Deletionists, inclusionists and delusionists

“When you come to a fork in the road,” Yogi Berra said, “take it.” Wikipedia has come to a fork in the road, and it should pay heed to Berra’s advice.

The rules that govern how the popular online encyclopedia works are set by its community of contributors – the so-called wikipedians – through a process of argument and consensus-building. But the community has begun to split into two warring camps with contrary philosophies about Wikipedia’s identity and purpose. On one side are the deletionists; on the other are the inclusionists. Between them is not a middle ground but a no-man’s-land. As one Wikipedia observer recently put it, “The inclusionist versus deletionist debate is as firm and strong as the abortion debate, gun control debate, or the death penalty debate.”

The adherents of inclusionism believe that there should be no constraints on the breadth of the encyclopedia – that Wikipedia should include any entry that any contributor wants to submit. An article on a small-town elementary school is no less worthy for inclusion than an article on Stanford University. The supporters of deletionism, in contrast, believe in weeding out entries that they view as trivial or otherwise inappropriate for a serious encyclopedia. Here’s how the encyclopedia itself describes the two camps:

Deletionism is a philosophy held by some Wikipedians that favors clear and relatively rigorous standards for accepting articles, templates or other pages to the encyclopedia. Wikipedians who broadly subscribe to this philosophy are more likely to request that an article that they believe does not meet such standards be removed, or deleted. Conversely, Wikipedians who believe that there ought to be a place for an article on almost any topic in Wikipedia, and that there should be few or no standards barring an article from it, are said to subscribe to inclusionism.

AIWThere is an Association of Inclusionist Wikipedians, with 207 members at the moment. (See logo.) Their slogan is “Wikipedia is not paper.” Because there are no physical constraints on the encyclopedia’s size, they see no reason to limit the number of entries. Let’s focus on making each entry as good as possible, they say, not on picking which entries should stay and which should be deleted. There is as well an Association of Deletionist Wikipedians, currently with 144 members. They have a slogan of their own: “Wikipedia is not a junkyard.” To them, Wikipedia needs to be seen as a whole, not just as a vast assortment of discrete entries. Deleting entries is, in their view, essential to improving the quality of the overall work.

To the inclusionists, Wikipedia is in essence a wiki. It’s an example of an entirely new form for collecting knowledge, a form unbound by the practices of the past. To the deletionists, Wikipedia is in essence an encyclopedia. It’s an example of an established form for collecting knowledge (albeit with a new production and distribution model), with traditions that deserve respect. The split between deletionists and inclusionists is thus a manifestation of an identity crisis that has always been inherent in Wikipedia. From the start, Wikipedia has pursued two conflicting goals: to be an open encyclopedia that anyone can edit, and to be a serious encyclopedia that is as good as the best print encyclopedia. In the early years of Wikipedia’s existence, when it was viewed mainly as a curiosity, the tension between those goals was easy to overlook. Nobody really cared. But as Wikipedia has become more popular – and as it has begun to be held to a higher standard of quality – the tension has reached the snapping point. The inclusionists’ desire for openness and the deletionists’ desire for seriousness are both worthy goals. But, as the diametrically opposed missions of the two camps reveal, they are also mutually exclusive goals. You can’t be a deletionist and an inclusionist at the same time.

At a deeper level, the split between the deletionists and the inclusionists is yet another example of the fundamental epistemological crisis of our time: the battle between absolutists and relativists. The deletionists are absolutists. They believe that some subjects are simply more significant than others, that absolute distinctions can and should be drawn among different kinds of knowledge. John Milton is more important than George Jetson. The inclusionists are relativists. No subject is inherently more significant than any other, they believe. It all depends on context. John Milton will be more important than George Jetson for some people. But for others, George Jetson will be more important. There are no absolutes; it’s all relative.

The tension between the inclusionists and the deletionists is not merely theoretical. Entries are being deleted and “undeleted” from Wikipedia all the time – as the recent dust-up over the deletion and reinsertion of the entry for “Enterprise 2.0” shows – and the practice of and criteria for deleting entries are sources of constant and often bitter debate among wikipedians.

