Feeds:
Posts
Comments

Posts Tagged ‘Main’

Wikipedia “hoax” not actually a hoax

By Nate Anderson | Published: November 06, 2006 – 03:55PM CT

A long-standing article on the little-known “NPA theory” of personality was recently deleted from Wikipedia after editors realized that the piece had been authored by the creator of the theory, a doctor named Anthony Benis. Critics are already pointing to the case as an example of Wikipedia’s problems, arguing that allowing a “hoax” article to persist for so long only shows how open Wikipedia is to “vandalism.” But a closer look reveals a more complicated story.

NPA theory suggests that human personality is a combination of three traits: narcissism, perfectionism, and aggression. Each one of these traits is coded for by specific genes, and the process of personality formation therefore follows the long-established rules of genetics. It’s a theory developed by Benis but not given much credence by the scientific community at large. Even Benis, when contacted by Ars Technica, acknowledges that his idea is not “conventional wisdom.” Still, it’s one that he believes in and wants to disseminate.

Back in May, Benis says that he was approached by a Wikipedia user who had seen Benis’ website and asked him to do an entry on NPA theory. Benis tells Ars that he agreed to do so and that he and his associates “tried our best to produce the finest possible article.” This article was posted to Wikipedia, where it remained undisturbed for many months.

In late October, though, a user brought attention to the fact that the article had been written and edited almost exclusively by user “ABenis,” and questions were raised about whether this was appropriate. Wikipedia has strict rules against vanity articles, and it didn’t help that an article was created about Benis himself that some argued was his own work. Benis says that a Wikipedia editor actually authored the entry, but acknowledges that it “took the appearance of a vanity entry.” After a lengthy discussion, both articles were eventually deleted—but on the grounds that they were not “notable” enough to merit inclusion in Wikipedia.

After the deletion decision was made, Benis took his article on NPA theory and put it up on its own website (it is also still available on Wikipedia France).

The entire affair seems hardly worthy of mention, except that it provides an illustration of how quickly stories like this one can be sensationalized. One of the bloggers who first described the incident eventually admitted that the writeup was “sensationalist”, and he awarded user ABenis a purple barnstar “for blogging carelessly.”

But the case also raises a pair of interesting questions. The first concerns the Wikipedian concept of “notability.” The proposed guidelines for academic notability no doubt keep thousands of academics from beginning brief vanity articles on themselves or writing up descriptions of their own pet projects, but Wikipedia is already stuffed with trivial topics—in fact, that’s part of its charm. Where else can you learn about the 1980 computer game The Prisoner, read up on sixteen-year old Indy rapper Grand, or find hundreds of words of plot summary for Robin Hood: Men in Tights?

The other question is whether interested parties ought to be contributing material to the site. On the one hand, people like Benis are in the best position to know about their own line of work. On the other, obvious conflicts of interest arise, and users would be rightfully skeptical of the information’s accuracy. But where does the line get drawn? People already contribute information to Wikipedia for topics they care about. Knowing when they are too involved in an industry/field of study/fan club to be objective can be almost impossible work.

Wikipedia in the news

Wikipedia continues to make headlines, but most of those seem to concern the site’s accuracy, massive size, or safety from viruses. That’s right—the BBC is reporting that the German Wikipedia was vandalized to include a link to a virus, though such events are simply inevitable outcomes of the Encyclopedia’s decision to let all users post material. Showing that the moderating system does a good job of catching this type of thing, Wikipedia users quickly removed the offending link.

Wikipedia continues to grow, but that growth may be slowing. In October, Wikipedia added 49,220 articles to its database, the fewest of any month in the last year. For the project as a whole (including other languages), last month was the lowest one since April.

The site has become a staple resource for students, often to the chagrin of teachers, but a (totally unscientific) look at three sample articles shows that Wikipedia can hold up quite well to scrutiny. The Chronicle of Higher Education asked three college professors to critique articles chosen from their fields of study.

The article on Aldous Huxley’s Brave New World earned a B-, with Dr. Peter Firchow noting that “it’s clearly lacking an editorial hand to smooth out the inconsistencies.” “African-American Civil-Rights Movement” was given a C by Dr. Doug McAdam, who notes that he “would not direct a student to this site,” but does say that “it isn’t bad in a lowest-common-denominator encyclopedia sense.” Finally, “Flow Cytometry” was examined by Dr. J. Paul Robinson, who gave the article an A. Robinson appreciated the main entry, but noted that the subsections were less accurate. “I went in and actually made some corrections to these,” he said, telling himself, “I can’t leave these and let the record stand like this.”

