SECs and other opponents of RTE, including publishers and journalists, saw RtE from a very different perspective. They argued that RtE would censor information, thereby restricting freedom of expression and affecting the overall quality of the internet by limiting the amount of online information available to users (see, for example, Tavani 2018). Based on these and other related concerns, they concluded that the public would be harmed by RtE. However, some proponents of RtE, including Bottis (2014, 2), have suggested that the privacy protection associated with RtE is necessary precisely because it would help repair the harms – namely the “psychological harm” (and perhaps even physical in some cases) – that users might otherwise suffer. Consider, for example, that victims of “revenge porn” websites would have explicit recourse regarding their requests to remove/de-index links linking their names to these controversial sites (see the report on revenge porn against the right to be forgotten, now RtE, in Kritikos 2018). We recognize that many of the ethical concerns raised in this review related to online recruitment. The use of the Internet to retain or regain subscribers was not as widespread, or at least not widely reported. This may be due to limitations in our search strategy, as articles may not explicitly refer to retention and tracking in their title, abstract, or keywords. Alternatively, this could reflect the smaller number of longitudinal studies (requiring retention and traceability) compared to cross-sectional studies, or that using the internet in this way is a new method that is not yet widely used. Nevertheless, it is surprising that so few retention and tracing studies have been identified, as longitudinal research has identified it as difficult to maintain research cohorts [137, 138]. It is not clear whether the Internet is not being used for this purpose because researchers and/or ethics boards consider it problematic and unethical, or because they are not familiar with such methods. Most articles reporting online retention used email to maintain contact with participants, which was not considered ethical.

Other online retention strategies using social media have been reported, such as: Researchers who “befriend” participants on Facebook [139, 140]. While these strategies may be effective in keeping adolescents engaged in research, they raise ethical questions regarding participant confidentiality, the relationship between participant and researcher, and perhaps access to information beyond what participants originally consented to. Most of the tracing studies identified in this review used social media to search for participants who were lost to follow-up, with many authors commenting on ethical and methodological considerations. Researchers interested in following participants via social media in the future will benefit from viewing these articles and other relevant documents [25, 141]. Given its effectiveness, tracking participants via social media is likely to become increasingly popular among researchers, but the lack of ethical guidance for this method is worrying. What is IoT and how does it raise ethical security concerns affecting search engines? IoT generally refers to the networking of “things” on the Internet; These objects (connected and “smart”) may include webcams, printers and other devices that are not easily identifiable, locatable or accessible to users via traditional search engines (see, for example, Burgess 2018). Thus, users might assume that since information about these objects would not be accessible via typical internet searches, the objects would be fairly safe from network hackers. Many users might also be excited about the conveniences they could enjoy if their “smart” objects were connected to each other.

Imagine, for example, a “smart home” where these objects could communicate both with each other and with the owner`s mobile devices (and possibly directly with connected devices and apps built into the owner`s car). On the one hand, we can imagine that the owner of this house would be happy if her “smart” refrigerator communicated with her during the return trip to make her aware that the milk production of the refrigerator is low. The refrigerator can also communicate directly with the owner`s smartphone to trigger an app that displays the lowest prices for milk in nearby grocery stores. Thus, ordinary users could easily enjoy the kind of amenities made possible by their interconnected “smart” objects. On the other hand, however, it is likely that these objects do not have an adequate level of network security. Thus, users could be vulnerable to hacking of their objects, and personal information about them (in relation to their interactions with these objects) could be obtained by unauthorized individuals and organizations if search engines were actually able to discover and locate their (not sufficiently secure) objects and devices. The grey literature review identified several universities, primarily in the United States and the United Kingdom, that offer publicly available guidelines on the ethical use of the Internet in research. The content of higher education policies varied considerably and included detailed guidelines (e.g.

[96, 109]), general procedures and guidelines for data recruitment, collection and storage (e.g. [104]) and resources that repeat the AoIR guidelines (e.g. [102]). Guidelines developed by universities were often aimed at researchers and were developed by ethics and review boards to provide step-by-step procedural guidance to support ethics review processes. The research also identified a guideline created by a teaching hospital [108] that includes ethical recommendations and protocols for social media-based research, including online tracking. While some guidelines specifically addressed ethical issues for online research with children, most did not. Two U.S. university policies reaffirmed the Children`s Online Privacy and Protection Act (COPPA), a federal law that requires verifiable parental consent when researchers collect personal information from children under the age of 13.

In that case, the courts were more rational with respect to the division of liability between search engines in relation to copyright issues. The debate revolved mainly around whether search engines are media or publishers. In 2009, Metropolitan International Schools Limited filed a defamation suit against Designtechnica Corporation, Google UK Limited and Google Inc, which sought to distinguish between search engines and other internet-based companies and set a precedent for future search engine cases. Another recurring issue is requests to ban certain word searches and remove certain content from a search result. Search engine operators cannot prohibit users from using certain search terms or remove search results for those search terms. Existing legal regimes for content and hosting providers are generally proposed to clarify the limits of search engine liability, as the liability of these internet actors is clearer. One visionary who saw the need for a new type of organization and retrieval system to handle the growing volume of information was Vannevar Bush, perhaps the most important figure in the history of information retrieval theory and search engines in the pre-Internet era. In his classic article “As We May Think” (Atlantic Monthly, July 1945), published about six months before ENIAC`s official announcement, Bush noted: The first step in addressing these legal concerns would be to recognize that search engine operators are not responsible for the content that appears in search results. This implies a parallel understanding that search engine operators such as Google cannot be held liable for illegal content disseminated by third parties on the internet while that content is still online.

On the other hand, search engine operators may run the risk of being held liable for illegal content that has already been removed by content providers, but still appears in a certain search engine result. According to various legal practices, this may require search engine providers to take the necessary technical and legal precautions to remove this content from their search results when that content is not online. Second, legal definitions that clearly define and set limits on the types of providers that are held liable for content on the Internet are necessary to ensure that the wording of the law is consistent with the spirit of the law. The separation of search engine operators from content and hosting providers is an opportunity for emerging markets to harmonize with courts around the world to recognize that search engines are an “intermediary” of information on the Internet. As a research tool, the Internet offers alternative strategies for reaching participants where they “live” and represents an apparently effective and cost-effective solution to the challenges associated with traditional methods of participant engagement [20].