Truth And The World Of Wikipedia Gatekeepers
NEAL CONAN, HOST:
This is TALK OF THE NATION. I'm Neal Conan in Washington. Wikipedia is both ubiquitous and irreplaceable, the go-to source for quick information on almost every topic imaginable. The online encyclopedia is written and edited by volunteers. Anybody can send in a new entry or update an old one, except sometimes they can't.
Case in point: Professor Timothy Messer-Kruse, perhaps the world's foremost expert on Chicago's Haymarket riot and the trials that followed, Wikipedia repeatedly rejected his repeated efforts to remove information he knew to be wrong. We'll find out why in just a moment.
If you've tried to edit Wikipedia, what's been your experience? Tell us your story, 800-989-8255. Email us, firstname.lastname@example.org. You can also join the conversation on our website. That's at npr.org. Click on TALK OF THE NATION.
Later in the program, "Undefeated," the next film in our series on Oscar-nominated feature documentaries, about a high school football team from Memphis. But first, Timothy Messer-Kruse joins us, a professor at Bowling Green State University, the author of "The Trial of the Haymarket Anarchists." He wrote about his experience with Wikipedia in the Chronicle of Higher Education, and he joins us from a studio in Perrysville, Ohio, and nice to have you with us today.
TIMOTHY MESSER-KRUSE: Thank you, Neal.
CONAN: So you have primary sources that contradict Wikipedia's account of that trial. You enter a change on the website. So what happens?
MESSER-KRUSE: Well, I tried to change what I thought was the most glaring inaccuracy in the page on the Haymarket. The page described the actual Haymarket bombing. It described the eight-hour movement leading to it. It described the trial that came from that event.
And in that article, the description of the trial began, saying the prosecution did not offer evidence connecting any of the defendants with the bombing. Well, my research has all been about showing what exactly went on in the trial, and there was an overwhelming amount of evidence. Now maybe it's not evidence that we today would find worthy of convicting these men and sending them to the gallows, but there was undoubtedly multiple kinds of evidence.
There was 118 witnesses called to testify, many of them involved in the anarchist movement themselves. There was forensic, chemical evidence. There was even some embarrassed admissions on the part of some of the defendants. So I thought that description in particular needed to be changed.
And I tried to simply delete that reference, and when I did so, within minutes, that page was restored, and I was instructed by whoever this volunteer editor was about some of Wikipedia's ongoing policies that prevented my making these changes.
CONAN: And you tried it again, and basically what they said was they don't rely on primary sources like transcripts of the trial but rather on the preponderance of secondary sources.
MESSER-KRUSE: That's right. So I was told that I needed to come up with some published sources that supported my point of view. Simply referencing the coroner's records or the trial transcripts or other sources that I'd uncovered was not sufficient.
So I actually bided my time. I knew that my own published book would be coming out in 2011. So I tried again and was told that I needed to represent a majority viewpoint, not a minority viewpoint, namely my own, and that Wikipedia was about verifiability, not necessarily about truth.
And if my account may have been truthful, the majority view still has to be represented on Wikipedia's website because it needs to be verifiable, it needs to represent what is the majority opinion.
CONAN: Well, let's bring another voice into the conversation, Andrew Lih, author of "The Wikipedia Revolution: How a Bunch of Nobodies Created the World's Great Encyclopedia." He's also an associate professor at the USC Annenberg School for Communication and Journalism, and he joins us from a studio at the University of Southern California. Andrew, nice to have you with us.
ANDREW LIH: Good to be with you.
CONAN: And, well, it sounds - that sounds ridiculous.
LIH: Well, it does, you know, I'm not unsympathetic to the professor's plight, and unfortunately I think a lot of new editors to Wikipedia do run into a lot of the issues that he talked about. And that's simply because now Wikipedia is such a popular resource that there are lots of volunteers putting their time in there just to make sure that information is accurate and there isn't vandalism all over the place. So certainly I think he ran into some of that culture of volunteers being maybe a little aggressive in enforcing things.
But I think this is a great situation which just falls in that perfect storm of problems that Wikipedia has a little bit of weakness on, which is what do you do with recent scholarship that isn't quite majority but is certainly cutting-edge and certainly valid and certainly legitimate.
