What Should We Do About Wikipedia?

Feature by Martha Nichols and Lorraine Berry

A Call for Information Activism



“Wikipedia Is Not Dying”: Jimmy Wales at Wikimania 2011 in Haifa, Israel  © Niccolò Caranti

Using Wikipedia is lusciously seductive. It’s convenient. It’s fast. Just whip out a smartphone, and you’ve got a universe of facts at your fingertips.

This nonprofit venture—“the free encyclopedia that anyone can edit,” Wikipedia’s home page declares—is barely a decade old. Yet, it’s now available in almost 300 languages. More revolutionary still, its open approach represents a new attitude toward knowledge: that it should be available to all, that information is ever evolving.

As writers and teachers, we’d like to embrace this collective vision. We aren’t exactly fans of the gatekeeping done by mainstream media. But the flap this April over certain Wikipedia editors removing female authors from its “American Novelists” page revealed this Internet behemoth’s serious downside.

A brief recap: Amanda Filipacchi, author of three novels, happened to notice that women’s names were vanishing from the page. In two New York Times op-eds, she reported that authors such as Amy Tan, Louisa May Alcott, and Harper Lee had been moved to a new “American Women Novelists” subcategory. Filipacchi wryly added:

I also noticed that Edwidge Danticat was plucked from ‘Haitian Novelists’ and dumped into ‘Haitian Women Novelists.’ So it seems, at least, that women from different countries are treated the same. It’s just too bad they’re not treated the same as men.

Even cynical magazine editors like us, well versed in the VIDA count and other bad news for female writers, were stunned by such disregard for gender bias.

How do we mobilize against an eight-headed monster that keeps ducking responsibility for unreliable information amassed by volunteers? That’s the quandary feminists and other Wikipedia critics now face. It’s not possible to boycott a ubiquitous cyber enterprise, is it? That’s a scary thought, whether you think a boycott is righteous or stupid.

By this point, many outraged commentators have weighed in about the female novelists scandal. Many journalists keep lauding Wikipedia’s idealism, anyway. Few have challenged the core premises of Wikipedia, a site that’s been plagued since its founding by a pattern of gender, racial, ethnic, and political bias.

At TW, we asked other writers to comment on a problem that’s almost as hard to pin down as the one with no name. We excerpt a number of their responses in this article and conclude with suggestions for what Wikipedia can do to right its veering course.

Some writers we contacted told us they hadn’t heard about female novelists being moved to a Wikipedia subcategory. Predictably, they sputtered at the news. But others—like Jodi Picoult, author of many bestselling novels—were well aware of this Wikipedia incident. Picoult connects it to the general bias against female authors in the literary world:

To move all the American female novelists to a subcategory is to create a list of American novelists that is all male—which is misleading and reflects a second-class status for the female author in America. What if the average student doesn't think to search for ‘American Female Novelist’ and instead just puts in ‘American Novelist’ in his search bar? His results would be sorely compromised.

Wikipedia is far from evil. Most of its thousands of contributors have good intentions, and many would probably call themselves information activists.

But when it comes to a global reference work that’s routinely used by schoolchildren, activists should not only strive to make knowledge freely available. They should fight hard to represent all voices—even if the truth is a messy, partial thing.


Wikipedia: The Buck Stops Nowhere

As Filipacchi discovered, taking on Wikipedia is like putting a big fat bull’s-eye on your forehead. Her own page was vandalized by unnamed editors, who cast aspersions on her work, her credentials, and her right to bring the site’s shortcomings into the public eye.

Who’s in charge there, anyway? Our first thought was that the Wikimedia Foundation, which operates Wikipedia and raises money for it, must provide oversight, especially during PR disasters. But according to Jay Walsh, Wikimedia’s communications director, the Foundation’s leaders have no control over content.

The encyclopedia’s volunteer editors—or “Wikipedians”—make all the editorial decisions. The Foundation estimates that there are about 80,000 active editors (those who make at least five edits a week) for all the Wikipedias around the world. It is true that anyone can click the “Edit” button at the top of a Wikipedia page, even if they’re not logged in. But most Wikipedians register for login accounts (often under pseudonyms) that allow them to start new encyclopedia pages or categories, create their own user pages, and vote in Wikimedia Board elections, among other benefits.

“There is no ‘editor in chief’ at the Foundation, nor are there staff who approve or review content on our projects,” Walsh told us by email. “With the exception of removing a very small number of articles or posts that might violate copyright or potentially harm someone, we do not remove or alter content. That’s all in the hands of our volunteers.”

Sue Gardner; photo by Victoria Wills, from Wikimedia Commons, CC BY SAWikipedians, of course, argue about many things, often in long discussion logs (or “Talk” pages) that are archived on the site. After a heated “Category: American women novelists” debate, Filipacchi’s page was restored. Wikipedia editors have moved female authors back to Wikipedia’s main American Novelists page and added “American Men Novelists” as a subcategory.

“I completely understand why Filipacchi was outraged,” says Foundation Executive Director Sue Gardner in a recent Wikimedia blog post. A self-described “feminist Wikipedian,” Gardner also points out that “[s]ince 2011, Wikipedia has officially discouraged the creation of gender-specific subcategories.”

Yet, her post stops far short of an apology. She doesn’t take responsibility, as executive director, for such policy being undermined by other Wikipedians. She and the Foundation stand at a distance, claiming that “Wikipedia belongs to its readers”:

It’s not always the case, but in this instance the system worked. Filipacchi saw something on Wikipedia that she thought was wrong. She drew attention to it. Now it’s being discussed and fixed. That’s how Wikipedia works.

The answer to bad speech is more speech. Many eyes make all bugs shallow.

Unfortunately, most of Wikipedia’s hundreds of millions of readers aren’t obsessive Wikipedians. For those willing to dig, the revisions to Filipacchi’s page can be tracked. But the whole point of Wikipedia for most of us—students, teachers, couples arguing about who wrote which novel, or anyone else who isn’t dedicated to doing hours of research—is to find the information fast.

Worse yet, those who saw the Filipacchi page while the malicious editing was going on would later find something else and perhaps wonder if their memories were deceiving them. That’s more than a little creepy. It’s Orwellian. Facts are here today and gone in seconds—or constantly open to interpretation, revision, mashing, deliberate trashing, or simply deletion from the public record.

“This should be our call to action,” says Julia Scheeres, author of the 2005 bestselling memoir Jesus Land and A Thousand Lives, a 2011 account of Jonestown:

Julia Scheeres
Anyone can edit Wikipedia pages. I corrected errors on my own page, for example. If there were more women writers editing Wikipedia, we could correct errors that relegate us to second-class status.

Gardner, too, urges feminists to get on Wikipedia and “fix it,” as she notes in her Wikimedia post. Since her 2007 arrival at the Wikimedia Foundation, she has made increasing the number of female editors a priority. The April 2013 version of the “Wikimedia Movement Strategic Plan Summary” lists the following as a “2015 target”: Support healthy diversity in the editing community by doubling the percentage of female editors to 25 percent.

Nonetheless, the response Amanda Filipacchi got from some Wikipedians was intimidating. If the immediate reaction to whistleblowing is to be denigrated by those who clearly have no interest in dissent—who instead swarm the naysayer, wasp-like—what is the likelihood that a hypothetical new army of women warriors will be brave enough to risk editing Wikipedia pages?

The site’s “About” page does discuss Wikipedia’s editorial oversight and management, noting that editors can become administrators (if approved by fellow Wikipedians), which gives them power to delete articles and block accounts. There’s also an Arbitration Committee of elected Wikipedians for handling disputes. But as the editors of this and other pages acknowledge, achieving consensus can be very slow.

By the way: Despite the brouhaha over female novelists, Wikipedia still includes a separate “List of Female Poets” but no equivalent page for male poets (at least as of this writing). It seems the “bad speech” continues, unless somebody kicks up a fuss.


Constantly Revising History—and Wikipedia

There’s good reason not to trust an information source that can erase its own mistakes as if they never happened. At least newspapers and magazines have to publish retractions—and stories are written under bylines—but at Wikipedia, mistakes are revised away by editors who often don’t use their real names.

Take the revisions to Wikipedia’s own story. Since its inception in 2001, it’s had plenty of critics—including cofounder Larry Sanger. His discomfort is clear in “Deconstructing Wikipedia,” a 2010 piece by Chris Lydgate in the Reed College alumni magazine. (Sanger is a Reed alum, as is Martha Nichols.) Sanger argues in that article:

Larry Sanger"An encyclopedia entry is not just a collection of facts. It’s the ability to construct a narrative of the subject, the ability to describe things in a way that does not supply a misleading implication. You have to have lived with a topic for a while to do that. This is the sort of thing that separates the real expert from the ersatz expert.

Not only did early Wikipedians turn deaf ears to his pleas for verification and peer review of articles, Sanger found himself deleted from the official story. His cofounder was the colorful, quotable Jimmy Wales, Wikipedia’s longtime entrepreneurial front man. (Wales still sits on the Wikimedia Foundation board.) At the time the feature about Sanger appeared in Reed Magazine, however, Wikipedia referred to Wales as the site’s sole founder. As of this writing, Sanger is once more listed as a cofounder on various Wikipedia pages, but many news stories about Wikipedia still only mention Wales.

The Wikipedian community has tightened up rules for entries and for the editorial verification given to articles about well-known people. Many of these changes have come under Gardner’s watch at the Wikimedia Foundation, although she announced this March that she’ll be stepping down soon.

Still, the continual revision of entries results in history being retold—and retold again. Wikipedia’s open-source approach also attracts trolls and sock puppets. For years, such vandals have inserted deliberate misinformation into pages about politicians, corporations, celebrities, and just about any other contested topic.

