The research process is in continual flux, even that of humanities scholars. The move toward broad digitization of materials has made many materials more accessible from more locations as well as connected various archives in new ways. Yet, historians often raise concern about what is gained and what is lost through digitization. One area that has sparked intense debate is the impact of digital environments on serendipity, a core experience in historical research. The present talk draws on a series of studies examining the changing nature of serendipity and examines existing similarities and differences between physical information environments like the library stacks and digital information environments like Twitter. How does the heavy reliance on digital sources affect serendipity and what opportunities exist for designing digital tools that support innovation, creativity, and resource discovery. While it is important to realize that the research process is a crooked path, it is also important to be aware of external constraints, including design features and affordances of digital environments, an increase in information sources, and time pressures. We argue that making serendipitous connections is a skill in which historians can be, and should be, trained in. We further argue that the role of physical and digital environments in serendipity is changing and future history pedagogies need to adapt.