MBTA’s “Transit Diary” a failure before it has even begun

By | September 17, 2014

MBTAThe MBTA is currently beta-testing a project & corresponding web application called “Transit Diary“. Its purpose is to collect data directly from riders about their experiences riding the T, presumably to help the T evaluate how well it is serving riders’ needs and where service-improvement efforts should be focused.

Unfortunately, Transit Diary employs obsolete, inferior techniques and technology. As a result, the amount of data collected by the project will be limited, it will be less accurate than it could have been, and it will come from an extremely self-selected, non-representative sample. At best, the resulting data will be far less useful than it could have been; at worst, it will be completely useless.

Here’s how Transit Diary works:

  1. You ride the T.
  2. At some point after you’re done riding the T, hopefully the same day, you visit a web site and fill out a survey, remembering as best as you can the details of your trip.
  3. At the end of the week, you visit the same web site and fill out a weekly survey, remembering as best as you can your experiences riding the T that week.
  4. If you don’t fill out the weekly survey by Sunday at might, it goes away and you can’t fill it out.

How… quaint.

Here’s how the T’s Transit Diary should work:

  1. The T pays the maintainers of the Moovit app to customize it to collect the data the T wants (Moovit already collects most of it! In real-time!), or pays to license the code and adds the customizations itself.
  2. The T releases the custom-branded app and aggressively advertises for users to install it, and free, accurate data pours in, in previously unimaginable quantities.
  3. There is no number 3.

Because of its dependence on obsolete, inferior technology, the MBTA’s Transit Diary is a waste of money (in particular, your money and mine, since we pay the fares and taxes that fund the T) that does a grave disservice to its riders.

Share

9 thoughts on “MBTA’s “Transit Diary” a failure before it has even begun

  1. Genevieve Reale

    I live in Salem mass where the bus leave from SALEM DEPOT TO haymarket @ 540am if there was any way the bus could leave 5-10 mins early I would not be late for work on adaily bases where I work At Boston Medical Center Taking care of sick people and Have to in the hospital by 7 am . Thank you so much for reading these e-mails and I hope to see some time changes to better help peolpe getting to and from their jobs

    Reply
    1. Mary

      Sure the bus will change their schedule because of you.

      Reply
  2. DailyTRider

    Totally agree. They’re still emailing me to fill it out, but every time I comply and start a leg, I forget to submit because I have to wait to submit until I arrive, by which time I’m doing something else. And then they email me again.
    Love the replies to the comments, too!

    Reply
  3. boris crunchy

    You’ve said the site is obsolete, flawed, inferior, and outdated. But you haven’t provided any specific criticisms, or evidence to back up your claims. Unless something was supposed to be implied by your 4-step description of how it works, but there’s no explicit criticism there.

    Reply
    1. jik Post author
      • In my opinion, the web site is poorly designed and difficult to use, though that’s not the worst problem.
      • Asking people to remember to go to a web site to fill out daily surveys guarantees that some people will forget, and you’ll lose data.
      • Asking people to remember to go to a web site to fill out weekly surveys guarantees that some people will forget, and you’ll lose data.
      • Removing weekly surveys from availability Sunday at midnight means many of them won’t be filled out because riders won’t get to them in time.
      • Even when people remember to go to the web site to fill out their surveys, the data they are entering, manually and after the fact, will contain many errors and inaccuracies.
      • Limiting participation in the program to people who are willing to take the time to fill out daily and weekly surveys drastically limits and narrows the demographics of the data collected.
      • Technologies exist to collect virtually all of the necessary information automatically in real-time using smart-phone apps. And for the information that can’t be collected automatically, the app can prompt the user to enter it in real-time rather than making the user remember and enter it after the fact.

      It is certainly true that using smart-phone apps to collect the data would also narrow the demographics of the data collected. But you’ve already done that by requiring the data to be entered daily through a web site, and the data collected with apps would be so much higher quality, and there would be so much more of it (because it will be so much easier, far more people will be willing to participate), that the end result would be far, far higher in overall quality and utility.

      Reply
  4. sunny22

    I don’t share your concern. I participated and I thought it worked great. I was able to fill in my surveys right during my commute. Why did you wait until you got home? I’m excited they are collecting this data. I like the simplicity, the design, and everything seemed to make sense. i imagine that a tool like this faces a lot more challenges than you know, maybe something like moovit wouldn’t be very easy to use for an older audience? jik, you sound like a savvy web user, why don’t you help them improve the tool instead of being so negative.

    Reply
    1. jik Post author

      I participated and I thought it worked great. I was able to fill in my surveys right during my commute.

      Wow, you were riding the T from Germany (where your comment was posted from)? That’s impressive! I’d be interested in hearing our explanation of how someone posting from Germany is a regular enough T rider to participate in this program. There are plausible explanations — perhaps you’re traveling right now but usually live in Boston — or perhaps you’re just making stuff up.

      I’m sure it’s a coincidence that the company that was paid a lot of money by the MBTA to build the Transit Diary web site is located in Germany, where your comment was posted from? Sheesh, dude, learn to use Tor or something, if you’re going to astroturf.

      I don’t see how you were “able to fill in [your] surveys right during [your] commute” when one of the things asked in the surveys is how long the commute took, which sort of implies that you can’t fill them out accurately until your commute is finished. Could you elaborate on that as well? I’m curious.

      I’m excited they are collecting this data. I like the simplicity, the design, and everything seemed to make sense.

      I tell you, as someone who has been involved in designing end-user software and building and maintaining high-traffic servers for decades, that the user interface is poorly designed and uses outdated technology. One anecdotal claim that one person liked it is not persuasive, especially when that one person probably works for the company that built it.

      i imagine that a tool like this faces a lot more challenges than you know,

      Or maybe I know what the hell I’m talking about and the tool is awful?

      maybe something like moovit wouldn’t be very easy to use for an older audience?

      So build a good data collection tool that is easy to use for an older audience, not a tool like what they built, which will collect bad data from everyone.

      jik, you sound like a savvy web user, why don’t you help them improve the tool instead of being so negative.

      1) The contract was already parceled out, the money was already spent, the tool was already built, and it isn’t going to be rebuilt, at least not for this year’s data collection.

      2) It’s not my job. It’s apparently your job, since you work for the company that built it. You were paid money to design and build a good, state-of-the-art data collection tool. You failed. I’m not going to do your job for you.

      Reply
    1. jik Post author

      First of all, I did, in fact, “give it a chance before bashing it.” I registered as a user of the site, I tried to use it to set up the routes I normally travel on, and I explored how the travel surveys work. How would I have been able to write all the details above about how it works if I hadn’t tried it?

      Second, the data collection process as they’ve designed it is obviously severely flawed and technologically obsolete. One need not “give it a chance” to know that, one need only look at the state of the art in transit apps, which make it incredibly obvious that there are much, much better and more accurate ways to collect data than the system the T has wasted time and money building and rolling out.

      Reply

Leave a Reply

Your email address will not be published. Required fields are marked *