So everything that he said about Wikipedia not encouraging original research, in other words citing secondary sources and not primary sources, is absolutely true. But I think probably the most instructive thing is to look at the - kind of the blurb for the professor's book, which I think has a lot of interesting keywords that tip us off to why this is so problematic and so tough to put into Wikipedia.
So in the description of his book, it says: In this controversial and groundbreaking new history, Timothy Messer-Kruse rewrites the standard narrative of the most iconic event in American labor history, the Haymarket bombing and trial of 19886. Using thousands of pages of previously unexamined materials, he demonstrates that contrary to long-standing historical opinion, the trial was not a travesty of justice, as has been commonly depicted.
So you can see all these words are kind of against the majority view. You know, it was published in August, 2011. It probably needs more time to steep in the community for it to bubble up to be the majority view, once a lot of scholars and historians look at it and accept that as the canon.
So I think we are in a very unique place in that this is recent scholarship, certainly solid scholarship. He's one of the experts in the world. And it just happens to not be, you know, steeped in the community long enough for it to be valid as a majority view in Wikipedia.
CONAN: So Timothy Messer-Kruse, you have to wait until enough of your rivals cite your work for it to become the preponderance of opinion.
MESSER-KRUSE: Well, and I wonder if citing is sufficient here in this case because the majority or minority opinion, the undo-weight policy, as they describe it, I mean, I think it could be enforced in a way that really blocks new scholarship over a longer period of time.
And this strikes me as one of the big differences between the Wikipedia culture and the academic culture. I mean, in the academic culture, we have very specific gate keeping. We have credentialing. We have professional associations. We have professional conferences. We have peer-reviewed journals. We have all this apparatus to make sure that only experts are really part of the discussion.
But that world, while being somewhat and admittedly exclusive, also allows for opinion and for research to change on a dime. I mean, if I went to a conference, and I have, and if I can convince the audience before me that my sources are more credible, and my interpretation is more accurate, they will leave that conference with my viewpoint.
CONAN: And that is, Andrew Lih, not the case in Wikipedia world.
LIH: Well, it's ironic that we're saying here that Wikipedia has to move slowly and deliberately when everyone kind of recognizes Wikipedia's value in the world as being so quick and so fast to, you know, document news events. So when Whitney Houston died, you had that, you know, updated right away in Wikipedia.
So in this case, you do have a situation where Wikipedia is not leading-edge but trailing-edge in terms of historical research. And it's part of the culture that is now on Wikipedia of trying to get it right rather than get it fast. And in cases of history, you know, this case is from the 1800s, and it's not being retried. They're just trying to be accurate.
CONAN: They're trying to be accurate, but they're saying we have these old secondary sources, as opposed to these new primary sources. Why don't you look at the primary sources and say on, those old secondary sources were clearly wrong, we'll update?
LIH: Well, I think what happens is if you look at the policy, there has been a real culture of not going to primary sources before - for the reason that Professor Messer-Kruse said, is that Wikipedians and editors are not experts in these fields. So I'm not sure you really would want average people who edit Wikipedia to pretend to be the person who can examine those documents in detail. We should be depending on professors like Messer-Kruse to actually examine them.
We just don't have enough secondary sources in February 2012. Maybe in April 2012, when enough folks look over his content and suddenly start to revise everything out there, we will have that preponderance of evidence. But right now, we are in a very strange special case here where the scholarship has just come out, maybe has not had enough time to become the majority view in mainstream sources.
CONAN: It could - of course, Wikipedia also had the opportunity to be leading-edge when he tried to correct it a couple of years ago, but we'll let that rest for just a moment. We want to hear from callers in our audience who have tried to edit Wikipedia. What's been your experience? Give us a call, 800-989-8255. Email email@example.com. We'll start with Bob(ph), and Bob's with us on the line from Fairbanks.
BOB: Hi, I had an interesting experience with Wikipedia about two years ago. I grew up in Montana, and I'm a geologist by training. I found an entry talking about geologic formations in Montana that was in error. I attempted to correct that Wikipedia entry about six times with substantiation of those rock formations by the Montana Geological Survey.
And my experience was the author of that Wikipedia article was very unaccepting of anything other than his original article. It took me roughly six times over about a month to get that change made.
