Isochrones 😍 ! This map shows you how far a 5h train ride will take you, departing from any city in Europe.
Inspired by Direkt Bahn Guru by
@juliustens
.
Si vous partez en train de ces villes aujourd'hui à 10h, jusqu'où pouvez-vous voyager avant 16h ?
L'API
#digitalsncf
(construite par
@navitia
) nous répond grâce à ses isochrones.
Notebook à explorer ici:
You can see the discrepancies in access to rail mobility between cities, how certain high-speed connections are favored, how certain cities are isolated...
A researcher from Cambridge,
@AlexisLitvine
, gave me access to train travel times from 1911.
Here's a comparison between now and then - a denser network with more small stations and no high-speed rail.
Their project
@ACommunes
reconstitutes historical networks.
Links below ⬇
Pour ceux qui ne veulent pas remplir leurs informations à chaque attestation, j'ai fait ce site mobile à partir du code open-source du ministère pour en générer en un seul clic:
Over the last months, I've made a lot of small improvements to the tool I built to replace Google Maps:
- Mapillary integration
- Better map style
- More consistent theming
It's now called , come check it out !
The data comes from , which sources it from
@DB_Bahn
.
It gives us the direct trains from any station in Europe (as long as it's reacheable from Germany).
From this we can build a graph of stations with travel durations between them.
Verdict des coiffeurs avec des noms rigolos.
Les blagues en "Hair" l'emportent avec un énorme 60%, suivies, évidemment, par les "Tif" (28%).
Belle densité dans les Hauts-de-France ainsi que la côte d'Azur.
Si vous partez en train de ces villes aujourd'hui à 10h, jusqu'où pouvez-vous voyager avant 16h ?
L'API
#digitalsncf
(construite par
@navitia
) nous répond grâce à ses isochrones.
Notebook à explorer ici:
Une autre petite viz de
@Paris
: L'ensemble de l'éclairage public disponible en Open Data.
J'ai tenté d'obtenir le rendu le plus proche de la vue satellite de la NASA (à droite).
Quelques mots sur la réalisation ⬇️
I was amazed by the
@nytimes
visualization of friendship locations, so I tried to copy it with
@Mapbox
feature-state capability. It's super fast!
demo:
the original article:
learn more about feature-state:
Since walking and local transit options are not included, we add some connections to the graph to account for buses, walking, biking...
For instance, you can easily walk from Paris Gare de l'Est to Gare du Nord in a few minutes.
Adding the assumption that interchanges / connections are 20 minutes, we can see all the stations that a reachable from a given city in 5h.
(since there is no guarantee that trains will connect perfectly, the map tends to be overly optimistic)
Paris bat au rythme des cyclistes : on compte jusqu'à 3x plus de vélotaffeurs sur les pistes depuis la grève. 🚲
Merci à
@opendataParis
d'avoir ouvert les données des compteurs !
I updated the data for chronotrains, including (finally!) the Paris Eurostar, missing stations (e.g. Millau), and many other schedules that I missed in the original dataset.
Anything you still see missing? Let me know!
La première dataviz du site: .
Une visualisation du trafic mesuré par
@Paris
au mois de juin 2019.
On y voit le volume de voitures (épaisseur) et la densité du trafic (couleur - violet = bouchons).
On y voit des choses intéressantes!
Working on a silly idea for creating custom "fog of war" maps of places you've visited.
The idea is that you can see the areas you've *not* visited so you can explore somewhere new next time. Heavily inspired by strategy games like Civilization 6.
Proof of concept in QGIS 👇
A daily map-based puzzle game where you try to find a city based on satellite images:
After each try, the map unzooms - letting you explore more area.
Can you guess this city from above?
I spent the last weekends building Data Pour Paris, a collection of dataviz for the city:
Check out:
- 🚗 the city's traffic patterns
- 🚃 all trains & metro, live
- 💡 every single public lightbulb
- 🏃♀️ all finishers of the Paris Marathon
and more!
here's the live map of the hospital accessibility (by car) in France. Aside from the mountains, there are empty spots in Médoc and Champagne (Sézanne). My guess is that winemakers don't get sick.
🚇 En la Ciudad de México hay 195 estaciones de metro.
¿Cuántas crees que puedes nombrar de memoria?
¡Pon a prueba tus conocimientos y diviértete con nuestro juego!
🔗
#MetroCDMX
#JuegoDeMemoria
#CDMX
Une carte de Paris pensée pour les piétons et les vélos:
- 🔸 Pistes, bandes, double-sens cyclables (dont les coronapistes)
- 🍏 Rues aménagées pour les piétons et zones d'intérêt à la Google Maps
- ⚓️ Basée sur l'open-data de Paris
The MapTiler tool I used to see tile boundaries and addresses is down due to unpaid Google Maps subscription - so I made a quick replacement: .
