Greater Greater Washington

DDOT: NextBus not the answer for Circulator

DDOT doesn't currently plan to integrate the Circulator into the NextBus system. Instead, they plan to develop a separate application for mobile devices to tell you how far away the Circulator is.


This number ought to yield real-time info. Photo by katerw.

DDOT agreed to release an email summarizing their reasons:

Many of the other regional systems disagree with Metro's approach to bus real-time information. NextBus is proprietary, expensive, rather inflexible, and difficult to administer. ... What we're trying to do with Circulator is use technology we already have (GPS tracking on all Circulator buses) and develop an extremely low-cost application to let people know how far away the next bus is. From a passenger standpoint, this can be just as useful as saying the bus is 5 minutes away.

We'd also have to deal with the fact that the system structure is different. Metrobus runs on a fixed schedule and the NextBus predictions use the fixed schedule to come up with a predicted arrival time. Circulator does not have a fixed schedule and is constantly monitored by the contractor to get the buses arriving as close to a 10-minute interval as conditions allow. The NextBus algorithm is not flexible enough to deal with how the Circulator service is tailored in real time.

Last I would argue that a NextBus type system is less important for an operation like the Circulator - its entire mission is to run frequently enough so that people don't have to check to see when it is coming. They should be able to just wait at a stop and know they won't have to wait any longer than 10 minutes. For some of the routes/systems that run at 20, 30, or 45 minute headways the actual arrival time is of much greater use to a customer.

The last paragraph, of course, assumes that the bus really does run that frequently. This weekend, that wasn't the case. Even so, when a UK system added real-time information, people thought the buses arrived more often and more reliably, even when they didn't.

The rest of the reasons in the DDOT statement ring true, however. A lot of tasks in technology are easy, but integrating different systems with different design parameters is often not one of them. I can certainly believe that NextBus has many specific design constraints, as any system would, and that the WMATA-NextBus contract is narrowly written to meet Metrobus's needs alone.

We should be able to see for ourselves. WMATA is a public agency, and supported by our tax dollars. We should be able to see the costs and the requirements of the NextBus contract. If the parameters of the deal could better facilitate including non-WMATA, more dynamic services like the Circulator, we should be able to suggest that for the next iteration.

The more important solution is open standards. If the NextBus system doesn't meet the Circulator's needs, let some other innovator build a better application. DDOT and Metro should both create data feeds where any developer can download the real-time bus arrival information. Let Google Transit integrate it into their trip planner. Let someone else build an even better application that not only tells you when your bus will come, but suggests alternate routes, and recommends a coffee shop along the way.

For now, something separate is better than nothing. A low-cost application is a good start. Ideally, we'd install real-time digital displays at the stops. In the meantime, DDOT already posts the phone number on the signs, and that number should tell callers how far away the next bus is. In the long run, both Metro and the other regional agencies should open up the real-time feeds, so future developers, whether at NextBus, Google, or in someone's garage, can build the integrated real-time system we really deserve.

David Alpert is the founder and editor-in-chief of Greater Greater Washington. He worked as a Product Manager for Google for six years and has lived in the Boston, San Francisco, and New York metro areas in addition to Washington, DC. He now lives with his wife and daughter in Dupont Circle. 

Comments

Add a comment »

"WMATA is a public agency, and supported by our tax dollars. We should be able to see the costs and the requirements of the NextBus contract."

Has anyone asked, or FOIA'd, for that information?

by lucre on Mar 17, 2009 11:15 am • linkreport

I was assisting another commenter with that request but I don't know whether he actually has initiated the request. If he hasn't, I'll be sure to start one. (Why not, I have three other requests pending already!)

BTW, does anyone have access to a high-speed scanner? I have 400+ pages of WMATA documents that I would love to share with everyone but I am not really willing to make everyone physical copies.

by Michael Perkins on Mar 17, 2009 11:38 am • linkreport

The DDOt statement is full of untruths. Let's start from the top:

"Many of the other regional systems disagree with Metro's approach to bus real-time information. NextBus is proprietary, expensive, rather inflexible, and difficult to administer. ... What we're trying to do with Circulator is use technology we already have (GPS tracking on all Circulator buses) and develop an extremely low-cost application to let people know how far away the next bus is. From a passenger standpoint, this can be just as useful as saying the bus is 5 minutes away."

Actually many of the area transit systems already use NextBus: Prince George's County The Bus, Howard County, Fairfax CUE, and Alexandria DASH. Three universities use NextBus in Baltimore: University of Baltimore, Towson University, Loyola College.

Difficult to administer? It's all done through the website www.nextbus.com. Looks pretty simple to me.

