progchan

i am doing a 2 yr computer programming diploma class at my local college and want to do something that can have a positive impact on my city.

my city has about 115k people. we have a decent transit system, but we only have a few spots to renew passes in the city and they all have horrible business hours. i am a daily transit user of this system.

we currently have a mobile app that only shows routes, schedules, and bus locations.

i am doing a 2 yr computer programming diploma class at my local college and want to do something that can have a positive impact on my city.

im in the brainstorming phase of wanting to develop an app that would either work alongside the existing app or replace it. i would like to get a working demo up and running this summer, and then start working with the transit system to implement it by 2027, 2028 at the latest.

i would want the mobile app that accomplishes a few things:

~let people purchase and renew passes or individual tickets via the app, and set up pass subscriptions via their credit card or debit card

~use the app to board the bus, ie tap your phone on a card reader, etc

~collect anonymous analytics data to look for spots needing improvement/upgrades


from what i can see, the system currently uses Trapeze terminals, MIFARE DESfire - based cards and possibly Genfare link.

at first, i thought that it would be possible to use a smartphone's NFC functionality, instead of needing a card. but after looking into it, ive realized that the current cards are RFID-HF and a proprietary system, and that i likely wont have any luck with this route.

but we have barcode scanners that are used to scan transfer passes, so i think the most viable angle would be to use a barcode on a phone screen in an app.

so i have some angles on this.

1. build a complete app system from scratch that does everything in-app or on the server, and generates a bar code that the existing system would scan. i would run it on a local hosting provider.

i am hoping this would require no hardware replacements or retrofitting for the bus fleet, and would let the elderly or people without data or phones continue to use the system "as-is" while letting riders who want to use a app have a system that works seamlessly.

2. based on my research, its likely that the system uses Genfare link, which has a developer api that i could hopefully leverage and might provide a more stable system with less potential issues caused by me.

i dont know this for certain and i will need to look into this.

3. replace the current fare collection system entirely, and replace the collection boxes with nfc/rfid readers.


i think this is unlikely because it would require retrofits and the buses atm dont have cellular connectivity.

i would like to collect anonymized user and ridership data for the purposes of finding out where chokepoints are in the system (as the city is in a semi-rural, conservative, car-centric area, and the bus system basically hasnt been meaningfully improved since the late 90s afaik)

i would like to try to collect info like:

~exact number of riders

~peak ridership hours of the day

~how often buses are late

~how often buses are late to get to stations/hubs, and making people miss transfers.

im hoping that the data eventually can be used to improve the system but plainly pointing out the choke-points and areas needing improvement.

ultimately, i am not hoping to make a ton of money off of this app. i would ofc charge for support and updates, but i would ultimately do it for free if someone would cover server and stripe costs, just because i do want to help the people in my city, as often the people most reliant on transit are often the most underserved and struggling.

so one of the issues with barcodes/qr codes is that people could take a photo or screenshot of them and share them.

so i would set it up that the barcode changes say every few minutes, to prevent copying and distribution. or set up some sort of TTL expiry for it.

atm i would write the backend with python and SQL. maybe once i learn C# and java i might want to use C# or java but im leaning towards python.

would probably set it up as a docker container and run it on stormweb or something (im canadian and having the backend hosted in canada is a must)


app would probably be some flutter or react thing, idk. maybe just a PWA if possible.

its a cool idea. if your city is already using the barcode scanners I say at least start there. at least with an app the usage would be tied to a user account. python and django or flask with mysql or postgres or mariadb for sql are fine. java / c# are a bit harder. if you're familiar with html/js and want to make a quick app there's cordova and then maybe the barcode could be an svg or something programatic. docker is fine too. you could also just do

apt install python3 postgres

on a vps or the cheapest hosts are usually php / mysql / cpanel and then you don't have to setup anything and you could basically just have raw PHP files for api endpoints if you want and you could do local dev with xamp, wamp or mamp. an endpoint for the scanners to log completed scans for analytics would be good. (linking scans back to the bus, user, and time) the stripe api isn't too bad. make sure to argon2id / bycrypt passwords. anyways I think your plan is good. good luck!