From an English legal perspective, the requirement of intentional and direct interference with the possession of chattels must be shown to found an action in trespass. Prominent authors1 take the view that trespass is actionable per se (ie direct, unauthorised contact with another’s goods will suffice), with no requirement that the claimant suffer damage, although there is little in the way of case law on the point.
Three American cases are examined here, in the order in which they were decided.
Ebay Inc. v Bidder’s Edge Inc. (District Court of Northern California, Case No C-99-21200 RMW) (Application for an interim injunction) Use of Web crawling software to scour, copy and retrieve information contained on a Web site
FACTS: The claimant, Ebay, is a well-known site, which facilitates person-to-person trading: sellers can list items for sale on the site and prospective buyers can search listings and place bids for items. The defendant, Bidder’s Edge, is an auction aggregation site – ie it does not host online auctions itself, but allows would-be buyers to search for items across a number of such auctions, including Ebay’s. Rather than conducting searches of the Ebay site when this was queried in a user’s search (and as Ebay had requested them to do during licence negotiations), Bidder’s Edge employed software robots (ie computer programs that operate across the Internet to search, copy and retrieve data from others’ Web sites) to trawl the site up to 100,000 times a day and Bidder’s Edge would compile its own lists from the data extracted by the robots.
The Court held that in order to succeed with a claim for trespass based on accessing a computer system, the claimant must establish:
- that the defendant intentionally and without authorisation interfered with the claimant’s possessory interest in the computer system; and
- that the defendant’s unauthorised use proximately resulted in damage to the plaintiff.
What did each of these elements mean in practice?
(1) The court did not accept that, because a Web site is publicly available and accessible, that precludes an action in trespass in the circumstances described above. Web servers are private property, to which the owners grant conditional access. Regarding the degree of interference that must be shown, the court applied an earlier Californian authority (Zaslow v Kroenert, 29 Cal. 2d 541, 551 (1946)), in which it was said that ‘Where the conduct complained of does not amount to substantial interference with possession or the right thereto, but consists of intermeddling with or use of or damages to the personal property, the owner has a cause of action for trespass, and may recover only the actual damages suffered by reason of the impairment of the property or the loss of its use’.
(2) It is enough for claimant to show that defendant’s activities have diminished the condition, quality or value of the property. Bidder’s Edge’s activities robots consumed a proportion of the claimant’s server capacity (albeit fairly small).2 The court was more concerned, however, by the deleterious effect of multiple aggregators following suit and denying effective access to bona fide customers of Ebay.
In the court’s view, the strongest argument in support of claimant’s application for an injunction was that to allow Bidder’s Edge to continue in its activities without restraint would encourage other auction aggregators to engage in similar intensive and repeated searching of Ebay’s site, causing the system to become clogged or unavailable to Ebay customers, with resultant data losses and harm to Ebay’s business. The court referred in this context to the fundamental right of a property owner to exclude others from its computer system – a violation of that right could of itself, potentially cause irreparable harm sufficient to justify an interim injunction.
Ticketmaster Corp v Tickets.com Inc. 2000 WL 1887522 (C.D. Cal) (Use of hypertext linking into pages beyond the Web site’s home page, ie “deep-linking”)
The claimant, Ticketmaster, is an online vendor of tickets to entertainment events (eg sports fixtures, music concerts, etc). The Terms and Conditions for use of the site prohibited copying the site’s contents for commercial purposes – however, there was no requirement on users to indicate acceptance of these terms and conditions prior to using Ticketmaster site, nor even to view them as a precondition to use.
The defendant, Tickets.com, sells tickets on its own account to certain events and also provides information on how to purchase tickets that it does not sell itself. In the latter case, Tickets.com provided a hypertext link on its site which when clicked upon, took the customer to the particular page in the Ticketmaster site for purchasing tickets to the event in question: obviously, this bypassed Ticketmaster’s homepage in the process and the advertising displayed there. In order to obtain the basic information on Ticketmaster events, it was alleged that Tickets.com had copied Ticketmaster’s interior Web pages and extracted the basic information about events (using Web crawlers to trawl through the contents), which it reformatted and placed on the Tickets.com Web site.
In this case, no evidence about the use of Web crawlers and any damage they may cause to claimant’s computer system was presented to the court. The court, having no difficulty in finding that the principles relating to trespass applied to the automated interference with a computer system by a software program, said this: ‘The computer is a piece of tangible personal property. It is operated by mysterious electronic impulses which did not exist when the law of trespass to chattels was developed, but the principles should not be too different. If the electronic impulses can do damage to the computer or to its function in a comparable way to taking a hammer to a piece of machinery, then it is no stretch to recognise that damage as trespass to chattels and provide a legal remedy for it. Judge Whyte in eBay found the damage in the occupation of a portion of the capacity of the computer to handle routine business and conjectured that approval of that use would bring many more parasitic like copies of the defendant feeding to that computer to a clogged level upon the information expensively developed by eBay, the net result likely being severe damage to the function of the computer and thus the business of eBay.’
