About Rationally Speaking

Rationally Speaking is a blog maintained by Prof. Massimo Pigliucci, a philosopher at the City University of New York. The blog reflects the Enlightenment figure Marquis de Condorcet's idea of what a public intellectual (yes, we know, that's such a bad word) ought to be: someone who devotes himself to "the tracking down of prejudices in the hiding places where priests, the schools, the government, and all long-established institutions had gathered and protected them." You're welcome. Please notice that the contents of this blog can be reprinted under the standard Creative Commons license.

Thursday, June 02, 2011

What do I think of Wikipedia?

by Massimo Pigliucci
Not that it matters, but the question keeps popping up in my own mind from time to time, most recently because of this article on college professors who came up with an intriguing idea: assigning their students to set up a new Wiki page on a topic pertinent to their course. One of the students found out the harsh way what it means to be a Wiki author: an entry she had worked on for close to an entire day was brutally edited by “Mean as custard” on the ground that much of it was (allegedly) trivial and promotional. Ouch.
I have my own Wikipedia page — which I did not set up or edit. I checked the page while writing this article (June 1st around 12), and it had been last updated on March 20, 2011 at 14:42. Everything in the entry is indeed correct.
Which brings us to the perennial issue with Wikipedia: how reliable is it? Well, you can begin by checking the Wiki entry on the reliability of Wikipedia, naturally, or do a Google search using the keywords “wikipedia reliability study,” which lists the Wiki entry at the very top of the results (immediately after the specially highlighted entries from Google Scholar). The second entry is from a reliable independent source, the BBC, but it dates from 2005, and refers to the famous Nature study that focused on a number of scientific entries and compared Wikipedia with the Encyclopedia Britannica, the reference librarians’ golden standard in these matters.
The BBC article quotes the Nature study directly: "Only eight serious errors, such as misinterpretations of important concepts, were detected in the pairs of articles reviewed, four from each encyclopedia ... But reviewers also found many factual errors, omissions or misleading statements: 162 and 123 in Wikipedia and Britannica, respectively."
What about Wiki’s own self-referential article? Well, I must admit that it was darn useful! It begins with a fair disclaimer to the effect that “The neutrality of this article has been disputed,” as well as a link to an ongoing discussion on a related “talk page.” The dispute has not been resolved yet (my bet is it never will be), so the disclaimer is still there.
Surprisingly, the most recent comparative study cited by Wikipedia about itself appeared in 2008 in PC Plus (obviously, not a peer reviewed journal). Though the article concludes that “the vast majority of Wikipedia is filled with valuable and accurate information,” the link is actually to a different article, published at TechRadar.com. Hmm.
The Wiki article goes on to discuss the opinions of librarians, academics, scientists, editors of other encyclopedias, and assorted others. There is even a table charting the increasing number of citations of Wiki articles in scientific journals (as uncovered via Elsevier’s ScienceDirect database). The numbers, however — assuming that they are accurate — are not that impressive, with only 614 citations for 2009. And of course, this doesn’t mean that scientists cite Wikipedia for its accuracy or as a reference source (I haven’t checked, but it’s possible that many of these citations are from social science articles about Wikipedia and other online materials).
So, given all of that (and much, much more), how do I use Wikipedia? I must admit that it has become my first stop for double checking things that I’m pretty sure I already know, basic facts or pieces of biographical information about uncontroversial topics or figures, or even occasionally as a general recap concerning subject matters I’m familiar with.
More interestingly, I go to Wikipedia at least some of the times to use it as a starting point, a convenient trampoline to be used — together with Google (and, increasingly, Google Scholar) — to get an initial foothold into areas with which I am a bit less familiar. However, I don’t use that information for my writings (professional or for the general public) unless I actually check the sources and/or have independent confirmation of whatever it is that I found potentially interesting in the relevant Wikipedia article.
This isn’t a matter of academic snobbism, it’s rather a question of sensibly covering your ass — which is the same advice I give to my undergraduate students (my graduate students better not be using Wiki for anything substantial at all, the peer reviewed Stanford Encyclopedia of Philosophy being a manyfold better source across the board).
The problem, as is well known, is intrinsic to the very idea of an anonymously edited system based on crowd-sourcing: the anonymity part. While I’m sure that the overwhelming majority of Wiki editors and contributors are conscientious and only wish to help, anonymity virtually guarantees that a small band of political partisans, pseudoscientists, quacks, or just weirdo hackers will chip away at Wiki’s reliability, particularly when it comes to hot spots like the Arab-Israeli conflict, or the entry for G.W. Bush. (The “Talk” page for the latter helpfully says that “this article is at increased risk of biased editing, talk-page trolling, and simple vandalism.”)
Are there alternatives to or significant improvements over the Wikipedia model? Yes, for instance, Scholarpedia. I know, this is probably the first time you've heard of it, and I must admit that I never use it myself, but it is an open access peer reviewed encyclopedia, curated by Dr. Eugene M. Izhikevich, associated with an outlet called the Brain Corporation, out in San Diego, CA. Don’t know anything more about it (even Wikipedia doesn’t have an article on that!).
Frankly, it seems inevitable to me that Wikipedia is here to stay and that the rest of us will have to do our best to cope with it. I do think that a major improvement would be to do away with anonymity. Just look at this very blog you are reading: death threats, spam and insulting comments disappeared once I moved from open to moderated comments, but the level of discourse — usually remarkably high — occasionally dips because, I suspect, people feel secure behind their pseudonyms, just like the “Mean as custard” character cited at the beginning. If you take responsibility for what you write by signing your own name to it, you put in play one of the most important assets human beings have ever owned since the dawn of our species: your own reputation. Anyone at Wikipedia willing to experiment with this alternative model? Anyone else wishing to try it and give Wikipedia a run for their money?


