Transportation Camp NYC

by on under conference-recaps
6 minute read

Recap of my recent trip to NYC to attend Transportation Camp NYC, a lovely unconference with about 500 attendees. Here’s the full session board. The following are notes from sessions I attended, sessions I wish I had a time-turner to attend, and sessions where others took really nice notes.

Session 1

Chatted to old friends

Session 2: Transit Scenario Planning

Led by Anson Stewart & Kate Chanba - Conveyal

A demo of different tools for creating and editing transit routes, updating frequency and seeing more or less how that would affect service.

Tools:

Enmodal

Link

  • Free and open-source

Remix

https://remix.com

  • Not free, hard to get for advocates
  • Population & Jobs layers
    • Poverty
    • Minority
    • Other demographic data
  • Very easy to use
  • Good sense of “What it would actually cost to make transit changes”
  • Used for communicating scenarios to the public

Conveyal Analysis

Link

  • Potential modifications:
    • Reroute
    • frequency
  • Compare accessibility to jobs
    • Based on LODES data
  • Open Source. Conveyal offers a hosted service.
    • Possible to run on a consumer laptop, analysis will be computed slower
    • Installation is a little trickier https://github.com/conveyal/analysis-ui
      • Requires Amazon Web Services S3 buckets

Brand New Subway

link

Session 3: NYT Subway Delays

Presented by Adam Pearce, Graphics Person at the NYT, @adamrpearce, GitHub

Adam pulled data from MTA’s real-time subway API, as well as went through old performance reports to look at the MTA’s current performance during “The Summer of Hell” and compare it to historical trends.

  1. Pulled data from the MTA’s GTFS-RT feed.
    • Processed with a node.js package
      • https://github.com/google/gtfs-realtime-bindings
      • https://github.com/1wheel/subway-parse
    • Dealing with missing data
      • If train sits at a station, the expected arrival time doesn’t get updated.
      • If train gets cancelled, feed gets updated with a garbage time
  2. Initially hoped to talk to analysts at the MTA, but press office got cold feet. Then NYT burned their political capital with the first SummerOfHell article
  3. But MTA had previously published research reports analysing their own data.
  4. MTA changed everyone’s developer keys
  5. Instead of individual days, ended up publishing aggregate details, demonstrated that MTA could never provide as much service as scheduled

Adam’s data wishlist:

  • Dwell time data
  • Historical data

There was discussion around Adam’s dream of comparing similar data feeds for other global subway systems, to further build the political will to improve the MTA. It sounded like a tough project to pitch to a newspaper. But maybe if Transit Center, or some other researchers were interested…

Session 4: Sources of Big Data for Road Ops (me!)

I led essentially the same session at Transportation Camp Toroonto, but I thought folks in NYC might be interested in learning about the sources of big data we use within Toronto Transportation’s Big Data Innovation Team. Thanks to a great pitch card from Lauren, the session was well attended.

I walked through our Data Sources Github repo, talking about where we get each data source and how we use it. This had a sedative effect on some.

One questioner heavily directed the session. He wanted more specifics relative how our work has led to tranformation in the department. The main point is that transportation policy evaluation can be more rigorous. Manual data collection is so expensive that it is generally of unreliably small sample size. With automatically collected data, and often passively collected data, we don’t need to be as deliberate with before-after data collection for small projects. So it’s easier to evaluate a multitude of interventions across the city. Our team is still only two years old, and we’ve only been 3 full-time staff since 2016, so I don’t think I have a good concrete example of internal change our philosophy has brought about.

This presentation made me Wonder whether we should have a Github pages Reveal.js presentation for our datasets. It would be a high-level overview of each datasst linkinng to its folder.

People seemed shocked multiple times that all our code was available on Github, guess that’s a kind of revolutionary government thing.

Honourable mentions

Presentations I kind of wish I attended

New Urbanist Memes for Transit-Oriented Teens

Basically, the dankest group on Facebook if you’re a transit nerd, the founder and admin Juliet Eldred gave a presentation at the same time as my talk. It’s more than just memes, promise.

Uphill Battle - Improving Transit From Within a Difficult Political Climate

Led by two MIT Transit Lab alums at the MTA (I think we still outnumbered Transit App employees), this discussion partially talked about how transit employees can lead change from within alrge poltiical organizations like the MTA. Nobody took public notes alas.

Transport for Cairo - Mapping Informality

Former McGill classmate of mine led a project in Cairo to map the informal transit network and create a public GTFS feed from it. Similar projects in Nairobi and Dhaka have gotten some press. Nobody took notes.

Transit Wiki

Zak Accuardi of Transit Center led a session teaching people how to edit The Transit Wiki, a wiki by and for transit professionals to facilitate knowledge transfer. More people should contribute to it and make staff and consultants do the same for research projects and use it for memorializing internal knowledge sharing materials.

Not to be confused with transit.wiki, which is a wiki of transit agency information for riders.

Governing Transport for City Regions

Nicole Badstuber a PhD researcher at University College London, and host of On Our Line, the London Reconnections podcast, led a session on Transportation Governance:

Someone(s) did a very good job of taking notes, and they are all here

Other sessions with notes

The organizers did not promote the crowdsourced notes angle well enough, but some sessions had exceptional note-takers. Here they are, in no particular order:
transit, scenario-planning, big-data, transportation camp
Made by Raphael Dumas
Carte-Noire theme by Jacob Tomlinson on Github