Last week, as we were completing this article, a Salon expose by Andrew Leonard reported that the Wikipedian known as Qworty has been systematically “revenge editing” authors’ Wikipedia pages (including Filipacchi’s). It turns out that Qworty is the writer Robert Clark Young. Among other spiteful edits of Southern author Barry Hannah’s Wikipedia page, Young changed Hannah’s cause of death from “natural causes” to “alcoholism”—possibly because his own work was criticized by Hannah at the 2001 Sewanee Writers’ Conference.

The Qworty fiasco came to Leonard’s attention, he writes, when members of Wikipediocracy, a site that details instances of Wikipedian fakery and bias, contacted him. Qworty isn’t a typical Wikipedian volunteer, Leonard notes, but, he adds grimly, “just as surely, there are others like him”:

We just don’t know. Nobody knows. Nobody watches everything that happens on Wikipedia; nobody can watch everything that happens. But Qworty’s example tells us that even when people call attention to a rogue editor, even when that editor’s temper tantrums come to the attention of the founder of Wikipedia, it’s quite possible that no action will be taken.

Repeat something often enough, and it becomes a “fact.” That’s the trouble with media of all kinds, whether it’s hard news or a supposedly neutral encyclopedia entry. And now that information rockets around the world in milliseconds on the Web, misinformation can be even harder to rectify.

Consider the “lone wolf” explanation that some Wikipedians bruited about after Filipacchi’s first NYT op-ed appeared and that was picked up by a number of media commentators. Even Salon’s Leonard promoted the idea that the removal of female novelists from the main page was the work of one editor.

Meanwhile, Filipacchi, in a followup Atlantic piece, documents that at least seven Wikipedia editors participated in the culling of the novelist herd.

For Wikipedians, it’s an article of faith that the crowd will get closer to the truth than standard top-down editing. They insist that their constant negotiations will expose errors and vandalism. Yet, prejudices such as sexism and racism are rarely spread by a single vandal. They flourish in packs, where it’s easy for individuals to remain anonymous, and a gang mentality eggs everyone on to new lows.


Why Wikipedia's Neutral POV Isn't Neutral

Wikipedia is far from alone in generating errors or bias in the media. As journalist William Wong, author of the 2001 book Yellow Journalist: Dispatches from Asian America, says of Wikipedia’s handling of female novelists:

William WongI find gender, racial, ethnic, and cultural hierarchies fascinating, whether in literature, politics, economics, academics, entertainment, sports…. I see that Amy Tan makes it onto the women's list. Is that an advance over her getting subcategorized as an ‘Asian American woman novelist’?

Plenty of other reference books, including the venerable Encyclopedia Britannica, have been riddled with sexist and racist assumptions—often in terms of the people they don’t include under various topics or in biographical entries.

One of Wikipedia’s oft-stated “five pillars” (or “fundamental principles”) is that “Wikipedia is written from a neutral point of view”:

We strive for articles that document and explain the major points of view, giving due weight with respect to their prominence in an impartial tone. We avoid advocacy and we characterize information and issues rather than debate them.

That’s a laudable goal for any media organization. But during the female novelists’ fracas, in which Filipacchi and others reported that author names were dragged back and forth on pages like skipping checkers, this pillar went by the wayside.

More troubling still, clinging to the neutral point of view (or NPOV) can obscure biases by sounding impartial when it really isn’t. Just as journalists have long argued over whether the objective news voice is really objective, Wikipedia’s emphasis on the neutral point of view in an encyclopedia entry may muddle the actual truth far more than a bylined article in which a writer’s biases are clearly stated.

In trumpeting the NPOV, there’s an unspoken assumption that writing from specific perspectives, especially minority viewpoints, can never be neutral. Kirsten Greenidge, an Obie-winning playwright, frames the conflict she feels this way:

Kirsten GreenidgeWhat I struggle with is the idea that I am very proud to be a female writer who is also a black writer. I wish the qualifier ‘woman’ or ‘black’ did not imply to the rest of the world that my work is lesser because of those associations, but far too often it does or can. Until this is no longer the case, it's far more equitable to refer to all those who write as writers, period.

In her Wikimedia blog post, Gardner argues that Wikipedia is just reflecting “the cultural biases and attitudes of the general society.” Yet here, the Foundation’s executive director sidesteps the need for advocacy when biases infiltrate the very roots of social institutions. Jim Crow laws didn’t change because a bunch of rational people sat down and discussed the problem. That required a social movement in which many put their lives—and their real names—on the line.

Grace Hwang Lynch, the News and Politics editor at BlogHer, says that moving women to a separate category reflects not only gender bias but also the way society in general interprets “American.” As she points out:

Grace Hwang LynchGranted, organization and architecture are factors for a website the size of Wikipedia, but the troubling part is not just Wikipedia…. Let's not ignore gender, but we should acknowledge it without creating this false idea of the ‘neutral’ American, which inevitably ends up meaning white and male.

Perhaps the most pernicious thing about crowdsourced information is the way it reinforces the biases that already exist. Because female writers are underrepresented in the pages of major magazines and literary journals, for instance, their work doesn’t garner as many book reviews or academic studies. So if Wikipedia editors rely on citations to construct lists of “the most important” novelists, then women's absence from literary culture is reflected in Wikipedia.


The Trouble With Wikipedian Categories

Joan Wallach Scott's 1986 essay, "Gender: A Useful Category of Historical Analysis," underscored decades ago that anything coded “masculine”—soldiers, scientists, plumbers, writers—has more power than words coded “feminine.” The opening to her article could be addressing Wikipedia today:

Joan W. ScottThose who would codify the meanings of words fight a losing battle, for words, like the ideas and things they are meant to signify, have a history. Neither Oxford dons nor the Académie Française have been entirely able to stem the tide, to capture and fix meanings free of the play of human invention and imagination.

In the discussion over what to do about the “American Women Novelists” page, many Wikipedia editors stuck to their guns about the need to clean up the original unwieldy alphabetical list of 3,000-plus authors. Shiftchange’s comment is typical: “The use of categories is not done in a normative or prescriptive manner, only for the purpose of simplifying navigation.”

But social reality is messy. It really is a moving target that can’t be quantified. Richard Zimler, author of the 2010 novel The Warsaw Anagrams and other books, calls such categorization “very haphazard and sloppy.” He told us that his most recent check of Wikipedia had him on the “American Men Novelists” page but not on “American Novelists.” He was categorized as a “Jewish American Writer” but not as a “Jewish Novelist”:

Richard ZimlerI find this sort of labeling absurd. Does the fact that Sidney Sheldon and Carson McCullers are both ‘American novelists’ mean that they have a great deal in common? ...How about Danielle Steel and William Faulkner? Or Anne Rice and Philip Roth? I suppose that grouping these writers together is supposed to help readers, researchers, and librarians find common ground in their books or lives, but it doesn't make any sense to me. It strikes me as specious.

In their lively discussion about categorizing novelists, some Wikipedians noted the need to group writers by genre rather than gender. Others were irritated by slapping national labels on writers. And the consensus was to “merge” the men and women again on the same American Novelists main page.

But the problem with the way basic categories of information are presented online goes much deeper than issues of genre or gender.

“Any writer who is not white and male has to deal with labels attached to their writing,” says Vietnamese American writer Aimee Phan, author of the 2012 novel The Reeducation of Cherry Truong. Whether it’s reviews or Wikipedia entries, she notes, the response to a minority writer’s work “has never been in our control”:

Aimee PhanAs writers, we're supposed to forget and transcend these categorizations. But it's hard to consistently ignore the reality that many, many more people will be reading your work, or about your work, through these lenses. If only our writing could speak for itself; but instead, it's what others are saying about our writing that gets more attention—and sometimes determines who will read our work.

The first question any information activist needs to ask, then, is why authors, artists, businesspeople, or plumbers have to be categorized by gender or race at all. "If it were up to me, the only categories that would include me would be ‘Novelist’ and ‘Writer,'" says Zimler, an American who lives in Portugal. "And I suspect that most writers would be happy being part of those two categories.”


Fix Wikipedia: No More Anonymous Editors

When we asked Joan Wallach Scott for a comment about what she thought of the Wikipedia incident with female novelists, she told us she was “appalled” that Amanda Filipacchi’s page had been vandalized. Scott adds:

I find that practice to be contrary to any notion of openness in the production of knowledge…. [N]o knowledge is free of subjective input, however well-masked the subjective input or how carefully it is presented as ‘truth.’

Technology itself is bias free, but not the people who use it. To date, high-tech users are still predominantly male, white, and childless (that describes 80 percent of Wikipedia’s editors, estimates one Forbes writer). The Wikimedia Foundation’s own strategic plan includes the following caveat: “We know that no one is free from bias.”

Nichole Bernier, author of the 2012 novel The Unfinished Work of Elizabeth D., reminded us that as recently as 2011, VS Naipaul said, “I read a piece of writing and within a paragraph or two I know whether it is by a woman or not. I think [it is] unequal to me.” Here’s Bernier’s take on why such sexism continues to dog the literary world:

Anyone who is inclined to think that way is helped along by a million small clues, starting with predictable things like the choice of cover (what self-respecting man is going to pick up a book with a woman spinning in a field in a swirl of taffeta?). And now apparently they can check [in Wikipedia] whether the writer is a novelist or a novelesse.

(Bernier has more to say about the ”Naipaul Test” in a recent post in Beyond the Margins.)

It’s time to hold Wikipedia’s editors, leaders, and backers responsible for continuing to scatter those small clues. Citing March 2013 stats, Walsh of Wikimedia calls the online encyclopedia and “its sister properties…the fifth most visited web property on the Internet.” (Wikipedia’s main page claims it’s currently “the sixth-most-popular website worldwide.”) Regardless, Wikipedia entries usually pop up at the top of Google searches.

