Ask HN: How many people would want to check-in/check-out in public transit?

[szcukg]

I have been thinking about if it's a feasible to have an app in which users check's in when they get into a bus. They give the bus stop information when they check-in and when the check-out. This can be used to predict bus timings and generate close enough real-time approximation. I'm not sure if this is already done, but if enough people check-in and check-out u can also determine the amount of people in a bus and determine which bus to maybe miss because of too much crowd.

In US public transit is probably not used as much as in other countries, so it might not be so useful in bay area as such. What do people think?

[mchannon]

The farebox already gathers this information in the case of buses; paper ticket and check-in/check-out systems are less precise (which train are you getting on?) but likewise gather this information. Most of the time this information is not passed along to a server. Duplicating this effort via a voluntary app doesn't seem particularly helpful.

Aside from the agoraphobics among us, most people are in way too much of a hurry to let standing-room-only transit deter them from making a connection. So your target market (people who want to ride an empty bus so much they're willing to wait for one) is already quite limited.

What this information is useful for (and is probably used extensively by some transit agencies) is planning routes, adding buses, trains, and train cars. Evidence (full buses!) suggests that many transit agencies could be doing a much better job coordinating larger bus sizes and more train cars during rush periods. Too bad there's not an app for that that the regular transit rider can employ.

[EllaMentry]

Here are my 0.02 pennies on why this won't work.

Most modern cities and towns have automated displays at bus stops which tell you when the next bus will be arriving. I have seen this in Canada, UK and across South East Asia so I assume it's a thing in many places or is getting there pretty soon.

Secondly, what is the incentive for me to do this? "Check onto a bus" what does this mean? The longest bus trips I take tend to be 20 minutes in length...why would I spend 1/20th of that time fiddling about with my phone "checking in" - there is no user benefit.

While not every idea is about money, I see no way for this to even sustain itself. In the era of self driving cars, smart cities etc. etc. having people to manual data analysis should be reserved only for cases where it is impossible to get a machine to do this...and you will have to pay people to do it - they won't do it for free (unless you can give them something in return...)

Keep the idea machine rumbling though!

[jameswyse]

The public transit system here in Brisbane, Australia kind of works like you're describing. We have 'Go Cards' which are just NFC top-up cards, when you board a bus/train/ferry you 'touch on' and when you alight you 'touch off' and it'll deduct your fare from the card automatically. Most buses still accept cash but it's much easier and cheaper to use the cards so I believe most people do.

It's a good system and I'm certainly glad it's NFC instead of having to find an app on my phone every time I want to jump on a bus. So the transit company does have this information but unfortunately they don't choose to share it. I really wish there was a public API :/

[ibudiallo]

Here is what is currently used in Los Angeles metro.

All metro buses (and recently trains) use GPS to record their location. And there is a service nextbus.com that allows you to get the nearly exact prediction on when your bus is coming based on your GPS location. It shows all the closes buses and and the timing in each direction.

Maybe if you want to build something you can start from there.

[LarryMade2]

Why not automate it and have butt sensors in the seats of the buses? (might have handrail/grip sensors too) Requiring the general public to consistently provide data is not the best method for accuracy. With the sensors you not only get a count but also weight distribution and other possible useful bits as well.