OWN WORDING | Concepción world dry run (own repository) | 2026-10-10 |
| c07.sf | In a week in May 2026 I also measured San Francisco's lots: 63,119 of 189,636 residential-zoned lots fit the small-lot band that single-s… | OWN WORDING | SF small-lot explorer, public repository | 2026-10-10 |
| c08.kicker | THE DEMONSTRATION · TABLE, SCREENS, PANEL | OWN WORDING | Chapter 08 summary of the installation (owner directions 4) | 2026-10-10 |
| c08.quote.demos | Demos are arguments. | OWN RECORD | My internship reflection (own document) | 2026-10-10 |
| c08.quote.minister | A minister doesn't care about your RMSE. A 3D-printed terrain model they can walk up to changes the conversation entirely. | OWN RECORD | My internship reflection (own document) | 2026-10-10 |
| c08.body | A model on a laptop is read by one person at a time. I built this one to be stood around: a terrain model of Andorra with the scenarios p… | OWN WORDING | Own project notes on the physical table; my photographs of the demonstration, 1 May 2026 | 2026-10-10 |
| c08.visit | On 1 May 2026 a delegation from Andorra's government came to the Media Lab. I ran the model for them live, at this table, with the panel … | OWN WORDING | Own record; photographs of the visit, 1 May 2026 (credit: gbila) | 2026-10-10 |
| c08.panel | The panel is one part of it, and you have been using it. In November 2025 it was an Arduino with four buttons and one potentiometer. By A… | OWN WORDING | Firmware ScenarioSlider.ino v1.4 and its earlier sketch | 2026-10-10 |
| c08.quote.translation | The hardest work is translation. | OWN RECORD | My internship reflection (own document) | 2026-10-10 |
| c08.table | The terrain model is about 1.2 meters square, with the dashboard projected onto it and calibrated to four corner coordinates. | OWN WORDING | Own project notes on the physical table | 2026-10-10 |
| c08.caption | PROJECTION-MAPPED TERRAIN MODEL OF ANDORRA · 22 APR 2026 | OWN RECORD | My photograph of the lit terrain table, 22 April 2026 | 2026-10-10 |
| c08.parts | The panel cost $109.18 in parts and was built for the delegation visit. | OWN WORDING | My parts receipts for the panel | 2026-10-10 |
| c08.spec.encoders | 2× EC11 ROTARY ENCODER WITH PUSH | DERIVED FROM DATA | Firmware ScenarioSlider.ino v1.4 header | 2026-10-10 |
| c08.spec.faders | 2× 10K SLIDE POT · YEAR 2010–2049 · ZOOM STREET → COUNTRY | DERIVED FROM DATA | Firmware ScenarioSlider.ino v1.4 header | 2026-10-10 |
| c08.spec.buttons | 6× PUSH BUTTON · TABS | DERIVED FROM DATA | Firmware ScenarioSlider.ino v1.4 header | 2026-10-10 |
| c08.spec.toggles | 5× TOGGLE · SCENARIO OVERLAYS | DERIVED FROM DATA | Firmware ScenarioSlider.ino v1.4 header | 2026-10-10 |
| c08.spec.mega | ATMEGA2560 CONTROLLER · 16 MHZ | DERIVED FROM DATA | Firmware ScenarioSlider.ino v1.4 header | 2026-10-10 |
| c08.spec.baud | JSON · 115,200 BAUD | DERIVED FROM DATA | Firmware ScenarioSlider.ino v1.4 header | 2026-10-10 |
| c08.protocol | {"type":"status","value":"ready"} heartbeat at start and every 5 s {"type":"year","value":2027} A0, 2010-2049, only on integer change {"t… | DERIVED FROM DATA | Firmware ScenarioSlider.ino v1.4, 115,200 baud over Web Serial | 2026-10-10 |
| c10.kicker | NOW · AS OF 18 October 2026 | DERIVED FROM DATA | Build date of this page | 2026-10-10 |
| c10.headline | Everything above this line happened. Everything below it is a model. | OWN WORDING | The record and model rule of this site | 2026-10-10 |
| c10.caption | R3C CORRECTED RUN · FIGURES UPDATED SEPTEMBER 2026 · CONDITIONAL SCENARIOS, NOT FORECASTS | MODEL · CONDITIONAL | Scenario engine, R3c corrected run, September 2026 | 2026-10-10 |
| c10.board.header | 2024 → 2049 · {{scenario}} | DERIVED FROM DATA | Generated from the scenario series | 2026-10-10 |
| c10.line.aria | {{scenario}}: 2049 population {{n}} | DERIVED FROM DATA | Generated from the scenario series | 2026-10-10 |
| c10.after | None of these lines is a forecast. They are there so a country can argue about which one it wants before it is on it. | OWN WORDING | The record and model rule of this site | 2026-10-10 |
| c10.section | WHAT I WANT TO TEST NEXT · DRAFT · TO BE PRE-REGISTERED | OWN WORDING | content/prereg.md | 2026-10-10 |
| c10.q1 | Do dispositions grounded in local survey data reproduce how residents rank their country's concerns in a second place, without retuning? | OWN WORDING | Round 2 issue-rank result and the second-place pipeline | 2026-10-10 |