As teachers and TW editors, we see increasing numbers of students and writers list a Wikipedia entry as a primary source without verifying the facts. Many writing and journalism instructors advise their classes to take anything gleaned from Wikipedia with a grain of salt. Even Wikipedia’s entry on “Researching with Wikipedia” says, “You should not use Wikipedia by itself for primary research (unless you are writing a paper about Wikipedia).” Still, far too many users pay no attention to such admonitions.

Here’s where a strict Wikipedia diet is in order, especially for students, writers, and anyone under the age of 50. Nibble sparingly; use it for informal overviews or as a pointer to primary references and statistics. Treat it as information fast food. It’s fine to use Wikipedia to win a bar bet, but it’s unacceptable as a research resource.

The stakes are high for generations to come. Writer and editor Kate H. Winter, author of the 2011 novel Lost Twain: A Novel of Hawai’i, says:

Kate H. Winter
Thinking teachers train the next generation to assess both the information and the source, to be critical thinkers. That will create a new generation that will subvert bias and rewrite the narrative. Sometimes a bad example is the best example.

Scott suggests that more female editors vet Wikipedia entries, which is in line with the Wikimedia Foundation’s goal of increasing the number of women who contribute to the site.

After noting that she’d just fixed a few small errors on her Wikipedia page, Scott also wonders what would happen if “lots of us” created our own entries about women, feminism, and gender studies. “If some of the omission is omission and not deliberate exclusion,” she says, “it could be rectified that way. And if the ‘boys’ reject the entries, then there will be lots of concrete things to expose.”

We’re all for keeping the pressure on in this fashion. But we call for more editorial leadership at the Wikimedia Foundation itself, despite Sue Gardner’s many good changes. Information activism is about more than reflecting and codifying cultural attitudes.

Sue Gardner at Wikimania 2011, Haifa, Israel; photo by Victoria Wills, from Wikimedia Commons, CC BY SAThe executive director of the Wikimedia Foundation has a bully pulpit. Gardner or her successor could certainly keep the public focus on eradicating gender and racial bias in Wikipedia. Why not? Reactions to that rhetorical question are bound to reveal more about high-tech culture—enough with the PC crap! women can fix their own pages, if they’ll bother to turn on a computer! biased? look at all the pages for female porn stars!—than Wikipedia’s idealism.

Some Wikipedians will bridle at the notion of top-down management of any sort. Indeed, foisting the accountability expected of for-profit media corporations on Wikipedia is neither realistic nor the right approach.

But true social change doesn’t happen without leadership. If you glance through Wikipedian discussion logs, you’ll find calls for leadership and vision there, too. Walsh notes that the Foundation funds projects such as “Wikipedia edit-a-thons, special workshops for students and those with unique education and experience, and efforts to promote our projects to new users.”

He confirmed that Gardner plans to give a keynote presentation at the annual Wikimania conference in Hong Kong this summer. A 2011 Fast Company profile of Gardner raises the possibility of mentorship programs for new Wikipedia editors. These are all opportunities for consciousness-raising among Wikipedians.

So, here’s a thought: As part of a welcome package whenever a new Wikipedian registers, email him or her pdfs of Scott’s 1986 essay about gender as a category; Peggy McIntosh’s 1989 article “White Privilege: Unpacking the Invisible Knapsack”; and Audre Lorde’s 1984 classic “The Master’s Tools Will Never Dismantle the Master’s House.”

Last but not least, the Wikipedian community could make two basic changes:

  • Require all Wikipedia editors to use their real names.

  • Don’t allow unregistered users to edit Wikipedia pages.

It may fly in the face of Internet norms. Sometimes it is politically dangerous to attach your name to controversial ideas, such as in writing about political movements in a repressive regime. But all too frequently, anonymity just allows online writers to be sloppy or craven, because they assume they’ll never get caught.

Writing under your own name rather than an avatar makes you publicly accountable in a way that need not be out of step with Wikipedia’s mission. And displaying the bylines of all who contributed to each Wikipedia page on that page—not as part of a separate “view history” link—will allow far more sunlight to shine on who’s making the decisions.

“Encyclopedia Genetica” (Pacific Science Center,  Seattle, Washington); photo by Ryan Somma of ideonexus.com


Publishing Information

Update May 21, 2013: Joan Scott's photo has been added beside her first quote.


© Hadley Langosy

Martha Nichols is editor in chief of Talking Writing. She's also a longtime contributing editor at the Women's Review of Books.

"If human beings were perfect, there would be no stories. Most narratives derive their tension from somebody making bad decisions or ending up in a scary place. But how can you overcome something if there’s nothing to overcome?" — "The Devil Made Me Write It" 


Lorraine BerryLorraine Berry is an associate editor at Talking Writing. On Twitter, you’ll find her @BerryFLW.

"Pity the female writer. Not only is she less likely to get reviewed in major magazines and short-listed for prizes, she often finds herself on the receiving end of interview questions that would leave most of us mouthing three little letters: WTF?" — "Women Writers and Bad Interviews" 



I think you have some good ideas, Obi, and they probably would go a way toward fixing some of the problems. I'm also pretty confident that if all the editors and writers contributing to the project had your ability to understand and explain things a lot of this would never have come up in the first place :)

Uhh, that was my experience. You really can't expect much better with anonymity as the rule. We don't need a study to prove it. Many people see anonymity as a license to be bad actors. Others have no intent to to do it, but lose their bearings when released from the internal strictures that the use of their names would impose. Having rules made by unknown people is just a bad way to go.

The comment about categorization being at the machine level rather than for end users interests me--it makes sense, and yet: Why did Filipacchi go to the American Novelists page and see no women under "A" and "B" on the main page? Perhaps she just happened to catch the page in flux, as it was being edited, but that's still a problem--and one of our core criticisms of Wikipedia: the constant revisions with no shoutout to users that something has changed or will be changing.

There's so much to discuss in all these comments. I'm just dropping in quickly now to approve those that were pending. In general, I would say that neither Lorraine nor I are calling for an end to Wikipedia. If we believed that, we would have written a very different kind of article. But we do think that WP has become so big that it now impacts the larger society of online users, many of whom have no idea that it's not a reliable information source. That's a big problem, from where I sit.

Question, UTCL, Obi, and other Wikipedians: When Lorraine and I do a followup of this piece, what do you think are the most pressing issues to focus on?

Interesting essay on a Wikipedian's user page:


I wouldn't perhaps endorse every word of it, but it is well worth reading. Among the highlights:

"... the level of intellectual aggression in Wikipedia due to the presence of anonymity is exceedingly high. Some of the most common and most disturbing forms of behaviour include angry, vengeful, overstimulated reactions to criticism, assaultive language and poor impulse control; good enough reasons to be wary. The attempts to prohibit trolling failed at the onset of Wikipedia likely because in an Internet world trolling is good for traffic, and traffic is the real life-blood of Wikipedia."

"The most damning part of open source format is that, by design, our goal-oriented community is forced to accept otherwise unacceptable revisionist viewpoints providing that they're verifiable. Partisan groups turn to Wikipedia to endorse their prejudices. Content disputes escalate. The socio-political coverage of countries, where adults do not easily access English Wikipedia, is left to young fanatics who perpetuate chauvinism. Controversial subjects are despoiled with opinionated agendas imposed by self-appointed wardens in contempt of policy guidelines. Scholarly literature is replaced with biased propaganda. Google books are intentionally obfuscated to avoid politically inconvenient facts. Many controversial articles contradict the opinions expressed in leading encyclopedias and quieten the viewpoints of rational thinkers."

You could have a hybrid, where people who edit BLPs *and* that aren't using declared/certified names are subject to greater scrutiny for their edits by some other trusted user group - sort of like autopatrolled, but more robust.

Most of the photos on Wikipedia are in the public domain or available via a Creative Commons license. There's usually plenty of info provided about the origin of a photo and how it should be credited. You can also track photos at Wikimedia Commons, which we do all the time at TW. I don't see a problem with the photos, unless they have been faked or misinterpret the text in some way. Is that what you mean, Amy?

While the mainstream media continued to report on the so-called "ghettoization" of women into gendered categories, they ignored several important points:
1) The "Women writers" category tree, which contains almost 400 sub-categories, was created in ~2006, initially as a way to highlight the contributions of women in this field.
2) The guidance around gendered sub-categories is that membership in such categories should not preclude membership in non-gendered alternatives - they should be non-diffusing. Thus, an African-American female poet should be in several categories, according to our current guidance: African-American women poets, American women poets, African-American poets, and American poets.
3) The existence of lists such as "List of female poets" (and not a male equivalent) is likely based on the perceived notability of those lists. Wikipedia reflects the outside world, and there are many published sources, courses, seminars, etc that look specifically at women poets - but the same volume of information does not exist for men poets. If you created a "list of men poets", it is possible it would be put up for deletion as a non-notable subject of study (though these days, that is starting to change). In the same way, we have a category for "Male feminists", but not an equivalent female category - perhaps because it is seen as more "defining" or worthy of note that someone is both a male and known as a feminist.
4) New categories can easily be created, but there is a process called CFD by which the community can decide to delete a category, due to it being a non-notable intersection. The "American women novelists" category, in spite of all of the scorn it received, was kept by consensus of the community by an overwhelming margin, including many female voters and first-time participants in the CFD process. However, rather than sticking around and helping to vote on other similar CFDs, 99% of those who participated have gone away, while bad categories still remain.

Your essay, while good, missed a key point - those editors who removed women from the American novelists category were in violation of the guidance developed by the community, and as soon as it was detected it was fixed. At the same time, the fix, which is having gendered and ethnic categories as "non-diffusing" (meaning, they remain members of the parent) is confusing to many editors - not because those editors are sexist, but because these cats behave differently than all other categories. I created a quiz, to ask editors to correctly categorize a bio in a "non-sexist", "non-racist" way - all 5 people who took the quiz failed - as you can correctly categorize someone in 32 categories, but if you miss #33 you have "ghettoized" them.

