Tag Archives: Reliability of Wikipedia

Wikipedia: Important enough to whitewash

This is an edited version of my weekly column for Loose Wire Service, a service providing print publications with technology writing designed for the general reader. Email me if you’re interested in learning more.

Wikipedia has gone through some interesting times, good and bad, but I think the last couple of weeks has proved just how powerful it is.

Powerful enough for those who feel denigrated by it to have been trying to spin, airbrush and generally rewrite how history — or at least Wikipedia — remembers them.

Take WikiScanner, cooked up by a young student, Virgil Griffith. WikiScanner does something very simple: It searches the Internet addresses of an organization — government, private, company or whatever — and matches them with any anonymous edit of a Wikipedia entry.

This means that while the edits themselves may be anonymous, the organization where the person is based is not. We may not know who did it, in other words, but we’ve got a pretty good idea of whom they work for.

The results have been surprising. Users of WikiScanner have come up with dozens of cases of companies, organizations and government departments apparently changing entries to either delete stuff they may not like, or making the text more palatable.

Some examples of apparent — none of these is confirmed but the Internet addresses match — self-interested alterations that have hit the news in the last few weeks:

* Diebold removes sections critical of the company’s electronic voting machines

* Apple and Microsoft trade negative comments about each other

* Amnesty International removes negative comments about itself, according to the Malta Star

(My own searches threw up no examples at all of institutions in my current home of Indonesia spinning on Wikipedia. Shame on them. What have they been doing with their time? One Indonesian embassy official seems to have spent most of his day editing an entry on rude finger gestures, but that’s about it. Clearly these people are not working hard enough for their country.)

The point about all this: Wikipedia is often derided as irrelevant and unworthy. Clearly, though, it’s important enough for these people, either officially or unofficially, on their own initiative or at the behest of higher-ups, to rewrite stuff to make themselves or their employer look better.

You might conclude from this that Wikipedia is not reliable as a result. I would argue the opposite: These edits have nearly all been undone by alert Wikipedians, usually very quickly.

(Wikipedia automatically stores all previous versions of a page and keeps a record of all the edits, and the Internet address from where they originate.)

The truth is that Wikipedia has come of age. Wikipedia is now important enough for ExxonMobil, The Church of Scientology, the U.S. Defense Department and the Australian government to spend time and effort trying to get their version of events across. If it was so irrelevant or unreliable, why would these people bother?

Of course, coming of age isn’t always a good thing. A recent conference on Wikipedia in Taiwan highlighted how Wikipedia is no longer an anarchic, free-for-all, but has somehow miraculously produced a golden egg.

It is now a bureaucracy, run by the kind of people who like to post “Don’t … ” notices on pantry walls. I’m not saying this is necessarily a bad thing. We all hate such people until our sandwich goes missing. Then we turn to them — or turn into them.

WikiScanner reveals that it’s probably good that such people take an interest in Wikipedia, because it’s clear that the site is under threat from people who would censor history and whitewash the truth to suit them.

Thanks to Virgil and the Wikipedians, that’s not going to happen anytime soon.

The Jakarta Post – The Journal of Indonesia Today

Encyclopedia Britannica Fights Back?

I don’t know whether this is the right response to the challenge of Wikipedia, but Encylopaedia Britannica seems to think so, according to The Boston Globe :

To respond to competitive challenges from Google, Yahoo, and the popular online encyclopedia Wikipedia, Britannica today will announce it is returning to an old practice after a lapse of a decade by naming an advisory board, whose 15 members top editor Dale Hoiberg calls ”some of the smartest people on earth.” The Chicago-based publisher hopes that the prestige and knowledge of the members — four Nobel laureates and two Pulitzer Prize winners among them — will help reassert the authority of an encyclopedia first published in 1768 but buffeted in an age when the Internet has loosened the definition of what is factual.

Librarians, teachers, and scholars say they are increasingly alarmed at the way students pull information from anywhere online and accept it as valid, without much consideration of the source. Wikipedia, for instance, allows anyone to make entries and yet draws 5 million visitors a month.

”You can’t do something so authoritative easily. It’s hard work,” said Hoiberg, a Britannica senior vice president.