  1. OK, Massimo, you've convinced me to do away with my anonymity here (although "jcm" are my initials).

  2. Great article, Massimo. I completely agree!

    One of the professors in my department gave an assignment to an upper-division undergraduate class that involved Wikipedia. He had the students divide into groups, choose a topic, and then write independent research papers about the topic. Once the papers were turned in and graded, the groups got together and chose two pages from each person's paper to make into Wikipedia entries. His stated goals were (1) to show students how easy it is to put stuff on Wikipedia, thus pointing out to always double- and triple-check information found there; and (2) to have students contribute to general public knowledge.

    The results were fascinating. Some groups had people take down entries or edit them within minutes (especially entires that mentioned brand names or companies). Other groups' entries hadn't been touched for months. Last I checked on one of the groups' entries, it still had not been touched over a year later.

  3. Besides the practical question of how to use Wikipedia and other related issues reviewed by Massimo in this most excellent post, I'd like to discuss the very nature of Wikipedia in the light of the most likely theoretical framework available for such a thing: the theory of evolution by natural selection (with the help, in this case, of a little bit of artificial selection).
    Processes of natural (or artificial) selection proceed by gradual improvement, but cannot be guaranteed to converge on any "global optimum". At best, they converge to "local optima" which may be inferior to other optima located elsewhere, and finding then difficult to evolve into a different (let alone superior) state except by some "catastrophic event" that completely re-scrambles the cards.
    As a biological example: perhaps having 6 or more fingers to a hand would be better than having our familiar five, but the five-digit solution was found many millions of years ago and has been shared ever since by a wide range of animals. Some of them may have some digits atrophied or fused into one big hoof, like horses, but no one has more than five. It would be very difficult to evolve a sixth finger, and it apparently has not happened. It would necessitate a major set of mutations in many genes for a species with 6 fingers to emerge and millions of years to impose its (supposed) advantages over run-of-the-mill 5-digit animals of the same species.
    At any given time, existing animals would be found already established at some local optimum, like 5 fingers, totally unaware of the advantages of having more, and totally incapable of getting more if they ever wanted.

