The Battle Over Who Owns Bus Arrival Times
from the questions,-questions,-questions dept
We’ve had quite a few stories in the past about various public transportation authorities trying to stop others from creating iPhone apps that indicate train/bus schedule information. Often, the transit organizations claim intellectual property over the matter, saying that they want the ability to license the data themselves, or to sell iPhone apps themselves. This strikes many as being incredibly short-sighted. The core business of the transit groups should be to get more people using the trains and buses. Having good iPhone apps out there (often for free) would seem to get more people to use public transportation, and that benefit likely outweighs any money received from selling $5 iPhone apps.
However, a few people have sent in a story from San Francisco, where things are a bit more complex. The basics do seem to be the same, with a guy named Steven Peterson creating an iPhone app called Routesy to tell you when your Muni bus is arriving, only to have it shut down after complaints that it was violating intellectual property. Where it gets a bit more complex, is that it’s not the public transit authority, Muni, that’s complaining. In fact, Muni claims that it owns the data and says the public is entitled to use it (in fact, it claims to encourage it).
The issue is that Muni teamed up with a private company to put sensors on its buses and trains to note where the buses are and to predict when the buses would arrive. To make things even more confusing, the original company, NextBus, appears to have been separated into two separate companies: NextBus and NextBus Information Systems (I have to admit, the link above is really not at all clear on this). In a separate post, it appears that NBIS (who sent the complaint to Apple and had originally threatened the creator of the Routesy app) is actually claiming that it owns the prediction data.
While this is silly, it is a bit more defensible. NBIS seems to be claiming that since it owns the sensors and the data they produce, and then run it through some sort of algorithm to predict actual arrival times, it owns the predictions. You can’t copyright facts… but you could potentially claim copyright over predictions on facts — and I think that’s what NBIS is actually doing here (no one arguing over this seems particularly clear). A big question: is a prediction a “creative expression”? Some could argue that it’s not, but that seems to be what the claim hinges on. In theory, Muni could be right that the data on where each bus is can be freely used by the public, but the predictive data may be a bit different. In the end, it probably hinges on what sort of contract Muni signed with NextBus (or NBIS), and then the real question might be why Muni didn’t make sure that the predictive data was made to be open to outside developers as well. My guess: when this deal was being done, no one even thought that outside developers would want to do anything like this, so it wasn’t even a point of discussion.
Filed Under: bus schedules, copyright, data, predictions, sf muni
Companies: nextbus, nextbus information systems
Comments on “The Battle Over Who Owns Bus Arrival Times”
while defendable...
Even if it’s an easier defensive position to take, I still think it’s a poor maneuver. It would make more sense to sue muni for publishing the data rather than the developer of this app.
Muni could at this point just release the factual data and not the predictions and let the developer go from there. While NBIS might have a claim for IP for the predictive data they can’t have a claim for the locations and speed of the buses.
Important data
I would think that the most important data would be where the bus is at the moment. I have to laugh at the “algorithm” claim. If the red line is at 5th and Main at 6:25am it typically takes 21 minutes to get to 38th and Garden. Add 21 minutes to current time. Complex, I know. But I don’t want to put the actual time since it may be copyrighted. ;]
Re: Important data
Metro systems (whether bus or rail) typically accordion. It’s not the fact that it “typically” takes x minutes to get from A to B, more than it is that if the last bus or rail running the same route took x minutes, this transporter will probably take x minutes +/- y minutes. You can then vary y based on the current progress of this transporter, time of day, rush hour snowballing (i.e. if this run was faster/slower than the last, it is more probable that the next run will also be faster/slower).
If the dev is scraping the current times, then 15 minutes later, those “current” times are historic times, and it takes a trivial amount of math to plot the history and compute his own prediction rather accurately.
On the legal side, this is a joke. Every lawyer knows that the takedown is bull when the complainant merely refers to “our intellectual property.”
Re: Important data
Actually no. It isn’t that simple.
You have to consider things like “how long did the last bus take to do the run”, “how long does it take on average on a monday”, “is there a significant event that changes the time”, etc.
What you are suggesting is a simplistic system that many busy companies run to make their schedules, not something that predicts where any given bus will actually be, they are very different animals.
I would say that, yes, this information is valuable and a work product, not just general information.
Awesome for Biotech!
I work in R&D in an academic setting, and can think of a few clever ways to copyright various “predictions” (ie a hypothesis) that would have commercial value in the sense you could potentially own the rights to disseminate a factual outcome.
Sounds like a great way to control information and silence “unauthorized” voices to me.