As a result, the ghettoization of American novelists is now fixed, but "ghettoization" across the whole tree is still endemic. There is now a gender bias task force attempting to address this issue at scale, but not many editors have signed up. Why don't you join?

There are many instances of true sexism at wikipedia, but I don't think it is the category system where you are likely to find it - it's really a bit of a red herring, and editors simply misunderstanding proper use of non-diffusing categories.

I never use Wikipedia for content. Evah. My husband is a librarian and also teaches writing at the university level and that's just a no-no. What this article made me think about, however, are the photographs which I use for clients all the time. Curious about what other folks think?

Excellent reporting, ladies. Very good post.

Fwiw, I realize many academics do today edit Wikipedia on their own nickel, and that is great. Just thinking outside the box!
Also, Andreas has leveled many other serious and cogent criticisms on things the wikipedia foundation could address, so this suggestion would be complementary to those.--~~~~

I know controversial topics (and people) would prompt the greatest fights. But the topics OBI lists would not ensnare (cause problems for) academics any more than non-academics. It would be interesting know if the alleged hostility toward academics (experts) disproportionately affects some disciplines more than others.

It is common to say that history is written by the winners. Who are the winners who are giving thousands of Americans "history" through Wikipedia? We know they are, in the main, young white males between 18 and 26. We don't even know their names. And they do, in my experience, often get belligerent when "anyone" tries to edit their version of American history. This is a problem that I don't think will be solved by having individual people try to fight their way through a system that seems bent on preserving a status quo.

you should know that a lot of the Wikipedia Forum folks are known for their vehemently pro-paid-editing stances. Some denizens of that forum take great pride in creating throwaway accounts in order to edit WP on behalf of their clients. I myself ran into a nest of such "sockpuppets" some months back, who were writing promotional copy about a particular law firm's lawyers, passing them off as far more notable than they were in obvious effort to draw in business. (WP is one of the most trafficked websites in the world, after all).

I mention this because i think it's related to Mr. Kolbe's comment about "second class citizens" above. There have been well-characterized incidents, most recently with the BP article, where paid editors or editors with conflicts of interest have rewritten, either directly or effectively, huge swaths of an article they had no business rewriting. It's a huge problem in medical articles, actually.

just to add to my last comment:

medical articles are what i know best, so that's where my analogies come from, but i think this example, of medical articles and medical information, really highlights why a real names policy will do no good.

in medical publishing over the last decade or so there have been innumerable scandals -- falsified data, ghostwriting, fake journals being put out by "legitimate" publishers for purposes of advertising, just really skeezy stuff. And this has all been done with people's real names and identities in the open, because there is so much money to be made by doing so. From my perspective, based on that example alone there would seem to be no indication that a policy that requires real names will provide enough accountability or opportunity for redress in the first place, let alone enough to outweigh the significant reduction in editing and participation that I expect one would see.

Now, a policy where conflict of interest issues take priority over outing/anonymity, i could get behind. If you're the subject of the article this will not be a problem for you. but maybe thats a bit too inside baseball for this particular forum.

Many thanks to all for the thoughtful comments and helpful resources. We look forward to continued discussion here.

As a technical note, comments are automatically put in the pending queue awaiting moderation when they have more than a certain number of links (not sure exactly how many) as a spam prevention measure. We're checking frequently to ensure posts get through. Andreas, I left your note above (#60) just so that people who might be periodically looking at the most recent comments will be sure to find your post on libel issues. In the context of this discussion, the links in all the posts are important to understanding the issues involved, and we do welcome them.

I did not know that Wikipedia had safe harbor status. I would imagine that if you are one of the folks who find themselves libeled by what is written there, one would feel powerless to change something. Is there an appeals process by which a piece of information can be appealed to be removed? Or is it left up to the person who has been libeled to go in and change the information? What happens if, as soon as one changes one's information, the person who is defaming you comes in and re-enters the original libelous information? At what point does someone higher up the food chain get involve in the process? Perhaps these questions have already been answered, but I'm curious as to what recourse someone has to change something that is written about oneself on Wikipedia?


There are mechanisms for dealing with that. They dont work all that well, but they exist, and could stand to be improved. I am well aware of how aggravating that must be, I've been through that particular wringer myself. I'd be happy to discuss further with you directly via email if that is of interest.

I think this goes back to the thing i said earlier about people wanting to work on the Product and not the Instructions -- many people simply give up after unproductive arguments on talk pages. It's easy to get discouraged.

Ok, here's an area: history of the colonization of the United States. Check out this article:

Now, it is decently sourced, informative, relatively neutral, etc. But the *whole* thing is written as a story of Europeans colonizing America. None of it is written as a story of indigenous peoples being invaded by these random white dudes with muskets and taking their land, and what happened to *them* as a result. Classic case of winners history. Here's how you could imagine a different telling of the story:

But, what role can and should wikipedia play here? The header of this article is "A call for information activism" - but should wikipedia be activist? Should it actively seek out and prioritize other narratives - e.g. narratives of indigenous peoples over europeans, or of women over men? What if 100 books reference the European history, and only one references the indigenous history? Should wikipedia have a sort of affirmative-action for sourcing, or take a normative point of view? Norms can be very tricky - for example, if there were norms established around the Israel/Palestine dispute article, as opposed to "neutrality" as the goal, it would change the whole dynamic.

While I agree that wikipedia reflects existing biases (it has to), due to its use of only published sources and its editor base, what practically can be done to overcome these biases? I can only think of two practical things: (a) changing the editor base and (b) asking academics to publish more works that tell new stories?

As I've already said elsewhere, this is a wonderful and thought-provoking article.

One idea I particularly like is the one of showing the article history (or at least a small part of it, like the most recent five edits) on the article page itself (say, in a field at the bottom). This would tell the reader whether the article version they're reading is five minutes old (because the article is undergoing an edit-war and is flip-flopping constantly), or whether it is a stable version that has remained unchanged for the past five months. This might even help draw in new editors, and attract more eyes to scrutinise the most recent edit.

As Obi says, biographies remain a significant problem. It is often said that things have changed since the famous Seigenthalter incident, but really, they haven't. The Qworty story and many other examples illustrate it, and the policy of allowing anonymous editing of biographies of living people is grossly irresponsible. Anyone from a disgruntled former lodger to a professional rival can edit your biography (see http://www.telegraph.co.uk/culture/art/8498981/Mayfair-art-dealer-Mark-W... or http://www.newstatesman.com/blogs/david-allen-green/2011/09/hari-rose-wi... or http://www.haaretz.com/opinion/in-a-web-of-lies-the-newspaper-must-live.... or http://www.time.com/time/specials/packages/article/0,28804,2042333_20423... ... and others). Moreover, when alarmed biography subjects turn up in Wikipedia and remove faulty or biased content, they are often reverted and hazed ( http://www.standard.co.uk/news/londoners-diary/rachel-johnson-in-the-gri... ).

This is all the more fatal as more and more biographies are created, while editor numbers have fallen. There are ever more articles to keep an eye on, and fewer eyes to go round.

A feature known as Flagged Revisions might have gone some way to address the problem (see http://wikipediocracy.com/2012/10/28/flagged-revisions-how-wikipedia-cou... ). Flagged Revisions allow anyone to edit an article, but edits by unknown users are only shown to logged-in users; the general public is not shown the edit until it has been approved by a "trusted" user. Flagged Revisions were implemented across the board in some Wikipedias, like the German and Polish ones. The feature was championed by Jimmy Wales, who announced several times in the press that implementation in the English Wikipedia was imminent, but in the end Flagged Revisions never found broad consensus in the English Wikipedia community (the feature is presently used in a few hundred articles only).

Another key problem, pointed out by Peter Cohen for example on Jimmy Wales' talk page ( http://en.wikipedia.org/w/index.php?title=User_talk:Jimbo_Wales&oldid=55... ), is that Wikipedia guidelines and policies are written and negotiated by Wikipedia editors among themselves. Other stakeholders, like the reading public and biography subjects, do not have a seat at the bargaining table. As a result, many of the guidelines and policies are primarily focused on meeting Wikipedians' own needs and desires – including their desire for anonymity. This enables editors like Qworty to flourish and remain undetected for years.

Wikipedians are an insular community with some rather narcissistic traits (well exemplified by Qworty's rants, as quoted in Andrew Leonard's first piece, http://www.salon.com/2013/04/29/wikipedias_shame/ ). The arrogance Qworty expressed in these outbursts is unfortunately quite characteristic of a good number of Wikipedians.

There is no social contract between Wikipedia and the wider world. This causes many problems, and not just in this area: another example is when Wikimedia Commons accepts anonymous uploads of adult images scraped from Flickr, without verification of (1) consent of the people depicted (sometimes in very compromising positions), or (2) copyright status. In many cases, the Flickr accounts vanish a short time later (Flickrwashing). This is one topic that was recently discussed at length on the Gendergap list ( http://lists.wikimedia.org/pipermail/gendergap/2013-May/date.html and http://lists.wikimedia.org/pipermail/gendergap/2013-April/date.html ).

The Wikipedia community has very skewed demographics. Close to 90 per cent of contributors are male. 67 per cent are single. 85 per cent are childless. The most common age, according to a United Nations University Survey carried out in 2010 ( http://wikipediasurvey.org/docs/Wikipedia_Overview_15March2010-FINAL.pdf ), is 18 (median age 22, mean age 26). It also has a well-deserved reputation as unwelcoming and socially inept. Kangaroo courts are a daily occurrence (examples can be found on this page, every day of the year: http://en.wikipedia.org/wiki/Wikipedia:Ani ). The social climate is often likened to Lord of the Flies (see comments on Wikipedia in this article published today: http://www.itworld.com/print/356932 ).