    Now, one should imagine that a given Wiki entry may evolve, through successive editing rounds, toward a certain "stable" state from which further progress is not achieved within a long time, but that does not guarantee that it does not contain errors. It just means that up to now no one has found any error, or having found one has not cared enough to correct it.
    Moreover, some articles will be at the beginning of the road, still in the process of reaching their stable state, while others are already stabilized.
    The mere amount of time since the article was created is not a reliable indicator of its stage of evolution: some article are on arcane subjects, with a small readership, and therefore the errors may survive for a long time undiscovered and uncorrected. Suppose, for instance, that Massimo's wiki page contains some misleading information about his background, mischievously introduced by Massimo himself for whatever egotistical reason. He won't correct it, and probably nobody will discover it. If it were my own (as yet non existent) wiki page, it would stay there forever just because fewer people know me, and fewer people will read my wiki than Massimo's.
    This suggests the possibility of fitting to Wiki articles a kind of score derived from two predictors: amount of time online and cumulative number of visits received. Both factors would predict the "fitness" or "evolutionary stability" or "relative optimality" of an entry. I do not know (this being far away from my usual pastures) whether such a study has ever been carried out, or such kind of index developed. But it is a question perhaps Massimo can make some comments about, from his twin qualifications as a biologist and philosopher, plus his (here revealed) inclination to think about Wikipedia.

  4. Well, being anonymous is not something bad intrinsically, per se. If anyone can modify a wikipedia entry, for the same reason one wants to manipulate being anonymous, another one can remove the manipulation being anonymous as well.

  5. Wikipedia is editable anonymously, but there is a body of supervisors overseeing the process (which is equivalent to the bit of "artificial selection" I mentioned in my previous comment). Thus hacking, error or tendentiousness may arise (and have arisen), but they would be eventually eliminated by overseers, if the issue is discovered and the attention of overseers is drawn to the issue.

  6. Hector, your analogy might have made more sense if you hadn't described the biological form of evolution as a stochastic process and the Wiki evolution as an intelligent process.

  7. I had not seen your comment before. The same I think can happen in any traditional encyclopedia. There are errors, and not all will be detected.

    On the other hand, adding some indexes will complicate things I think.

    I have seen that polemic entries are tagged as such, and usually, you know when you enter.

  8. The wiki evolution, Baron P, as I said in my comment, is part the result of "natural" selection via spontaneous participants, and part "artificial" or "purposeful" selection via experts in an overseeing committee, who monitor the process and intervene by improving, deleting or modifying specific contributions.

  9. My general take is that Wikipedia is pretty accurate for most things NOT covering current history, politics or big business.

    Big businesses and government agencies have been caught "scrubbing" Wiki entries. And, setting aside the egregious wrongs on the John Siegenthaler page, other living historic figures, especially politicians, have had incorrect info "dumped" on their pages.

    Other than that? Older history, natural sciences, philosophy, etc., Wiki is a good starting point, and more, in all these areas.

  10. Hector, your so-called spontaneous participants have nevertheless come there with intelligence and the purpose to apply it. So your analogy still fails.

  11. Sorry, Baron, I have been apparently not clear enough. Of course each spontaneous wiki-contributor acts with intelligence, but not necessarily with the best information, and some may even be malicious (deliberately misleading, for whatever reason). Natural selection operates through such interventions as actually arise, correcting older entries in several ways and directions, until an entry "stabilizes"; after such point, the entry may be updated or enriched but it is generally stable in its fundamental facts or concepts. Sometimes a visible bias or error persists and has to be corrected by expert monitoring staff.
    This is similar in natural selection. Individuals may act intentionally, e.g. by choosing a specific mate, or pursuing a certain survival strategy, but some of these decisions are validated by success whereas others end in failure, and thus natural selection results (irrespective of the existence or non-existence of "intelligent purpose" in the actions undertaken). In another example, when people undertake a thousand new businesses, all with a profit intent, and all supposedly based on an understanding of the market and expectations about the future, some of them will succeed and thrive while others go bankrupt in the course of a few years (as a majority of new corporations do in the US, in less than 5 years). The fact that the entrepreneurs used available information and their intelligence to take decisions does not impede "natural selection" to operate.

  12. Hector, I know how the Wikipedia selection system works and I know it evolves by an intelligent trial and error process. If you are proposing that natural selection is also an intelligent trial and error process then the analogy works.
    However I am at a loss to understand why you've had to add that intelligence to "take" decisions does not "impede" natural selection. Because then if the situation were analogous, lack of such intelligence would not have impeded Wikipedia from evolving.