Whether the deletionists or the inclusionists gain the upper hand will determine Wikipedia’s future scope and quality. If the deletionist philosophy prevails, the inclusionist Wikipedia will be lost forever; we will never know what a truly open encyclopedia – a truly wikified encyclopedia – would ultimately look like. If the inclusionist philosophy prevails, the deletionists’ ambitions for Wikipedia will go unfulfilled. We’ll never know how good, by traditional standards, an encyclopedia created by volunteers might have been.

signThe best way forward in this case – the way that creates the least harm – may not be through the process of consensus-building. Trying to find common ground between the deletionists and the inclusionists seems a futile exercise – in fact, those who seek compromise between the two camps are known as “delusionists.” The time may have come to form two competing Wikipedias – to “fork” the encyclopedia, as software programmers would say. Let the deletionists and the inclusionists pursue their separate ideals separately – and let users decide which version best suits their needs.

UPDATE: More here.

20 thoughts on “Deletionists, inclusionists and delusionists

  1. michael webster

    Count me as a delusionist. Why should there only be one wikipedia? There is more than one fork in this road: from tightly edited, to not edited, to entries voted upon, to entries voted on and only the popular ones kept, etc.

  2. Seth Finkelstein

    [insert tongue into cheek]

    The solution is obvious – METADATA!

    The deletionists should propose a W3C/PICS ratings system of encyclopediaousity, which is then applied. Each user can configure their browser profile to best suit their own philosophy, and thus is diversity preserved in an absolute world.

    Mrs. Wormwood: “Calvin, can you tell us what Lewis and Clark did?”

    Calvin: “No, but I can recite the secret superhero origin of each member of Captain Napalm’s Thermonuclear League of Liberty.”

    Mrs. Wormwood: “See me after class, Calvin.”

    Calvin: (speaking in retrospect) “I’m not dumb. I just have a command of thoroughly useless information.”

    [But, honestly, sometimes knowing the secret superhero origin of each member of Captain Napalm’s Thermonuclear League of Liberty is more useful than what Lewis and Clark did :-)]

  3. Mickeleh

    Seth is right, no matter where his tongue is.

    There’s a simple technical bridge between the deletionists and inclusionists which is unavailable to the factions in the abortion, gun control, or death penalty debates. Turn the deletionist filter on, and you have an encyclopedia. Turn it off and you have a wiki.

    Let the deletionists rate the articles. Let the user decide whether to see the full set or the approved set.

  4. Mila

    Isn’t the point of Wikipedia to give all readers the opportunity to contribute to the wealth of knowledge? What is considered “relevant” or “important” information is far too subjective. As another poster mentioned, ratings by, perhaps, expert authors might be a good idea, but changes should be open to all authors.

  5. JohnO

    I agree wikipedia is not paper – therefore the deletionists should be able to have their own wikipedia, and the inclusionists should have their own as well. Obviously the only real difference would be the filter as to what gets seen.

  6. The Rub

    Why not have a certification process? If an entry is up to snuff, it is given a snooty deletionist-approved stamp. If not, then there’s no guarantee.

    That seems the easiest way to include both sides in the development of the single encylopedia.

  7. pwb

    Wikipedia has two other major problems: article length and citation proliferation.

    Counter-intuitively, both lengthy articles and citation proliferation are making article quality worse, not better.

    Long articles are more controversial, contain more inaccuracies and more porrly written.

    Citation proliferation leads to false authority and articles consisting nothing more than a string of disjointed citations.

  8. Chris_B

    Yep. Seth is spot on.

    I suspect Seth is a deletionist. This comes from having been through a few rounds of edit battles with him.

    I personally believe that Wikipedia as it stands now is a junkyard. A bucket of trivia and truthiness. Worth what they charge for it. I think weve already seen what a truly wikified encyclopedia looks like. Only a fork can restore the utility of the ideal from either perspective.

    Like Communism, the Wiki ideals look great on paper but dont work at all in the real world.

  9. Dragos

    “The deletionists should propose a W3C/PICS ratings system of encyclopediaousity, which is then applied. Each user can configure their browser profile to best suit their own philosophy, and thus is diversity preserved in an absolute world.”

    nice tehnical patch BUT

    who will build the filters? Maybe there are several layers of “deletionism”. We will just move the issue to who controls the filter mechanism.

    In the end I think we will move to model already existing in the financial world. It is called “rating agencies” like S&P, Moody’s etc.

    Several groups of people (organized in agencies or not) will develop their own filters. The user will then have to just select the filter that suits him best (maybe for some filters he/she would even have to pay a subscription) or no filter at all

  10. Morgan Goeller

    This sounds a lot like the battle between coherentism and foundationalism, and I would agree that this is one of the primary conflicts of our time. To me, the most interesting thing about this is that a fundamentally human problem is being expressed through technology and information.

