Why this Site?

  • Our Mission:
  • We exist to shine the light of scrutiny into the dark crevices of Wikipedia and its related projects; to examine the corruption there, along with its structural flaws; and to inoculate the unsuspecting public against the torrent of misinformation, defamation, and general nonsense that issues forth from one of the world’s most frequently visited websites, the “encyclopedia that anyone can edit.”
  • How you can participate:
  •  Visit the Wikipediocracy Forum, a candid exchange of views between Wikipedia editors, administrators, critics, proponents, and the general public.
  • 'Like' our Wikipediocracy page on Facebook.
  •  Follow Wikipediocracy on Twitter!

Press Releases

  • Please click here for recent Wikipediocracy press releases.

Wikipedia needs to abandon ADAM

By Andreas Kolbe

Wikipedia contains over half a million biographies of living persons, for people ranging from Barack Obama to local shopkeepers. Yet the core editors whose job it is to ensure that all of these biographies comply with Wikipedia policies number only a few thousand. And while the number of biographies rises daily, editor retention rates in Wikipedia are currently plummeting. (The slide showing editor retention data is known within the Wikimedia Foundation as the “Holy Shit slide”.)

Biographies of major figures like a US President are usually fairly well watched over, and in quite good shape. But for little watched biographies at the other end of the notability spectrum covered by Wikipedia, things look very different.

What happens with minor biographies is that they are often only frequented by anonymous people who have a dislike for the biography subject, or at any rate have neither the interest nor the writing skills to produce a balanced biography. Such contributors will typically add derogatory information – sometimes true, sometimes not – or random stuff they read and found “interesting”.

The results are not pretty – 50 per cent of a journalist’s biography about an argument he had with Mitt Romney? Or take this addition to the “free encyclopedia”, made by an anonymous person wishing to be known as just “Shylocksboy”: “Jayne Middlemiss called Titmuss ‘a big, fat slag’ because of her flirting with Sharpe, whom Middlemiss had an eye on.”

Dirt

Dirt

 

 

These examples are typical of the anonymous dirt accretion method (ADAM) of biography writing. It results in biographies whose actual biographical content becomes overwhelmed by tangential material inserted by anonymous people with a grudge against the biography subject, or an interest in random trivia.

Here is another example, turning a sexual harassment accusation, denied by the accused and a high-profile eye-witness (okay, we’ll name-drop: George Clooney), into established fact. Even the tabloid sources the snippet was based on did not assert that the claim was true, and the court case seems to have sunk without trace; but in the court of Wikipedia, an accusation is enough to convict.

Wikipedia has traditionally defended the existence of unbalanced and slanted articles by putting forward the concept of “eventualism”. According to this philosophy, articles will often start out as an unbalanced mess of loosely connected information, but “eventually” “someone” will come along and put all of it together in a way that makes beautiful sense. In essence, this treats a Wikipedia biography as a scratch pad, to record things that might or might not come in useful when someone gets round to writing a “real” biography.

This may have been a workable, even sensible, concept in the encyclopedia’s early days, when Wikipedia biographies came on page 50 in Google search listings, and Wikipedia editors were happy to have any information at all on any given page. But today, over a decade later, with Wikipedia ranking as the number one Google link for almost anybody who is in the public eye, it is no longer an acceptable way of working. A Wikipedia biography needs to be balanced, and fair to its subject.

Wikipedia needs to abandon ADAM.

 

See also the essay at Wikipedia, WP:ADAM.

Photo credit: © Supertramp | Stock Free Images & Dreamstime Stock Photos

13 comments to Wikipedia needs to abandon ADAM

  • Rd232

    Good post – but half of it seems to be missing! It analyses the problem well, but “abandoning ADAM” is not an implementable solution! How can ADAM be abandoned? Any ideas?

  • Andreas Kolbe

    My personal preference is that changes made by unknown or inexperienced editors should be held in abeyance and only shown to the general reading public once a more experienced and trusted editor, who has demonstrated a reliable understanding of Wikipedia’s “Biographies of Living Persons” (BLP) policy, approves them.