Discuss Print
Filed under: , , ,

Read Full Post »

Britannica begs to differ on Wikipedia’s accuracy

By John Timmer | Published: March 23, 2006 – 10:57AM CT

Wikipedia has the potential to bring unprecedented detail and accuracy to its articles, given that it can be peer reviewed by an entire world’s worth of experts. But there is a gap between that potential and reality. Ars has covered a variety of potential and actual issues with Wikipedia in the past, ranging from hoax pages to manipulation of entries for political ends. Do these problems represent exceptions or systematic flaws in the material there? A potential indication of the quality of Wikipedia came when the respected scientific journal Nature submitted articles from both Wikipedia and the Encyclopedia Britannica to experts in the appropriate fields for evaluations.

The results of the evaluation were very positive from a Wikipedia perspective:

However, an expert-led investigation carried out by Nature — the first to use peer review to compare Wikipedia and Britannica’s coverage of science — suggests that such high-profile examples are the exception rather than the rule. The exercise revealed numerous errors in both encyclopedias, but among 42 entries tested, the difference in accuracy was not particularly great: the average science entry in Wikipedia contained around four inaccuracies; Britannica’s, about three.

Based on Nature’s description (in Word format) of how the evaluation was performed, everything seems largely above board. Potential bias may creep in from Nature’s news staff considering the reviews from the perspective of a “typical encyclopedia user,” but the evaluations themselves are included to allow readers to get a sense of how big a problem that is. All appears to be reasonable.

But the Encyclopedia Britannica went through those evaluations and, based on their analysis, is now suggesting that such appearances are deceiving. And they make that suggestion in language that, for the generally sedate publishing world, is rather sharp:

Nature’s research was invalid. As we demonstrate below, almost everything about the journal’s investigation, from the criteria for identifying inaccuracies to the discrepancy between the article text and its headline, was wrong and misleading. Dozens of inaccuracies attributed to the Britannica were not inaccuracies at all, and a number of the articles Nature examined were not even in the Encyclopedia Britannica. The study was so poorly carried out and its findings so error-laden that it was completely without merit.

Accusations include Nature having used articles from publications other than the encyclopedia and, in one case, material that wasn’t even produced by Britannica. Nature claimed that they matched the size of entries by deleting references only, but Britannica’s response indicated that many submissions were either fragmentary excerpts or subjected to extensive editing. Nature’s chosen evaluators also get criticized for getting facts wrong in their evaluations, and being unable to recognize simplifications that are reasonable for a publication targeted at a general audience.

Britannica is basing all of this criticism on the excerpts from the reviews provided in the Word document linked above, but are calling for Nature to release all the material involved in the article for public evaluation. This may help clarify some issues, but in the end, much of the finger pointing comes down to a “he said/she said” matter of how to interpret the seriousness of a given error. There is really no objective way to determine whether an editorial decision represents an appropriate simplification or a glaring omission. Nature’s response, however, has the potential to be interesting for reasons that go well beyond this controversy. In a very real sense, they and Britannica are kindred publications, facing increased pressure to maintain subscription-based publication in the face of open-access journals. How hard they press an issue that is seen as an embarrassment for their potential allies may be very telling.

Read Full Post »

Britannica versus Wikipedia heads to the WSJ

By Ken Fisher | Published: September 12, 2006 – 06:45PM CT

The Wall Street Journal is fond of hosting minidebates, and we’re fond of reading them. This week they pit Dale Hoiberg, editor-in-chief of Britannica, against Wikipedia’s Jimmy Wales. Rather than summarize the debate, I’ll offer some observations on what’s really at stake in this debate and what both sides are presupposing in their arguments. I recommend checking out the debate, though. It’s nothing if not funny in a feisty way.

Jimmy Wales, as usual, is confident and unapologetic. He’s also arguing that quality, neutrality, and balance are most achievable when large numbers of contributors are involved. The argument, pieced together, is something like this: “with a population as large as the Internet, an open encyclopedia has the greatest potential for quality and quantity over and against a closed, proprietary encyclopedia.”

In many ways this view is opposite of the academic approach, which takes a dim view of mass participation when it comes to the production and codification of knowledge. The academic world, for better or for worse, is very tiered. It would be wrong to describe it as a pyramid because it’s far more dynamic than that, but generally speaking, there are fewer experts at the top producing and reviewing knowledge than there are at the lower rungs of the educational system. This is not to say that there is not a positive flow from bottom to top. There is indeed, and when academics are honest, they celebrate and praise this as opposed to pretending as if their views come from On High. The point here is that the so-called knowledge producers are not in complete isolation, but they are powerful gatekeepers. This is true in general throughout the academic world, but Britannica represents a microcosm of that, even if all of its contributors are not strict academics.