| c10.q2 | Can trips derived only from profiles, census, streets and timetables come within a pre-declared tolerance of held-out phone data? | OWN WORDING | Telecom protocol (drafted, request not sent) | 2026-10-10 |
| c10.q3 | Does operating a simulation together, on a physical table, change how a group reasons about a policy, compared with reading the same resu… | OWN WORDING | My own stated belief that demos are arguments | 2026-10-10 |
| c10.hash | sha256 54ec8e42 · DEPOSIT: NOT YET | DERIVED FROM DATA | SHA-256 of content/prereg.md, computed at build | 2026-10-10 |
| c10.quote.apply | Applying to the MIT Media Lab master's programme to keep building synthetic societies. | OWN RECORD | My Media Lab application progress deck, slide 15 (own document) | 2026-10-10 |
| c10.quote.tools | building tools that help imagine and test better futures, and then helping turn the best ideas into reality | OWN RECORD | My fall 2025 reflection (own document) | 2026-10-10 |
| c10.landing | MIT MEDIA LAB · CAMBRIDGE | DERIVED FROM DATA | MIT Media Lab | 2026-10-10 |
| c11.research | City Science, MIT Media Lab. Supervision: Kent Larson, with Luis Alonso. | VERIFIED | MIT visiting-student letters | 2026-10-10 |
| c11.paper | Marcel Bartumeu Ramentol, Parfait Atchade-Adelomou, Adrian Mora-Carrero, Kent Larson, Luis Alonso-Pastor, Marta Domenech (Andorra Researc… | VERIFIED | Urban Science 10(8):450 (2026), author list | 2026-10-10 |
| c11.habm | City Science team (TechRxiv preprint: Mora, Atchade-Adelomou, Alonso-Pastor et al.). My part: the coupling, identification checks and out… | VERIFIED | h-ABM preprint (TechRxiv); my coupling commits, August 2026 | 2026-10-10 |
| c11.partner | Andorra Research + Innovation. | VERIFIED | Urban Science 10(8):450 (2026), affiliations | 2026-10-10 |
| c11.data | Official Andorran statistics and inventories (Departament d'Estadística and the other sources cited per series in the source drawer), Obs… | VERIFIED | Source strings of scenarios.json; survey data agreement | 2026-10-10 |
| c11.maps | Street geometry © OpenStreetMap contributors (ODbL). Globe land: Natural Earth 1:110m (public domain). Globe idea after Three.js Interact… | DERIVED FROM DATA | public/data/agent-trips.json and roads.json attribution; Natural Earth terms of use | 2026-10-10 |
| c11.kendall | A synthetic population of the City of Cambridge, Massachusetts, which I built in 2026; every person in it is fictional. Census geography:… | DERIVED FROM DATA | Attribution strings of public/data/kendall-cells.json, kendall-trips.json and kendall-transit.json | 2026-10-13 |
| c11.site | Design and code: Marcel Bartumeu Ramentol. Type: the system's SF Pro where the device has it, otherwise Inter; Syne and Instrument Serif … | OWN RECORD | Font licences and library notices | 2026-10-10 |
| c11.colophon | The colophon, the changelog and the claims count are on the credits page. | DERIVED FROM DATA | This site | 2026-10-10 |
| contact.email | marcelbr@mit.edu | OWN RECORD | My MIT address, shown on my current site | 2026-10-10 |
| contact.email2 | marcel@bartumeu.com | OWN RECORD | My personal address, on my CV | 2026-10-10 |
| contact.github | github.com/marcelbartumeu | VERIFIED | GitHub profile | 2026-10-10 |
| contact.orcid | ORCID 0009-0005-9195-8787 | VERIFIED | ORCID record | 2026-10-10 |
| contact.mit | media.mit.edu/people/marcelbr | VERIFIED | MIT Media Lab people page | 2026-10-10 |
| work.about.2 | Since 2025 I have worked with the City Science group at the MIT Media Lab, and I am first author of The Andorra Scenario Engine (Urban Sc… | OWN WORDING | MIT Media Lab people page; Crossref record of the paper | 2026-10-10 |
| work.about.2c | I am first author of The Andorra Scenario Engine (Urban Science, 2026), published in August 2026. | OWN WORDING | Crossref and MIT DSpace record of the paper | 2026-10-10 |
| work.about.3 | I built a synthetic population of 90,000 people for my country, and the projection and Arduino control panel for a terrain table of Andorra. | OWN WORDING | Population run P04; own panel firmware | 2026-10-10 |
| work.about.4 | Before that I built things: radio-controlled boats, cars and planes, an electric car at fourteen, and Mevot, an electric-longboard compan… | OWN RECORD | Own print files, kart photographs, Mevot archive | 2026-10-10 |