CONAN: And are you sure it was the author of the original article who was disputing your correction, or was it an editor somewhere?
BOB: No, it was the author, and the author rejected my attempts to correct the article. That was obvious in the editing part of that article on Wikipedia, that he was the one rejecting my input.
CONAN: Andrew Lih, in cases like Bob's, is there a court of appeals you can go to and say wait a minute, I think I'm being treated unfairly here?
LIH: Yeah, if it really gets down to it, there is an arbitration committee, and, you know, folks in Wikipedia's community really hope it doesn't get to that point. But those are for really extreme cases where there's multiple parties who just cannot agree, and there's disruption within the community because of that.
But unfortunately, you know, all too common, a lot of editors start to see this phenomenon of Wikipedia now having 3.8 million articles in the English language edition. That's a lot of articles. A lot of people feel that it's pretty darn good the way it is. And unfortunately, a lot of the policies and a lot of the community members who act in editing are working in kind of a defensive mode and trying to prevent bad things from going in rather than encouraging good things from being there.
So that's really something that the Wikimedia Foundation, which oversees the project, really is trying to combat, is to try to be more welcoming of newcomers because they know this is a problem.
CONAN: Bob, was the entry eventually changed?
BOB: It was. I finally communicated with the original author, and he did change it. But I tell you, it really colored my opinion of the authenticity and accuracy of the Wikipedia articles after that experience.
CONAN: Thanks very much for the phone call, appreciate it. We're talking about editing Wikipedia. If you've tried to, how'd it go? 800-989-8255. Email firstname.lastname@example.org. It's the TALK OF THE NATION from NPR News.
(SOUNDBITE OF MUSIC)
CONAN: This is TALK OF THE NATION from NPR News. I'm Neal Conan. When Wikipedia first launched, it seemed like a wild idea: Let the world write an encyclopedia. It was enough to send purists running to their shelves of World Books and Britannicas, but the experiments worked for the most part, and in 2009, founder Jimmy Wales came on this program to talk about the editing process, which they have oft refined.
I think we should make Wikipedia as good as it can possibly be, he said, but also warned users should recognize it has both strengths and weaknesses. Well, if you've tried to edit entries on Wikipedia, what happened? 800-989-8255 is our phone number. The email is email@example.com. You can also join the conversation on our website. That's at npr.org. Click on TALK OF THE NATION.
Timothy Messer-Kruse is with us. He's an expert on the Haymarket riots, which he ran into a problem trying to edit the Wikipedia entry on the trials that followed. Andrew Lih is also with us. He's the author of "The Wikipedia Revolution: How a Bunch of Nobodies Created the World's Great Encyclopedia." And let's see if we can get another caller in on the conversation. This is Mark(ph), and Mark's with us from Tulsa.
MARK: Hello, I feel like I won the lottery. I got on.
CONAN: Oh, congratulations.
MARK: Yeah. I had - now several things about Wikipedia. One of the things is, I see frequently on things that I look up that it's asking for more sources, and it's asking for more submissions on what it is that they have. That said, I have - and I have used Wikipedia not as it, but I have used their sources that they quote and gone in and put them in a reading list on a research paper I was doing if they had something informative.
CONAN: An interesting point, Mark. Go ahead.
MARK: The one time I did submit something, I found an error, and it was something reasonably trivial, but I found an error, and there - do you remember an actor named Ty Hardin?
CONAN: I sure do, yes.
MARK: PT-109. OK, "Bronco Lane" from the - OK, I was reading his biography, and they misquoted and said that he was - that he played Sugarfoot and not...
CONAN: Will Hutchins, everybody knows Will Hutchins was Sugarfoot.
MARK: Well, that's - yeah, everybody our age, OK. So at any rate - also, oddly enough, how I thought to do this was I found the same mistake in the Smithsonian Institution in the 50-year history of television, and I got them to change it. But I submitted the deal, and my source, the source they used was a news release from ABC that who knows when, and it might have been a typo. But I used his - I went to his website.
He has a website, and I decided, just like you do a citation, you know, with APA rules, and I sent it into them and changed the one line in a paragraph, and I mean it wasn't a week, they changed it.
CONAN: And so everything went pretty smoothly, then?