Loosened the definition of what is factual? I’m guessing this is not the same Encyclopaedia Britannica that recently acknowledged one schoolboy was able to spot five errors in two entries. The issue is not whether people are redefining what is factual, it’s whether they so readily accept something as more authoritative than other sources because it has a long history behind it, or because the author of the piece, and those editing it, have titles before their name and initials after it. As I wrote in a recent column on the accuracy of Wikipedia, it’s not about the contributor’s qualifications, but about their contribution. The folk putting together Wikipedia don’t sit around making stuff up — if they did, they’ll quickly find their entries altered, deleted, or put into some sort of side-channel while the matter is earnestly debated. It’s peer review on steroids.

There’s some good recent history of EB in The Boston Globe’s piece. And the author, Eric Ferkenhoff, talks to Jimmy Wales as well for balance. But I think he may have allowed EB to put a bit too much of their spin on the story. The hiring of an advisory board sounds to me less about ‘reasserting authority’ and more about wondering how the hell to create a product that can compete with the collective wisdom of thousands of Wikipedians. No one is claming that Wikipedia is perfect (at least as far as I know), and that it should not be used merely as one of several sources, but experience should probably teach us that the same is true of other reference books, including Encyclopaedia Britannica.

Of course, there is concern more generally about students taking stuff from the Internet and accepting it as fact. As Eric writes:

It is such uncertainty about the accuracy of Web-based information that troubles many traditionalists. ”I think it would be impossible to find a librarian, or even a teacher, in the country that’s not concerned,” said Jo Sommers, head librarian at The Latin School of Chicago, an elite private school. ”The kids, they gravitate to the Googles, and they don’t understand the issue of going to sources that are authoritative, sources that have been vetted. They just assume that everything they find on the Internet is right.”

But that’s a problem for teachers to explain to their students and to help them distinguish from rubbish and fact (and the importance of confirming information elsewhere). But it’s disingenuous to imply from that that there is EB on one side of the fence and then there’s Googe/Yahoo/Wikipedia and all the rest. After all, EB is on the Internet too. Does that make it rubbish?

Lucian George, Encyclopedia Britannica and Wikipedia

Into this debate about the reliability of Wikipedia leaps the 12–year old figure of Lucian George from north London, who found five errors in the latest edition of the Encyclopedia Britannica. BBC reports that

A schoolboy has uncovered several mistakes in the latest edition of the Encyclopaedia Britannica – regarded by readers as an authority on everything. Lucian George, 12, from north London, found five errors on two of his favourite subjects – central Europe and wildlife – and wrote to complain. The book’s editor wrote back thanking him for “pointing out several errors and misleading statements”.
A Britannica spokesman said the company was “grateful”.

His father, Gabriel George, told BBC News: “Lucian told me he had found a mistake. Then, a few days later, he found another. Then there was another. “By the time he had found five, I said to him that he should write to the editors to complain about it.”

So what did he find?

  • Chotyn, in which two battles between the Poles and the Ottoman Empire were fought, is said to lie in Moldova. (It’s not, it’s in Ukraine.)
  • The Polish part of the Belovezhskaya Forest, according to the encyclopaedia, lies in the Bialystok, Suwalki and Lomza provinces. (Suwalki and Lomza provinces have not existed since 1998. And, even when they did, the whole Polish section of the forest – which extends into Belarus – was in Bialystok, the BBC reports.)
  • The terrain of the European bison: Poland, according to the encylcopedia. (Actually it encompassed parts of Russia, Lithuania, Ukraine, Slovakia and Belarus – not just Poland.)

Of course, errors in the Britannica are nothing new (here’s a list from one guy alone). And I’m no Lucian George but (that’s what we’re going to say from now on, isn’t it? ‘I’m no Lucian George, but isn’t that use of the semi-colon incorrect?’) I’ve spotted a few errors in my time in both the Britannica and Encarta. (If one 12–year old can find five, how many more errors must there be in there?) But maybe we shouldn’t be so apologetic about Wikipedia’s eccentricities and errors when we realise that no single encyclopedia can get it right? The academics who do these entries are, after all, just academics, and probably don’t have time to have any peer review of their entries.

In a way Wikipedia is like having dozens of Lucian Georges looking over your shoulder when you amend or add an entry, which may end up being a better way of keeping out errors. Not least, of course, an error in Wikipedia will get fixed on the spot, but how many copies of EB are going to sit on shelves, riddled with mislocated East European data, for years to come, misleading the public and irritating the good burghers of Chotyn?