| work.about.5 | I study at Babson College and Olin College. I race alpine skiing for Andorra (FIS-licensed), raced two seasons for Babson's men's ski tea… | OWN RECORD | FIS profile; Babson roster; eTower board report | 2026-10-13 |
| start.1.abstract | Four conditional futures for Andorra to 2049, calibrated on fifteen years of official statistics. I am the first of six authors. | OWN WORDING | Urban Science 10(8):450 (2026) | 2026-10-10 |
| start.2.abstract | A synthetic Andorra of 90,000 people, tested against real surveys twice. Round 1 failed and Round 2 missed 12 of 24 items; both are repor… | OWN WORDING | Round 1 and Round 2 findings (own repository) | 2026-10-10 |
| start.3.abstract | A projection-mapped terrain model of Andorra and the Arduino panel that drives it, built for a government delegation's visit. | OWN WORDING | Own project notes and parts receipts | 2026-10-10 |
| start.2.chip.hash | ROUND 2 HASH-LOCKED BEFORE RUN | OWN RECORD | Round 2 specification and its hash (own repository) | 2026-10-10 |
| start.2.chip.misses | MISSES REPORTED | OWN RECORD | Round 1 and Round 2 findings (own repository) | 2026-10-10 |
| start.3.chip.hw | HARDWARE V1.4 · APR 2026 | DERIVED FROM DATA | Firmware ScenarioSlider.ino v1.4 | 2026-10-10 |
| pub.1 | Bartumeu Ramentol, M.; Atchade-Adelomou, P.; Mora-Carrero, A.; Larson, K.; Alonso-Pastor, L.; Domenech, M. The Andorra Scenario Engine: A… | VERIFIED | Urban Science 10(8):450, published 5 August 2026 | 2026-10-10 |
| pub.1.published | PUBLISHED 5 AUG 2026 | VERIFIED | Crossref, MIT DSpace and ORCID records | 2026-10-10 |
| pub.1.first | FIRST AND CORRESPONDING AUTHOR (1 OF 6) | VERIFIED | Urban Science author list | 2026-10-10 |
| pub.1.oa | OPEN ACCESS, CC BY 4.0 | VERIFIED | Crossref licence record; MIT DSpace | 2026-10-10 |
| pub.1.issue | URBAN SCIENCE 10(8):450 | VERIFIED | Urban Science 10(8):450 | 2026-10-10 |
| pub.1.zenodo | CODE ARCHIVE (V1.0, MAY 2026, THE SUBMITTED VERSION) | VERIFIED | Zenodo archive 10.5281/zenodo.20225094 | 2026-10-10 |
| pub.prep.pipeline | Synthetic Agent Generation Pipeline (working title): the methodology and architecture for generating realistic synthetic populations. | OWN RECORD | Paper in preparation, working title (owner direction, 12 October 2026); not submitted | 2026-10-10 |
| pub.prep.mobility | Mobility Validation Against Telecom Data (working title): validating simulated agent movements against real-world telecommunications data… | OWN RECORD | Paper in preparation, working title (owner direction, 12 October 2026); not submitted | 2026-10-10 |
| pub.prep.opinion | Public Opinion Validation Against Surveys (working title): comparing synthetic agents' opinions and responses against actual survey data.… | OWN RECORD | Paper in preparation, working title (owner direction, 12 October 2026); Round 1 and Round 2 findings (own repository) | 2026-10-10 |
| pub.prep.status | IN PREPARATION | OWN RECORD | Paper in preparation, working title (owner direction, 12 October 2026); not submitted | 2026-10-10 |
| pub.prep.mobility.status | COMPARISON NOT YET RUN | OWN RECORD | Paper in preparation, working title (owner direction, 12 October 2026) | 2026-10-10 |
| pub.3 | Catalan translation of the paper, built on a 455-entry Termcat and IEC glossary. Unpublished. | OWN RECORD | My translation glossary (own document) | 2026-10-10 |
| press.1 | Diari d'Andorra, 27 September 2026: Imaginant l'Andorra del futur. The feature reports the published version. | VERIFIED | Diari d'Andorra, 27 September 2026 | 2026-10-10 |
| press.2 | Andorra Research + Innovation presented the work publicly in September 2026. | VERIFIED | Andorra Research + Innovation announcement | 2026-10-10 |
| cv.edu.babson | Babson College, BSBA, concentration in Technological Entrepreneurship; graduating May 2027. | OWN RECORD | Own record, CV | 2026-10-10 |
| cv.edu.olin | Olin College of Engineering, General Engineering certificate (in progress); coursework includes Modeling and Simulation. | OWN RECORD | Own record, CV | 2026-10-10 |
| cv.edu.gmvs | Green Mountain Valley School, Vermont, a ski academy: high school diploma, 2021–2023. I moved there alone at 15. | OWN RECORD | Own record, CV; school enrolment | 2026-10-10 |
