Show HN: Public transit systems as data – lines, stations, railcars, and history
publictransit.systems37 points by qwertykb 10 hours ago
37 points by qwertykb 10 hours ago
This is interesting, for sure.
Thinking about this, I think what happens if we take it to the extreme - a system that lets a single user enter his route to work - and compare how various transportation systems, including innovative ones, would impact his day to day.
It's really hard to build such tool, and if it's done well and made appealing to users, it could be the base of a political tool for people to pressure their representatives to implement new and better systems and improve voters lives.
This is interesting! However, can't it switch the unit to meters, not miles? From the link you provided, it appears the API doesn't include information about its unit in the responses. So, it's confusing.
So interesting that Baltimore’s very subpar public transit system made it into the first batch of systems — not complaining at all I love that city & love to see it in random places on the internet :)
>a unique visual style that appeals to transit enthusiasts
But it seems to be just an API with a website sporting a terminal font. What am I missing?
The ⌘K command palette is a great touch for this terminal-style UI. It would be even better if we could search by railcar models (like 'R211') or specific line colors directly through it. Excited to see this grow!
Nice site but there's so few systems in it.
yep, its still and will always be WIP, every system exposes their data a bit differently, so it's not completely plug and play. you're welcome to contribute to the existing system data however! https://github.com/WaffleThief123/publictransit-systems/
Very nice! I think we're confusing route length and track length, though.
Can you elaborate further on this so I might be able to fix it? As best I am able to discern, system track length for all current ones is accurate.
Could track length include sidings and double-tracks and extensions and other maintenance track needs that aren't useful for the passengers themselves?
Looks good.
London, Paris, Copenhagen, Rome, Madrid?
I guess it takes a while to add each new one in.
tysm! Yeah I've got a bit of a backlog, as each set of data for each system needs a deep-dive on how the system exposes it programmatically, then need to do tests and some studying on each to build cars,timelines, etc. Toronto is on the list, if there's any in particular you want, file an issue on gh, and I'll do my best to get it added in a timely manner. https://github.com/wafflethief123/publictransit-systems/issu...
Working in the industry: this is a can of worms that you might regret opening, because of simply how completely insane so many of the systems there are. I've seen cities repurpose text fields in otherwise perfectly standard formats because they needed a way to indicate that the bus was {insert any absolutely insane situation}, and not parsing that field means that the data doesn't make sense.
Your best bet for most of Europe are the open data platforms. Example for France: https://transport.data.gouv.fr/. There's soooome standardisation around a few formats:
GTFS, Netex, SIRI along with their various extensions (like GTFS-RT for realtime data), etc. Just parsing these (which is already a large undertaking in and of itself) should get you covered for a bunch of networks.
Oh, also, much of the data you'll find, especially from smaller cities or regions, is awful. You're going to be told that the line icon is white, on white text, and that's actually perfectly normal because actually their bus header is from an obscure system from former Yugoslavia that actually interprets "0xFFFFFF" as black when on layer 1. Good luck!
[dead]
[dead]
[dead]