    Corresponding wiki software solutions exist of course – they are called “flagged revisions” or “pending changes”, and one version of this has long been implemented across the entire German Wikipedia (as well as the Wikipedia projects for several other languages).

    A couple of years ago, the press were told, with great fanfare, that the system would also be introduced for the English Wikipedia. But implementation was first delayed, and then stalled altogether because of objections raised by the editor base. (This closely parallels the situation with the adult image filter that was announced last year, and then unceremoniously abandoned.) There are currently no plans to introduce pending changes in the English Wikipedia either globally, as in other Wikipedias, or at least for all biographies.

    As it is, Wikipedia allows anyone to edit people’s biographies, and makes their changes visible to the public immediately. (Rapper will.i.am recently complained on Twitter that his biography stated he was a million years old …)

    The important point to remember is that ADAM is already against Wikipedia’s “Biographies of Living Persons” (BLP) policy, as written today. The policy demands that biographies be balanced and neutral at all times. But there is a gap between policy and practice.

    Many people clicking the Edit tab on a Wikipedia biography have never even heard of Wikipedia’s BLP policy. Others know it, but have no intention of following it. The temptation to deal someone they don’t like a subtle or not-so-subtle blow is too great, and Wikipedia is making itself – and its biography subjects – irresponsibly vulnerable here.

  • Eric Barbour

    Rd232:

    With all due respect, this situation has been discussed, re-discussed, debated, mined for manganese nodules, and practically bound in leather since the flagged-revisions idea came up long ago. There are numerous threads discussing it in Wikipedia Review’s archives, literally too many to list. And the subject was beaten to death on Wikipedia, again too many links to list.

    It was agreed, repeatedly and by many people, that flagged revisions would be a major step forward in keeping BLPs “clean” (or at least “cleaner” than they usually are). Yet your precious community rejected them. And there are now journalists and Wikipedia users who think Wikipedia biographies are “neutral”, because flagged revisions were implemented (which they were not!!)

    There are other ways to help keep BLPs from becoming defamation targets. Special watchlists, more stringent requirements for notability, and more. None of these is being done on en-wiki.

    I have already estimated there are about 12,000 overly negative or hostile BLPs on Wikipedia. The number may be growing, it may be shrinking–it’s impossible to tell without a tracking system like flagged revisions. The German Wikipedia has been using flagged revisions for FOUR YEARS, with great success. What’s the excuse for en-wiki dragging its feet??

    Feel free to explain why this is happening.

  • Oz

    A lot of subjects of BLPs are called “conspiracy theorists” based on one or two questionable sources such as “Conspiracy Theories in American History: An Encyclopedia.” This typically will show up in the article lead. I think there ought to be some standards that regulate this (although WP seems to have a major preoccupation with stamping out anything called a “conspiracy theory.”)

  • David Wainwright

    The flagged revision policy has not been implemented on English Wikipedia for several reasons. There are logistical concerns that reviewers will take too long to implement valid changes. There are also concerns that flagged revisions could allow one reviewer or a few reviwers to become gatekeepers for a given article. Just as the major BLP problems are with low-traffic articles, the big problems with flagged revision will be with the low-traffic articles.

    An imperfect, but more realistic proposal would be to ban anonymous IP edits, and to strictly enforce Wikipedia’s rules. From what I read, 80% of all Wikipedia vandalism comes from anonymous editors. Wikipedia policy prohibits a lot of disruptive behaviors (e.g., edit warring, sock puppets, biased editting), but from my experience as a former editor, enforcement is poor. Also, I think we need to think beyond just BLP articles. While BLP is the subject of much scutiny because of concerns about defamation, there have a number of cases where non-BLP articles have been use to libel people.

    • Andreas Kolbe

      David,

      The German Wikipedia has operated flagged revisions for four years now, I think. Since installing it, they have never looked back. And it is my impression that they generally suffer from fewer content problems.

