Slashdot is powered by your submissions, so send in your scoop


Forgot your password?

Is Wikipedia Failing? 478

An anonymous reader writes "A growing number of people are concerned about where Wikipedia is heading. Some have left Wikipedia for Citizendium, while others are trying to change the culture of Wikipedia from within. A recent essay called Wikipedia is failing points out many of the problems which must be solved with Wikipedia for it to succeed in its aim of becoming a reputable, reliable reference work. How would you go about solving these problems?"
This discussion has been archived. No new comments can be posted.

Is Wikipedia Failing?

Comments Filter:
  • Editorial board... (Score:5, Interesting)

    by BWJones ( 18351 ) * on Wednesday February 14, 2007 @09:44AM (#18010720) Homepage Journal
    What can be done to change the system?

    Now that Wikipedia has reached a critical mass, the time has come to establish a trusted editorial board that can vet articles to established experts in the field of subjects. This board could then also solicit articles by experts and find other wikis that host specialized information to link to the common Wikipedia. This will prevent much of the vandalism and uninformed disasters that seem to befall certain subjects or topics when they are edited by people who are not competent to be making edits in certain topics. As a professor in the biosciences, I've seen more than one article/entry on Wikipedia, written by an expert in that field that has been absolutely, shamefully and quite inaccurately edited or altered by well meaning individuals that absolutely have no idea what they are doing/saying.

  • by geoffspear ( 692508 ) on Wednesday February 14, 2007 @09:55AM (#18010820) Homepage
    Or maybe they already know exactly what will happen if they tried this because they already did. Nupedia was an unqualified disaster.
  • by Dogtanian ( 588974 ) on Wednesday February 14, 2007 @10:14AM (#18010994) Homepage

    Just edit the Wikipedia is Failing article to say it's fixed.
    You're obviously joking, but it's my sincere belief that one of the dangers facing Wikipedia is where the community, and the defence of Wikipedia from criticism becomes more important than the integrity of Wikipedia itself. This is an inherent risk with anything community-based; superficially, the effort is to support and protect the project (and those taking part may well still believe this), but in reality the loyalty is to the community or team, even at the risk of the stated aims of the project.

    Another problem is edit decay, often exacerbated by Wiki-masturbation. What do I mean? Basically, edits are normally on a small scale. Lots of individual small-scale edits do not make a big article; on the contrary, I've copyedited at least one article that was fine on a sentence-by-sentence level, but messed-up, disorganised, verbose and unreadable because no-one had bothered to step back and look at the article as a whole. Thus many small edits (even if individually useful) tend to increase the structural decay of an article, and make it hard to see when something useful is being lost.

    A problem occurs when minor edits are made, or an article changed several times, with little ultimate point (hence "masturbation"). It's in these sorts of pointless changes that good work gets lost for no real purpose. In such cases, it may make sense to go back to an earlier version, compare any major changes, find out why these have happened, and if there seems to have been no justifiable reason for them, to revert some or all of the article.

    Should the aim of Wikipedia be change? No. The aim of Wikipedia should be changability; a subtle but very important difference. Unlike evolution in nature, we can go back as far as we like if an earlier version is better, and there's no reason we shouldn't do this. Some subjects inevitably date, necessitating change; but many do not. Changeability is about having the choice, and that includes the choice of saying "actually, the earlier version *was* better".

    The WP article actually covers some similar ground to the above, but both are issues that had been on my mind for a long time beforehand.
  • by Stumbles ( 602007 ) on Wednesday February 14, 2007 @10:14AM (#18011000)
    The day they allow the "Everywhere Girl" to remain posted is the day I will change my mind about them.
  • No (Score:3, Interesting)

    by Kohath ( 38547 ) on Wednesday February 14, 2007 @10:15AM (#18011012)
    The article is just a bunch of complaining.

    Wikipedis is failing to be exactly what the article writer wants it to be. It's succeeding perfectly in being what it is.

    The article writer values his opinion more than reality. He's undoubtedly disappointed a lot.
  • by hey! ( 33014 ) on Wednesday February 14, 2007 @11:01AM (#18011550) Homepage Journal
    Actually, since the system keeps all the edits of the articles, why not make it possible for an editorial board to cryptographically sign an article version?

    In fact, why have a single editorial board? Why not let anybody set up an editorial board, and create a virtual wikipedia over the wikipedia? You could search only the RNC blessed versions if you wanted.
  • by Nelson ( 1275 ) on Wednesday February 14, 2007 @11:05AM (#18011594)
    So the relatively low number of well written articles compared to the total is that sign of this failure?