For Wales, “openness” at all levels is the sine qua non of the highest levels of quality, and the openness he has in mind is one wherein anonymity is both cherished and respected. Part and parcel with this, Wales argues that the open nature of Wikipedia means that it can draw better contributors than can Britannica. Wales’ arguments, at their core, rest upon the age-old idea that two heads are better than one. Alongside this, there’s a notion that convenience and ideology both draw people to contribute to Wikipedia. With regards to the latter, there can be little debate. The rest is open season.

Hoiberg is certainly more traditional in his views, clearly suggesting that accuracy is more important than anything else. He is also not ready to surrender anything to Wikipedia, reminding everyone that the jury is still out on the question of whether or not an open system like Wikipedia truly delivers better content than a closed one like Britannica. He speaks as though the issue is unresolved, while Wales speaks about potentialities. Neither are calling the game just yet, to be sure.

Hoiberg’s arguments regarding community are perhaps the most indicative of academic arguments in this circle. Surprising though it might seem, Hoiberg, too, believes that his encyclopedia is produced by a community. The academic community that directly contributes is 4,000 strong and, according to him, they can revise their work online any time they need to. From the viewpoint of many academics, Britannica is a community product, but it is controlled by a small number of elites who essentially call the shots. In this way, the project is much like everything else in academia: there are clear lines of authority not just for administration, but also for knowledge.

Furthermore, the academic mindset is firmly represented in Hoiberg’s constant focus on getting things right the first time. He says that the system is designed to produce strong final products, and he seems to sparkle at the idea that Wikipedia publishes works in progress and rough drafts.

There’s not much more to say about the debate other than the fact that I think it mostly turned into a negativity fest with jabs flying both directions. In the broader discussion, however, I think there’s a core bit of difference that can be gainfully considered, but the biggest question I have after reading it is this: what metric can be used to decide which is better? Is doing so even worth the effort? Is there even an objective standard that applies to both encyclopedias? It’s not like Britannica is going anywhere, and Wikipedia is definitively here to stay.

Discuss Print

Read Full Post »

New Citizendium to correct Wikipedia’s wrongs?

By Ken Fisher | Published: September 19, 2006 – 09:18AM CT

Wikipedia co-founder Larry Sanger is starting a Wikipedia competitor. His new Wiki-based Citizendium project (a “citizens’ compendium of everything”) is expected to launch within a few short weeks, and in many ways it is aimed squarely at fixing Wikipedia’s weaknesses. At least, such is the tone of the site’s project page.

Citizendium will begin life as a “progressive fork” of Wikipedia, taking all of its launch content from the “infamous” online encyclopedia. Yet the project will distance itself from Wikipedia to focus on becoming “the flagship of a new set of responsibly-managed free knowledge projects.” There’s no ifs about it: Citzendium is being created to “outdo” Wikipedia.

How? The big difference between Citizendium and Wikipedia will be the presence of “experts” on Citizendium who will have the final say in editing disputes. Just who are the experts, though? The plan is for the site’s operators to post the ideal “credentials” of an editor, and then contributors themselves will decide if they fit the bill. In other words, one essentially appoints oneself to the position of expert. An editor must publish their own credentials online, and these credentials must be verifiable. I’m sorry Mbutu Ngangi, but you can’t be an editor just because you say you have millions of dollars in Laos and five doctorates.

What happens if even the expert editors can’t solve a problem? If things get unruly, Citizendium will also feature “constables” who will have the power to ban troublemakers. The project’s homepage reminds you, however, that “those who want the option of working anonymously and in a wild-and-woolly atmosphere in which rules are not necessarily enforced should always be able to do so on Wikipedia.”

Starting from not-scratch

Citizendium will indeed begin life as a full mirror of Wikipedia, and may carry Wikipedia content throughout its tenure. Citizendium will continue to refresh its articles with updates from Wikipedia, with one major exception: articles that have been modified at Citizendium will not be updated. There are plans for editing tools that will allow Citizendium users to track and see changes on the Wikipedia version of an article, but the general idea is to take Citizendium articles in a new direction with the help of experts, where experts are needed.