MARK: Yeah, but - now that was probably seven, eight years ago. I don't know if today - and I also don't know if something very scholarly like these gentlemen are talking about, I can't honestly say how they might react there. But every time I've used one of their sources in a paper where I needed a good reading list, I needed a good book cited list on what I put together in the paper, I never found anything that I considered to be misinterpreted or not quoted properly.
And I'm talking science papers, you know, like on statins and that kind of stuff.
CONAN: Mark, thanks very much, and I think that's probably most people's experience. But I wonder, Timothy Messer-Kruse, is your book now among those cited on the reading list on the Haymarket trials?
MESSER-KRUSE: It is now, and in fact there's been quite a few revisions made since my article in the Chronicle of Higher Ed came out a couple weeks ago. There's been a lot of activity and a lot of behind-the-scenes discussion. And I think I've kicked loose a lot of information that wasn't there before.
I'm not sure that everybody has that ability to sort of, you know, lobby for changes outside of Wikipedia and have Wikipedia then respond to that lobbying, but there have been a lot of changes made just in the last couple weeks.
CONAN: Let's bring Andrew - let's bring Steven Walling into the conversation, community organizer at the Wikipedia Foundation, the nonprofit that supports Wikipedia - excuse me, it's the Wikimedia Foundation that supports Wikimedia - I'll get it right, the Wikimedia Foundation that supports Wikipedia. He's a longtime editor himself, has made thousands of edits and started hundreds of articles. He joins us from Youth Radio in Oakland, California. And nice to have you with us today.
STEVEN WALLING: Thank you very much.
CONAN: And is it fair to say that Timothy Messer-Kruse's article did cause some people to rethink?
WALLING: Yeah, absolutely. I mean, Timothy talked about sort of the process of credentialing and peer review that, you know, certifies expertise among academics. And what both the callers and his experience have pointed out, I think, is that while Wikipedia doesn't have a body of experts in every single subject writing every single article, what we have is a public, transparent discussion and revision history of what's going on and what the arguments are and what sources should be included.
And as you can see, that often leads to a successful outcome; not always. Sometimes in cases where we have breaking news like this, it can be difficult to work out over time, but I think it's overall a successful process.
CONAN: So successful process, yet a policy that goes for the preponderance rather than the truth, is that inaccurate, and is that acceptable?
WALLING: So I would say the most interesting thing about Wikipedia, other than the fact that it's a collaborative process for creating the content, is that it's a collaborative process for creating the rules. So one of the interesting results of the argument here is not just that people are talking about the Haymarket article, but people are talking about the line in the policy that affected his revisions, which led him to believe that he couldn't be a part of the conversation.
So everything in Wikipedia is open to revision and is open to a collaborative discussion among peers about what's right and how we get it right.
CONAN: Here's an email we have from Daniel: What's the standard here? Wikipedia is an encyclopedia, not the cutting-edge journal of American labor history. Would Encyclopedia Britannica have responded any more quickly? I want Wiki to be as accurate as possible, and caution seems to be merited. I wonder, Timothy Messer-Kruse.
MESSER-KRUSE: Well, I've actually been commissioned to write several encyclopedia articles in the last few years that incorporated some of my research. I've actually been publishing in this field, in this area, since 2003, my first article came out. So it's not as if the very first information is just in the last year.
And one of the things about academia is that, you know, academic encyclopedias operate by identifying experts and asking them to contribute what they understand to be the cutting-edge state of the art. So in some ways, sort of the old hide-bound academic encyclopedias can get ahead of something like Wikipedia.
CONAN: Let's go next to Deborah(ph), Deborah with us from Berkeley.
DEBORAH: Hi, I'm wondering about a comment about the transparency of this editing process to the regular user. I'm a scholar, and I use Wikipedia a lot, and I'm wondering if this either-or conversation is part of the problem here. And it would be very helpful to have a revision, such as this new information, as part of the entry so that it doesn't have to necessarily replace what's there right away, until there is a preponderance of agreement about it, but could become part of the information and with the explanation that this is a new, you know, breakthrough research.
CONAN: Steven Walling, might that be possible?