    Don't get me wrong but that really misses the point. Take, for example, Voltron. I can plug that into Britannica and Wikipedia. Britannica doesn't know who or what Voltron is. Wikipedia has a fairly detailed explanation. Accurate? Well written? I'd be shocked if that article fell in to the 2000 or so "well written" articles. I doubt it's verifiable in any credible way. Also, I don't see Britannica ever having an article that talks about Voltron. It's not a scholarly article because it's not a scholarly subject. That doesn't change the fact that when I couldn't remember the names of the pilots of the lions and for whatever reason I wanted to remember them, wikipedia provided an answer and a whole lot more where most other sources wouldn't provide anything. That's the beauty of it.

    I don't know that you should read a candidates wikipedia article and decide off of that alone if you will vote for them. I don't know any single sources that you should use for that. I also don't know that I'd read about global warming on wikipedia and use it as an exclusive guide to your own beliefs on it; again, there is no good single source on such an important subject. However if you do want to look up who's driving for each F1 team next season or Voltron, or what looks like well over a million other articles, wikipedia is probably ok. The alternative is either nothing or you scour the web for some hobbiest that cares enough about Voltron or whatever to put up a webpage of his own and provide a detailed document on it.

  • by cyclomedia ( 882859 ) on Wednesday February 14, 2007 @11:07AM (#18011624) Homepage Journal
    Like in software there's usually a Stable version, even if it's quite old and a Beta version, i'd go so far as to suggest that Wikipedia pages should have three versions

    1. The Stable Page - and THIS should be the default at .../wiki/The_Page
    2. The Candidate Page - The candidate to become the next stable page
    3. The Current Page - Up to the minute revert war free for all

    Both [1] and [2] are essentially historic versions of the page but linked to from handy labelled tabs and some kind of moderation/voting system can elevate a page from current to beta to stable.

    obviously newly created articles would only have one or three versions and these would filter across all three until a moderator/vote decides to split the article into the aforementioned modus operandi
  • by Anonymous Coward on Wednesday February 14, 2007 @11:19AM (#18011792)

    Neutral point of view confounded by fact that most people here are fairly left wing.

    That is an understatement and raises an interesting point. I don't know why Wikipedia has so many editors who are literally communists but I can observe that once the people putting pressure on an article's direction reaches a critical mass of one political orientation there is no hope of "neutral point of view". All articles relating to my country reflect the views of the most left wing 3 to 5% of the population because their kind make up around 80 to 90% of editors of those articles. They are even able to force the naming of articles in styles which literally nobody uses in order to conform to their adgenda.
  • by ronanbear ( 924575 ) on Wednesday February 14, 2007 @11:24AM (#18011870)
    Karma would be doled out by an algorithm that would assess your edits based on how much, how often and by whom (their karma) your edits get edited in turn. If you troll you end up with karma in the toilet and your edits are brought to the attention of other editors who "metamod" your down. If other people who watch the page give positive feedback on your edits then your karma improves and other editors will be quicker to trust you. The system wouldn't be foolproof but it wouldn't need to be. All it would need to be is more efficient at correcting bad edits and retaining good edits than the current system. Trolls would be caught quicker and it would take them longer to do less damage. At the moment a large number of trolls just replace entire articles with a single line. This sort of edit could be fixed automatically so editors don't have to waste their time doing it. As articles become more "mature" it should become harder to make big edits to them.
  • by squizzar ( 1031726 ) on Wednesday February 14, 2007 @11:34AM (#18012020)
    I agree with the parent in that the reason for problems is the equal importance of each potential editor. In order to become a reliable source of information I think a model similar to that used by a lot of open-source software projects can be used, whereby a topic (or group of topics depending on size) is overseen by a maintainer - someone who is an expert in the field, or a group of knowledgeable persons. This would allow for articles where the majority of the work is contributed by one author to be maintained primarily by that author, and articles with greater controversy to be maintained by a committee.

    The trick to making it work would be to ensure that the committee members are relatively unbiased towards their topic, and that they will accept and reject comments based on the quality of their supporting arguments, research and references. They would also frame them responsibly within a topic - presenting arguments or opinions as such, and not simply stated as fact.

    Similarly to the Slashdot moderation system, the actions and opinions of committee members would be meta-moderated by the community. It would be important that for controversial cases (if not all cases) a statement with the reasons for a decision made by each committee member is presented, and based on that community members can vote on whether they feel the approach taken by each members is appropriate and correct. Possibly this would be undertaken by a select group of respected policy makers in order to ensure an objective decision is made.