| cv.edu.escola | Escola Andorrana, Ordino, 2017–2021. | OWN RECORD | Own record, CV | 2026-10-10 |
| cv.research.babson | Babson independent research on the Andorra model, advised by Ken Accardi, 2026. | OWN RECORD | Babson independent research record | 2026-10-10 |
| cv.ventures.mevot | Mevot: started at 15; built and sold electric longboards under my own brand; grew to six figures in annual revenue; exited in 2025 throug… | OWN RECORD | Own Mevot archive; own record | 2026-10-10 |
| cv.ventures.apparel | Helped on the backend of a clothing brand doing over seven figures in ARR. | OWN RECORD | Own record; my current website | 2026-10-10 |
| cv.ventures.frame | Frame Creators: an extreme-sports film studio I co-founded with my brother (film archive 2015–2024). | OWN RECORD | Own record, CV | 2026-10-10 |
| cv.leadership | eTower, Babson's founder community: President for one year (2025–26), now Advisor. | OWN RECORD | eTower board report, spring 2026; Babson page | 2026-10-10 |
| cv.athletics | Andorran Ski Federation team (FAE) from 15 to 18. FIS-licensed alpine racer for Andorra; two seasons on Babson's men's ski team. Skied 10… | OWN RECORD | FIS athlete profile; Babson roster; own record | 2026-10-13 |
| cv.tools | Python; JavaScript and React; Arduino; system dynamics and agent-based modeling; Blender, Fusion 360, Shapr3D; FDM printing; DaVinci Reso… | OWN RECORD | Own record, CV | 2026-10-10 |
| about.portrait | CAMBRIDGE, MA · 24 JUL 2025 | OWN RECORD | My own photograph | 2026-10-10 |
| about.card | The real one | OWN RECORD | Fields of my own card, each with its own source | 2026-10-10 |
| ase.problem | National plans are built from indicator trajectories with no mechanism a decision-maker can pull, and a small country cannot learn from a… | OWN WORDING | Urban Science 10(8):450, introduction | 2026-10-10 |
| ase.question | Can a small team build a data-grounded, auditable national model that compares conditional futures instead of predicting one? | OWN WORDING | Urban Science 10(8):450, abstract | 2026-10-10 |
| ase.calibration | The model is calibrated on official statistics from 2010 to 2024 and run to 2049 under four scenarios. | OWN WORDING | Urban Science 10(8):450, methods | 2026-10-10 |
| ase.status | Published 5 August 2026. Conditional scenarios, not forecasts. | OWN WORDING | Urban Science 10(8):450 | 2026-10-10 |
| ase.role | As listed in the publisher's Author Contributions, I did the conceptualization (with Kent Larson), methodology (with Luis Alonso-Pastor),… | VERIFIED | Urban Science 10(8):450, Author Contributions | 2026-10-10 |
| sp.problem | A scenario engine has no people in it, and policies land on people who support, reject or adapt to them. | OWN WORDING | My Catalan pitch, February 2026 | 2026-10-10 |
| sp.system | A language model writes 123 archetypes ($9.55 in total); deterministic code expands them into 90,000 people (76,321 adults, 13,679 childr… | OWN WORDING | Run metadata of populations P04 and P09, frozen September 2026 (own repository) | 2026-10-10 |
| sp.ev.households | Household structure matches the official register (band TVD 0.0043). | OWN WORDING | Household reconciliation against the official register (own repository) | 2026-10-10 |
| sp.ev.overall | Across all items, 12 of 24 missed. | OWN WORDING | Round 2 scorecard, 8 September 2026 (own repository) | 2026-10-10 |
| sp.limits.telecom | Mobility is unvalidated: a telecom-based validation protocol is drafted, the data request has not been sent, and the comparison has not b… | OWN RECORD | Draft telecom data request, not sent (own document) | 2026-10-10 |
| sp.role | I designed the pipeline, the validation harness, the pre-specifications and the coupling contracts. | OWN RECORD | Own record; repository history | 2026-10-10 |
| sp.credits.survey | Survey data: Observatori d'Andorra, aggregates only. Supervision: Kent Larson, Luis Alonso. | VERIFIED | Survey data agreement; MIT visiting-student letters | 2026-10-10 |
| tp.quote | A minister doesn't care about your RMSE. | OWN RECORD | My internship reflection (own document) | 2026-10-10 |
| tp.question | Can a national model be operated by hand, in a room, by the people who decide? | OWN WORDING | My statement that demos are arguments | 2026-10-10 |
