8 comments

[ 3.6 ms ] story [ 33.2 ms ] thread
Is it reasonable to believe Pigeon could only add value if the time delay between a new delay-causing issue and some verifiable / consensus info getting posted by users is faster than similar updates by an official source like MTA?

(I am assuming that other information from Pigeon users would matter far less than info material to delays or service disruption. Knowing if there’s an annoying street performer on a platform, the example in the article, might have secondary value, but seems far less important. I could be wrong though.)

There may be issues that go unannounced. Not sure if this was a regular thing, but I got a train from Grand Central a while back and the trains & platform were so packed I could only get on the 3rd or 4th train that came.

I'm also not sure if all lines have real time info yet, my line only got it within the last 12 months, and when I played with their external API, the data was of dubious quality.

I'm not sure how much lasting value this app really has, but that's probably why it is an incubator project, rather than part of Maps.

I think the type of information you can get from other Pigeon users about delays would at least be more granular than what the MTA provides. The notices from the MTA are often late, sometimes wrong, and cover large areas. It might be useful to know where the bulk of the delay is on a line (at some interchange? in an area with close-together stations? river tunnels?).
The MTA's own SubwayTime app provides real-time train times based on the train's current locations. (This is new in the last year.) It even works when trains are running on different lines, changing which stations they stop at, etc.

I'm struggling to imagine what "live alerts from other riders" can provide on top of that?

Even for things like if a subway station is particularly crowded, or a train is particularly crowded... seems like that could be inferred from the train locations and average activity per-station for that day of the week.

I don't think the API gives actual train locations? It just gives an estimate of how long until the next train arrives. I've found that it is pretty accurate on some lines, but it feels like just a guess on others. I recently waited 45 minutes for a train which the sign and app perpetually said was 12 minutes way. I generally look at the sign or app now and think of it as a binary thing - the train is either probably coming soon, or it is going to be a while.

I heard a story from a colleague who had recently traveled to Japan. He leaned over the edge of the platform to see if a train was coming, as is common in NYC, and found that this surprised everyone around him. They trusted the signs and the schedule.

The numbered trains have more accurate data than the lettered lines. Here's some more detail from this article: http://www.metro-magazine.com/rail/article/715387/nyc-transi...

"Over the past few years, however, MTA New York City Transit (NYC Transit) customers using the numbered lines (except 7) and the L have benefitted from the installation of countdown clocks that alert them to which trains are coming and how many minutes away they are. Countdown clocks were first introduced along the L in conjunction with the installation of the Communications-Based Train Control (CBTC) moving block signaling system. Second-generation clocks, facilitated by Automatic Train Supervision (ATS), were installed along the 1 through 6 lines.

Countdown clocks take the guesswork out of waiting for a train, offering a reduction in stress levels for harried New Yorkers. The clocks are arguably the most welcome innovation to the New York City subway since the arrival of dependable and effective subway car air conditioning. Instead of leaning over the platform, all they have to do is look up and the arrivals of the next trains are displayed in real-time.

With stations along the numbered lines (A Division) nearly all covered, that unfortunately still left about two-thirds of the system in the dark, so to speak. Given the size of the B Division and the technical complexity of the implementation, customers using the A, B, C, D, E, F, G, J/Z, M, N, Q and R Lines were looking at 2020 before they could begin relaxing those neck muscles.

Then, prompted by New York State Gov. Andrew M. Cuomo, innovation combined with the arrival of a new technology to create a solution. That solution involves the use of beacon technology. In description, the system being piloted is relatively simple: Installed on platforms are Bluetooth receivers that communicate with beacons placed on the first and last cars of a subway train. As the train enters a station, the system uses its arrival and departure times to estimate the time at which the train will reach the next stop. That information is them transmitted via Wi-Fi to the LCD screens."

Oh joy, I get to read all the crabby comments of New Yorkers late on their commute.