  13. Baron,
    it is not exactly an "intelligent trial and error process". As a matter of fact, when you look up some info in the Wiki you do not know whether it is intended to be true or a deliberate lie or misrepresentation (besides involuntary error). It all depends on someone else (or yourself) discovering mistakes or misrepresentations and edit them out (or replace them with correct --or supposedly more correct-- information). The idea is that the many potential contributors will end up with something approaching the truth, even if the initial intention was malicious or the initial information wrong. And even if the contributions do not converge on the truth, there is a back up: the experts monitoring the entries and exercising "intelligent" (and to some extent coercive) power.
    What I meant in the phrase you failed to understand is that natural selection can operate on actions undertaken for "intelligent" or "rational" reasons, as well as on "malicious" or "deliberately misleading" ones, or on totally unconscious events as it happens in biological processes. No matter the mechanism causing the actions, natural selection (in this case with the help of "artificial selection" or "intentional breeding" in the shape of overseers) weeds some out and allows others to proceed.
    My main point is that even with "natural" and "artificial" selection at work, you probably end up with a "passable" entry, i.e. able to pass muster and (so far) survive editing, but you do not know whether other better versions could be available in a traditional encyclopedia.

    To use a Popperian term, and in line with Massimo's post regarding how to use Wikipedia, its entries may play a role in the context of discovery, but in the context of justification they play a very limited role (chiefly pointing to more reliable sources such as peer reviewed journals).

  14. There's an interesting documentary called "The truth according to Wikipedia" that tries to discuss the issues that are involved.

    P.S. I often use duboisist as my signature instead of my name because there are more than one Clinton Freeman, but I've only seen one duboisist on the internet.

  15. As I argued with a younger friend over Wikipedia, you must read everything critically, even peer reviewed sources. The friend was arguing that Wikipedia was useless, mainly because she was instructed not to use it as a source for term papers, in of itself, good advice. But just because info is in Wikipedia doesn't mean its not true (or untrue).

    Great post.

  16. Massimo, I started writing a long response but ended up posting it on my own blog instead:


  17. Oscar,

    the problem with anonymity is that is seems to encourage a race to the minimum common denominator, as I've seen from time to time on this blog (especially before I decided to moderate the entries).


    I do see a problem with your analogy, and I think Baron has a point. Yes, in a sense Wikipedia does work by artificial selection, but since it is selection on ideas - not organisms with standard transmission genetics - it is really much more of a Lamarckian than a Darwinian evolutionary process. This is related to the reason I don't think memetics is very useful: it is a forced attempt to talk about cultural evolution in Darwinian rather than the obviously more appropriate Lamarckian terms.

  18. Massimo,
    I am using "natural selection" in a more general way, as is common for instance in evolutionary economics or evolutionary epistemology, in which the mechanism of transmission is not necessarily genetic as it may include "contagion", cultural transmission and other paths. I am not thinking of memes or memetics in this regard.
    Regarding "natural" versus "artificial" selection in the context of Wikipedia, I used quotes to emphasize the restricted analogical value of those words. I was simply distinguishing between the uncoordinated contributions of a myriad potential editors acting spontaneously (and possibly in a malicious or biased manner), and the more purposeful contributions of the Wiki "council of elders" who have the goals of Wikipedia in mind and try to achieve quality and accuracy on the basis of expert knowledge. In the first category of "cloud-sourcing", even if each contributor does his/her best trying to be accurate and unbiased, there would often be among them contradictory or inconsistent views. From that point of view spontaneous contributions are "anarchic" and may contain an indeterminate amount of error, lack of information or internal inconsistency.
    The cloud, left to itself, would possibly improve the wiki articles over time, even without the intervention of overseers. In an unmoderated wikipedia each article, I surmise, would "evolve" over time until reaching relative stability. This relatively stable state would be a function of (a) the time elapsed, and (b) the attention attracted by the subject (reflected perhaps in the number of visitors). To this, the occasional intervention "from above" would correct some detected errors. These interventions would probably be few because of the huge size of the wikipedia and the constraints (time, expertise) operating on the overseers. Thus, the evolution of each article would converge to locally-optimum relatively stable states, even if some superior versions could be possible.

