A few months ago, I spoke with Alex Bell, an engineering student at Columbia and the brother of an old friend of mine, about his transit app. His idea was simple: crowdsource train locations through user-submitted messages. Unfortunately, the app never reached the critical mass of users it needed to br successful, but Bell isn’t giving up.
As The Times reports today, Bell has signed up with Densebrain, a mobile company that wants to use passive cell signals to triangulate train locations. With the approval of each user, Densebrain’s app reads when the cell signal is lost and notes when and where service is restored. For instance, if someone loses signal just south of 161st St. in the Bronx and resurfaces at Grand Central, the app knows that this user took a 4 train, and it can provide real-time info on that train’s location. With over 600,000 users of its free NYCMate subway map app, Densebrain thinks it has the user base to support such a project.
Of course, concerns over privacy remain to be tested. Will users consent to anonymous location tracking? And how will the app distinguish between different trains that run the same route? For now though, Densebrain’s plan is another in the effort to tell us just where our trains are and when, and that sounds promising to me. [New York Times]
11 comments
I love this idea!
Since its part of NYCMate now I had no problem enabling it. Just hope NYCMate has enough users for it to reach critical mass.
Stupid question, How do the countdown clocks determine how far away they are and is that data stored in the same manner? Because, I would think that just by using the countdown clock data an app could be created based on what time the trains are supposed to be coming in.
Also, what is the deal with 51st street countdown clock, it’s been a minimum of 9 months since the clocks went up (I think closer to a year) and they still haven’t been turned on, I don’t remember nearly as long a lead time at any of the other stations. Does anyone know of reason for this?
How will this work with transfers, I wonder?
How does this work on lines with long elevated segments?
It currently won’t work at all with elevated trains. It’s based on losing cell phone service at a specific point and figuring out what train you took based on where your phone regains service.
Seems to me the current system is more of an approximation of when trains will arrive than an exact count. I often see displays saying “Next Train to *** in 5 mins” and then that same train comes in 2 mins later. I’m not complaining that the train came 3 mins “early”, but it’s not exactly accurate.
The countdown clocks receive their information from the ATS (Automatic Train Supervision) system. This system provides dispatchers with precise train locations and associated information. My guess is that the MTA would not provide that data due to operational and security concerns.
Prior to ATS (and still on the lettered lines other than the L) the dispatchers can only tell when a signal block of track is occupied. Since these signal blocks are several hundred feet long, this is not very precise because you don’t know if a train is at one end of the block or in the middle. The older system also can’t tell you what train occupies a block on track segments shared by more than one line – e.g. an A vs. C.
The L uses data from it CBTC (Communication Based Train Control) system. Moving forward, I don’t know if the MTA intends to install ATS system wide or just wait for CBTC to be installed or something else.
The system can tell where the train entered the block from, which means it can track it backward to its scheduled origin.
Systemwide ATC (not just ATS) is a good thing – for one, done right it can increase track capacity to 40 tph – but not because of train tracking.
I don’t know why the MTA won’t publish the arrival estimates on the web. CTA has no problem providing real-time train arrival information publicly and is even set to release an API for developers to use the Train Tracker data. Security seems like a silly concern.
But this way, it can only be used with a compatible smartphone with the app.