Some people who contribute to Wikipedia are very smart; but many others are not: they simply have lots of time. This can make participation in Wikipedia very tedious. As your article points out, Jimmy Wales has some responsibility for this state of affairs – to Wales, it was always more about keeping as many people as possible on board, while Larry Sanger was prepared to kick plainly unsuitable people out. This also underlies the lack of leadership on the gender gap issue, for example. The Wikimedia Foundation could make the kind of friendly space policy that is common in professional workplace environments part of its terms of use, but it does not do so, for fear of alienating the large number of contributors who are comfortable with a sexist online environment (see http://lists.wikimedia.org/pipermail/gendergap/2013-May/003616.html and other recent threads on that list, also see http://geekfeminism.wikia.com/wiki/Benevolent_dictator_incident and http://commons.wikimedia.org/wiki/Commons:Deletion_requests/Template:The... ). Instead, the Foundation tends to focus on PR measures and feel-good stories designed to cover up the site's worsening internal problems.

One thing I am certain of is that Wikipedia will not improve by playing out its internal dynamics. It is the underlying system which attracts a certain demographic and creates the social climate.

Outside criticism, as offered by this article, as well as Filipacchi's op-ed a few weeks ago and Andrew Leonard's phenomenal article in Salon, is key. Anyone wishing to improve Wikipedia should do so from the outside, if that is an avenue open to them. As long as the Wikimedia Foundation's priorities do not change, trying to work on the inside is a losing battle.

The customers need to make their voices heard.

That's a great angle, UTCL. I've tried to do something similar in my area, computer science. It's a tough sell for a lot of reasons, but what with MOOCs, the current NSF emphasis on STEM, it might be getting easier.

I have to say I was stunned by the ferocity with which my attempts at editing were met--not even about really controversial things. But things that are considered conventional wisdom in scholarship in my area. I thought, "Geez, is this really necessary?" It was also stunning to see that the most current works prominent in the field were nowhere in sight, while works superseded by decades worth of new material were presented as definitive.

Oh, an Obi. I'm sorry I misidentified your gender. But since I didn't mean anything by it in terms of the superiority of one sex or another, I just want you to know it was an honest mistake. I'm happy to have you--and everyone else--here.

I'm going way back in the comment thread for this. obi writes,

The difference between consensus on wikipedia and consensus in academia is consensus in academia is driven by expertise, past experience, and stature, whereas consensus in wikipedia is judged solely on the merit of the argument. Please note, all of the above is an idealistic description, and violations of all of that happen – but that is at least how it is *supposed* to work.

Actually, if we're talking about ideals, then past experience and stature don't matter in academia either. Expertise does, of course, but that's implied in being able to make a good argument. Double-blind peer review is standard for archival publications in my area.

Lorraine and Annette raise a related issue. Some kinds of secondary sources are typically not peer reviewed, especially popularizations of some scholarly area. It's an interesting question how quality can be maintained. Some of the controversies I've been reading about (thanks for the links, Andreas) are Kafkaesque arguments between non-expert Wikipedians and expert non-Wikipedians. Some of the boundaries on Wikipedia are becoming clearer to me.

I have tried to be involved, but gave up arguing with people wedded to outdated sources and resistant to changes in the historiography. The works of white male authors were favored over the works of women and blacks who were experts in the field. Another Wikipedian who had been writing in the area wrote to me out of the blue telling of her long term struggles with editors who just refused to acknowledge the enormous shifts in scholarly consensus on various matters. Things were cast in political/culture war terms.

You're all pointing to a fascinating question: Who are--or should be--the stakeholders in an enterprise like Wikipedia? I don't discount the need for academics getting involved, but the cultural shifts required are major, as Rob notes. Ditto for the Wikipedian community reframing its priorities. Big, big hurdles--but doesn't this matter enough to try?

I think a discussion of stakeholders is useful. If we grant that Wikipedia is an important source of information for everybody, then who has a vested interest in making that information reliable? I have some answers myself, based on my own biases. In essence, I'm asking who should be responsible for fixing Wikipedia, which is an extension of the question Lorraine and I started with.

Obi, I do think that WP should impose a form of affirmative action to counter the prevailing white male bias. That is what I mean by being an information activist, although I also believe that feminist stakeholders like me need to get involved.

But something has to change in the rhetoric of the Foundation (which is quite activist in some of the free-information causes it preaches) and in the presentation of WP to end users before I feel like anything I add will be more than wasted effort. Sue Gardner and the Foundation have been quite concerned about diversifying the "editing community" -- that is, drawing in new editors. Well, I need to be satisfied that WP is not misrepresenting the nature of its information on every page, especially to school kids, before I'm drawn in.

So, there's a gauntlet thrown down, although even if I saw a few nods in this direction, I'd be more likely to throw in with WP.

As for Nupedia, there's a good account of why it failed in the Reed Magazine article I cite, "Deconstructing Wikipedia." In an earlier go-round, I was in touch with Sanger, who told me that this Reed article is one of the best he'd seen to date about Wikipedia.

I see that Martha and I responded at the same time. I hope this gives you all the information you need!

Amy, thanks for commenting.

Here at TW, we also use Wikimedia Commons photos--such as the ones of Larry Sanger and Sue Gardner in this article. We verify that they're available under a Creative Commons license and adhere to any requirements of the license. If the ownership is questionable (for example, a photo is posted of a piece of artwork owned by a museum), we wouldn't use the photo but instead request permission from the original source (in this case, the museum).

Thank you, Rob. That's a wonderful explanation: "The world is a messy place. Because Wikipedia is widely viewed today as being authoritative, I think its conceptual foundations are worth looking at."
Yes. Exactly.
I think if you are going to become an authority on the collecting of knowledge, your theoretical understanding of what knowledge is and how it works in the world is crucial to the running of the site.
Obi--you are clearly someone who has thought about these issues, and I'm delighted that we have someone of your astuteness here to discuss these issues with us.
In fact, there are so many outstanding comments here that I want to thank you all. This feels salon-like, a virtual salon, where we are sitting around a living room, talking into the night about what is the nature of knowledge and how do we know what we know.
Heady stuff.
Andreas--I agree with you that this can't just be an issue that gets discussed for a day, but, rather, needs to be an ongoing discussion in which the pressure is kept on to continue to work at the problems with Wikipedia.
I have no desire to tear Wikipedia down. I just want it to work better. And while I know that the information on Wikipedia is reflective of the messed-up world from which it draws its information, I guess what disappoints me is that in my utopian vision of Wikipedia, its authority can be used to help to break down some of the arbitrary categories in which things are placed. It's all well and good to be an accurate representation of the way the world is, but I would love to see open-source encyclopedias also be a model for the way things could be. If the encyclopedia continues to assign categories to its entries, it remains descriptive, when I think it has the potential to be prescriptive--but only if these matters have been thought through to a point that someone like you, Obi, is at. If I were convinced that all Wikipedia editors were as invested in epistemology as you are, I would feel better. But a Wikipedia that merely reflects the biases that exist in our culture doesn't feel nearly as valuable to me. (I have no idea whether any of this makes sense. I am sleep-deprived.)

Hi Rob - I see your point, but the software doesn't allow different types of relationships. You're either in a category, or not. If you're an article, that is inferred to be "instance of" for set categories like "American novelists" or "related to the topic of" for topic categories like "France". The eponymous cats are bizarre in that way, as they sort of act (and people treat them) as if they were articles, but they're categories. You can read the current guidance here: http://en.wikipedia.org/wiki/Wikipedia:EPON#Eponymous_categories

Also, the women novelists category was not created to reduce the size of the parent -the NY times columnist got that wrong and it continues to be repeated: because gendered cats should never reduce the size of the parent (being non-diffusing).

As to your more general point - I think all ontologies, like all models, are wrong in a fundamental sense. Some, however, are useful. But you should join up and start a discussion about categorization guidelines and suggest ways to make it better - your insights would be valuable!

(I'm a little paranoid about spambots/harvesters, so not putting address in plain text). string the phrase "use the command line" together, without spaces or underscores, for the username, and it's at gmail.

Cultures evolve all the time, at least vibrant ones do. It is very often the case that an organization or entity starts with a set of rules and goals and, over the course of time, weaknesses in the model reveal themselves. Adjustments are made, sometimes radical ones. Although I strongly believe in substantive expertise, Wikipedia serves a purpose that I think could be better carried out with greater transparency.

Ok, Martha wants us to think outside of the box, so let's do so. As noted before, the WMF runs their whole operation on around $30m/year. Where does that go?

(all from http://meta.wikimedia.org/wiki/Wikimedia_Foundation/Annual_Report/2011-2...)
HALF A BILLION PEOPLE use Wikipedia and our other free knowledge projects. Today, Wikipedia is the FIFTH MOST-VISITED website in the world.
All of the other top-40 websites are private sector companies; we are the only non-profit on the list.
Each month, we generate 19 BILLION PAGE VIEWS to more than 23 million articles in 285 LANGUAGES. More than 80,000 volunteer editors regularly contribute content to Wikipedia and its sister projects.
Financial contributions 2011–12
A total of 1,130,700 people donated the equivalent of more than $30 million US dollars in over 80 currencies.
Volunteer contributions 2011–12
Individual contributors made 139.4 million edits, added 3.3 million Wikipedia articles, and uploaded 2.9 million images, audio files and video files.
Wikimedia foundation spent around $11.7M in salaries in 2011, which works out to around $93k (~$46/hour) average per employee (assuming 125 employees). On the other hand, the volunteers, if we assume that each edit took on average 30 seconds of "thought" (a low guesstimate), or roughly 1.16M hours of work, at ~$46/hour (similar to average wikimedia staff cost), this is an additional ~$54M worth of work done by the volunteers. Now this is totally back of the envelope and full of assumptions, it's just rough numbers - but the point is, even with a low estimate of time-per-edit, the contributions of the volunteers completely dwarf the WMF contribution, and the WMF isn't in the position financially to do much more, given that they are dependent upon donations. If the world gives them $30M, and they have to spend $15M to keep the site running, how many professional editors could they hire with that remaining money, to watch over how many different fields and how many different articles? I just don't see the MATH working out, unless WMF suddenly wins the lottery or something.