WALLING: Absolutely. It's actually a really core part of our policy on neutrality, that when we get such a serious debate in the sources, that it's clearly up for grabs that what Wikipedia should do as a neutral source of information is actually talk about the debate rather than trying to decide what the fact is...
DEBORAH: Right, so was that a problem with the editor not understanding the policy in this case?
MESSER-KRUSE: Well, in my case, there were several editors I dealt with. I could tell by their handles, I guess it is. And I received a very similar response from all of them.
DEBORAH: So maybe that policy is not being implemented quite effectively. It concerns me as a scholar because I'm hearing about some new, important research that if I was able to go to the conference I would hear, but if I'm not in that situation, then I may not have that information if people are not including it as part of part of the debate and the discussion on Wikipedia, which would be sort of mimicking the conference and the peer review and conversation process of the scholarship.
CONAN: Which leaves you in the situation of unknowingly regurgitating incorrect information.
DEBORAH: Well maybe, but if it's put out as a position by certain - you know, one point of view...
CONAN: Oh no, I'm not talking about your revision but going back and being unaware that this new research is available. Hey, you're saying, well, that's what everybody says, maybe that's what happened, there was no evidence put forward in this trial. Deborah, thanks very much for the call.
DEBORAH: Thank you, and it's an interesting conversation.
CONAN: Let's see if we can go next - this is an email that we have from Catherine(ph) in Ann Arbor Public Schools: A few years ago, I tried to enter an article about a relatively obscure documentary film and its Canadian author-director. I was told it could not be published because essentially it was not famous enough to be included in Wikipedia.
The irony is that I wrote the article because I saw the film and wanted to know more about who made the film but couldn't find that information without a fair amount of digging. I wished there had been a Wikipedia about this film, so I made one.
I was slapped down. I found the experience to be really discouraging, and I've never tried to contribute again. And, Andrew Lih, I think that's happened to more than a few people.
LIH: Yeah. That's one of the main things in Wikipedia called notability. It's something notable enough to be in the pantheon of these 3 million, 4 million articles. And that's a real problem, and that's something that when they did poll of folks who tried to edit Wikipedia but were rebuffed, you know, the Wikimedia Foundation is trying to measure what the audience feels and what their experiences are like. And they've been very serious about this in the last few years, to fund these types of studies, because they know that ever since 2007, the number of editors who are active on Wikipedia has steadily declined. And it's flattened out, but it still cause for some worry because if you still want a healthy-growing Wikipedia, you still need contributors to be there.
CONAN: You need the contributors. And you need various types of contributors. There had been criticisms in the past that the people who edit and write for Wikipedia are overwhelmingly white and male and a little bit older than the nation's demographic.
LIH: Well, the numbers are pretty startling, even for folks who kind of knew it was a problem. I don't know if you want to guess how many - what percentage of editors were active are male?
CONAN: I would guess something along the lines of 75 percent.
LIH: Ninety-one percent.
LIH: It was even more than most people thought, so we're talking about 9 percent female editors in terms of being active in Wikipedia. That's a huge imbalance. That's a huge monoculture. That's a problem. So one of the big directives and one of the big things that they've tried to do recently is engage the galleries, libraries, archives, museums sector, you know, really go for folks who really already have a mission that's in synch with what Wikipedia has in terms of bringing knowledge to the world. That's been very successful. I was just at a conference last week in D.C. at the National Archives, which focused on this exact thing, on how to galvanize more galleries and libraries and museums to work with Wikipedia. And there's already been some great examples.
Your caller brought up the Smithsonian Institution. They've been a big fan of working with Wikipedia. The National Archives' head archivist of the United States is a huge fan of Wikipedia, as well. So as many problems as we might have in terms of getting users to edit Wikipedia, part of this is also extending what we're calling kind of an invitation culture, going out there and actively engaging museums and folks who have a volunteer community to work with these folks to improve Wikipedia and improve museums and libraries at the same time.
CONAN: Andrew Lih, author of "The Wikipedia Revolution: How a Bunch of Nobodies Created the World's Greatest Encyclopedia." Also with us is Timothy Messer-Kruse, professor at Bowling Green State University and author of the "Undue Weight of Truth in Wikipedia." And with us as well is Steven Walling, community organizer for the Wikimedia Foundation. You're listening to TALK OF THE NATION from NPR News.