| tp.system | A terrain model of Andorra (about 1.2 m square) with the dashboard projected onto it, calibrated to four corner coordinates and masked ou… | OWN WORDING | Own project notes; firmware ScenarioSlider.ino v1.4 | 2026-10-10 |
| tp.evidence | This site speaks the panel's own protocol: plug a real panel into a laptop and it drives the whole page. | OWN WORDING | Firmware ScenarioSlider.ino v1.4; this site's bus | 2026-10-10 |
| tp.limits | Built and shown to a delegation in May 2026. The table size is approximate, and the panel is a prototype at hardware version 1.4, not a p… | OWN WORDING | Own project notes | 2026-10-10 |
| tp.role | I designed, wired and programmed the panel, and built the projector view and the dual-screen sync. | OWN WORDING | Own repository and firmware | 2026-10-10 |
| tp.credits | The tangible-table tradition is City Science's (CityScope); the 2017 CityScope Andorra table was the group's. | VERIFIED | MIT Media Lab article, October 2017 | 2026-10-10 |
| sf.problem | Single-stair code reform would make small lots buildable for small apartment buildings. How many lots qualify? | OWN WORDING | SF small-lot explorer README | 2026-10-10 |
| sf.system | The minimum rotated rectangle of every parcel in the San Francisco Assessor map (about 226,000 parcels), in US feet; an interactive Leafl… | OWN WORDING | SF small-lot explorer README | 2026-10-10 |
| sf.ev.sf | 29,701 (15.7%) ideal-band and 63,119 (33.3%) flexible-band lots out of 189,636 residential-zoned lots. | DERIVED FROM DATA | SF small-lot explorer README | 2026-10-10 |
| sf.ev.culver | In Culver City only 14 of 7,401 residential parcels (0.2%) fall in the flexible band. | DERIVED FROM DATA | SF small-lot explorer Culver City report | 2026-10-10 |
| sf.limits | The counts are a geometric upper bound. | OWN WORDING | SF small-lot explorer README | 2026-10-10 |
| sf.role | I wrote the analysis and the viewer; the first version took a week in May 2026. | OWN WORDING | SF small-lot explorer repository history | 2026-10-10 |
| sf.chip.live | LIVE | VERIFIED | SF small-lot explorer, live map | 2026-10-10 |
| sf.chip.code | PUBLIC CODE | VERIFIED | GitHub repository | 2026-10-10 |
| sf.chip.dates | MAY–JUN 2026 | VERIFIED | Repository history | 2026-10-10 |
| mevot.chip.founded | FOUNDED 2021 | OWN RECORD | Brand on grip tape, 29 May 2021; my earlier CV | 2026-10-10 |
| mevot.chip.exit | EXITED 2025 | OWN RECORD | Own record | 2026-10-10 |
| mevot.evidence | A 29-page case study of how Boosted Boards collapsed (June 2024); a launch plan for a 100-unit first batch with no paid pre-orders. | OWN RECORD | My case study of 15 June 2024 and my launch plan (own documents) | 2026-10-10 |
| making.printer | My first 3D printer, bought at twelve. | OWN RECORD | Own record; my current website | 2026-10-10 |
| making.rc | Radio-controlled boats, cars and planes built from scratch, and an automated device for my laundry. | OWN RECORD | Own record; my current website | 2026-10-10 |
| making.climb | A climbing-device prototype with electromagnets, the Spider-Man dream, printed in two halves. | OWN WORDING | G-code headers and project photographs | 2026-10-10 |
| making.masks | During the first COVID lockdown, at 14, my older brother and I made face masks, 3D-printed on my printer (print files from 31 March 2020)… | OWN RECORD | G-code file MASK-European-Union-V2, 31 March 2020 (own file) | 2026-10-10 |
| making.kart | A 30 hp electric go-kart, built in our basement: a welded steel frame, a motor, a chain drive and printed drivetrain parts. | OWN RECORD | Own record (owner direction, 12 Oct 2026); June 2020 print jobs | 2026-10-10 |
| making.press | Diari d'Andorra covered the masks. The article features my brother. | VERIFIED | Diari d'Andorra: Material sanitari 3D | 2026-10-10 |
| making.unity | I built my first game in Unity, Space Cleaner, and put it on the App Store. | OWN RECORD | Own record; own Unity project files | 2026-10-10 |
| making.gmvs | At Green Mountain Valley School: an iPad stand and a cardboard chair that held 200 lb. | OWN RECORD | Own engineering class files | 2026-10-10 |
| making.arduino | An Arduino door lock with keypad, servo and LCD, and a temperature alarm. | OWN RECORD | Own Arduino project files, 7 and 13 January 2023 | 2026-10-10 |
| making.safe | A fingerprint safe. | OWN RECORD | Babson report of 20 April 2024 (own document) | 2026-10-10 |