    In the cases where a single maintainer is responsible then the editing decisions should be made by that person. Clearly a complaints and resolution system is required so that lazy, biased or ineffective maintainers can be replaced, and where sufficient controversy exists, an article is upgraded to requiring the committee based approach to maintenance.

    The idea is to allow user's additions a fair chance of being added to the encyclopedia, whilst reducing the amount of unwanted additions and edits. Of fundamental importance is ensuring that the moderation system itself is open, allowing a clear decision making process that can be scrutinised for unfairness. Primarily that a proposed addition put forth by a user, supported by references,research, expertise etc. is given priority over the opinions of a committee moderator if the committee moderator cannot provide suitable reasons (again, research etc.) that refutes the addition. The meta-moderation system would be in place to highlight bad decisions if they are made, and possibly to suggest resolution (perhaps the committee-moderator has to stand down, or perhaps both arguments should be presented in the article to provide completeness). A system of voting in new committee members (perhaps for expressed expertise or substantial contribution) to a topic would be required. Again this should probably be performed by a meta-moderation group that will objectively decide whether that person should be included at the top of the decision making process for an article.

    This in itself would provide a fairly comprehensive system, hopefully filtering out the unwanted information from that which should be included. By putting someone's name (or a group's) to an article, there is an element of integrity that the persons themselves will wish to maintain - courting controversy through bias and unjust decision making will reduce the influence of that person on the article, and in the community as a whole, a form of self-moderation in that in order to contribute at a top level, the contribution must be seen as appropriate by the community.

    I also feel that the first initial hurdle (of getting an article past a maintainer) would reduce a significant amount of the damage done to articles.

    Simply allowing everyone to edit where they feel is not a viable model for long term success - a more comprehensive approach that pays more attention to the relevant facts and research behind an article is needed.
  • by realisticradical ( 969181 ) on Wednesday February 14, 2007 @11:57AM (#18012318) Homepage
    I would expect that giving users with high karma the ability to censor Wikipedia would actually cause more bias because power would be more limited to a specific group.

    Also, I looked up George W. Bush, christianity, and the Republican Party and they're all very good, factual articles. There seems to be a lot of misplaced criticism of wikipedia for being biased, incorrect, and misinformed.
  • Re:Agreed (Score:5, Interesting)

    by bcrowell ( 177657 ) on Wednesday February 14, 2007 @11:59AM (#18012344) Homepage
    in fact the science articles are some of the best.
    Generalizations are always dangerous, but IMO science articles on WP tend to be some of the worst. I've worked on a lot of the physics articles. (I teach physics at a community college.) Typically they fail to put things in context, use too much math too early, and focus on irrelevant equations and derivations rather than the important concepts. I think this is symptomatic of what's wrong with WP in general: articles tend not to rise above a certain (low) level of quality, because of random, disorganized edits. Also, although many people on WP are good writers and explainers, and many are knowledgeable about their subjects, there aren't as many people who are good at both, and the structure of WP doesn't work well to help them cooperate.
  • Re:Well... (Score:2, Interesting)

    by bcrowell ( 177657 ) on Wednesday February 14, 2007 @12:03PM (#18012404) Homepage
    Well, since I have to create an account with Citizendium just to look at the articles, I'm not too worried about it overtaking the Wikipedia just yet.
    Yup. From the information I've gleaned from the outside, it's having all the same problems as Sanger's original Nupedia project (in which I participated for a while), including a dysfunctional server/software setup, and a lack of transparency.
  • by MagnaDoodle666 ( 783430 ) on Wednesday February 14, 2007 @12:07PM (#18012486)

    A Slashdot-like karma system where editors with high karma can block those without from editing thei stuff? I actually really like this idea... A system where expertise can have a karma ranking system through either qualifications or community mediated promotion through contribution. This would allow experts in their fields to contribute without fear of having their contribution savaged by those who may not know what is going on.
    How about creating a system where experts get special accounts, recognizing their credentials. Wikipedia could have a system where they confirm that you are indeed a professor at this university, working on this field. The text that is contributed by experts could then perhaps be marked a different color on Wikipedia's edit page, so that everyone knows that they are actually editing out an expert. It's ridiculous that there's no way mechanism right now to recognize experts.
  • by Garse Janacek ( 554329 ) on Wednesday February 14, 2007 @12:08PM (#18012498)

    Well, its use is phenomenally widespread, and in many fields it is one of the best places to look for a general survey -- even in highly technical fields (for example, there are many times I've gotten better explanation of some topic in higher mathematics from Wikipedia than from my textbooks). I'm almost certain some of these were not included in the count of 1700 "good articles," just because if you only have 1700, having dozens of them on areas of math that 99.99% of people will have no interest or need for seems unlikely (how many people do you know who need to read about higher cohomology?). Thus, the "good article" status is almost certainly not a real measure of how many good (in the English sense, i.e. the opposite of bad) articles there are on Wikipedia. While having the "good article" distinction is useful since it can direct people to especially polished material, it is not at all a good idea to make the logical leap and conclude that all the other articles are bad.