  19. Hector, I wasn't accusing you of doing memetics ;-)

    The issue, though - that whether one thinks of it in a Darwinian or Lamarckian fashion- is what the fitness function is. Why do we think that anonymous contributors would (necessarily) seek to improve accuracy? What the currency of selection here?

  20. Massimo,
    my very point is that we DO NOT know whether anonymous contributors are really trying to improve accuracy, or are simply mistaken, biased or malicious. But they are subject to the criticism of others, and in most subjects (excepting perhaps some very polarized themes, say Marxism) the majority of the cloud would make constructive contributions.
    If one abstracts from an external "objective" view, which one may see embodied in the overseers, the "fitness function" is mere survival (controlling for length of exposure and popularity of the subject). Survival would be threatened by the number and disparity of views about the subject (akin to the concept of "selective pressure"). "Verisimilitude" (in a Popperian sense) is not per se a factor in the selection process, at least not much more than ideology or other biases. It could only be measured with respect to a widely uncontested agreement of peer-reviewed encyclopedias and other scientific material, if it exists.
    The distinction between local and global optima, which is well known, refers here to the fact that once a "reasonable/agreeable" version has been achieved, probably the text would stabilize, even if potentially better versions could be extracted from the scientific literature. To this "spontaneous" process the exceptional intervention of wiki overseers is sometimes added, operating mainly when some gross error or bias is discovered and reported, and also by inserting requests for sources or additional information.

  21. Hector, the fact that contributors are subjected to other people's criticisms solves nothing unless we have reason to believe that the majority of people follow a fitness function that increases accuracy. Suppose, for instance, that it turns out that the majority of Wiki authors align themselves with liberal political positions. Then we would see a liberal bias naturally emerge from the corrections, a bias that doesn't necessarily bring the entries closer to being factually correct.

  22. Hector says, "my very point is that we DO NOT know whether anonymous contributors are really trying to improve accuracy, or are simply mistaken, biased or malicious."
    But my very point is that the contributors are acting with intent to effect the nature of the body they are building (admittedly not living but an artifact of culture nevertheless).
    And while Lamarck doesn't say much about the connection between evolution and culture, he does allow for a connection there with the experience that we gain from culture.
    Yet Hector's somewhat skewed Darwinian views of evolution seem not to account for intention or experience as contributors to a body that lacks ability to contribute these elements on its own.
    Especially one as dead as Wikipedia would remain without our use of its resources.
    And yes Hector does seem to be turning to something akin to memetics to give life to Wikipedia where blind information givers couldn't.

  23. Massimo, I can imagine cases in which the accuracy of Wikipedia would be improved by allowing anonymous contributions. For example, subjects of oppressive governments may find it difficult to post accurate information to Wikipedia without remaining anonymous.

    Of course, that's a necessarily speculative claim; it would be difficult to gauge the actual effects of government censorship on the content of Wikipedia. Nonetheless, it seems plausible enough to be used in defense of Wikipedia's policy of allowing anonymous contributions.

    However, even anonymous contributions can be assessed by reputation if they are attributed to a consistent pseudonym. If Wikipedia required contributors to register, it would become possible to rank contributions according to the track-history of the contributor's account. This wouldn't eliminate practices like "sockpuppetry," but I think on balance it would motivate contributors to maintain a higher level of discourse, especially if (for example) it were easier to undo contributions from accounts with low reputations. (One excellent implementation of a comparable system can be found at http://stackoverflow.com/.)

  24. Scott, anonymity for those cases could be granted by the owners of Wikipedia, with real names been kept on file. It seems to me that it is impossible to assure consistent use of pseudonyms, considering how easy it is to register / unregister from online accounts.

  25. Yes, I agree with Massimo: the benefits of an open source, collaborative, web-based encyclopedia such as Wikipedia can be had without anonymous authorship, and, should one require it, anonymity could be granted.

    Lastly, one must support Wikipedia if for no other reason than the creators of Conservapedia have expressed as their motivation for countering Wikipedia the liberal bias of Wales et al.

    (Ironically, Wales is a self-described Objectivist and libertarian which, though by no means constitutes 'conservative', is far from being a liberal.)

