New York MTA changes stance on open data

Transit

by David Alpert • October 8, 2009 12:46 pm

The New York MTA has a new policy: Transit data is more valuable to the agency when released publicly at no charge than when hoarded as a potential source of revenue.

Photo by Baptiste Pons.

The MTA wasn't getting any money from Google in exchange for including it in Google Transit, but was demanding payment from smaller developers like StationStops, one guy who built a iPhone app with Metro-North schedules in his spare time. MTA even sued StationStops and got Apple to pull the app.

But following widespread coverage in the press, a letter from New York Councilmember Gale Brewer, and a new MTA chief, the agency decided to reverse course. They dropped the complaint against StationStops and fellow iPhone app The Next Train, a different programmer's app for the Long Island Rail Road. This week, Apple finally reinstated StationStops to the App Store.

MTA officials told the New York Times that they're "trying to evolve" to address this "emerging area." Colin Durrant of Massachusetts' Office of Transportation, which released data freely in August, told the Times:

We felt it was an essential role of government to open up our data and our information to developers. Rather than having a consultant develop a tool or an application or some sort of software, why not put the data out there and have people compete to develop products that we might not have the time nor the money to create? It's a win-win for everybody.

At the recent Metro board meeting, Alternate Director Gordon Linton argued that WMATA shouldn't release data without ensuring they get a cut, and even though Google wasn't paying MTA or anyone else for data, that means nothing because they hadn't asked. MTA's policy decision here changes that. They didn't ask, but they asked from others, and have decided officially that they won't and shouldn't ask. That's because the value to riders far exceeds the paltry revenue impact of this issue, and even Google isn't "lining their pockets." WMATA may now be the only major transit agency without a policy encouraging innovative applications that help riders. It's time to stop being the last holdout.

10 comments — Share	— tags: data openness, Gordon Linton, WMATA

Innovation resistance at Metro, part 1: The value of "bottom-up" (Sep 29, 2009)
Google: We don't pay MTA for data (Aug 24, 2009)
Google paid for NYC data (or not); Chicago removed indemnification (Aug 20, 2009)
New York MTA threatens blogger, asserts copyright over schedule (Aug 19, 2009)
Nextbus IP, Metro data quality dissuades one developer (Aug 13, 2009)

Comments

If I knew WMATA was going to have been such a poor TA which has a bad stance on data, I wouldn't have moved to the DC area. I live in a world class city with third world transit.

Who wants to help start a rider revolt? Examiner, if you're reading this, I BEG you to do a story on this!!

by Jason on Oct 8, 2009 12:51 pm

Not that this has anything to do with data collection, but I have to admit, I've never understood "station stop" as a phrase. Isn't it redundant?

Why not just say "station stop thing type situation"? Will there be automated ATM machines installed at every station stop?

by J.D. Hammond on Oct 8, 2009 1:05 pm

There are plenty of other examples of groups and organizations who realize that they end up better off by letting others develop products for them. My favorite recent example was the NetflixPrize, where Netflix offered up a $1 million prize to whoever could improve their recommendation accuracy by 10 percent. The team who won - Bellkor's Pragmatic Chaos - made out great because they got the money, and more importantly, the street cred. Netflix won, because they acheived more accuracy and saved money by not dedicating their own staff to the project.

So yeah, these kinds of agreements are a win-win for everybody. For the life of me, I don't understand WMATA's blind obstinancy on this one. Too much old blood there.

by JTS on Oct 8, 2009 1:06 pm

Maybe there's a layered strategy here:

Accurate live MTA data: $50.
Random guesses: $5.

PS Not that many of us aren't jealous now, but how many years ago did Nokia run their ads about Finnish kids live-tracking a bus so they could pelt it with snowballs?

by KadeKo on Oct 8, 2009 1:10 pm

I think Metro needs to understand that TripPlanner is an inferior product to Google Transit. And secondly, there are plenty of other avenues that Metro can examine when it comes to generating revenue. This happens to be the wrong one.

by Arch on Oct 8, 2009 1:38 pm

@Arch: How much more can we do to make them understand? They don't want to understand, they're too stupid to want to do so.

I really wish the press would pick up on this. That and the idea someone else had of doing a protest at the next Metro board meeting. How about getting some of the other planning agencies (ACT, etc.) on board? Just putting it out there...

by Jason on Oct 8, 2009 1:53 pm

I think that the only way we will be able to change Metros mind about this is if we can get a strong public reaction. This will probably mean attending the public meeting en mass, and getting coverage in the press. As much as the press seems to love to kick Metro when theyre already down, Im surprised we havent seen more articles on this. But theres not a lot we can do about that. However, everyone here that cares about this issue can show up at the next Board of Directors meeting and Riders Advisory Council meeting and make themselves heard. I intend to be at both.

@ J.D. Hammond: from Wikipedia the term station stop may be used in announcements, to differentiate a halt during which passengers may alight from a halt for another reason, such as a locomotive change.

by James on Oct 8, 2009 6:22 pm

Why is it that Metro is willing to spend $15,000 per month on NextBus, but insists that Google pays for access to their schedule data? It doesnt make any sense. The hardware required for NextBus is trivial, cheap and low maintenance. The real magic is in software that can learn traffic patterns and use this to deduce arrival times. Once this is implemented it, it is also low maintenance. It doesnt really make sense to license it for $15,000 per month. You could make the same argument for Google Transit. Again, the hardware is cheap and trivial. The magic is in the software (excellent parsing engine, ubiquitous, excellent platform integration and excellent data source integration).

I think the results both system provide sway the argument even more towards Metro paying Google. NextBus is a nice feature, but not necessary if you use Metro often, say for commuting, but is less useful to infrequent users such as tourists or maybe the occasional weekend trip to someplace you wouldnt normally go. After all we can more afford to waste 30 minutes once than to waste 30 minutes twice a day, five days a week due to a late bus during our daily commute. Google Transit on the other hand fills a big void. It instructs riders how to make the best use of Metro, and is probably the first place riders not familiar with the system will look (to be honest, Ive been riding Metro for 3 years and I only found out about Trip Planner six months ago when I went to Metros webpage for the first time looking for other information). Without a tool like that there is just too much data to sift through to efficiently figure out how to get from point A to point B.

Now why is Metro willing to pay for a nice to have, but insists on getting paid for a necessity?

by James on Oct 8, 2009 6:42 pm

@James - I think you are a bit out of touch with the bus riding community. I ride the bus all of the time and rely on NextBus. I would rather have NextBus over Google Transit anyday.

by True Bus Rider on Oct 9, 2009 11:17 am

@True Bus Rider: I ride the bus twice a day everyday. For me NextBus is a crap shoot. Half the time it is within a minute, the other half of the time it is off by 10 minutes. This morning I was watching NextBus for 45 minutes. I saw the prediction for my bus go from on time to 15 minutes late. 3 minutes before the actual arrival time, NextBus was predicting 18 minutes. Then the prediction jumped to 1 minute, which was correct, but by this time it was too late for me...I watched the bus go by from my window.

Regardless I think my argument still stands. Knowing (potentially) very accurately when the bus will arrive is a nice to have. Efficiently disseminating route and schedule data is a necessity for any transit agency. Let me put my argument another way: does Metro charge the company that prints their schedules or does Metro pay them? So why charge Google.

I only ride one bus, and I will agree with you, Google Transit is useless to me on that route because I know it like the back of my hand. The reason I don't use other bus routes is because of the difficulty in figuring out how to get where I want to go. This is problem very effectively solved by Google Transit.

by James on Oct 9, 2009 6:59 pm