Earlier this spring, Public Engines, Inc. sued ReportSee, Inc. in federal district court in Utah. Both companies maintain websites that publish local crime statistics and information gathered from law enforcement agencies. Public Engines gathers crime data by contracting with law enforcement agencies across the country to provide software and data management services. In contrast, ReportSee primarily gathers its information from public data feeds, police departments and, according to allegations in Public Engine's complaint, by scraping Public Engine's CrimeReports.com website. The complaint alleges breach of contract because scraping violates the terms of service of the website and hot news misappropriation, among other claims. (Techdirt has a good outline of the claims in the complaint and our legal threats database entry on this case has more background along with links to the complaint and other legal documents.)
This strange case got another wrinkle on June 24, 2010, when both parties stipulated to a preliminary injunction that bars ReportSee from any further scraping activity, making any commercial use of crime report data from Public Engines websites, and from contacting or communicating with law enforcement agencies for the purpose of obtaining Public Engines' data feed.
Hot news misappropriation -- Everyone wants in on the action.
The case is a little bizarre because from the outset, the facts aren't a great fit for a hot news misappropriation claim. While Utah hasn't had a hot news misappropriation case, Public Engines pled facts in the complaint that track the elements of the now-familiar test laid out in the influential case NBA v. Motorola in the Second Circuit, 1
(1) plaintiff generates or gathers information at a cost; (2) the information is time-sensitive; (3) a defendant's use of the information constitutes free riding on the plaintiff's efforts; (4) the defendant is in direct competition with a product or service offered by the plaintiffs; and (5) defendant's free-riding behavior would reduce the incentive to produce the product or service that its existence or quality would be substantially threatened.
Why is the hot news misappropriation doctrine a stretch here? For starters, while Public Engines argues that "[o]ne of the main purposes of the CrimeReports.com website is to make information about crimes available in near real time," it's not clear how the information is time sensitive. What makes this crime data "hot"? And hot to whom? In the recent case Barclays v. FlyOnTheWall, the financial information was hot because it ceased to be valuable after a short period of time and customers were purportedly paying to have early exclusivity. In this case, the crime data is made publicly accessible on a website - for free. Public Engines claims that the data is made available "in near real time," by which it means it updates its database at least once every 24 hours. But is the nature of the information "hot" enough to really be time-sensitive? Just how "hot" is "hot" anyway?
The bigger problem for Public Engines' hot news claim is its assertion that competition from ReportSee will "reduce Public Engines' incentive to produce this product or service, and the existence or quality of the information reported on CrimeReports.com will be substantially threatened." However, Public Engines is being paid by government agencies to compile the data and display it on its CrimeReports.com website. As its own complaint points out, "each law enforcement agency pays a fee for the CrimeReports.com service" because "[w]ithout Public Engines' technology, few law enforcement agencies would be able to make this information available in this form." The agencies pay Public Engines for these services because they "have a keen interest in making available to the communities they serve current and accurate information about crimes and criminal activity in the communities they serve[.]"
In the classic fact pattern for hot news misappropriation, a provider of news is paid by consumers of news only after the investment in newsgathering has been made (thus competition that undermines consumer demand through free-riding would negatively impact upstream incentives to invest in newsgathering activities). The Public Engines case has that fact pattern turned inside-out and backwards. The consumer in this case pays nothing. The information is available on a public website that does not charge a fee for anyone to use its site, does not advertise and does not solicit business from users.
So what's Public Engines' business model? It's paid upfront for its services by a party interested in investing in the very newsgathering activities at issue, irrespective of consumer demand. So where is the threat to the incentive in putting the data online? Even Public Engines' complaint underscores how it serves the agencies' interests in paying for newsgathering activities. Indeed, in its complaint, Public Engines brags that "CrimeReports.com is able to serve as an ‘official' crime information portal for the law enforcement agencies."
What, you thought making public records available online meant better access to government information?
What makes this case really crazy interesting however is that the "news" at issue is government data. Public Engines contracts with police departments and law enforcement agencies in order to get crime information out to the public. Open government data initiatives generally promote the use of government data to improve reporting, transparency and accountability. Access to data feeds is the point of websites like Data.gov and similar websites for state and local jurisdictions.
But those initiatives may be in tension with the fact that government agencies often pay third parties to collect, compile and maintain public records data in useful formats, and who may retain rights over the data. This isn't the first time a third-party data contractor has stepped in the way of a commercial use of data feeds. In the Bay Area a few years ago, Routsey's iPhone app making use of data feeds with bus and train arrival times got in a jam when the contractor providing the data to MUNI, the public transportation agency, asserted its rights to the data.
Under the terms of the license agreement between Public Engines and law enforcement agencies, the agencies own all of the raw crime data, which Public Engines was contractually bound to keep confidential. Public Engines however, owns the data generated from removing sensitive and identifying information that was not suitable for release to the public, what it called "De-Identified Data." Furthermore, the law enforcement agencies were "prohibited from disclosing the intellectual property associated with the [software] . . . as well as the ‘De-Identified Data.'"
Third-party ownership of processed government data, like in Public Engines' case, probably doesn't raise a public records issue because other parties (like ReportSee) in theory can still obtain public records from government agencies themselves. Their rights would not have been abridged by the existence of a separate contractual arrangement with a third party. Parties like ReportSee would have to cull through voluminous crime reports and blotters, just as they would have in the old days. In practice however, there could be a problem if the agency denies public records requests and instead directs requesters to third-party contractors, who then stand in the way of access.
The bottom line is that this sort of dispute could be avoided if government agencies are more proactive and farsighted when negotiating terms with third-party providers of data management services. In particular, government agencies should maintain control over the resulting data, or at a minimum, require that the contractor permit a wide range of uses of the data. It's not just in the public interest of promoting government transparency and accountability. It's also in the agencies' interest to streamline its public records requests. The agencies are already paying for the data management services anyway, why spend even more government resources in order to respond to redundant public records requests?