| making.racing | Motor mounts, a brake disc and spacers for the kart, designed and printed; racing in karts and Andorran hill-climbs. | OWN RECORD | Own CAD and print files; Automòbil Club d'Andorra events | 2026-10-10 |
| making.apparel | Helped on the backend of a clothing brand doing over seven figures in ARR. | OWN RECORD | Own record; my current website | 2026-10-10 |
| etower.chip.pres | PRESIDENT · 2025–26 | OWN RECORD | eTower board report, spring 2026 | 2026-10-10 |
| etower.chip.adv | NOW ADVISOR | VERIFIED | Babson eTower page, 10 October 2026 | 2026-10-10 |
| etower.intro | I was President of eTower, Babson's founder community, for one year (2025–26), and I remain its Advisor. The house was founded in 2001 an… | OWN RECORD | eTower board report, spring 2026 | 2026-10-10 |
| etower.ev.ops | A new VP of associates and an OKR system; The Pitch and Megaphone initiatives with Babson's president. | OWN RECORD | eTower board report, spring 2026 | 2026-10-10 |
| etower.ev.beta | In spring 2026 every Babson Beta Challenge finalist was an eTower resident; the wins are theirs. | OWN RECORD | eTower board report, spring 2026 | 2026-10-10 |
| wrong.intro | Every model here misses somewhere. These are the misses I know about, with their numbers. | OWN WORDING | Round 1, Round 2 and audit findings (own repository) | 2026-10-10 |
| wrong.r1 | Tourism as a top-three concern: 81% of agents against 1% of residents (July 2026). | DERIVED FROM DATA | Round 1 findings, July 2026 (own repository) | 2026-10-10 |
| wrong.r2 | Round 2 equivalence: 12 of 24 items missed (8 September 2026). | DERIVED FROM DATA | Round 2 scorecard, 8 September 2026 (own repository) | 2026-10-10 |
| wrong.r3 | Economic self-assessment: 6 of 6 items missed against a survey of 750 respondents (8 September 2026). | DERIVED FROM DATA | Round 2 findings, ECONOMIA survey block (own repository) | 2026-10-10 |
| wrong.r4 | Unemployment: 9.63% synthetic against 2.0% official (2026). | DERIVED FROM DATA | Validation schema and unemployment check (own repository) | 2026-10-10 |
| wrong.r5 | Population back-cast: 5.5% error against 1.7% for a linear trend (September 2026 update). | DERIVED FROM DATA | Scenario engine back-cast, R3c corrected run | 2026-10-10 |
| wrong.r6 | Engine wiring: four scenarios ran one branch, and p10 equalled p50 equalled p90 (July 2026). | DERIVED FROM DATA | Engine audit, July 2026 (own repository) | 2026-10-10 |
| wrong.r7 | Digital twin in the title: a reviewer called it an overclaim (June to July 2026). | DERIVED FROM DATA | Peer review of the paper | 2026-10-10 |
| wrong.r8 | Kendall Square: the reference run misses the work-leg mode split by 29.4 points against the census. | DERIVED FROM DATA | Kendall Square world README and held-out benchmark v2 (tier2, the reference arm), frozen commit e94f175 (private repository) | 2026-10-13 |
| wrong.r9 | Kendall Square: 9.6% of work legs walked against 28.2% in the census. | DERIVED FROM DATA | Kendall Square world README (tier2, the reference arm), frozen commit e94f175 (private repository) | 2026-10-13 |
| wrong.r10 | Kendall Square tier 3.1 (timetabled transit): the work-leg mode split miss grew from 29.41 to 30.11 points, so it was not adopted. | DERIVED FROM DATA | Kendall Square world README and docs/TIER3.1_DEMO.md, frozen commit e94f175 (private repository) | 2026-10-13 |
| wrong.r11 | Andorra mobility: unvalidated, telecom protocol drafted, request not sent. | DERIVED FROM DATA | Draft telecom data request, not sent (own document) | 2026-10-10 |
| credits.colophon | Type: the system's SF Pro where the device has it, otherwise Inter; Syne and Instrument Serif (both SIL OFL, self-hosted). Built with thr… | DERIVED FROM DATA | Font and library licences | 2026-10-10 |
| credits.changelog | v1.0, built 18 October 2026: first version. | DERIVED FROM DATA | This site's build | 2026-10-10 |
| credits.claims | This build checks 365 claims against their sources and data files; wording I drafted for myself is marked until I approve it. | DERIVED FROM DATA | This site's claims register | 2026-10-10 |
| footer.line | v1.0 · built 2026-10-18 · commit 7405113 · 365 claims checked · model figures updated September 2026 | DERIVED FROM DATA | This site's build | 2026-10-10 |
| seo.title | Marcel Bartumeu Ramentol | DERIVED FROM DATA | Page title | 2026-10-10 |
