Screen-scraping and Harmful Cybertrespass After Intel - George H. Fibbe

• Publication year: 2004

Screen-Scraping and Harmful Cybertrespass After Intelby George H. Fibbe*

I. Introduction

The topic for this Symposium, "The Internet: Place, Property, or Thing—All or None of the Above," touches on a debate that has existed since the early days of the Internet. There is no question that people commonly understand their experience using the Internet with the help of spatial metaphor—e.g., "sites" and "addresses" that we "visit," and programs called "robots," "crawlers," and "spiders." Leaving metaphor aside, many of the constituent parts of the Internet, especially computer servers, are items of private personal property.

Indeed, the debate over metaphor is reminiscent of the scene from the movie Field of Dreams¹ in which Shoeless Joe Jackson looks out over the magical baseball field and asks, "Is this heaven?"² The response is, "No, it's Iowa."³ Similarly, whether we consider our interaction with the Internet broadly—"Is this cyberspace?"—or more narrowly—"No, it's just a group of computers"—is a particularly timely topic for courts deciding disputes over "screen-scraping."

Many writers have criticized the "cyberspace-as-place" metaphor and its effect on judicial reasoning. Essentially, critics contend that the cyberspace-as-place metaphor has driven courts down the wrong path. These writers often object to courts' use of this metaphor because of the public policy implications of strong property rights on the Internet. This is especially true of screen-scraping cases applying traditional trespass doctrine to the Internet.

It is unlikely, though, that courts are blindly riding the cyberspace-as-place metaphor to reach results dramatically different from those they would otherwise reach. Enough room exists in the still-unsettled trespass doctrine and in available analogies for courts to reach differing results. Indeed, courts both enjoining cybertrespasses and refusing to enjoin them have openly resisted certain metaphors. Thus, judicial decisions may be driven more by a mundane balancing of interests than by confusion regarding the appropriate cyberspace metaphor. Specifically, courts are more likely to favor property owners when trespassers cause substantial commercial harm, which is common in screen-scraping cases.

The California Supreme Court's recent decision in Intel Corp. v. Hamidi⁴ may be an example of this balancing. In Intel the court acted against a general willingness among courts to apply the doctrine of trespass to cyberspace misbehavior.⁵ The majority even criticized the dissent's use of various spatial analogies,⁶ and some may contend that the court in Intel simply avoided confusing metaphors. In terms of the interests of the litigants that were at stake, however, Intel was different from most cybertrespass cases because it was not a purely commercial dispute. Intel lacked evidence of a threat of serious commercial harm that generally weighs in favor of owner protection, and it concerned somewhat unique, individual free-speech claims.⁷

Because of its noncommercial setting, Intel does not signal a reversal of courts' willingness to enjoin cyberspace trespasses by screen-scrapers. At the least, Intel is a poor vehicle for assessing the interests of website owners against commercially harmful scrapers.

II. What Courts Have Done

The path courts have taken in applying trespass-to-chattels to protect website owners from screen-scraping has been well noted but is worth a brief review.

Screen-scraping, also called data aggregation or indexing, encompasses technologies variously referred to as robots, spiders, crawlers, or automated devices. Disputes over screen-scraping tend to be between business actors and involve unauthorized access to websites for profit, all to the commercial detriment of the website owner.⁸ Like cases involving spam, screen-scraping disputes commonly include efforts by website owners to block the unauthorized access, as well as scrapers' evasion of such protections.

The first court to apply trespass-to-chattels to computer misbehavior was a California court in the 1996 case of Thrifty-Tel, Inc. v. Bezenek,"⁹ which concerned teenagers hacking into a telephone company's computer system. In the process, the teenagers were trying to find authorization codes that would allow them to make free long-distance phone calls. The teens managed to overburden the system and denied some phone customers service. The trial court found the teens liable for conversion.¹⁰ The appellate court affirmed but avoided the question of whether intangible computer access codes could be the subject of conversion by holding that the judgment was correct as a trespass-to-chattels claim.¹¹

After Thrifty-Tel, Internet service providers brought a series of cases against senders of unsolicited commercial e-mail, commonly known as spam.¹² The most influential of these cases was CompuServe, Inc. v. Cyber Promotions, Inc.¹³ In this 1997 case, the court found that Cyber Promotions's spam used up CompuServe's storage capacity and processing power, thus diminishing the value to the company of its computer systems.¹⁴ Also, CompuServe was losing customers because of spam, and the court found that CompuServe's lost goodwill was actionable under the trespass-to-chattels theory as well.¹⁵

