by Raizel Liebler
Digital Confidential: The Secrets of Studying Behavior Online (MIT Press 2015), edited by Eszter Hargittai and Christian Sandvig, is a very different — and unfortunately unique — guide to conducting research online, ranging from looking for supposedly static materials to engaging with research subjects (read: actual, real, breathing people). So what makes this book so different? It engages with the actual messy difficulties of doing research, rather than the usual — reporting of cleaned and tidy results.
As someone who engages in online research and is dedicated to the importance of digital archiving, I am so pleased that this book exists. Online research studies is difficult, in part because the ground continues to shift, constantly.
All of the chapters are well-written and intriguing, but these are some of the most important insights, plus my responses:
Megan Sapnar Ankerson’s Read/Write the Digital Archive:
- As good as the Internet Archive is, there are significant limitations, including “depth, consistency, and accuracy”, with few sites completely saved — and clicking through lots of broken links is extremely frustrating. (38) And there isn’t really a good way to see a snapshot of a website on a particular day.
- Content created in propitiatory forms, like Flash, are much more likely to disappear than plain text.
- Websites sometimes flicker on and off, without knowing when a researcher will need to keep their own copy: “I was pleased that I had used my screen-recording software to create videos of these sites when I did because by the time I started work on my dissertation a year later, many were already gone.” (44)
- For-profit companies aren’t interested in archiving older versions of software.
Ankerson’s insight is similar to what I’ve found in my own research, so it’s great to have a citable source that demonstrates why doing research about websites (or similar) that have fluidity in their access and information is so difficult — and why it is so important to lock that down for researchers.
Virág Molnár & Aron Hsiao’s Flash Mobs and the Social Life of Public Spaces:
- Searching for visual items rather than text is super difficult in online research.
- Compiling longitudinal data at any point, but especially time series data from the internet retrospectively is very difficult. “Hence, the importance of gathering data about online phenomena as they occur” (74)
How are researchers who usually get funding for something that is intriguing retrospectively supposed to appropriately plan (and get funding to study) something that no one may know is on the horizon? This is especially true for studying Twitter and other in-the-moment social media services — or in response to major events.
Eric Gilbert & Karrie Karahalios’ Social Software as Social Science:
- Through creating an app for Twitter, the anonymity standard for computer science peer-review was broken. “Everyone who qualified to review my paper already knew who wrote it and probably had a personal experience with the software–they had tested it on their own personal networks.” (150)
- Extreme sudden popularity for a new service, like this app, causes crashes and annoys users. But outlaying cash, can be a choice of tech success over paying for food. Read all of pages 151 and 152 for some real talk: “My back-of-the-envelope calculations suggested that buying enough capacity to handle the South Korea spike would have cost more than our rent and grocery bill combined. … I was lucky enough to be funded on a fellowship, but that also meant no grant money to spend on cloud computing. … I wish I had budgeted for this time.”
Structural changes need to occur to help create new types of cutting edge research, including funding them properly — and understanding that traditional norms of “hiding the birdie” until it appears for publication need to be changed, if new researchers are going to take risks with their research.
A final note, only touched on in this book: the law. Copyright, not only as law, but as practice, make it nigh impossible for present and future researchers to effectively research online culture. As Ankerson notes:
“I was reminded by several sources that the copyright circumstances were either unknown or beyond their control. I could like at these files for my own research, but few felt they had the authority to grant official rights to share these publicly, even in cases when the original production studio folded or merged.” (50)
Summary: Highly recommended for researchers trying to figure out the ethical and practical difficulties of conducting online research. May be even more useful as a guide to those that fund or supervise those doing online research to explain why it is so much more complicated than just pressing that button!