| seo.description | Technologist from Andorra, based in Boston. Works with City Science at the MIT Media Lab and is first author of The Andorra Scenario Engi… | OWN WORDING | MIT Media Lab people page; Crossref record of the paper | 2026-10-10 |
| seo.work | The index to Marcel Bartumeu Ramentol's work: research on national scenario models and synthetic populations, hardware, and ventures, eac… | OWN WORDING | This site | 2026-10-10 |
| seo.wrong | Where I'm wrong: every miss in my research that I know about, with its numbers, what I did about it, and its status. | OWN WORDING | This site | 2026-10-10 |
| seo.about | About Marcel Bartumeu Ramentol, from Andorra: research with MIT Media Lab City Science, building, and racing. | OWN WORDING | This site | 2026-10-10 |
| seo.credits | Credits, sources, colophon and changelog for the portfolio of Marcel Bartumeu Ramentol. | OWN WORDING | This site | 2026-10-10 |
| tour.0 | Marcel Bartumeu Ramentol. Technologist from Andorra, based in Boston, with City Science at the MIT Media Lab. | OWN WORDING | Chapter 00 | 2026-10-10 |
| tour.1 | 2017: at 11, I was at the City Science Summit in Andorra. Today I work in that group, at MIT. | OWN WORDING | Chapter 01 | 2026-10-10 |
| tour.2 | Four futures to 2049. Under Overgrowth, electricity demand doubles and hospital beds per 1,000 people served nearly halve. | MODEL · CONDITIONAL | Scenario engine, R3c corrected run (570 to 1,146 GWh; 1.35 to 0.69 beds) | 2026-10-10 |
| tour.3 | My first synthetic population: 81% ranked tourism in their top three. Real residents: 1%. I wrote it up as a fail. | OWN WORDING | Round 1 findings, July 2026 (own repository) | 2026-10-10 |
| tour.4 | The second test was pre-specified and run once. 12 of 24 gates missed, on the record. | OWN WORDING | Round 2 scorecard, 8 September 2026 (own repository) | 2026-10-10 |
| tour.5 | Same pipeline, the Media Lab's neighborhood: 29.4 points off on commute mode, and the miss is printed. | OWN WORDING | Kendall Square world README (tier2, the reference arm), frozen commit e94f175 (private repository) | 2026-10-13 |
| tour.6 | You have been using my control panel. Plug in the real one. | OWN WORDING | Chapter 08 | 2026-10-10 |
| tour.7 | Everything above this line happened. Everything below it is a model. | OWN WORDING | Chapter 09 | 2026-10-10 |
| tour.8 | Index · CV · Paper · Contact | DERIVED FROM DATA | This site | 2026-10-10 |
| tp.chip.built | BUILT | OWN RECORD | Own project notes and photograph of 22 April 2026 | 2026-10-10 |
| mevot.hero | MEVOT BOARD · RENDER · JAN 2025 | OWN RECORD | My own render of the Mevot board | 2026-10-10 |
| making.hero.1 | THE PRINTER, MID-PRINT · 7 JAN 2020 | OWN RECORD | My own photograph, 7 January 2020 | 2026-10-10 |
| making.hero.2 | MEVOT PROTOTYPE CASES · 19 JAN 2020 | OWN RECORD | My own photograph, 19 January 2020 | 2026-10-10 |
| media.delegation-demo | DEMO TO THE ANDORRAN DELEGATION · MIT MEDIA LAB · 1 MAY 2026 | OWN RECORD | Photograph supplied by me, cleared 12 October 2026 (photo: gbila) | 2026-10-10 |
| media.delegation-panel | THE CONTROL PANEL IN USE · MIT MEDIA LAB · 1 MAY 2026 | OWN RECORD | Photograph supplied by me, cleared 12 October 2026 (photo: gbila) | 2026-10-10 |
| media.delegation-roads | EXPLAINING THE ROAD NETWORK · MIT MEDIA LAB · 1 MAY 2026 | OWN RECORD | Photograph supplied by me, cleared 12 October 2026 (photo: gbila) | 2026-10-10 |
| media.delegation-terrain | THE TERRAIN MODEL, SEEN BY THE DELEGATION · 1 MAY 2026 | OWN RECORD | Photograph supplied by me, cleared 12 October 2026 (photo: gbila) | 2026-10-10 |
| media.delegation-phone | A VISITOR FILMS THE TABLE · 1 MAY 2026 | OWN RECORD | Photograph supplied by me, cleared 12 October 2026 (photo: gbila) | 2026-10-10 |
| media.mit-table-2026 | CITYSCOPE TABLE · MIT MEDIA LAB · 2026 | OWN RECORD | Photograph supplied by me, cleared 12 October 2026 | 2026-10-10 |
| media.gs-colour-andorra | RACING FOR ANDORRA · GIANT SLALOM | OWN RECORD | Photograph supplied by me, cleared 12 October 2026 | 2026-10-10 |
| media.gs-bw-canada | GIANT SLALOM · CANADA | OWN RECORD | Photograph supplied by me, cleared 12 October 2026 | 2026-10-10 |
| media.sf-lots | MY SMALL-LOT EXPLORER · SAN FRANCISCO | OWN RECORD | SF small-lot explorer, live map (my public project), captured 13 October 2026 | 2026-10-10 |