The irony is that both sites will both share the GFDL license, which means that there’s nothing to stop Wikipedia editors from copying the best of Citizendium. For his part, Sanger seems fixated on the idea of making a welcoming place for experts to contribute, and it is his view that Wikipedia isn’t that place. Nevertheless, the work of these comforted experts may end up right back at Wikipedia. It is truly hard to predict how the “experts” will react, for while many people are willing to talk about “experts” and “academics” as though they were a monolithic culture, they’re just everyday people, and there are more than a few contributing to Wikipedia already.

To begin, Citizendium will launch in English, but the project is open to forking off into other language if it appears worthwhile.

Discuss Print
Filed under: , , ,

Read Full Post »

New Citizendium to correct Wikipedia’s wrongs?

By Ken Fisher | Published: September 19, 2006 – 09:18AM CT

Wikipedia co-founder Larry Sanger is starting a Wikipedia competitor. His new Wiki-based Citizendium project (a “citizens’ compendium of everything”) is expected to launch within a few short weeks, and in many ways it is aimed squarely at fixing Wikipedia’s weaknesses. At least, such is the tone of the site’s project page.

Citizendium will begin life as a “progressive fork” of Wikipedia, taking all of its launch content from the “infamous” online encyclopedia. Yet the project will distance itself from Wikipedia to focus on becoming “the flagship of a new set of responsibly-managed free knowledge projects.” There’s no ifs about it: Citzendium is being created to “outdo” Wikipedia.

How? The big difference between Citizendium and Wikipedia will be the presence of “experts” on Citizendium who will have the final say in editing disputes. Just who are the experts, though? The plan is for the site’s operators to post the ideal “credentials” of an editor, and then contributors themselves will decide if they fit the bill. In other words, one essentially appoints oneself to the position of expert. An editor must publish their own credentials online, and these credentials must be verifiable. I’m sorry Mbutu Ngangi, but you can’t be an editor just because you say you have millions of dollars in Laos and five doctorates.

What happens if even the expert editors can’t solve a problem? If things get unruly, Citizendium will also feature “constables” who will have the power to ban troublemakers. The project’s homepage reminds you, however, that “those who want the option of working anonymously and in a wild-and-woolly atmosphere in which rules are not necessarily enforced should always be able to do so on Wikipedia.”

Starting from not-scratch

Citizendium will indeed begin life as a full mirror of Wikipedia, and may carry Wikipedia content throughout its tenure. Citizendium will continue to refresh its articles with updates from Wikipedia, with one major exception: articles that have been modified at Citizendium will not be updated. There are plans for editing tools that will allow Citizendium users to track and see changes on the Wikipedia version of an article, but the general idea is to take Citizendium articles in a new direction with the help of experts, where experts are needed.

The irony is that both sites will both share the GFDL license, which means that there’s nothing to stop Wikipedia editors from copying the best of Citizendium. For his part, Sanger seems fixated on the idea of making a welcoming place for experts to contribute, and it is his view that Wikipedia isn’t that place. Nevertheless, the work of these comforted experts may end up right back at Wikipedia. It is truly hard to predict how the “experts” will react, for while many people are willing to talk about “experts” and “academics” as though they were a monolithic culture, they’re just everyday people, and there are more than a few contributing to Wikipedia already.

To begin, Citizendium will launch in English, but the project is open to forking off into other language if it appears worthwhile.

Discuss Print
Filed under: , , ,

Read Full Post »

Experts rate Wikipedia’s accuracy higher than non-experts

By Nate Anderson | Published: November 27, 2006 – 04:01PM CT

A new salvo has been fired in the perennial war over Wikipedia’s accuracy. Thomas Chesney, a Lecturer in Information Systems at the Nottingham University Business School, published the results of his own Wikipedia study in the most recent edition of the online journal First Monday, and he came up with a surprising conclusion: experts rate the articles more highly than do non-experts.

This less-than-intuitive finding is the conclusion of a study in which Chesney had 55 graduate students and research assistants examine one Wikipedia article apiece. Each participant was randomly placed into one of two groups: group one read articles that were in their field of study, while group two read randomly-assigned articles. Respondents were asked to identify any errors that they found.

Those in the expert group ranked their articles as generally credible, higher than those evaluated by the non-experts. Chesney admits that this is unexpected, but has a possible explanation: “It may be the case that non-experts are more cynical about information outside of their field and the difference comes from a natural reaction to rate unfamiliar articles as being less credible.”