Another way to think of it is this: they have around 23 million articles total, and around 4 million in the english wikipedia. If you spread out the money they have, they can spend around $1.30 on each of the 23 milliion, or if they spent it all just on english-language articles, they could spend $7.50 per article.

Here's a different idea, thinking out of the box again. If wikipedia is seen as an important knowledge resource, and *academia* sees it as an important resource - not for research, but rather "it's important that wikipedia reflect currently up-to-date scholarly consensus on every field of study" - then why doesn't *academia* pony up and help fix it? For example, in the world of open source software, the major contributors to Linux and Apache are not geeks in their basements, but paid programmers, hired by IBM and RedHat and Novell, etc - because these corporations see it in their interest to contribute to the common shared platform of Linux - so they pay their programmers to contribute to an open source project.

What if something similar obtained in academia? What if Stanford paid its top history professors 20% of their salary to edit wikipedia, what if they were judged by the quality of the articles they produced (linking to individual diffs, of course, accepting that others might tweak). These professors could edit with their real name, declare their bona-fides on their userpage, and declare the same thing on the web-page of the university, thus giving other editors a sense that Prof. So-and-so may actually know what she is talking about. If every university department from every school appointed one prof (who would, as we all know, farm it out to her grad students - ack!), then we would have thousands of new, highly qualified professional "editors", working to clean up sourcing and able to provide cogent arguments around why paper X from 20 years ago is now discredited and paper Y should be used in its stead. One could ask editors and publishing houses and newspapers and so on to do the same thing - if you already use (and cite! urg) wikipedia, then you should pay some people on staff to help edit articles, not about *your* newspaper, but about the NEWS.

Lorraine complained about not editing until late at night, given other demands - but what if editing wikipedia was your job? Not as a paid corporate shill, but as a pursuer-of-truth - e.g. an academic, a researcher. If wikipedia is that important, then shouldn't someone be willing to pay for it?

Meanwhile, we could badger the WMF to put together a "professors' welcome course", that would be used to ease profs into editing wikipedia and explaining how it is different than other forms of academic writing, rules around sourcing, consensus, etc.

That way, instead of academia throwing rocks at wikipedia while dismaying that so many people keep *using* it, academia would instead put their $ where their mouth is, and start paying people to try to fix things...

This has been a great discussion. I am glad I happened upon it. I don't see a strong justification for anonymous editing. I get why people who voice opinions have reason to fear how others may respond to them. But it is odd to think that people who want the power to make supposedly factual claims, in a venue that encourages people to rely on those claims, would have no responsibility to stand by their assertions. Appending your name to something is the most common and surefire way to take responsibility for what one writes.

Wikipedia may lose editors who do not want their names known, but they may gain other editors who can better function in an environment that would, very likely, change once the cloak of anonymity was lifted. I, too, have heard horror stories--and had my own--about editing with people who become belligerent and nasty in the way that Internet anonymity notoriously fosters. After a long bit of this stuff, good people walk away. The names of editors should be readily available on each entry. Wikipedia is on the Internet, but it is more than a blog or magazine, and certainly the editors of Wikipedia should not be in the same category as those who post comments on blogs and magazines. Someone above said it well, the rules seem to be made with the comfort and convenience of editors in mind, not what is best for the users of the site. That is exactly backwards.

Excellent Rob - thanks - I never knew who first said that - I think I was working on non-linear models to understand fishery declines, and while stuck on a tricky partial differential, someone said that quote and it has stuck with me...

It also plays on Lorraine's utopian vision, that she'd like Wikipedia's categories to be a bit better than the real world sources they reflect. Would a "more useful" ontology, that is somehow better than the ones we've built before, lead to greater insight?

The whole thing also reminds me of Borges, as another wise editor at WP pointed out to me: http://en.wikipedia.org/wiki/Celestial_Emporium_of_Benevolent_Knowledge

I'm perhaps in a mood because of a category that has (thankfully) just been nominated for deletion, called "Villages in Norfolk connected with houses of historical interest" - it's such a wonderful example of the sort of thing cooked up by the hive mind. What if such a scheme was allowed to blossom? What categories would be created next? "Towns with blue windmills near the center"? "People who once visited villages containing old stone bridges"? The mind reels.

But that one example, and this (https://en.wikipedia.org/w/index.php?title=Special%3ANewPages&namespace=...) which lists all new categories as they are created on a daily basis, may give you a taste of what we're dealing with.

Great post Lorraine. I just want to respond to two bits:
1) "And if you think about it, distinguishing people by their genitalia is as arbitrary as distinguishing them by the color of their eyes.
Wikipedia, in theory, could be revolutionary, but it can’t be if it still carries forth the same categories that we are making now–and which continue to contribute to inequality–and enshrines them on the page.
Why do we need to distinguish if someone is a male or female writer in the first place? Why can we not simply refer to them as writers?"

Because wikipedia reflects reliable sources. If there were anthologies and professorships and seminars and conferences about blue-eyed novelists, you can bet there would be such a category in wikipedia. But there aren't. There *are* such sources for women writers (and now, more and more, for men writers), and African-American writers, and gay writers, etc, so these categories are accepted by the community. Wikipedia is a mirror - fractured, broken, flawed - but nonetheless a mirror of society. Even the library of congress has a category for female authors (but not one for blue-eyed authors)

So those cats are as arbitrary as the course list at your local liberal arts college. Which groups get studied and written about, and why?

If you created a list of blue-eyed novelists, it would be deleted with prejudice as an example of original research. Such things get created all the time, and are quickly killed.

For a live example of same, look at the debate around the American men novelist category - many people have called for its immediate removal, while others are arguing that this is now a notable subject- there are actually books about *male* novelists these days...

Wikipedia *can't* be a revolutionary in terms of content. We can't just decide that we have a better way to organize and discuss knowledge. The world has to change first - sad, but that's the current set-up, and quite fundamental - it's one of the core policies of wikipedia - No original research.

"If a young woman goes to the site to look up novelists, and the only names she sees are male names, what does that tell her about female novelists?"

Again, the guidance in place for a long time has said this should _not_ happen. It did, in violation of that guidance, and was corrected - but this was never the intent - all such cats were supposed to be non-diffusing.

You can add as many cats as you like to an article - check out Maya Angelou - here are some of her cats: 20th-century women writers;20th-century American writers; African-American memoirists;American memoirists;African-American writers;American activists; American dramatists and playwrights;African-American dramatists and playwrights;American people of Sierra Leonean descent;American television actresses;African-American television actresses; African-American women poets;African-American poets;American women poets; American poets;Grammy Award-winning artists;Lecturers;People from St. Louis, Missouri;People of Mende descent;Wake Forest University faculty;Women writers from Arkansas;Writers from Arkansas

Anyone who looks at a list of poets, or playwrights, or writers from Arkansas should see her there. But they will also see her in the more specific cats, as the community has accepted that these intersections like "African-American + playwright" is a notable intersection of study. Unlike a library, you can put a wikipedia article on multiple shelves.

Now, as I stated earlier, ghettoization outside of American novelists remains endemic - it's only in the past few days that another editor finally de-ghettoized the African-American novelists - but this is tedious, time-consuming work, and most editors, frankly, aren't going to be bothered - it's a SEP (somebody else's problem) I've considered whether bots could do this work, but it's actually quite complex in the general case, so I'm not sure if an algorithm could do it - it still requires a human touch. Perhaps we could ask the Wikimedia foundation to provide short-term grants to people to go through the whole tree and de-ghettoize it? When I've done such work, it takes me several minutes, sometimes up to 10, to properly do the job for a single bio like Maya Angelou. 10 minutes * tens of thousands of bios in the tree = not gonna be solved anytime soon - which is why I've taken a different tack, which is to try delete some of these categories that only serve to ghettoize. But I'm not in charge, and some well-meaning editors vote to keep those same cats I wanna kill. Such is the way of things.

Your points about how to increase recognition of women writers are spot on, but that is *also* the reason categories such as American women novelists and lists like the List of women poets were created in the first place - not to ghettoize, but to highlight. It just wasn't (and isn't, still), always done correctly.

Hi, obi,

Thanks for pointing out the issue of eponymous categories. I think the traditional approach (in artificial intelligence knowledge representation) is to allow for different kinds of relationships. "English spy novelists" is-a-subcategory-of "English Novelists", but "Ian Fleming" is-an-instance-of "English spy novelists" (and, indirectly, all of the ancestor categories). The Big Six is-a-collection-of the publishers Hachette, Penguin, and so forth. Ian Fleming wrote books, but there's also a collection of books written about Ian Fleming.

Difficulties in mapping real-world things and concepts into an ontology arise pretty quickly. For example, Wikipedia has a page for zebras, but Stephen Jay Gould asks in a famous essay, "What, if anything, is a zebra?" That is, IIRC, if you walk up the evolutionary tree to find a common ancestor species for the three modern zebra species, you get a subtree that includes modern horses. Our intuition that zebras are a well-defined category isn't supported by biology; it's based on surface characteristics, like the concept of human races. We might ask, "What, if anything, is an American Women Novelist?" It's a category, but does it have more than surface meaning? And the motivation for creating it seems misguided in general. "The American Novelists page has too many entries" may be a technical problem or a user interface problem, but it's not a knowledge representation problem. There are just a lot of American novelists.

