I saw this tonight:
https://www.septastats.com/train/9212
At first, it looked like a train was horribly late. But while digging around in the data, something didn't look right, as the train had been in that state for HOURS.
I then looked at the schedule for the Newark Line (http://www.septa.org/schedules/rail/w/WIL_1.html) and saw that the train was scheduled to arrive at Temple University over 12 hours prior!
My next step was to check SEPTA's API, and well... this happened:
$ curl -s http://www3.septa.org/hackathon/TrainView/ |jq .[21]
{
"lat": "39.89061",
"lon": "-75.30364",
"trainno": "9212",
"service": "LOCAL",
"dest": "Temple U",
"nextstop": "University City",
"late": 38,
"SOURCE": "Newark",
"TRACK": "",
"TRACK_CHANGE": ""
}
So it looks like we have a "zombie train". For whatever reason a train failed to finish its journey, but still shows up on the API as being in service.
Once I get a contact a at SEPTA, I'll be sure to let them know about this particular failure mode.
I saw this tonight:
https://www.septastats.com/train/9212
At first, it looked like a train was horribly late. But while digging around in the data, something didn't look right, as the train had been in that state for HOURS.
I then looked at the schedule for the Newark Line (http://www.septa.org/schedules/rail/w/WIL_1.html) and saw that the train was scheduled to arrive at Temple University over 12 hours prior!
My next step was to check SEPTA's API, and well... this happened:
So it looks like we have a "zombie train". For whatever reason a train failed to finish its journey, but still shows up on the API as being in service.
Once I get a contact a at SEPTA, I'll be sure to let them know about this particular failure mode.