However, the lack of evidence of harm, or of obstruction to the functioning of Ticketmaster’s computer system, prevented it from making a case in trespass against Tickets.com. The taking of data did not operate in competition with Ticketmaster – any sales were concluded between the customer and Ticketmaster. The only possible detriment to Ticketmaster, said the court, was loss in advertising revenue occasioned by users skipping the home page and going directly to the part of the site to make the transaction.
In this respect, it is difficult to see a real distinction between the facts of this case and what happened in Ebay – all Bidder’s Edge did was to collate information about what was on offer at auction Web sites and present these findings in response to a search by a prospective purchaser – any sales that were made were concluded with Ebay, and not Bidder’s Edge.
The differentiating factor in Ticketmaster was the lack of damage (or the lack of evidence presented to the court as to potential damage) caused by Tickets.com’s activities in linking through to Ticketmaster Web pages beyond the home page: establishing a link to a site is plainly less invasive than sending hoards of Web crawlers to repeatedly access and copy a site’s data.
Oyster v Forms Processing Inc N.D. Cal., No C-00-0724 JCS, 12/6/01. (Decision on defendant’s application for summary judgment – ie asking the court to dismiss a number of the plaintiff’s claims). Application of the tort of trespass to the use of robots which scour a Web site to find meta-tags3 which are then copied and used on defendant’s Web site.
The claimant owned the rights to a software system called Formspro, which allows users to process documents and convert them into computer-legible data.
The defendant was a data-entry and document management service company. In 1999, the claimant discovered that some meta-tags from its Web site had been copied onto the defendant’s Web site. The claimant brought an action against the defendant, alleging (amongst other things) trespass and copyright infringement.
The defendant made a pre-trial application to dismiss the claims. In relation to the claim for trespass, it alleged that the claimant had failed to adduce evidence that the use of robots or “web-crawlers” (owned by a third party, whom the claimant said were acting on behalf of the defendant) to access the claimant’s meta-tags and copy them obstructed the function of the claimant’s computer system.
While the court agreed with the decision reached in Ebay, it also opined that whether the interference was substantial or not was not the point – it was the unauthorised use of the claimant’s computer that constituted the tort (developing the point made by the court in ebay about the importance of the right of property owners to say who may use their property, and how).
CONCLUSIONS
The legal analysis adopted in these US cases could be applied just as well to a matter before the courts in England.
1. Stringent Terms and Conditions of Use of the site, and a requirement that the user signify his acceptance (usually by clicking a button) to those terms before use, should define the scope of the permission granted by the Web site owner for the use of its site. The ebay site had specific conditions dealing with site use, whereas Ticketmaster did not even require users to read its terms and conditions prior to using the site.
2. An important factor deciding whether interference is unlawful may well be how discerning the webmaster is in granting licences to third parties to send robots and other web crawlers to trawl its site for details of its products or services. A clearly drafted set of terms and conditions, together with a selective webmaster, indicates that a degree of control over access to and use of the site is retained. The court in Ebay noted the essential proprietary right, namely the ability to exclude others from one’s property. Web site owners should carefully consider the extent to which their practices are consistent with this right to exclude. Unless the site generally permits the kind of automated access performed by a webcrawler, robot or similar software that repeatedly interrogates and copies information from a site, such access will constitute an actionable trespass.
3. Potential deleterious effects on a claimant’s business were viewed with some consternation in the above-mentioned cases. Indeed, it was the lack of any (or any sufficient) evidence of this that led the court to its conclusion in the Ticketmaster case. While the extent of the damage suffered will obviously affect the amount of compensation, the paucity of English authority on the need for damage to be shown in order to establish the tort would make it advisable to have cogent evidence that webcrawling has (or may) result in identifiable harm to business.
Helen Barnard is a barrister employed by the Treasury Solicitor’s Department. She advises on all aspects of e-law (including Public Key Infrastructure, secondary legislation under the Electronic Communications Act 2000, and web sites) and data protection.
© Helen Barnard, 2002
1 Salmond and Heuston on the Law of Torts (21st edition, 1996), p.95; Street on Torts (10th edition, 1999), pp.69-69; Winfield and Jolowicz on Tort (15th edition, 1998), p.585-596
2 According to evidence adduced at the hearing, the load on eBay’s servers incurred by the activities of Bidder’s Edge’s robots represented between 1.11% and 1.53% of the total.
3 Meta-tags are Hypertext Markup Language (“HTML”) code that describe the contents of an Internet Web site to a search engine. There are two types of meta-tags: 1) “Description” meta-tags, which are intended to describe the website, and 2) “Keyword” meta-tags, which are keywords relating to the contents of the website. So, the more frequently a word appears in the meta-tags and the text of a web page, the more likely it is that the web page will appear in search engine results (i.e. will be “hit”) – the greater the number of hits, the higher up the list of search results the web page will be.