Find a quadkey or tile address quickly, and copy it to the clipboard with a single click!
🚄 From Hamburg, Germany, 12h in trains will take you to Montpellier 🇫🇷, Budapest 🇭🇺, Plymouth 🇬🇧.
What other cities would you like to see animated like this? Tweet back and I'll reply with the video.
Today is my last day
@Mapbox
. It's been great working alongside so many talented and curious people. I'll be joining Shone, helping to make cargo ships autonomous! ⛴
@_benjamintd
Could you now create a heat map showing best/worst connected places in Europe? That would be ace. But also: I'm definitely doing a
#RailNatter
where I just tour about in this!
I've made a few adjustments to the Scratch Map, a fog-of-war map of all the places you've been, according to your Google location history.
Here's the link, if you want to try it out and give me some feedback !
I've used plenty of bike shares before, but
@jumpbikes
really stands out.
- The bikes feel safe and sturdy
- Dockless, yet you can only attach them to bike racks (ie city infrastructure)
- Electric assist makes you faster than a car in many situations
- Bonus, they use Mapbox :)
I'm soon going to have to pay my first
@Mapbox
invoice 🥲
Unfortunately 5$/1000 visits is unsustainable for an ad-supported website.
Thinking of moving to Maplibre. What's a cheap way to serve a simple background map?
Frequent updates and detail at high zooms unnecessary.
chronotrains had over a million visits and around 80 million isochrones requests in the past couple weeks.
thankfully, after some help from
@vercel
to get the infrastructure right on launch day, I never had to think about load, response times, etc.
An isochrone map depicts the area accessible from a point within a certain time threshold.
This map shows you how far a 5h train ride will take you, departing from any city in Europe.
[interactive map: ]
On entend beaucoup parler de
#DansMaRue
depuis le
#79inter
de
@Anne_Hidalgo
.
Visualisez son utilisation depuis son lancement en 2012:
Quelques observations en thread ⬇️
@ayoungmanriver
@juliustens
I'll clean it up before open-sourcing. The geometries are pre-computed for hours 1-5, I could do more but I need to find a way to make it compute faster :)
How far in time would you like to see ?
12h in trains from Prague take you to 🇳🇱🇫🇷🇱🇺🇧🇪🇩🇪🇮🇹🇭🇺🇷🇴🇺🇦🇦🇹🇨🇭🇩🇰🇵🇱🇸🇮🇨🇿 (flags in no particular order except trying to confusingly alternate horizontal and vertical stripes)
An isochrone map depicts the area accessible from a point within a certain time threshold.
This map shows you how far a 5h train ride will take you, departing from any city in Europe.
[interactive map: ]
Damn, data processing on
@felt
is ✨fast.
Drag and drop any GIS format of any size and it'll just pop ready in a minute.
Tippecanoe's doing the magic I guess.
J'ai été interviewé par
@OuestFrance
au sujet des déserts ferroviaires en France.
À ce propos, voici une carte qui illustre les zones dont il est le plus difficile de sortir en train, en métropole.
We've released the beta version of our video export feature on
@PelicaLabs
. It's a beta in a beta. Betaception. We're looking for feedback if you want to map a run, a trip, or just a nice location :) -->
Obsessed with these low-angle satellite images from
@planetlabs
1: Doha, Qatar. November 11, 2017
2: Houston, Texas. March 14, 2018
3: Riyadh, Saudi Arabia. March 10, 2018
4: Osaka, Japan. March 13, 2018
@botic
@oler
@juliustens
You can see any station when zooming in, my mouse just happened to hover over this one :)
I tried ordering the stations somehow but sometimes you'll need to zoom in a little.
I'll try adding a search bar next.
@juanleitonm
@juliustens
it's a next.js app hosted on
@vercel
with a database on
@supabase
.
Quite a bit of pre-processing to compute the isochrones beforehand.
Mapbox-gl-js is doing the work on the front-end. Client-side caching with vercel/swr, server-side caching with
@vercel
's built-in cache.
Lisbon is... disappointing in comparison 😅 Leaving Portugal by train takes a while!
But you can get to Barcelona in 12h 🏖️ if you think the Atlantic is too cold.
Great piece by the NY Times.
I'm lucky to have worked with people like
@morganherlocker
with strong values on privacy. With location data, privacy is the product - the added value is storing as little data as needed to derive insight.
It's all cached at the edge thanks to the Vercel CDN. The base map is also super fast with the tiles hosted at
@Mapbox
.
the interactions feel as fast as when developing locally, that's insane.