A common way for businesses to extract and use data is screen scraping. This process can provide significant commercial benefits to businesses, but it comes with a legal risk. Depending on the circumstances scraping data from websites without consent can be unlawful.
Screen scraping (otherwise known as web scraping, data scraping, or web harvesting) is the action of using a software program or application to copy and extract data from a website to make use of it for another purpose. It is an automated process usually implemented using software tools known as bots or crawlers.
A common example is aggregator sites, which use screen scraping to collate data from multiple sites and present it to consumers in a consolidated form, such as an airfare comparison website. Some popular aggregator websites you would be familiar with include:
Attribution: By Skyscanner Ltd (http://www.skyscanner.net) [Public domain], via Wikimedia Commons; and By Booking.com BV (http://www.booking.com/) [Public domain] via Wikimedia Commons.
Screen scraping can provide significant commercial benefits. For example, it can allow companies to:
But those benefits can come at a cost to the owners of the sites being scraped. The company scraping data may take sensitive information owned by another company, or divert customers away from a scraped site, or cause a scraped site to slow down or crash.
Depending on the circumstances, screen scraping presents a range of possible legal risks.
In Australia copying data or substantially copying data from a third party website without the authority of the owner may infringe copyright.
It is a question of whether the data qualifies as an original “literary work” and is therefore capable of copyright protection.
The legal position in Australia is that copyright generally won’t exist in basic databases or lists unless a human author was involved and there was some intellectual effort in the creation of the database or list. For example, copyright generally won’t exist in a TV guide or phone directory. However, this needs to be considered on a case by case basis.
In Nominet UK v Diverse Internet Pty Ltd  FCA 1244, Nominet UK provided a central registry for UK-based Internet domain names and provided a search function using a database derived from its register known as the WHOIS Database. Australian companies, including Diverse Internet Pty Ltd, used data mining techniques to extract and collate names and other details of registrants on the WHOIS Database and then send misleading notices to domain name registrants.
The court held that the data mining constituted an infringement of copyright in the Registry Database and the WHOIS Database.
Screen scraping without consent could be in breach of website terms and conditions. Many website terms and conditions now specifically prohibit any form of screen scraping.
In Australia clickwrap agreements on websites, where a consumer checks the box to agree to terms and conditions, are valid agreements. The position in Australia in relation to hyperlinked website terms and conditions (or browsewrap agreements), where a set of terms and conditions is available via a hyperlink on the website but the user is not forced to open the relevant web page, or click “I agree”, is less clear. But it is arguable that these terms and conditions are enforceable and they should not be ignored.
In the European Union, the Court of Justice of the EU (CJEU) ruled in Ryanair v PR Aviation that a website operator can contractually prohibit other businesses from scraping information or data from a website.
The case involved airline operator Ryanair and a Dutch price comparisons business, PR Aviation. PR Aviation extracted flight information from Ryanair’s website and put that information on its own website, where consumers could compare prices of airline companies and, on payment of a commission, book flights.
To gain access to the Ryanair flight information, PR Aviation had to agree to Ryanair’s terms and conditions which prohibited the use of an automated system or software to extract data from the website for commercial purposes.
It is an offence under State and Federal legislation in Australia to access restricted computer data without authority to do so.
For example, section 308H of the Crimes Act (NSW) and Div 478 of the Schedule to the Criminal Code Act 1995 (Cth) prohibit unauthorised access to restricted data.
In most cases data scraping won’t breach these provisions, because the data being scraped is not “restricted data”, i.e. it is not data protected by a password or other security or access control system. However, significant penalties apply to those who intentionally access restricted data, the maximum penalty is 2 years’ imprisonment.
It is established in Australia that the tort of trespass applies to physical property or goods. Trespass occurs when someone intentionally acts to directly cause an injury to another person’s property or goods without lawful justification.
There are international cases where courts have applied the tort of trespass to online activities. In the US, a preliminary injunction was granted to eBay, who had complained that the data scraping activities of another company, Bidder’s Edge, constituted trespass under US law. It was held by the court that the eBay website was personal property and that Bidder’s Edge trespassed when it intentionally interfered with eBay’s personal property.
It is unclear whether Australian courts will apply the tort of trespass to an instance of screen scraping, but the international guidance suggests that it is a possible risk.
The legal principles that may apply to screen scraping in Australia will be of equal interest to website owners and users of screen-scraping technology.
Website owners may consider implementing greater security measures and amending their terms of conditions to include restrictions on screen scraping.
Businesses that use screen scraping technologies should be mindful of the possible legal risks in Australia and consider reviewing their strategies and operations.
If you are uncertain how screen scraping may affect your business, contact us.