| media.portrait-bw | MARCEL BARTUMEU RAMENTOL · PORTRAIT | OWN RECORD | Photograph supplied by me, cleared 12 October 2026 | 2026-10-10 |
| data.pop | Resident population: official statistics (SIGDDE), 31 December, 2010 to 2024. | DERIVED FROM DATA | Source string of scenarios.json, history.Pop (repository paths removed) | 2026-10-10 |
| data.gdp | GDP per capita, nominal euros, as used in the model inputs. | DERIVED FROM DATA | Source string of scenarios.json, history.GDPpc (repository paths removed) | 2026-10-10 |
| data.afford | Housing burden, percent of income: 2010 to 2017 back-extended, 2018 to 2024 from the household survey (ECV). | DERIVED FROM DATA | Source string of scenarios.json, history.Afford (repository paths removed) | 2026-10-10 |
| data.co2 | CO2: national inventory (BTR1) for 2010 to 2021; 2022 to 2024 are fuel-based central estimates, not official. | DERIVED FROM DATA | Source string of scenarios.json, history.CO2_total (repository paths removed) | 2026-10-10 |
| data.elec | Electricity demand: no observed series; 2024 is the model's own 2024 state, and later years assume 2024 use per resident. | DERIVED FROM DATA | Source string of scenarios.json, history.ElectricityDemand_kWh_year; CLAIMS.md Andorra model numbers | 2026-10-10 |
| data.tour | Visitors per year, tourists and day-trippers (2012 methodology), from the tourism series of the model. Not overnight stays. | DERIVED FROM DATA | Source string of scenarios.json, history.Tour; CLAIMS.md Andorra model numbers | 2026-10-10 |
| data.natcov | Natural land cover: an estimate from observed land cover, 2010 to 2024, joined to the model's 2024 state with a 0.034 percentage-point step. | DERIVED FROM DATA | Source string of scenarios.json, history.NatCov (repository paths removed) | 2026-10-10 |
| data.pm25 | PM2.5 in Escaldes-Engordany: European Environment Agency TEOM series, 2018 to 2024; the gravimetric years 2014 to 2017 are left out becau… | DERIVED FROM DATA | Source string of scenarios.json, history.PM25 (repository paths removed) | 2026-10-10 |
| data.beds | Hospital beds per 1,000 service population: observed in 2017, 2019, 2023 and 2024. | DERIVED FROM DATA | Source string of scenarios.json, history.HospBedsPer1000SP (repository paths removed) | 2026-10-10 |
| data.wei | Summer water stress (WEI): no observed series; 2024 is the model's own 2024 state. | DERIVED FROM DATA | Source string of scenarios.json, history.WEI_gross_Q3 | 2026-10-10 |
| sp.title | A 90,000-person synthetic Andorra | DERIVED FROM DATA | Population run P04, frozen September 2026 | 2026-10-10 |
| kendall.question | Without tuning any constant to the outcome it is scored against, how close does the same pipeline get in another city? | OWN WORDING | Kendall Square world README, frozen commit (private repository) | 2026-10-10 |
| kendall.system | The pipeline that built Andorra's population, pointed at the City of Cambridge, Massachusetts, around Kendall Square: 118,403 residents i… | OWN WORDING | Kendall Square world README and tier2 run_meta, frozen commit e94f175 (private repository); public/data/kendall-cells.json totals | 2026-10-13 |
| kendall.evidence | On the held-out benchmark, the reference run misses the census work-leg mode split by 29.4 points, and walks 9.6% of work legs against 28… | OWN WORDING | Kendall Square world README and held-out benchmark v2 (tier2, the reference arm), frozen commit e94f175 (private repository); public/data/kendall-cells.json facts | 2026-10-13 |
| kendall.limits | A variant with timetabled transit (tier 3.1) made the mode-split miss worse, 29.41 to 30.11 points, so it was not adopted. Any trips draw… | OWN WORDING | Kendall Square world README, docs/TIER3.1_DEMO.md and tier2 run_meta limitations, frozen commit e94f175 (private repository); public/data/kendall-*.json | 2026-10-13 |
| kendall.credits | Census geography: U.S. Census Bureau. Street and path geometry © OpenStreetMap contributors (ODbL). Transit timetables and shapes: MBTA G… | DERIVED FROM DATA | Attribution strings of public/data/kendall-cells.json, kendall-trips.json and kendall-transit.json | 2026-10-13 |
| kendall.chip.progress | IN PROGRESS | OWN RECORD | Kendall Square world repository (private) | 2026-10-10 |
| kendall.chip.heldout | HELD-OUT BENCHMARK | OWN RECORD | Kendall Square world README (private repository) | 2026-10-10 |