NextBus uses the GPS tracking devices already in place for WMATA and presumably could do the same for the Circulator.

Proprietary? Umm, so what? Every company that does this offeres their own version of the software. They are all proprietary, just as the Circulator's in-house software will be proprietary.

"We'd also have to deal with the fact that the system structure is different. Metrobus runs on a fixed schedule and the NextBus predictions use the fixed schedule to come up with a predicted arrival time. Circulator does not have a fixed schedule and is constantly monitored by the contractor to get the buses arriving as close to a 10-minute interval as conditions allow. The NextBus algorithm is not flexible enough to deal with how the Circulator service is tailored in real time."

This is just a flat out lie. NextBus uses GPS locations for arrival predictions and NOT the published timetable. In addition, NextBus provides predictions for both timetable based routes and routes that run on continuous loops without a timetable. Someone at DDOT should actually talk to NextBus before releasing a statement like this.

I doubt their system runs like clockwork on a 10 minute frequency. Just as you folks did at WMATA, you should demand NextBus at the Circulator.

by DDOT Statement wrong on Mar 17, 2009 1:05 pm • linkreport

Michael: not exactly "high-speed", but I do have access to a scanner...

by Froggie on Mar 17, 2009 1:53 pm • linkreport

Froggie: Well, I have access to a flat-bed scanner too, but it would take me many hours to scan 400 pages.

by Michael Perkins on Mar 17, 2009 2:05 pm • linkreport

Understood, was just putting the offer out (since I don't have your E-mail)

by Froggie on Mar 17, 2009 2:26 pm • linkreport

I have access to a high speed scanner at work.

Shoot me an email (I know one of contributors of this site)

by Matthew on Mar 17, 2009 2:39 pm • linkreport

BTW, does anyone have access to a high-speed scanner? I have 400+ pages of WMATA documents that I would love to share with everyone but I am not really willing to make everyone physical copies.

I would ask WMATA for the docs in electronic format one more time, in writing -- email is fine.

If they don't help, then tell them you're going to hold a 'Scanning Party' with 20 of your closest activist friends. You will scan these pages and make sure everyone knows that WMATA leadership is 'petty' and 'recalcitrant' and 'obstructionist' and is 'just not very interested in helping us improve public transit in DC'. Name names.

And then we scan the pages. And then we win. :)

Side note -- I've scanned lots of pages before, and it _is_ time consuming, but it's immensely fulfilling, too. The best is to use it with OCR software if you can, so you can pick apart all the text.

by Peter on Mar 17, 2009 4:41 pm • linkreport

@Peter: As always, I asked WMATA for all documents electronically, and their policy is that they are supposed to maintain documents electronically, and provide them electronically if possible.

That being said, they have not decided to get software to redact things electronically. Therefore, every document I request that has to be redacted (and almost all contracts do because WMATA insists on redacting contractor employee names) has to be printed and then redacted with a "Marks-a-lot" marker.

It's frustrating. Some of the documents I request are Excel spreadsheets, which WMATA has printed out, then scanned to PDF (no text layer) before they're provided to me. They claim Excel spreadsheets contain metadata which would be a security risk for their network. I have to re-OCR them to be able to sort different columns.

Just to get everyone's interest, I have the contract for WMATA's "mystery shopper" service as well as the contract for the promised Smartrip upgrades circa 2005.

by Michael Perkins on Mar 17, 2009 5:48 pm • linkreport

Screw the prediction accuracy and administration. Give us one server with access to the raw, live GPS feeds updated on a 10-second frequency. We'll perform analysis. We'll even archive it so that we have historic timing data - it costs about $1/year in hard drive space for every bus in the fleet.

So give it to the public space. Let them develop applications. And as a carrot:

Announce that after a year of the data being publicly available, you will be adopting the best prediction system and plastering it on the official WMATA site in return for a $50,000/year contract as technical consultant.

by Squalish on Mar 19, 2009 12:03 am • linkreport

@DDOT Statement Wrong: Alexandria DASH actually dropped the NextBus system because of their poor customer support and issues with inaccurate prediction times at the Pentagon. I'm not sure what WMATA did to solve that problem, or if they actually did anything at all (I wouldn't be surprised if they didn't, WMATA's NextBus implementation is a mess to be honest), but they're a bigger agency with a bigger contract that means more staying power to get things fixed, versus smaller agencies which are more or less expendable, in NextBus's view at least. DASH is currently working on implementing a system from a Canadian company called Strategic Mapping.

by NextBus ain't the best on Aug 7, 2009 2:58 pm • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.

or