      The fact that with flagged revisions, IP changes are not immediately visible to the public, but are only visible to logged-in Wikipedians, in itself discourages a lot of the type of malicious edits that we see in the English Wikipedia. The people who remain to make edits are those who are serious about contributing, because they are the ones who see their edits approved.

      The English Wikipedia has twice the burnout rate of the German Wikipedia. I posted some editor retention figures to the WikiEN-l mailing list the other day:

      http://lists.wikimedia.org/pipermail/wikien-l/2012-May/110172.html

      Of all people who have ever registered an account at the German Wikipedia, almost 1% are still around to make more than 100 edits a month. In the English Wikipedia, that percentage is half that, less than 0,5%.

      That is a stunning difference.

      Both projects started in 2001, only three months apart. And while core editor numbers in the English Wikipedia have dropped from 5,000 in 2007 to less than 3,500 today, the German Wikipedia’s core editor numbers have remained stable throughout that period.

      There may be other cultural factors involved here, and it would be worth doing a more wide-ranging comparison of editor retention figures across Wikipedias that do or do not use flagged revisions, but I feel quite certain that the implementation of flagged revisions, and the change in culture it brings with it, generally has a positive effect on editor retention. There is simply less nonsense to deal with on a daily basis.

  • Hersch

    The biggest threat to the subjects of BLP articles is not vandalism by anonymous IP editors, but carefully calculated biased editing by pseudonymous established editors who know how to game the system.

    • David Wainwright

      Hersch, I agree with you that IP editors are not the only problem, but I think that if Wikipedia actually enforced its many rules, much of the crap caused by a few established editors could be stopped. From what I saw, Wikipedia was pretty good at enforcing copyright violations (i.e., to avoid lawsuits), mediocre at enforcing edit warring rules, and lousy at enforcing anything else. Wikipedia has definite policies against tendentious editting (i.e., intentional bias), but I have never seen them enforced.

  • Andreas Kolbe

    Hersch is right in that it often takes several years, and an arbitration case, to stop pseudonymous editors pursuing vendettas against living people via Wikipedia.

    Flagged revisions could be of some help, by temporarily or permanently removing BLP reviewer rights from people found to exhibit a consistent pattern of BLP policy violations.

  • Jeremy Sparks

    Besides flagged revisions (German style, not the watered down pending changes), Wikipedia should also introduce a 5-year delay between any events and its inclusion. The current drive to include every single minute event the moment it happens removes all sense of perspective.

    A 5-year delay would allow, on the contrary, to start discerning whether any overblown media / tabloid / internet scandal or outrage had actually any impact at all on the topic at hand. Some exceptions would have to be granted in certain cases (it’s probably worth including that Mitt Romney runs for president in 2012 while he is running) but limited to listing key current events while leaving out the minutiae (“On the campaign trail, candidate said X while his opponent said Y, calling candidate epithet in the process”).

    You can’t have a broad perspective when you only keep looking at what is immediately before your eyes.

  • anonymous ip editor

    “These examples are typical of the anonymous dirt accretion method (ADAM) of biography writing.”

    I’ve seen something different at work, and yet, I’m not sure the problem is ADAM itself.

    I’ve seen several cases where some person already had an entry in the wikipedia, then yes, some aspect of that person’s life or behavior came under scrutiny (an arrest, a firing, a blog post, a history), and then descriptions of the issue were added into the Wikipedia.

    The ensuing edit wars often looked like this:

    1) Derogatory information is added
    2) Removed
    3) Added back
    4) Complaints about accuracy, and then removal
    5) Accurate, well sourced, but derogatory information is added in.

    6) WP:DELETE NOTABILITY
    7) SWIFT DELETION BASED ON 5 EDITORS VOTING

    As one example, maybe you can see what happened to Hugo Schwyzer in the past 90 days or so.