There's no easy way to resolve these kinds of problems, even in a technical sense, and of course they're not specific to Wikipedia. The world is a messy place. Because Wikipedia is widely viewed today as being authoritative, I think its conceptual foundations are worth looking at.

1) I really encourage you guys to read about Citizendium - it was started by one of the founders of wikipedia, to fix many of the issues he saw and uses many of the solutions that you are all identifying/proposing above.

They have tried this non-anonymous, editorial control model, and maybe their articles are better, I don't know - how often do *you* go there to read them? From the Wikipedia article on same: "As of January 2013, it had 16,356 articles, of which 165 had achieved editorial approval, and around 10 contributors making at least 20 edits a month"

To me, Citizendium is a cautionary tale. Now, maybe it's because they forked, and wikipedia had a first mover advantage, and the editors didn't move over - but I think we should still study carefully what they did and why they haven't achieved more.

2) "when a man does take our arguments seriously, we figure it has to be a woman." hehe... I take anyone's arguments seriously, if they are well thought out - which yours are. I think this whole piece is one of the better pieces of writing (and nice discussion too) on this whole mess, which is why I'm responding here and not elsewhere.

3) If you want wikipedia, or even just wikipedia's category system, to be better than our sources, or more enlightened, there has to be some mechanism to decide what that newer/better way is. For wikipedians, when it comes down to it, we usually go back to published, reliable sources - that themselves have all of the editorial control that you list above.

If we are no longer relying on outside sources and creating our own better universe, then it's just a discussion amongst ourselves, and if ourselves is mostly 18-year old males, then I'm not sure we'd be happy with the result. Take a look at wikimedia commons, and the absurdly detailed category system they have developed for naked pictures over there, like http://commons.wikimedia.org/wiki/Category:Photographs_of_nude_females_f...

I think an appeal to a higher power of reliable sources, no matter how bad those sources still are and what flawed problems they present, is better than just the wholesale creation of a new world which is what we see over at commons today - they don't have any restrictions on reliable sources AFAIK to inform their ontology.

http://www.salon.com/2013/05/21/wikipedia_cleans_up_its_mess/ I must admit to being more troubled by this Andrew Leonard article. While it's great that Qworty has been banned from the site, the stipulation, as noted above by Andreas, that mentioning an editor's real name is enough to get you banned surprised me.
More and more, Wikipedia appears to be some secret cabal that is capable of defining for the general public what something is, without having to reveal their own identities. Do we really want to trust people who can't be held accountable for the information that they publish?
I spent years in academia, and knew of many feuds between historians who were fighting over interpretations of events, but at least no one was anonymous. Books are not published anonymously; scholarly articles are not published anonymously. What is the point of anonymous postings on Wikipedia? Is the world of the Internet really such a dangerous place that we have to be wary of the trolls who have the ability to harass us?
I'm feeling depressed at the moment. I'm wondering whether there are those who spoil it for the rest of us--that the dream of the exchange of information with real people's names is impossible because there are too many crazies on the Internet. I'm feeling discouraged about how Wikipedia can really change if anonymity is the rule that cannot be broken, and as a consequence, people are not held accountable for their anonymous posts.

While I understand the dismay of academics who themselves rely on primary sources, there are very good reasons for *not* allowing primary source material in Wikipedia, and that reason is editorial oversight (or lack thereof) (primary sources are allowed but only for very simple things that can be verified by a non-specialist).

Suppose you are an academic, and you do primary source research on the witch trial in 15th century Italy. You then share it with colleagues, who work in the same field, for their comment; then you send it around for publication, and *other* experts vet it, comment on it, argue with it, and so on. And then you get it published, meaning someone *other than you* decided this was worthwhile, and then that paper now gets cited, or it doesn't. The point is, in a reliable secondary source, there has been editorial control by, in most cases, paid professionals and experts in their field.

OTOH, suppose you arrive at Wikipedia and start editing the entry directly, adding your own personal analysis of the primary source documents, which only exist in a small archive in a small town very far away. How are other editors supposed to respond? What if your claims conflict with the claims of all other sources already published, vetted, and subject to editorial control? Your analysis would be discarded - it has to be. Original research is not allowed exactly because there is no chief editor - the goal instead is to neutrally represent, as a tertiary source, the already vetted/accepted analysis of secondary sources that themselves are subject to editorial control and citation. Also the claim that all secondary sources are treated equally by Wikipedia is patently false - there are regular and long debates about suitability of given secondary sources and which ones should be given more weight than others.

Obi, one thing we have to confess here is that the problem of non-diffusing gender categories being diffused (against guideline advice) does not just affect American women writers. It affects women of all nationalities, and all professions. Getting it right is a Sysiphean task – and one I don't realistically see being done, from the present baseline situation, especially as categorisation is not an issue many Wikipedians care about. You are an exception.

The English Wikipedia contains around 600,000 biographies. For every biography whose categorisation you correct, ten biographies will be miscategorised by well-meaning editors blithely unaware of the difference between a diffusing and a non-diffusing category.

You'd be better off writing an article on categorisation for The Signpost, or creating an online workshop, than by getting stuck in recategorising biographies yourself. Get in touch with Ed (User:The ed17); given the prominence of the issue, I am sure The Signpost would run an op-ed on diffusing and non-diffusing categories, describing the mistakes that were made and how to avoid them.

Meanwhile, outside commentators should keep up the pressure, or the issue will simply be forgotten, and everything will return to business as usual.

Hi Rob - per your example from 2007, I'm not sure what the intent was of that line, but I think at least non-diffusion of gendered categories was already understood (the section says "Both male and female heads of government should continue to be filed in the appropriate gender-neutral role category (e.g. Presidents, Monarchs, Prime Ministers, Governors General.)", which is essentially an argument for non-diffusion).

From the ethnicity/race categorization guidelines, c.2007:
"Race and sexuality cross-categories are typically used to split larger categories (Category:LGBT sportspeople is used to reduce the size of Category:LGBT people)."
This example brings up an interesting if rather geeky point that I'm not sure we've discussed. In some cases, a category can be both diffusing *and* non-diffusing, at the same time! How is this possible? Let's take LGBT Sportspeople as an example. It *should* diffuse LGBT people, because it is a more specific intersection of LGBT + job. However, it should *not* diffuse "Sportspeople", as you don't want all of the gay athletes *only* in that one category - so for Sportspeople, it is non-diffusing.

Another example is Women writers. This is diffusing on Women, but non-diffusing on Writers. It can get more complex, but I don't want to over-category-geek too much on this page. Just pointing this little trickiness out.

Then Annette has this zinger:
"But it is odd to have people doing this who cannot be trusted to do analysis and draw conclusions about primary sources, but are expected to do that for secondary sources."
Ouch! I don't have a great defense for that one, except to say, I'd rather have rookie editors choose a bad secondary source, than have rookie editors coming up with their own analysis of the Nuremburg trials from the transcripts. So it's sort of a lesser-of-two-evils thing, vs a "OMG use of secondary sources is such an awesome and unassailable solution" thing. Again, if you have a way to make this better - for example, suggestions for how to change the guidance to make it more likely that we'll choose good sources, please make suggestions!

Then Annette sez:
"Why not stick with experts who get their stuff peer reviewed?"

Well, experts who get their stuff peer reviewed usually put that stuff into journals which are protected by copyright and behind a pay-wall, except the awesome PLOS. I think one thing we haven't talked about sufficiently here is that wikipedia articles are *all* covered under a creative commons license, meaning you have permission to take the article, publish it, tweak it, modify it, do whatever you like basically, as long as you give appropriate credit. This is different from 95% of academic journals and probably 95% of books, and is fundamental to the functioning of the site.

Wikipedia has no license fee, no registration fee, no use fee, no subscription fee, and the content itself is freely licensed to permit re-use without need to ask permission.

That distinguishes it from almost every other source of information on the planet.

One point that deserves emphasising with respect to anonymous contributions vs. some form of real-name registration, or open editing vs. a system like flagged revisions (see above), is that there are halfway houses. There is probably no pressing need for Wikipedia or the reading public to know who edits the article on Venus, Swahili, or Barack Obama. But Wikipedia has hundreds of thousands of biographies of people of marginal notability – scholars, journalists, minor actors, regional politicians etc. – that sometimes see appalling anonymous defamation and revenge editing. Wikipedia should be a little more responsible with their reputations.

Excellent brainstorming, obi. I'll offer an academic perspective.

What if something similar obtained in academia? What if Stanford paid its top history professors 20% of their salary to edit wikipedia, what if they were judged by the quality of the articles they produced (linking to individual diffs, of course, accepting that others might tweak)... That way, instead of academia throwing rocks at wikipedia while dismaying that so many people keep *using* it, academia would instead put their $ where their mouth is, and start paying people to try to fix things...

Unfortunately, to my knowledge, research universities don't directly pay their professors to do more than teach. Roughly speaking. That is, a professor's work life is divided between academic research, teaching, and service. The incentives are heavily weighted toward research---publishing, getting grants, maybe policy making, and so forth. I think contributing to Wikipedia would count as service, at the very bottom of the average professor's list of priorities.

I think for the ideas you lay out to work, there would need to be a shift on the academic side, not money but some recognition that editing Wikipedia has value in the academic world. That would involve a big cultural shift. In my area, non-peer-reviewed writing tends to be discounted heavily. I think Wikipedia is more important as a whole than it's given credit for in academia, and changes in our perspective could improve both sides.

This is why my own outreach efforts (in the context of medical articles) have tended to focus on the patient education aspect of things. Theres a much clearer social need there than there is for, say, 18th century French literature. While there have been increasing calls for engagement with WP by, e.g. the society for neuroscience (link), other academic fields have, to my knowledge, not by and large attempted to engage in this way.