  26. Massimo remarks: "Suppose, for instance, that it turns out that the majority of Wiki authors align themselves with liberal political positions. Then we would see a liberal bias naturally emerge from the corrections, a bias that doesn't necessarily bring the entries closer to being factually correct."

    But this is also true in other contexts, in which we trust to tend toward accuracy, e.g. peer review of subjects with high political or academic stakes such as climate change or (some decades ago) tobacco smoking or systematic zoology. Cliques of like-minded authors, peer reviewers and journal editors, often exchanging roles over time, may produce the same effect, even more effectively since there would be no open-sourced criticism as in wiki environments.

    In my view, open or cloud environments tend to incorporate a diversity of views and information, often criticizing and correcting previous contributions, and thus unsupported claims are often (though not always) driven away by weight of contrary evidence (if the issue is not resolved, overseers intervene).

    It is most frequently the case that wiki pages "stabilize" without further contributions trying to modify them, which is an indication that nobody has found further defects in the current version, at least so far. This belief is further reinforced if (a) a long time has elapsed since the item was created or last modified in any substantial way, and (b) the wiki item has a wide readership as evidenced by continuing and numerous flow of visitors. Hence my hunch that time elapsed and flow size are concurrent "internal" indicators (albeit not a guarantee) of "robustness" in wiki content. "External" indicators might be for instance those reflecting the utilization of wiki content, though this is difficult to assess since citing Wikipedia is not easily accepted in other contexts, e.g. the context of peer reviewed science.

    This process of gradual convergence to a steady state also operates in a "journal cloud" within specific scientific disciplines or fields. Diverging views compete among them (albeit in different journal following different "schools") until the controversy is substantially over with the victory of one side or another, or (more likely) their common replacement by a superior and more comprehensive framework.

  27. But there are major differences with the cloud of scientific journals, including the quality of expertise (and the variance in that quality), as well as the above mentioned problem with anonymity.

  28. Of course there are differences, Massimo. Nobody says they are the same.
    Anonymity certainly encourages a sense of impunity, but so does the anonymity of peer reviewers, who may for instance obstruct the publication of a possible refutation of their own work. This may be checked by an honest journal editor, but see my previous clique hypothesis (of which several examples can be cited). Such "conspiracies" do not last forever, but can derail the progress of knowledge for a considerable time.

  29. Massimo,

    I still do not see the problem. You moderate, so that's it. You moderate the messages, not the messenger, right? If the message (content) is not suitable, you do not post it. That's right. But you do not ban a message because your archenemy wrote it.

    Wikipedia actually works like you do, moderating it. I don't see the problem.

    In practical situations, you can track the messenger up to some extent at least, provided that you have knowledge about internet.

  30. What this kind of Darwinian analysis leaves out is policy. It's not just raw plugging away and reverting leading to a natural bias. If someone is editing to "push POV" in Wikipedia terminology, there exist ways of bringing them to book.

    I also fail to see how this kind of 'natural bias' type analysis is solved by lifting anonymity. Biases don't suddenly go away if everyone uses their real name. There are plenty of people who edit under their real name who are biased, problematic, push their point of view (often completely insane or pseudoscientific).

    Real name policies are a solution looking for a problem. You can have real names only and you still end up with all the same problems you have without real names only: you still need to deal with those problems. And Wikipedia seems to do so reasonably well.

  31. Hector,

    the more you develop your analogy the more it becomes a disanalogy. Peer review is never entirely anonymous, as you point out, which is what I'm suggested for Wikipedia. Moreover, it continues with public after-publication criticism, which is always signed. Indeed, in Victorian England people could publish anonymous reviews of books and pamphlets, and the discourse got nasty and not too useful as a result.


    moderation is different. That's right, I don't edit entries, but the whole point of Wiki is that there is constant editing and counter-editing, so the two systems are not really comparable.


    as I said, there is plenty of evidence that anonymity simply makes people reckless and nasty. What exactly is the argument *for* it, other than those specific cases where someone's life may be threatened, and which I've covered above?

  32. Massimo,

    The argument on Wikipedia for anonymity is simply that it works and that without it there wouldn't be as many contributors.