  • Andreas Kolbe

    Well, Wikipedia did indeed delete an article on Schwyzer earlier this year. The deletion discussion which led to that result is here:

    http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/Hugo_Schwyzer

    It’s relatively brief; sometimes these discussions attract dozens of people commenting:

    —o0o—

    This BLP apparently asserts the subject’s notability as a writer and a professor. The subject is clearly a prolific writer about himself, sex, and feminism, mostly in the blogosphere, but fails to as to WP:BASIC substantial coverage by multiple independent reliable third parties to show significance as a writer or professor. He unambiguously fails the notability requisites that seem most applicable – WP:WRITER – and the alternate WP:ANYBIO. The subject is also on this side of stellar as to WP:PROFESSOR. JFHJr (㊟) 01:15, 18 February 2012 (UTC)

    Note: This debate has been included in the list of Authors-related deletion discussions. • Gene93k (talk) 02:47, 18 February 2012 (UTC)
    Note: This debate has been included in the list of Academics and educators-related deletion discussions. • Gene93k (talk) 02:47, 18 February 2012 (UTC)

    Delete. After doing a search, I just added one short paragraph to the article about Schwyzer being banned from a feminist blog. At least it was reported in a real secondary source (the Atlantic), as opposed to self-reported. Even with that addition, although the guy gets a lot of notoriety, I don’t see any real notability, as an author, a teacher, or a commentator.–Bbb23 (talk) 03:14, 18 February 2012 (UTC)

    Delete. I am familiar with his work, and was not expecting to vote delete on this AfD, but after a pretty thorough search for sources, I haven’t found any besides the atlantic article that meet our standards. 98.248.194.216 (talk) 08:21, 18 February 2012 (UTC)

    Comment: “He has also written of the many times he has had sexual encounters with his students, including a class trip he was charged with chaperoning in which he had sex with four of his students.” What? This is legit?–Milowent • hasspoken 05:31, 19 February 2012 (UTC)
    Yep. No BLP issues here. It’s a large part of what has made him a controversial figure – he’s written about that and more in his personal blog, and in articles he’s published elsewhere. 98.248.194.216 (talk) 06:26, 19 February 2012 (UTC)

    Delete, possible speedy G10. The combination of primary sources cherry-picked to show the subject in a bad light (however much he may deserve it) looks like a BLP violation to me. —David Eppstein (talk) 00:25, 20 February 2012 (UTC)

    The result was delete. Wifione Message 16:30, 25 February 2012 (UTC)

    —o0o—

    FWIW, it sounds like a typical case. You have a borderline notable person, a controversy in the blogosphere, and anonymous contributors with a partisan interest in that dispute. The importance they assign to the issue is based on the intensity of their feelings of approval or disapproval towards the biography subject, rather than actual media or scholarly interest in the issue, and their aim is to increase the visibility of the dispute through Wikipedia.

    The result is invariably an unbalanced biography. It typically resorts to primary sources (in this case, the subject’s self-published blog confessions about his amoral behaviour in his past life as an alcoholic) when the issue has not attracted the attention of mainstream publishers – i.e. the sources Wikipedia claims to reflect.

    When allowed to go on unchecked, this results in a novel biographical narrative that bears little or no resemblance to what media or scholarly publications have to say about the person (if they have anything to say about them at all).

  • DrPh

    I think another issue that we have is that notability guidelines seem extremely fuzzy, if not in writing but in the way they are interpreted. I was working on an article about an Australian scientist and creationist advocate. While at first, it looked like his work was notable and generally well-reported on in his field, as we went through to do further research the other editors and I found that nearly every claim about him was either sourced to an unreliable source or sourced to a source written by himself. Not only that, of the three things that he was allegedly notable for (being a food scientist, being a chocolate advocate, and being a creationist), there were zero reliable sources *about him personally* that talk about any of those things except for the 2nd thing. It was very frustrating, because everyone except for 1 or 2 other editors seemed to think that the subject was unnotable but we had to keep going anyway even though it was a live person bio and it was filled with things that we couldn’t prove with an RS were true!

    The article survived one AfD on notability grounds but was deleted a few weeks later after a second admin entered and managed to pull consensus to by rereading the fractious, convoluted, and almost incomprehensible debate on the AfD page.