Interestingly, CompuServe tried to block Cyber Promotions's spam, but the company managed to evade the blocking efforts by disguising the identity of the sender.¹⁶ This "masking" or "aliasing" of one's Internet address is a common practice of spammers and screen-scrapers alike and turns up in almost all of the major cases on the subject.¹⁷

One early and important screen-scraping case was EF Cultural Travel BV v. Explorica, Inc.,¹⁸ decided in 2001. Former employees of EF Cultural Travel Bv ("EF"), a tour company, started a competing venture. The new company, Explorica, hired software firm Zefer to develop a screen-scraping tool, which Explorica used to compile EF's tour prices from its website. Explorica then undercut EF's prices. EF obtained a preliminary injunction against the scraping under the federal Computer Fraud and Abuse Act.¹⁹ The court of appeals upheld the injunction because Explorica developed the scraper tool with EF information that was covered by a confidentiality agreement signed by a former EF employee.²⁰ Thus, the scraping was without authorization and was prohibited.²¹

The first major application of trespass theory to screen-scraping was Judge Whyte's now-famous opinion in eBay, Inc. v. Bidder's Edge, Inc.²² That dispute arose after the well-known auction website could not come to licensing terms with Bidder's Edge ("BE"), an auction "aggregator" that wanted to list eBay auction items on its own website. When eBay's technological efforts to block BE's access to its site were unsuccessful,

BE's software robots accounted for approximately 1.5 percent of the traffic on eBay's site.²³ The court recognized the reputational harm eBay might suffer due to BE's scraping, especially when BE may not have been displaying accurate or up-to-date information.²⁴ More central to the court's ruling was its recognition that eBay had a valuable property interest in its server capacity, and that the use of that capacity by BE deprived eBay of the use of some amount of that capacity.²⁵ Based on BE's scraping and the threat posed by BE's potential competitors accessing eBay's website, the court issued a preliminary injunction.²⁶

Shortly after eBay, a different federal court in California refused to grant a preliminary injunction against a company engaging in screen-scraping in Ticketmaster Corp. v. Tickets.com, Inc.²⁷ Tickets.com used a spider to scrape event information from Ticketmaster's site so that Tickets.com users could be routed directly to individual Ticketmaster event pages if Tickets.com could not offer tickets for the event.²⁸ In Ticketmaster the court focused more on the information at issue and seemed less concerned about how Tickets.com accessed that information.²⁹ Because Ticketmaster failed to produce any evidence of harm or potential harm to its computer system, the court declined to issue an injunction.³⁰ No evidence was before the court that Tickets.com's use interfered with Ticketmaster's regular business, and, unlike in eBay, there was no specter of "dozens or more parasites joining the fray."³¹ Also, the court did not recognize a substantial commercial harm caused by Tickets.com's activity where ticket buyers were sent to Ticketmaster's site to make their purchases.³²

Six months after eBay, a federal court in New York decided Register.com, Inc. v. Verio, Inc.³³ Register.com allowed users to register Internet domain names. As part of its business, it sent to registrants commercial marketing communications if they chose to receive them. Register.com also sold its list of recent registrants to other businesses on a weekly basis but put restrictions on how buyers of the lists could solicit the registrants' business. In a plan it called "Project Henhouse," Verio scraped registrants' contact information from Register.com's website so it could solicit them for Verio's website support services. Verio's business strategy included both screen-scraping and sending spam.³⁴

Verio admitted its scraping used some portion of Register.com's computer capacity, and the court noted that there was evidence that without an injunction, Verio's competitors would join in the scraping.³⁵ In enjoining Verio, the court found that part of Register.com's irreparable harm included "lost opportunities to sell competing services to its opt-in customers,"³⁶ and loss of "reputation and goodwill with customers and co-brand partners."³⁷

In late 2001, in Oyster Software, Inc. v. Forms Processing, Inc.,³⁸ defendant, Forms Processing, Inc. ("FPI"), copied Oyster's "metatags" such that search engines would show a description of Oyster's products and services but would then divert the searcher to FPI's website.³⁹ Relying on eBay, the court found...