This book puts forward a methodological approach to researching the Internet. It starts from the premise of the Internet as an object: itself the subject of study – not what it reports but what it is; similar to the study of the book, as distinct from the study of literature. Its aim is to establish the “methods for the medium”, by which the author means doing research using the inherent methods of the individual Internet technologies.
The chapters follow the various principles put forward by Rogers. The first principle is to think like the objects and learn what web scraping, page crawling and page ranking can tell us directly. Understanding the methods is the way to understand the medium.
The second principle is to understand the nature of digital objects: Twitter has tags, Facebook has likes, web pages have hyperlinks. Digital methods use these and their like as research tools in themselves. Instead of doing a survey of web page influence, a digital method way is to analyse the net of hyperlinks for information about influence. Each technology offers its own digital method.
The third principle is to build upon the devices as a research tool itself. A list of Google results is not merely a ranking; it is also a reflection of the weights of the underlying search algorithm and of the values of the domain being searched. Comparing different results for the same search terms in different countries offers a unique way of sensing the cultural values of each country. One can also use the objects in different ways. The collection of Facebook friends and their interests associated with one person will be different from those of some other person, and this too is a valid research result.
The fourth principle is to accept digital methods as valid without reference to other research methods. Internet research is often seen as needing to be triangulated with data from the real world, but today the Internet is the real world. The example given is Google Flu Trends, which maps flu outbreaks by analysing the number of queries about flu from different places. The more queries from a given town, the more likely it is that its inhabitants have flu. This data are always checked against “real” data from medical sources. But examining the incidence of queries for Thanksgiving recipes from different places shows which places eat what at Thanksgiving. This principle says that this finding does not need verification: the Internet is the data source.
Each chapter presents one online technology and shows how it can be exploited as a research tool conforming to the four principles. The examples are wide ranging, surprising and well chosen. The presentation is clear and uses colour diagrams to get its points across. Overall the book is an excellent guide to how to adapt your research to the new media and how to look at familiar objects with new eyes. Properly used, “we look at Google results and see society, instead of Google”.