And, Steven Walling, going out and finding new people to enter this process and make the process friendlier, that sounds like that's your job, right?
WALLING: Yeah, partially. There's a team of both volunteers and staff that work together on this in a variety of ways at the Wikimedia Foundation. So we focus a lot on the developing world, places like India and Brazil where people are coming online really rapidly. We focus on mobile - so how do people contribute from their mobile devices. And then I hooked in to focus on sort of policy and how do we level the playing field for people who want to be a part of the discussion on Wikipedia.
CONAN: Let's get another caller in. This is Zach. And Zach, with us from Charlotte.
ZACH: Hey. Well, back when Wikipedia very first started, I was just actually describing it to my brother, this new thing called Wikipedia. And he was an undergraduate as an archaeologist and he looked over an archeology article and said it's pretty good, but doesn't say anything about, you know, X. So I was like, well, why don't you add an article about it? So we took a paper he had written as an undergrad, loaded it up on Wikipedia pretty much. And years later, we're talking about it, he's like, I wonder what ever happened to that thing? That was so terrible. I was an undergrad, didn't know what I was talking about. And we looked - we checked on Wikipedia and he just did all the little changes and additions people had made over time. It was great. He was amazed, you know, just a few little turns of phrases and (unintelligible) article at. That's just so much better just from the crowd helping out.
CONAN: So crowd editing and crowd writing improved an undergraduate paper to an A-plus paper maybe.
CONAN: Well, Zach, thanks very much for the call. And, Andrew Lih, I think that's the way it's supposed to work.
LIH: Yeah. It's important to remember that on balance, Wikipedia has much more in terms of success stories than fails. And it's kind of ironic, if you look at the origin story of Wikipedia, it started off as a project called nupedia.com that actually required a pedigree. You had to prove you had a higher degree in order to edit nupedia.com. It was only after one year and only 12 articles to show for it that they realized this kind of credentialed editing system just wasn't working, and they opened up the world to edit this encyclopedia and they called in Wikipedia. And that's what we see today; is that the masses do a much better job, in most cases. But you still need some situations to bring in experts and to really polish off certain articles. And I think that's what the community is learning now, kind of a hybrid approach is what's needed.
CONAN: Timothy Messer-Kruse, I wanted to go back to you. Given what your - you've experienced and what you've learned about this process, what might you suggest as an improvement?
MESSER-KRUSE: Well, I think one thing is to make new contributors more aware of sort of the Wikipedia culture because I think one of the obstacles I ran into was that I was too easily deterred from trying to persist and make these changes, although I, you know, I try it a dozen times over two years. I sort of gave up after I was scolded and told to look at the civility policy at one point. At one point, I was branded a vandal for trying to change a page after someone had changed it back. And I kind of slunk away. And in the last week, I've been reading some of the comments to my article and some people have been suggesting that I was not persistent enough. So it seems like a catch-22. Either you persist and resist against these policies and accusations, or you don't. In academia, of course, if I submit an article to an editor and I get it returned to me and rejected, I don't then call up the editor and yell at them and insist that it be published. I just go somewhere else. So there's that difference in culture, I think, that maybe many academics like myself would find an obstacle to really contributing.
CONAN: Thanks very much for your article. It really opened an interesting area of discussion. Thanks very much for your time as well.
MESSER-KRUSE: Well, thank you for having me.
CONAN: Timothy Messer-Kruse, the author of "The Trial of the Haymarket Anarchists: Terrorism and Justice in the Gilded Age," which is, well, groundbreaking research on the trial that followed the Haymarket riots. Andrew Lih, thank you for your time today as well.
LIH: You're welcome.
CONAN: Andrew Lih is the author of "The Wikipedia Revolution: How a Bunch of Nobodies Created the World's Greatest Encyclopedia." Steven Walling, community organizer for the Wikimedia Foundation, joined us as well. Thank you for your time.
WALLING: Thank you very much.
CONAN: Coming up, we'll talk with the directors of the fourth film in our Oscar docs series for this year. Dan Lindsay and T.J. Martin from "Undefeated" join us after a short break. Stay with us. I'm Neal Conan. It's the TALK OF THE NATION from NPR News.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.