Wikipedia is a fascinating corner of the web—a font of knowledge that leads to expected places. But as any teacher or professor will tell you, it’s not a primary source. Use it as a jumping-off point, but scroll to the bottom and seek out original sources for the “truth.”
Is that fair? Is Wikipedia indeed a repository for half-truths? It’s a topic that Professor Amy Bruckman from the Georgia Institute of Technology’s School of Interactive Computing has researched extensively and examines in her book Should You Believe Wikipedia?, out in 2022 from Cambridge University Press.
Her conclusions may surprise you. Ahead of a September keynote at IntelliSys 2021, we spoke to Professor Bruckman, a Harvard grad who holds a PhD from the MIT Media Lab, about how to test assumptions—and the definition of truth and existence—in an era of misinformation.
Before we get to Wikipedia, your wider research focuses on the field of “social computing,” which includes ethics, research, content creation and moderation, plus social movements. When did you first encounter web-based communities?[AB] Around 1990, I was a grad student at the MIT Media Lab and my friend Mike Travers showed me a model of MIT in a multi-user, text-based virtual world. He had programmed a bot of his advisor, Marvin Minsky. Virtual Marvin would automatically start off in his office in the Media Lab, walk across campus to a classroom, and deliver a lecture at the correct time Tuesdays and Thursdays, reading a chapter of his book, Society of MindSociety of Mind. It was magic. I was hooked.
And was that when you built your first multiplayer real-time world? [AB] Yes, that was when I built MediaMOO, a multi-user text-based world designed to be a professional community for media researchers. Then my dissertation project was a virtual world for kids called MOOSE Crossing, where kids built the world together and learned object-oriented programming and practiced their creative writing.
Many people have fond memories of using MOOSE Crossing as kids. In fact, there was something on NPRon NPR about it last year. But these were early days in collaborative computing. What were you running MOOSE Crossing and MediaMOO on? [AB] Well, this was before the invention of the web, and we were using computers running the UNIX operating system. The internet wasn’t yet a mass medium, but we could see that it would be, and the potential was exciting.
Which brings us to Wikipedia. Many of us consult it, slightly wary of its bias, depth, and accuracy. But, as you’ll be sharing in your speech at Intellisysyour speech at Intellisys, the content actually ends up being surprisingly reliable. How does that happen?[AB] The answer to “should you believe Wikipedia?” isn’t simple. In my book I argue that the content of a popular Wikipedia page is actually the most reliable form of information ever created. Think about it—a peer-reviewed journal article is reviewed by three experts (who may or may not actually check every detail), and then is set in stone. The contents of a popular Wikipedia page might be reviewed by thousands of people. If something changes, it is updated. Those people have varying levels of expertise, but if they support their work with reliable citations, the results are solid. On the other hand, a less popular Wikipedia page might not be reliable at all.
Because few people access that page, or know/care enough about the subject to correct/challenge them? Which brings us to the big ideas behind what is truth, and how we reach it. [AB] In my book and my talk at Intellisys, I try to teach everyone a bit of basic epistemology, and show how that helps us better understand the internet. I believe ideas like virtue epistemology can help us to improve the quality of the internet going forwards.
Okay, virtue epistemology is definitely a big idea. Give us a working definition, and how it applies to Wikipedia. [AB] Virtue epistemology suggests that knowledge is a collaborative achievement, and we all can work to achieve knowledge (justified, true belief) by aspiring to epistemic virtues: “curiosity, intellectual autonomy, intellectual humility, attentiveness, intellectual carefulness, intellectual thoroughness, open-mindedness, intellectual courage and intellectual tenacity.” Being someone who is careful with knowledge is a lifelong quest, and trying to embody those virtues helps.
So if someone embodies those virtues, we expect them to be in pursuit of noble truth. But how do we know what is true? [AB] The real world exists, but is only knowable through our fallible senses. But that doesn’t mean that reality is subjective. Am I sitting on a chair? You see it with your senses and I with mine, but we agree that there is something called a “chair,” and I am sitting on one. The high degree of correlation between my subjective perceptions and your subjective perceptions is caused by the fact that the world exists—there’s really a chair. The more people agree on something, the more we can be sure of it. And the more those people possess what we would call “reliable cognitive processes,” the more we can be sure of it. So let’s pick a harder example than my chair: Is human activity changing the climate? We know the answer is yes because a large number of people with reliable cognitive processes agree. Truth exists independent of the knower, but social consensus is our best way of figuring out what that truth is.
Most of us exist inside a bubble of similarly minded folks, which shores up our confirmation bias. Can you explain that concept with regard to Wikipedia too?[AB] I’m not actually a climate scientist. I know that human activity is changing the climate because I have chosen sources I trust. And I interact with a community of people (in person and online) who share my views. When everyone around me believes that human activity is changing the climate, it’s easier for me to decide that it’s worth extra money to buy a car with a hybrid engine. I live in a bubble of like-minded folks. That’s good most of the time. I don’t have to go get a degree in climatology before I go car shopping. But there are growing numbers of bubbles of people who share false beliefs, and reinforce those beliefs in one another. That’s a problem for the internet in general.
What happens with false beliefs on Wikipedia? [AB] Maybe the biggest surprise of the internet to me is that false bubbles generally are corrected on Wikipedia. Even if you pick a controversial topic like climate change or vaccination, the Wikipedia page typically reflects mainstream scientific consensus. Something about Wikipedia is working better than the rest of the internet, and I think we can learn from it as a positive model.
Can metadata help? [AB] Metadata is critical to the future of the internet. We all need help deciding what to believe. It would help a lot if information came with a reliability rating. But there’s nothing easy about creating those ratings. We need both a method of judging what is reliable, and a financial model to pay for the process of creating those labels.
If Wikipedia is a good example of mass peer review, then it can also incorporate testimony to establish a baseline of truth, right? For example, I was invited by the USC Shoah Foundation, which was founded by Steven Spielberg, to see their recording of Holocaust testimoniestheir recording of Holocaust testimonies for future generations. But eyewitness accounts often don’t pass into ‘truth’ or are considered too subjective on Wikipedia. Is that when we have to urge people to look to wider sources? [AB] You need an intermediate layer—interpretation of primary sources by a Holocaust scholar. That’s the difference between a work of history and an encyclopedia. The job of a historian is to synthesize primary sources and form an interpretation. The job of an encyclopedia is to summarize work by historians and give you a list of links to go read if you want to learn more.
Wikipedia also asks us to educate ourselves, and then share that knowledge, as subject matter experts. Can you talk about your personal non-academic experience here?[AB] Editing Wikipedia can be a lot of fun. I used to help with the page on trash cans. The group of people working on the page had a long conversation about the words “bin” versus “can” and how the name for a waste receptacle varies around the world. There’s a mini-golf course near my parents’ house, and the trash can is shaped like a dolphin. I added a picture of it to the article, and the next time I visited I showed the owner that his trash can was famous. The photo was there for a decade or so. Someone has since taken it down, sadly. But I enjoy contributing to something meaningful. Sometimes even things less silly than trash cans.
Finally, and this is going pretty deep, is any of this true? That’s where, as you’ve pointed out, metaphysics comes in, and where we get to look up at the sky and wonder ‘am I dreaming this life?’ Discuss.[AB] One thing you learn quickly when you hang out with epistemologists is that truth exists. How we agree what that truth is can be tricky. But the affordances of internet technology are a surprisingly good fit for how knowledge is constructed. I think we can learn a lot from success stories like Wikipedia. And maybe apply those lessons to the design of the rest of the internet.