    There, that's a (credible, I hope) argument that Wikipedia is not failing, followed by a partial refutation of the article that it is (I don't have time for a more thorough discussion). So the answer to your question is yes -- now let's get back on topic and leave aside the FUD :-P

  • by KalvinB ( 205500 ) on Wednesday February 14, 2007 @12:33PM (#18012852) Homepage
    MediaWiki is slow and therefore demands more resources than are actually necessary to do what they are doing.

    This new project (Citizendium) is being developed on a fast server which hinders the ability to optimize code. Smart people start with low cost equipment and optimize the heck out of it to make it work for as long as possible. Only then do you start spending more on faster systems and more bandwidth. You don't spend rediculous amounts of money up front for resources you have no use for. You first find ways to reduce the amount of resources needed and only then do you increase resources.

    People don't understand these simple concepts and that's why money is wasted and projects go bankrupt.

    Cubia is starting simple. The goal is to see how complex it can get before a $7 GoDaddy account is insufficient to run it. The next step is user submitted articles.

    Citizendium has the oppositite goal: see how much money they can waste until the demand matches the resources and then blow more money on more resources.
  • Re:Agreed (Score:5, Interesting)

    by jpflip ( 670957 ) on Wednesday February 14, 2007 @12:36PM (#18012910)
    I'm a physicist as well, and I'd say that Wikipedia's science articles are generally quite good, though not always very pedagogical. I find that Wikipedia is among the best places to get an up-to-date introduction to (or at least the basic gist of) to some topic that I'm not fully familiar with, even a very technical one. I agree that far more work is needed to make Wikipedia's science articles as complete and pedagogical as they should be and that authors sometimes get a little too pedantic or sidetracked. Nonetheless extensive contributions from experts make it a surprisingly good starting point for real science. Again, in general - there are certainly plenty of exceptions.
  • by omnilynx ( 961400 ) on Wednesday February 14, 2007 @01:16PM (#18013526)

    At the moment a large number of trolls just replace entire articles with a single line. This sort of edit could be fixed automatically so editors don't have to waste their time doing it.

    That actually does happen largely automatically. I've done some counter-vandalism work, and there are bots that compile lists of likely vandalism and scripts that automatically roll edits back. All editors have to do is scan the list to double-check the bot and click roll-back on the proper entries. It takes about five seconds per revision. Two or three causal editors can easily combat normal levels of vandalism.

    I think the real problem with Wikipedia at the moment is that the articles are in a constant state of flux, so readers can never be sure that the particular version they're reading is accurate. I'd recommend a 'canonical' version that's been examined and decided by consensus to be accurate. The main page would be the current version, but there'd be a prominent link to the latest canonical version. This has already been partially tried, but the goals and methods were different.

    I also like the user karma system. But it would be a big change, technically and social, from the current system.

  • No change required (Score:3, Interesting)

    by sugarmotor ( 621907 ) on Wednesday February 14, 2007 @01:16PM (#18013532) Homepage
    I submit wikipedia is doing just fine. After six years why would you expect more? 65,000 Britannica Micropeadia articles of size ca. 700 words, compare quite well to over a million wikipedia articles. Also look at how long the Britannica took: First edition of ca. 2400 pages after 3 years in 1771.

    If anything, the wikipedia community should take a break and relax for a while.


    References tannica [] ca []
  • Re:Agreed (Score:5, Interesting)

    by 0rionx ( 915503 ) on Wednesday February 14, 2007 @03:21PM (#18015064)

    I'm amazed it took this long for someone to point this out! Many of the articles in fields such as biology, geology, history, philosophy, etc that tend to have political/religious controversy surrounding them are often not of the highest caliber. Articles in non-controversial fields, especially computer science and mathematics (IMO), are often, as the previous poster stated, extremely well written and highly detailed. Want to learn about the traveling salesman problem? The related Wikipedia article [] is almost ten pages long with graphs and detailed explanations, cites 16 qualified sources, and provides more than a dozen external links for further reading. How exactly is that trivial?

    I wish I had saved some mod points for a +1 Underrated...

The shortest distance between two points is under construction. -- Noelie Alito