    Without anonymity, people would be less able to edit articles about controversial topics: there are still plenty of people who wouldn't want others to know they have been editing the article on sexual fetishes or unpopular historical figures (Hitler?). Perhaps you are the only atheist in Smalltown USA and don't want your neighbours knowing you've been editing the article on Richard Dawkins. Or perhaps you want to sneak on at work and edit articles on Pokemon without your boss knowing that it was you. There are plenty of reasons for anonymity.

    Anonymous users aren't treated with the same respect as logged-in users, and logged-in users with a fair degree of history get treated better. The real name of a user doesn't actually affect much other than their job. Take someone like Newyorkbrad, who is a prolific Wikipedian and has some important positions of responsibility on the site (he's a member of the Arbitration Committee, a checkuser and an oversighter - he can basically wipe edits off the face of the planet so nobody can see them anymore). For many years, he was just Newyorkbrad but was outed: he's a lawyer in New York called Ira Brad Matetsky. It changes things for Newyorkbrad, but it doesn't really change much for how good faith editors deal with him. If he screws up, he will still end up being taken through the same dispute resolution processes.

    As for the people being "reckless and nasty": there are policies and a community of enforcers in place to deal with that. If you go around being mean to people, whether you are a logged in user, an anonymous user, or a real name, you will get blocked fairly quickly.

    The argument for anonymity is that it works for Wikipedia: of the completely anonymous edits on Wikipedia made by users who aren't logged in, over 80% of them are legitimate. Restricting editing to the registered or those with real names only will reduce good faith, positive editing far more than it will stop vandalism or negative editing. Attempts to build Wikipedia-like projects with real name only editing have been tried and failed.

  33. Massimo,

    I get your point. But I stated my argument previously. If you find it nasty, reckless, unpolite, uncorrect, not accurate, etc. nothing stops you from changing it. You can change it. As simple as that.

    I can even accept your points as good ones, but me and other people have made another good ones.

    Further, you say that there is already an alternative. So, why both ways can not coexist?

  34. For computer science stuff, I've found wikipedia quite good. The rationale, I guess, is that technical things tend to be done by people who are really into it, and they're seldom vandalised. (who would know how?)

    As for topics where I'm unfamiliar (pretty much everything else), I like wikipedia for a first glance into the topic, but from there I use things like SEP or iTunesU or the library to find better material on the matter. It's my first stop, but usually not my last.

  35. Massimo, you said "anonymity for those cases could be granted by the owners of Wikipedia, with real names been kept on file." That solution, it seems to me, would not scale well, and scale is exactly what Wikipedia is designed to leverage. According to these statistics, more than a million accounts have contributed 10 or more edits; if even one percent of those require special handling by moderators, that special handling will likely become a bottleneck, limiting the efficiency of Wikipedia.

    This is not to say they shouldn't do it, but only that I doubt the result would really be open any longer, unless the working budget of Wikipedia were to increase by an order of magnitude or so.

    Also, I think what you say about "consistent use of pseduonyms" applies just as well to consistent use of real names. False names, email addresses, and physical addresses are far easier to invent than to vet. I suppose you could require contributors to link their wikipedia accounts to bank or credit card accounts; that would be one relatively sound way to automatically vet identity. But again, at that point, haven't you radically altered the constitution of Wikipedia?

  36. "...the link is actually to a different article, published at TechRadar.com. Hmm."

    If you actually look at the bottom of the second page of that article, you will notice this: 'The full version of this article is published in PC Plus magazine, issue 268.' Also, PC Plus is affiliated with TechRadar, so I think this isn't much of a problem. They just linked to the incarnation which was/is available on the web.

    As for anonymity; it cuts both ways. If you have to have a reputation to be taken seriously, there is a higher barrier to entry than there would otherwise be. Aside from that, it's in my opinion a much better rule to just judge the statements made on their own merit as opposed to the reputation of the author.

    I think, however, that on a different note, this sort of discussion has had the positive affect of making people more aware of a reliability problem inherent in all media. If you question the reliability of wikipedia today, you have a larger chance of questioning the reliability of a news channel tomorrow and of a newspaper a day after that and in the end be forced to critically evaluate the never absolute reliability of any information source.