All about money
Corporations only think about money. The revenue from the information would be trivial. Worse is transit utilites not giving out the info – wouldn’t you think they would want people to have the info so they could use the transit service? Of course not.
It is all in the contract...
The predictive data belongs to whoever the contract says it belongs to. If the bus company has the right contract, they own the data and can allow anyone to use it as they please. If NBIS is the “owner” of the data, than they can license it as they see fit.
I think Apple's action is the more questionable issue...
Obviously there exists an argument about who owns the data. However, Apple made that argument moot. They refused to update the Routesy application, and when it quit working (after a modification to the Muni web site) it was removed from the app store.
Apple simply took the word of NBIS as they must I think given the receipt of a take down notice. Nobody has ruled in favor of NBIS, but NBIS wins by default. If this data is really in the Public Domain (despite their arguments about owning predictions), then the Public loses by default. We (the Public) are denied the use of technology simply because “someone” asserts they own something, even if the courts, if pressed, might very well rule that this “someone” does not.
Predication
> “A big question: is a prediction a “creative expression”?”
In general, I’d say “Definitely, yes!”. But let’s change that up a little bit:
“Is a predication produced entirely by data and an algorithm creative expression?”
The answer here is, I think, “No.”
facts and creativity
As the comments by CNE and Awesome imply, the simplicity of the likely “algorithm” (a mystification of “calculation”) makes it seem unlikely and dangerous to suggest that it is not an unprotectable “fact.” And anyway, why then doesn’t the transit authority just make the location of the buses public and allow people to create apps with their own “algorithms” that will provide the predicted times. Or just to rebroadcast the locations of the buses so individuals can apply their own minds to the task of taking that fact and making their own predictions on when the buses will arrive at the locations they care about.
So...
If I predicts things like… the next presidential candidate, the weather, etc…
Those all become my IP? That is so sweet!
I predict that the next president will be either a man, or a woman.
From this point on, if you refer to the next president’s gender… you are violating my IP!
I also predict that there will be children born and that they will have names.
If you, henceforth, refer to said children by said name, you are violating my IP!
Re: So...
What if we look at it like a book. NextBus has a book of data on where each bus is, and how long it will take to get to each stop on the route.
If you copy one or two numbers, it’s fair use. If you copy the whole book, then what?
There’s a difference between the concept of “owning who the next president is” and owning a book of specific numbers. Nobody said that NextBus owned the data point “The next bus on the 4 route arrives at Embarcadero in 8 minutes”. You could estimate it at 8 if you want. But if you use their entire book, you are using their IP.
For example, if some guy is the oldest guy in the world, and Guinness publishes it. They don’t own that fact, and you can also publish the fact that the guy is the oldest guy in the world. However, if you publish an entire book that is a copy of Guinness Book of World Records, well, then would you see that they might have a grievance?
– Disclosure: I consulted for NextBus a few years back. I don’t currently, and can’t speak for them.
Hmmm
So if I work for a company as a contractor and develop software for the company I retain all rights to control the software and any results made by the software?
I though when one gets hired one forfeits the rights to company one is working for. This would be irrelevant of the contract – especially if the entity that does the contract actually publishes the software.
In the case the data is publish by Muni for free when they obtain it through NBIS through a contract. This suggests you cannot be using the published data. If you cannot be using the published data what is the point of publishing it?
Predections are protected
A big question: is a prediction a “creative expression”?
I used to work for a private weather forecasting firm, and I can assure you that predictions are creative (at least when done by a human) and are protected by copyright. They are different from facts which are not protected by copyright.
Re: Predections are protected
“I used to work for a private weather forecasting firm, and I can assure you that predictions are creative (at least when done by a human)”
Isn’t that the truth. Most of the weathermen I’ve heard have been very “creative” with their predictions as well…
Getting nextbus here in DC soon
Great…we’re getting Nextbus here in DC as well. Not that it makes much difference whether the data is available or not. The buses will still run late and be an hour apart.
predictions and creativity
I’d be shocked if Nextbus’s “algorithm” is anything as complex as what goes into a weather prediction. There are predictions and there are predictions. One way to tell the difference between protected predictions and unprotected predictions is to ask how much individual or institutional investment there is in the method developed. As I wrote above, the location at any given moment of a bus is indisputably a “fact.” Thus, even if in the unlikely event Nextbus’s investment in developing and applying their “algorithm” is anything comparable to the investment organizations put into weather prediction, I have no doubt that a lot of useful information could be garnered by customers themselves, not to mention app developers (who could develop their own “algorithms”) by broadcasting in real time the locations of all buses within the system.