Whatever the reason for the results, they will cheer defenders of Wikipedia’s accuracy, though Chesney urges caution in extrapolating too generally from his study. For one thing, the sample size was small. For another, 13 percent of those in the “experts” group reported finding mistakes in their assigned articles.

Whether this is better or worse than traditional, expert-based encyclopedias depends on who you ask. Nature did a highly-publicized comparative study between Wikipedia and the Encyclopedia Britannica last year in which they found that the two were similar in terms of accuracy. Britannica disputed those findings and still claims to offer a more reliable product.

Chesney’s study was not intended to settle the debate. He notes that, whatever Wikipedia’s comparative accuracy, plenty of people (academics included) are using it, and he simply wanted to see whether Wikipedia could be considered accurate enough to be worth using. His study suggests that it can, but that caution—and further research—needs to be used before citing anything learned from Wikipedia as a fact.

Discuss Print

Read Full Post »

Britannica versus Wikipedia heads to the WSJ

By Ken Fisher | Published: September 12, 2006 – 06:45PM CT

The Wall Street Journal is fond of hosting minidebates, and we’re fond of reading them. This week they pit Dale Hoiberg, editor-in-chief of Britannica, against Wikipedia’s Jimmy Wales. Rather than summarize the debate, I’ll offer some observations on what’s really at stake in this debate and what both sides are presupposing in their arguments. I recommend checking out the debate, though. It’s nothing if not funny in a feisty way.

Jimmy Wales, as usual, is confident and unapologetic. He’s also arguing that quality, neutrality, and balance are most achievable when large numbers of contributors are involved. The argument, pieced together, is something like this: “with a population as large as the Internet, an open encyclopedia has the greatest potential for quality and quantity over and against a closed, proprietary encyclopedia.”

In many ways this view is opposite of the academic approach, which takes a dim view of mass participation when it comes to the production and codification of knowledge. The academic world, for better or for worse, is very tiered. It would be wrong to describe it as a pyramid because it’s far more dynamic than that, but generally speaking, there are fewer experts at the top producing and reviewing knowledge than there are at the lower rungs of the educational system. This is not to say that there is not a positive flow from bottom to top. There is indeed, and when academics are honest, they celebrate and praise this as opposed to pretending as if their views come from On High. The point here is that the so-called knowledge producers are not in complete isolation, but they are powerful gatekeepers. This is true in general throughout the academic world, but Britannica represents a microcosm of that, even if all of its contributors are not strict academics.

For Wales, “openness” at all levels is the sine qua non of the highest levels of quality, and the openness he has in mind is one wherein anonymity is both cherished and respected. Part and parcel with this, Wales argues that the open nature of Wikipedia means that it can draw better contributors than can Britannica. Wales’ arguments, at their core, rest upon the age-old idea that two heads are better than one. Alongside this, there’s a notion that convenience and ideology both draw people to contribute to Wikipedia. With regards to the latter, there can be little debate. The rest is open season.

Hoiberg is certainly more traditional in his views, clearly suggesting that accuracy is more important than anything else. He is also not ready to surrender anything to Wikipedia, reminding everyone that the jury is still out on the question of whether or not an open system like Wikipedia truly delivers better content than a closed one like Britannica. He speaks as though the issue is unresolved, while Wales speaks about potentialities. Neither are calling the game just yet, to be sure.

Hoiberg’s arguments regarding community are perhaps the most indicative of academic arguments in this circle. Surprising though it might seem, Hoiberg, too, believes that his encyclopedia is produced by a community. The academic community that directly contributes is 4,000 strong and, according to him, they can revise their work online any time they need to. From the viewpoint of many academics, Britannica is a community product, but it is controlled by a small number of elites who essentially call the shots. In this way, the project is much like everything else in academia: there are clear lines of authority not just for administration, but also for knowledge.

Furthermore, the academic mindset is firmly represented in Hoiberg’s constant focus on getting things right the first time. He says that the system is designed to produce strong final products, and he seems to sparkle at the idea that Wikipedia publishes works in progress and rough drafts.

There’s not much more to say about the debate other than the fact that I think it mostly turned into a negativity fest with jabs flying both directions. In the broader discussion, however, I think there’s a core bit of difference that can be gainfully considered, but the biggest question I have after reading it is this: what metric can be used to decide which is better? Is doing so even worth the effort? Is there even an objective standard that applies to both encyclopedias? It’s not like Britannica is going anywhere, and Wikipedia is definitively here to stay.

Discuss Print

Read Full Post »

Older Posts »

%d bloggers like this: