| dc.description.abstract | Many of us go to the Internet to gather information relevant to our questions. Yet, much of the information on the Internet is irrelevant for answering them. How, then, can the Internet be a source of relevant information? How can it become a better source of relevant information? My thesis explores two questions: What explains successful and unsuccessful information gathering online? And how can information gathering online best be improved? I distinguish between two positions: epistemic individualism and epistemic structuralism. Epistemic individualist accounts explain (un)successful online information gathering primarily in terms of Internet users, such as their open-mindedness (or lack thereof). I call this “epistemic individualism about explanation.” Relatedly, epistemic individualist accounts can also maintain that improvements to information gathering online should focus on Internet users—e.g., making them more open-minded. I call this “epistemic individualism about amelioration.” However, structural factors beyond users arguably impact information gathering online—e.g., how biased a tool is towards providing irrelevant information. I call accounts primarily emphasizing structural factors in explanations and improvements “epistemic structuralism about explanation” and “epistemic structuralism about amelioration,” respectively. The thesis assesses whether epistemic individualism or epistemic structuralism provide the best explanations and suggestions for improving information gathering online. In Chapter 1, I take an account by Richard Heersmink as a starting point. Interpreting it as an epistemic individualist account, I take the main claim to be that a user’s good character traits are crucial for successfully using search engines to gather information online. However, drawing upon work by C. Thi Nguyen, I argue echo chambers can make members with good character traits engage in bad information gathering practices by manipulating them to direct their trust badly. I suggest the epistemic individualist account should respond by adding a virtue of well-directed trust. The response’s success rests on a condition: well-directed trust is most plausibly cashed out in terms of an individualist virtue. In Chapter 2, I argue this condition fails because search engine bias exerts a crucial causal influence on how well a user directs their trust online. The failure leads me to conclude that epistemic structuralism is the most plausible account about explanation. But it is not necessarily the most plausible account about amelioration. One might focus primarily on improving users for pragmatic reasons. Yet, assuming epistemic structuralism about amelioration, I argue in Chapter 3 that reorganizing and relabeling information in an online environment might be more important improvement for information gathering than fixing flawed information gathering tools. | eng |