    The real problem is that both viewpoints have merit and neither viewpoint can be proved to be absolutely correct and useful all of the time.This causes a huge ammount of cognitive dissonance with the people who are trying to make sense of it all. This tension is broadcast out to others, which turns into conflict.

    As far as a solution goes, I agree with Dragos. For inspiration, we should look to systems that place a premium on the ability to discern good information from bad in a chaotic, fast-paced, high-risk environment. The financial markets are a great place to start, much better than some of the ivory tower solutions and theories that we hear coming out of academia and committee-land.

  11. Bill Higgins

    I agree with the folks who suggest some sort of metadata to indicate whether an article is part of the encyclopedia proper or everything on Wikipedia. Wikipedia already has a system in place for attaching metadata to articles (e.g. “the neutrality of this article is disputed” or “it has been suggested that this article be merged with article X”).

    From a user experience point of view, the way I find articles is by:

    1. doing a term search using either Wikipedia search or Google search scoped to
    2. following a hyperlink in one Wikipedia article to another Wikipedia article (wiki 101)

    In the case of 2 (following links) the nature of the article will determine what other articles show up as links. E.g. if the John Milton article (encyclopedia proper) links to the George Jetson article, it’s likely a “bug” in the Milton article that will be handled by the normal editorial process.

    The case of 1 (Wikipedia or scoped Google search) is a bit more problematic. The Google search, because of its page rank algorithm, will tend to bring the most relevant articles to the top. I’m not familiar with Wikipedia’s search implementation, but it’s possibly not as good as Google at producing relevant results. In this case it may be nice to have a “show only results from the encyclopedia proper” filter option, though I won’t be so bold as to suggest a default setting :-)

  12. Brad

    “the fundamental epistemological crisis of our time: the battle between absolutists and relativists.”

    Don’t be a big tease, elaborate a little. It almost seems like this crisis you speak of may be more important than whether or not the biography of George Jetson gets a URL.

  13. Fletch

    I like thae idea of assigning a filter that the user can turn on or off.

    But I think I’m more in the inclusionists camp with some limits.

  14. Nick Carr

    It almost seems like this crisis you speak of may be more important than whether or not the biography of George Jetson gets a URL.

    It’s all one.

    What’s truly amazing, by the way, is that George Jetson’s grandfather, Montahue Jetson, gets his own entry. Now, that’s cracking the egg at the little end, relatively speaking.

  15. Kevin Kelly

    This grand dichotomy also resembles the ancient and huge gulf between “lumpers” and “splitters” in the biological taxonomic world (and somewhat in the library classification world). The lumpers tend to want to shoehorn a new-found organism (or subjects) into the existing categories (too many categories become a junkyard), while splitters tend to want to fork categories and create new species or subjects (to increase distinction and precision). In the current context of article inclusion and deletion, those who want one sprawling mega-wikipedia (with filters) would be lumpers, and those who want divergent ‘pedias would be splitters.

    While it seems as if this polarization (“you are either a lumper or a splitter”) would be destabilizing (with no one in the middle), in the end, after hundreds of years of back and forth, the extremes are roughly balanced (because there are good arguments for both sides) and steer the system right up the middle.

    That’s my prediction for the wikipedia: that the middle — the “delusionists” prevail.

  16. rifter

    It’s been clear to me for a long time now that Wikipedia will have to be forked. Deletionists have too many excuses to get rid of articles regardless of how much work has gone into them, and articles are routinely deleted out of spite, or because of censorship, or for political reasons, none of which strictly are within the guidelines of wikipedia even though wikipedia itself claims deletionism is an adherence to its guidelines.

    The statement deletionism is a strict adherence to wikipedia guidelines in itself is a condemnation of the way those guidelines are written and followed.

    As has been noted before in other articles, deletionists often delete articles about which they know nothing, or edits that experts have placed there when the deletionist has no expertise, sometimes for the above, and sometimes in a misguided call for notability.

    All this leads to a lot of hard work people put into the encyclopedia being thrown into the recycle bin, and as wikipedia stands there is no cure for it beside forking. The deletionists own wikipedia. Let them delete it all to a stub and let someone else build what wikipedia claims to be (even while stating in its guidelines that it is not): a repository for all human knowledge.

Comments are closed.