I think readers always benefit from knowing who has edited a piece of writing purporting to convey factual information. And I think it would also help to order behavior--an important thing given the absence of an editor. But starting with relatively minor figures is a good idea.

obi, you are right that the media missed many things in this category discussion, but in our piece we did note that (1) the "American novelists" category problems had been resolved by the Wikipedian community, and (2) that the community had an official policy re: non-gender specific categories (we mentioned that in reference to Sue Gardner's Wikimedia blog post).

Regarding categories diffusing from the parent and the confusion over this, I guess I'd ask a more basic question: why is there a need for such a confusing system? If all the editors who took your quiz failed, then it's a setup for misinterpretation. Red herring or not -- and I have no doubt that this category problem is just one tiny instance of the sexism in Wikipedia -- it's not working.

Should I or Lorraine join the gender bias task force? Hmmm. Maybe. As a non-Wikipedian, I suspect I do more good pushing from the outside as a journalist. I also think that my basic premises are different than Wikipedia's five pillars. In fact, I'm not a fan of the NPOV, although I do appreciate the emphasis on civility.

Hi Lorraine - Thanks for welcoming us to your salon. I'm not going to address right now the many excellent points both you and Andreas have made, as I want to leave space for others to participate - I just want to leave you with one last thought on the area most interesting to me, which is the category system ==> let's move to category intersections.

If we had category intersections, there would be no more gendered, sexuality-based or ethnic categories anywhere in the tree (even nationality categories could go away), except at the very highest level. Every person would be in high-level categories like "Men", "African-Americans", "Novelists", "Journalists" - and then we could let *users* define which combinations of those they were interested in looking at. It would be a radical shift, but it would also be in line with our principles, as it would not involve original research, and it would remove a number of categories which are always going to be problematic (and never complete). This might help address the issue you highlight, of "arbitrary" categorizations that tend to divide us.

There is one potential disadvantage to intersections - it would really be ALL-inclusive - meaning, if so-and-so is a famous gay interior designer, and named as such in all the articles about him, but is also an writer, and as an writer he is never referred to as 'gay writer', he's just called a great writer- in the category intersection world, he would still show up in the list of gay writers. That's a tradeoff we'd have to accept - whereas today, we could *remove* him from the category 'gay writers', saying "he's not really known for that particular intersection, there are no sources to support it". But regardless, I think intersection is the way to go, I know Andreas agrees with me, and I'm in discussions now with a developer about this, but more pressure on WMF from the outside would certainly help.

This feature has been requested for, I think, 8 years now. In some sense, it would be easier than the cleanup Andreas notes, of 600,000 bios. In the category intersection world, there will literally be no ghettos.

If you want to see how difficult it is to categorize in today's system, check the complexity of the algorithm I proposed for correct, non-sexist/racist categorization:
http://en.wikipedia.org/wiki/Wikipedia:WikiProject_Countering_systemic_b.... Using this algorithm, a bisexual, African-American woman, who is a journalist, poet, and writer from Chicago, ends up in 17 different categories. If we had cat intersects, she would be in 7, and you could assign them with your eyes closed.

Wikipedia guidelines and policies are generic and don't cover in detail which specific sources are "reliable" or not. The guidelines around sources have been the subject of several years of discussion, revision, and refinement - there are literally hundreds of talk page discussions that cover the material we are discussing here around sourcing. But, it can always get better, so please join, read the guidance, read the policies, follow them for a while, and then see if you can help get them to be even better.

Just to cover a few other points:
1) This page clearly lays out the policy around usage of primary sources. http://en.wikipedia.org/wiki/Wikipedia:No_original_research#Primary.2C_s...
"A primary source may only be used on Wikipedia to make straightforward, descriptive statements of facts that can be verified by any educated person with access to the source but without further, specialized knowledge... Do not analyze, synthesize, interpret, or evaluate material found in a primary source yourself; instead, refer to reliable secondary sources that do so."

Lorraine, I don't think anyone ever said that primary sources were *inferior* to secondary sources; instead, it's more that they often require *analysis* and *drawing-of-conclusions* that we don't expect or want wikipedia editors to undertake, esp in the absence of a formal peer review process that can vet those conclusions. So rather than a fact-generating machine, think of it as a fact-organizing machine.

2) Annette, I'm sorry that you ran into problems getting better scholarship accepted. As noted in http://en.wikipedia.org/wiki/Wikipedia:RS#Some_types_of_sources, wikipedia has a preference for scholarly/peer reviewed articles over other types of published material, so you can use that guideline when making arguments. There are also many methods open to content dispute, if editors aren't seeing your way - 3rd party resolution, various notice boards, etc. Ultimately, if you encourage other academics who know the latest material to join wikipedia as editors, I think you can change things - you just need to bring people to the table who both understand the material and are willing to learn wikipedia's rather odd culture.

3) Wikipedia reflects publication bias, which itself mirrors wider societal bias (for example, perhaps less women or people of color get published in field X?). Well, that's a gauntlet you've thrown down - yes indeed wikipedia is biased towards publications (and thus inherits any bias in that process) in fact we *only* accept already published material as a source for both notability of the article (so we know it should exist in the first place), and the content in the article.

While I'm sympathetic that this itself may lead to bias, how would you specifically suggest fixing it? Allowing non-published work to be used as a source? That's a guaranteed pandora's box. The poor mathematicians would have a heart attack, as there are massive amounts of crazy math papers out there that no-one will ever publish - shall we now put them in wikipedia? There are many other forms of bias in wikipedia - for one, we have 4 million articles, but since our users are english-speakers, we mostly publish on things that are written about in the English language! Most of our users are from the US and the UK, so we mostly write articles about shows and movies and pop stars from those countries. Most of our users are young men, so we have an absurd amount of content on video games.

I see your point, but aside from letting unpublished work in, or some sort of affirmative action for publishing by minority groups, I'm not sure how wikipedia itself can correct a bias that academe hasn't. Remember, Wikipedia is a mostly-volunteer project, run on about $30M/year. Something like PLOS, which is ONE journal, runs on ~$20M/year. Now multiply that by *all* of the other journals, and you have a sense of the amount of resources available to wikipedia vs the amount available to the publishing industry. If the publishing industry can't fix this bias, with 100x the resources, why should we ask Wikipedia to try to?

Again, I think the best solution is, a MORE DIVERSE editor base.

4) Finally, on the issue of neutrality. Several commenters have stated that neutrality is not useful, or not attainable, etc. My response would be, yes, true neutrality is unattainable. But the act of editing a wikipedia article is fundamentally different than the act of writing a paper for a journal, which is where people get mixed up. When you're writing an article for publication, you have a hypothesis, you have a POV you want to promote, and your article will be added to the many hundreds of other articles on the same subject, as another contribution to the literature. There's almost always room for one more theory, or one more refinement - publication of an article doesn't supplant or displace *all* the other articles that have been written in that same field.

Wikipedia is fundamentally different, even different than a textbook, because there is ONLY ONE wikipedia page per topic, and it can't be 20 pages long. So you have to find a way to distill all of the disparate arguments, sources, materials, points-of-view, and represent *all* of them, fairly, without undue weight, in a single article. If you're doing this, and you don't also at least *attempt* to be neutral, you are doing the reader a disservice.

Perhaps there really is nothing to be done without a fundamental change in the character of the enterprise. The demographic profile of the editors suggests that Wikipedia has settled naturally where one would expect it to: under the control of mostly young white males with lots of time on their hands who are, at times, overly aggressive in protecting their turf and comfortable putting up with what seems like a needlessly complicated editing apparatus and myriad rules that can be used to circumvent change. It appears a club; with exhortations to pay heed to the club’s existent culture as if the culture, rather than the product, was of principal concern. None of this rests comfortably with the claim that “anyone can edit” Wikipedia. I sill maintain that anonymous editing contributes to the problem. This is particularly so regarding living subjects and historical figures who are extremely popular.
Yes, it is free. And we cannot expect too much from free things. Wikipedia has its strengths. But it is up to teachers and professors to make its limitations clear. That could become part of the educational mission.

Obi, I knew about the Wikipedian view of primary sources. Lorraine and I worked on different parts of this article, and a section in which she raised the issue of primary vs. secondary sources ended up cut by our editor because we already had a lot of fish sizzling in the pan. So, I'm not shocked by the WP emphasis on secondary sources, but, after all this discussion, I still don't agree with it. I doubt most journalists or academics could--which raises what I think the key issue is for a followup piece, whenever we get there.

UTCL and Rob get closest in talking about the difference between what end users believe about the information in a WP entry and the Wikipedian community's assumptions. There's a big gap here, and yes, it is a presentation problem. But now that WP is as large as it is, I believe that Wikipedians and the Wikimedia Foundation need to take responsibility for how WP information is perceived. This is not just a "dumb user" problem; it's a real difference in POV and purpose.

As for Annette's zinger about the need for expertise, I'm in her corner. That's probably obvious, but I have to say again that referring to any Wikipedian as an "editor," rookie or otherwise, tramples on the meaning of what editors do. I'm an expert on very little, but as an editor and journalist, I know how to question where information comes from and to vet what I see.

In many respects, I believe very much in the postmodern nature of Wikipedia. The truth is an ever-evolving, variegated, oddball thing, and I don't think mainstream media or anybody else has a lock on it. But however suspicious I am of anybody trotting out their expertise, I don't deny that experts exist. If anything, this wide-ranging conversation has convinced me that WP does need a corps of salaried "top editors" to manage disputes. While that may just seem like more criticism rather than a realistic solution, I'd say that's our role as writers: to question, to push outside the WP box, to propose radical shifts.


Add new comment

More Like This

Oct 26, 2012 | Journalism, Media Debate
May 2, 2011 | Advice, Self-Publishing
Jan 9, 2012 | Talking Art, Blogging