While not beyond the realm of possibility, the likelihood of copyright law coming into play seems remote given the distinction between “facts” and “expression”. If there is a copyright law issue, it would likely be associated with how information is presented to the end user, i.e., is the data presented in such a way that it qualifies as a “compilation” under 17 USC 101?
A more likely possibility is that the “hot news” doctrine (as expressed by the USSC in INS v. AP (1918)) is involved. This is not a copyright doctrine, but one based upon a common law tort known as “misappropriation”.
It would be helpful in the context of this matter to understand:
1. Where is the alleged misappropriator obtaining the data displayed in his app?
2. What is the business model being used by the entity asserting “foul”, and how does the use of its data (assuming it is its data that is being displayed in the app) cause it financial harm?
how it happened
Michael Stonecker: from the stories it looks as if Nextbus provides the predictions to the public transit authority and that the app developer got them from their. Maybe Nextbus is claiming that its deal with the transit authority gives the transit authority a limited license with respect to how it can use the predictions. As to business model: the stories make plain that Nextbus charges for its predictions, though they also make it appear Nextbus has not been entirely consistent in its assertions regarding the price.
Re: how it happened
What makes this matter confusing, to me anyway, it that it is not clear how the app developer is getting the data in the first place (again assuming that the data he is using is in fact the data allegedly being misappropriated). It seems unlikely the app developer is sitting in a terminal, or has others sitting in a terminal(s) copying the data and forwarding it immediately to wherever it needs to go in order to be shown in real-time via the app. The only thing that seems to make sense is that somehow that data is being displayed in real-time via an on-line, and from there is being scraped for inclusion into the app display.
What also make this matter confusing, at least to me, is that it is not altogether clear how the data originator is making money in the first place. One possibility is, of course, that the transit authority is paying the company some money for his real-time data since it does help the transit authority better inform the public so that it better informs its ridership. Of course, the wider dissemination of this information via other sources does help the transit authorith’s riders…but it is not at all clear how this alternate source of data (again assuming it is in fact copied from the original data) negatively impacts the income stream to the data originator.
If it does have a negative finalcial impact, the “hot news” doctrine could come into play. How this might happen, however, is not altogether clear.
Importantly, no matter the answers to the above, the “hotnews” doctrine is not a copyright issue, but rather one rooted in unfair competition under state law. The INS v. AP case made clear that factual information is not covered under copyright law given the “fact” versus “expression” distinction central to copyright law.
MTAs doing it right
The mass transit authority of Portland OR, TriMet, offers a full-featured API including GTFS (Google Transit Feed) feeds and XML interfaces for scheduled arrival times, live predicted arrival times (TransitTracker), locating stops close to coordinates, and even finding a route between two points (TripPlanner). This can all be accessed by any software with an API key, and registering for an API key is free. As a result there are a plethora of third-party apps available (I know of at least 4 iPhone apps for arrival times).
It seems like this is good for everyone. TriMet’s riders are happier because they have a range of options (TriMet’s own automated hotline for arrival times, portable apps, desktop apps, both TriMet and third-party websites, etc…) to get the info they need on schedules and routes. this further opens things up for mashups that integrate TriMet’s data in to other systems (perhaps the best example being Google Maps), which draws all the more riders.
Even if they can claim copyright to stop other people using the predictions, it seems like a losing move.
All your bus are belong to me
I predict that if a bus passes stop A, it will still be a bus when it reaches stop B. So combining Muni’s logic and quantum mechanics, I own their bus. Bwaahaa haa haa!
mooove along
Reminds me of a bus driver in San Francisco I encountered a few years ago, idling the bus at the bottom of a hill (I wished to go to the top w/the bus). When I asked when she (and the bus) would be leaving the stop, I got nothing but attitude distilled to “…when I feel like leaving,” but otherwise would not say.
Would not have mattered a wit as to what a schedule might have indicated. It was about this time I decided that the Bay Area – after having lived there for 30 years – was becoming more and more unlivable (transit/bums/attitude/greed… yada yada), so I took a hike (figuratively) – feeling that SF Bay (and Calif if general) was/is doomed. I now reside in another (more polite/livable) state (good move on my part).
Re: mooove along
Your style of commenting is really peculiar. I feel like I just read a comment and commentary for it at the same time!
works for hire
Isn’t this considered “works for hire” which makes it exempt from copyright?
NextBus, Inc. was formerly known as NextBus Information Systems, Inc. and changed its name to NextBus, Inc. in July 2005. The company was founded in 1997 and is based in Alameda, California. As of July 1, 2005, NextBus, Inc. operates as a subsidiary of Grey Island Systems International, Inc.