Web scraping returned to center stage in the Sept. 9 Ninth Circuit decision that affirmed a preliminary injunction in favor of hiQ Labs, Inc., holding that LinkedIn cannot prevent hiQ, a web scraping company, from harvesting data from publicly available LinkedIn profiles.
Overall, the Ninth Circuit’s decision should not be taken as a green light to scraping. The decision was not a full review of the merits and scraping cases are very fact specific. The safest way to avoid scraping is to use technology, which identifies scrapers, blocks them, and alerts the website owner.
“Web scraping” is the collection of data from computer servers through specialized software or “robots.” Such software simulates human web browsing to collect information from scraped websites. Collected data is either used by the scraper for internal purposes, to provide its services and products, or sold in one form or another to the scraper’s clients.
hiQ used robots to gather information about employee skills and sold the information to its customers, such as eBay, Capital One and GoDaddy. hiQ also scraped information about client employees in order to assess which employees are most likely to leave their job.
After LinkedIn served hiQ with a cease-and-desist letter, hiQ sought a preliminary injunction for LinkedIn’s tortious interference with hiQ’s contracts. LinkedIn used a claim under the Computer Fraud and Abuse Act (CFAA) as a defense. The CFAA prohibits “intentionally accessing a computer without authorization, or exceeding authorized access, and thereby obtaining information from any protected computer.”
Significantly, circuit courts have been split in interpreting “unauthorized access.” The Second, Fourth and Ninth Circuits held that the CFAA prohibits unauthorized access by means of hacking. This means that a scraper would not violate the CFAA, so long as the access to information was authorized.
LinkedIn’s key argument was that after it had sent a cease-and-desist letter to hiQ, hiQ was no longer authorized to scrape any data from LinkedIn profiles. The court disagreed. It interpreted “accessing a computer without authorization” as the action of circumventing a target website’s technological access barriers, such as usernames or passwords.
Other key legal theories potentially apply to web scraping claims: the tort of trespass to chattels, breach of contract, and copyright infringement.
Trespass to chattels was successfully used in eBay Inc. v. Bidder’s Edge Inc., 100 F. Supp. 2d 1058 (N.D. Cal. 2000). However, since then, courts have been reluctant to accept this theory without proof of tangible damage to, or interference with, the proper function of the target website resulting from scraping.
Copyright law protects creative expression and may, therefore, protect the manner in which information is arranged on a website. However, web scraping is often only a collection of data, rather than a collection of data arranged in an original manner. A mere collection of data not arranged in any creative manner cannot be protected under copyright law, because it will not meet the originality requisite for copyright protection.
Moreover, with public websites like LinkedIn, into which users input their information, the owner of the website often does not own the scraped data in the first place. Therefore, copyright likely does not effectively protect computer servers from scraping.
Overall, the Ninth Circuit’s decision should not be taken as a green light to scraping. However, the decision is merely a grant of a preliminary injunction and was not a full review of the merits. Moreover, all scraping decisions are both fact-intensive and specific, rarely representing scenarios identical to one another.
A Roadmap for Scrapers and Scrapees
Nevertheless, several general themes have emerged that provide a useful and practical roadmap for scrapers and scrapees alike.
Scrapers can eliminate their exposure by entering into, and strictly abiding by, a license agreement with the targeted website owner. If this option is not available, then scrapers should be careful not to damage, slow down, or interfere with the scraped website to avoid tort claims.
Instead of relying on the courts as a first line of defense, the safest way to avoid scraping is to use technology, which identifies scrapers, blocks them, and alerts the website owner.
Website owners can also avoid making the information public. Understandably, however, this might not be a viable solution for many businesses, such as LinkedIn, whose users rely on the information being public.
In the meantime, interested parties should closely monitor the hiQ Labs litigation for a decision on the merits and to see whether a future Supreme Court decision will ultimately resolve the current circuit split.
This column does not necessarily reflect the opinion of The Bureau of National Affairs, Inc. or its owners.
Harry Rubin is a partner and chair of the Technology and IP Transactions Group at Kramer Levin Naftalis & Frankel LLP in New York. He is a globally recognized authority on IP-driven transactions and the development and implementation of global IP protection, monetization and commercialization strategies.
Karolina Ebel is an associate in the Technology and IP Transactions Group at Kramer Levin Naftalis & Frankel LLP in New York. She works on corporate matters, assisting in the representation of clients in mergers, acquisitions and other strategic transactions, financing transactions, and corporate governance issues.