    Rok Us

  37. Anonymous editors are good for fixing occasional dumb little mistakes, like spelling or Wiki syntax errors. Someone might not consider it worthwhile to make an account just to delete some errant double brackets. I know that I've fixed minor stuff like that on Wikipedia, and I've never bothered to get a full Wiki account.

  38. A thought about Wikipedia vs Stanford Encyclopedia of Philosophy

    1) Stanford misses articles on many important concepts or theorists - for example, Emile Durkheim

    2) Stanford articles are good, but Wikipedia ones are getting better. Which is to say, articles on Wikipedia are rated from stub, start, C, B, to Good Article, A and finally, Featured classes. Majority of Wikipedia articles are around start-C classes, in my experience. But how good are Stanford article? I'd say somewhere around B/GA class.

    Recently, I improved the content of Wikipedia's articles about Marx, Weber and Durkheim from C-ish to GA class. I used Stanford as one of my sources, but at this point, I'd very much direct my students to Wikipedia articles about those theorists than to Stanford's, which I think are now (for those theorists) less helpful than the Wikipedia one.

    And of course, for Durkheim, they'd have no choice anyway.

    Eventually, once Wikipedia's volunteer editors read, digest and incorporate (cite...) Stanford's articles into Wikipedia, Stanford product will become a historical memento to the web of 1990s-2010s. "This is a website from the days Wikipedia was not as good as it is now". I wonder when I will be able to say that about Stanford? 2015? The day is coming...

  39. I doubt it. Academics will always put a (justified, in my opinion) premium on articles written by experts and peer reviewed by an expert community. Imagine applying the same logic, for instance, to the medical literature. I really don't want a bunch of anonymous people with unknown qualifications to be the source of my doctor's knowledge about my health...

  40. "Are there alternatives to or significant improvements over the Wikipedia model? Yes, for instance, Scholarpedia. I know, this is probably the first time you've heard of it, and I must admit that I never use it myself ..."

    See, that last bit sorta belies the first bit. You like the idea of Scholarpedia but your revealed preference is for the reality of Wikipedia.

    It's like comparisons of Wikipedia to Britannica - the trouble with Britannica is that approximately no-one ever actually opened it past high school, if their high school even had a copy. But Wikipedia is used by everyone every day.

    So the apposite comparison is not Wikipedia versus existing encyclopedia approximately no-one uses, or Wikipedia versus hypothetical notion, but Wikipedia versus nothing.

  41. > but Wikipedia versus nothing. <

    Not exactly. As I mentioned, for philosophy articles I *never* look at Wikipedia, I much prefer the Stanford Encyclopedia.

    Besides, let's not forget that Wikipedia isn't the result of inevitable and immutable laws of nature, it didn't even exist until a few years ago, it may not exist in a few years...

  42. "Not exactly. As I mentioned, for philosophy articles I *never* look at Wikipedia, I much prefer the Stanford Encyclopedia."

    Yes, the SEP is damned fine. Be better if it was proper free content, though. CC by-sa includes due credit. I wonder if anyone from WMF has approached Stanford. I know one of the big motivations for mathematicians on Wikipedia was to have a good source for mathematics without Wolfram - not being freely reusable is the sort of thing that motivates free culture partisans to replace you.

    I wonder how long before philosophers decide they have to get their area's Wikipedia articles into shape, even despite Wikipedia's chronic inability to keep idiots out of experts' (or anyone's) faces. I say "how long" rather than "whether" from observing academic disciplines one by one (mathematics, computer science, most of the hard sciences, and just now psychology) realising they have to come to the mountain, idiots or no.

    (For those seeing a need to do this in their field, but daunted by the inexhaustible reserves of stupidity on the Internet and thus on Wikipedia: a good way to do this is (1) get a high-calibre academic posse together (2) contact the Foundation (3) get it in the WMF blog. May alleviate the worst of it. We love this stuff.)

    "Besides, let's not forget that Wikipedia isn't the result of inevitable and immutable laws of nature, it didn't even exist until a few years ago, it may not exist in a few years... "

    True, and true :-)


Note: Only a member of this blog may post a comment.