The Fine Print of Terms of Use
The United States Court of Appeals for the Ninth Circuit (in California) recently held that the scraping of data from public websites without any prior consent is lawful, even if it contradicts the provisions of a website’s terms of use. This is an innovative decision that deviates from or even contradicts past decisions by the same court regarding similar issues.
The act of scraping data refers to the process of collecting content that appears on websites, even without the owner’s consent, using algorithms designed to identify information on websites by categories defined in advance.
Collection is conducted using software that collects and sorts through large quantities of data from public online sources. These tools are used to index and present search results by search engines, by price comparison sites, and many more, whether for commercial, academic, or journalistic use.
However, scraping is frequently listed among the acts prohibited by websites’ terms of use. Oftentimes, site owners are also protected by additional legal, non-contracts-based doctrines, such as copyright or computer laws, which they can rely on to enforce this prohibition.
The Ninth Circuit decision was specifically concerned with the interpretation of a US computer law, the Computer Fraud and Abuse Act (CFAA), enacted in 1986. The court examined whether the scraping conducted by hiQ from the public profiles of LinkedIn users constituted an act of hacking that violated the CFAA.
In this context, it seems the court distinguished between publically available information and information protected by passwords or other barriers by the websites where scraping was performed. This distinction was the basis for the difference is this ruling than in previous ones, as the password-protected areas were entered through technological means or by expanding the access granted by the password holder, and the access was to public information that under copyright law may possibly not be owned by LinkedIn at all.
The issue of scraping raises complex questions of copyrights as well, as it is not always clear who the owner of the content is, especially if the service offered by the website is a platform service and users are provided with the possibility of uploading content and commenting. Thus, when a website is a platform, the ownership of the platform, the platform’s look and feel, the organization of the data, the display mechanism, and the manner of public accessibility are owned by the platform.
However, the content itself belongs to the creators, i.e. those uploading it to the platform. In this context, websites may argue that had they not existed, such data would not have become public and accessible. Still, the question remains: Is information publicly displayed by a platform, such as LinkedIn, and its content owned by LinkedIn? And is LinkedIn permitted through its terms of use to prevent the scraping of such information?
In this matter, it seems the court favors freedom of competition and information, and has determined the act of scraping data in this case to be lawful.
It remains to be seen whether this decision will be further challenged, as well as how the decision will affect players in the market. While this decision is good news for startup companies wishing to challenge media giants, but who require the information found on their websites, it also burdens website owners, as it casts doubt on their ability to rely on the contractual provisions such as the terms of use to prevent actions that compromise their economic interests.
***
Source: barlaw.co.il