Tweet
Just a small mention regarding the Armenian Genocide in bold
Full Comment
Colby Cosh: The big Wikiquestion
Caltech graduate student Virgil Griffith has been called a "mad scientist about town." He says that one of his goals as a computer programmer is to "To create a cornucopia of minor public relations disasters for companies and organizations I dislike." He is the gadfly behind WikiScanner, the new Web application that automatically probes Wikipedia for the originating IP addresses of anonymous edits. Since its Aug. 13 launch, WikiScanner has been used to identify computers at hundreds of organizations whose users are making underhanded or self-serving changes to the ever-controversial online encyclopedia. The staff and readers of Wired.com have used the tool to catch PepsiCo deleting critical paragraphs about the nutritional content of Pepsi, the Turkish treasury department scissoring out references to the Armenian genocide, Dow and ExxonMobil cutting out references to Bhopal and the Exxon Valdez and even questionable changes originating from the FBI and the CIA.
Canadian organizations, it seems, have not been entirely blameless. Last week, The Globe and Mail revealed that changes to various Web pages about members of Parliament had been made from IP addresses belonging to the House of Commons network. Controversial material was hacked from the article about Steven Fletcher, large deletions were made to the entry on Joe Volpe and someone has been gardening Pierre Poilievre's entry with a degree of fidelity bordering on obsession. The entry on the 2006 Liberal convention was the subject of what must be at least 30 edits over a two-month period. It seems certain some parliamentary staffers have not quite absorbed Wikipedia's strong "neutral point of view" ethos. But it's not just Parliament. Scan for Bombardier IP addresses and you'll find changes to entries about the company's aircraft; look for Canadian Tire, and you'll spot deletions from the entry about the retail chain's ubiquitous "money." (At the Globe itself, someone seems to have been preoccupied with monitoring entries about media personalities connected to the paper's corporate sibling, CTV.)
It is not easy to use WikiScanner to find changes that rise to the level of outrageousness of the cases being collected over at Wired. Many ostensibly "non-neutral" edits appear to be harmless efforts to fix mistakes. They probably shouldn't be happening anyway, but as Wikipedia is used more and more, one can sense the pressure on corporations and public individuals to have their say in the great hullabaloo of edit and counter-edit. And there are plenty of borderline cases: should a staffer for a Cabinet minister ignore wrong information about the order of precedence, even when correcting the information might make his boss look a little better? The safest answer is "Yes" but it comes with a cost to the accuracy of the encyclopedia.
Mr. Griffith's WikiScanner, even though it represents good news for the long-term reliability of Wikipedia, awakens the epistemological issues that have lingered around the site since it came to the attention of the public. Call it the Big Wikiquestion: How is it possible for an encyclopedia that can be edited by anyone to be trustworthy? It induces the same natural nervousness that we feel when we hear that there are towns in Europe and China that have no traffic signals, and learning that such places sometimes have safer roads than our own doesn't help. We possess stubborn biases in favour of centralized planning. It has taken decades to mitigate those biases just a little when it comes to politics, and it will take decades more for us to learn that anarchy or near-anarchy can be a functioning organizational style in narrower social and economic contexts, even though the Internet itself is an obvious everyday example.
No single printed source should ever be taken as gospel, and that goes double for the actual Gospels. But the best answer to the Big Wikiquestion might be "Trustworthy compared to what?" In empirical tests Wikipedia has compared quite favourably to traditional reference books. At one time people had relatively low expectations of neutrality from printed encyclopedias; the great Eleventh Edition of the Britannica (1911) is remembered because its articles were written by pre-eminent thinkers, not necessarily coldly objective ones. Would Kropotkin's survey of "Anarchism," T.H. Huxley's of "Biology" and Einstein's of "Physics" be thought admirable for lack of bias today?
Other great reference books turn out to have a lot in common with Wikipedia when their history is studied; the corpus of citations that still buttresses The Oxford English Dictionary was collected in a rather Wiki-esque fashion, with the editors soliciting the learned public for contributions by means of printed advertisements in literary and scholarly magazines. Is the OED "trustworthy"? After generations of error-checking and revision, it is as trustworthy as works of the human mind can hope to be. But it still depends, as always, on exactly what you need it for and how much time and ability you have to double-check.
[email protected]
Full Comment
Colby Cosh: The big Wikiquestion
Caltech graduate student Virgil Griffith has been called a "mad scientist about town." He says that one of his goals as a computer programmer is to "To create a cornucopia of minor public relations disasters for companies and organizations I dislike." He is the gadfly behind WikiScanner, the new Web application that automatically probes Wikipedia for the originating IP addresses of anonymous edits. Since its Aug. 13 launch, WikiScanner has been used to identify computers at hundreds of organizations whose users are making underhanded or self-serving changes to the ever-controversial online encyclopedia. The staff and readers of Wired.com have used the tool to catch PepsiCo deleting critical paragraphs about the nutritional content of Pepsi, the Turkish treasury department scissoring out references to the Armenian genocide, Dow and ExxonMobil cutting out references to Bhopal and the Exxon Valdez and even questionable changes originating from the FBI and the CIA.
Canadian organizations, it seems, have not been entirely blameless. Last week, The Globe and Mail revealed that changes to various Web pages about members of Parliament had been made from IP addresses belonging to the House of Commons network. Controversial material was hacked from the article about Steven Fletcher, large deletions were made to the entry on Joe Volpe and someone has been gardening Pierre Poilievre's entry with a degree of fidelity bordering on obsession. The entry on the 2006 Liberal convention was the subject of what must be at least 30 edits over a two-month period. It seems certain some parliamentary staffers have not quite absorbed Wikipedia's strong "neutral point of view" ethos. But it's not just Parliament. Scan for Bombardier IP addresses and you'll find changes to entries about the company's aircraft; look for Canadian Tire, and you'll spot deletions from the entry about the retail chain's ubiquitous "money." (At the Globe itself, someone seems to have been preoccupied with monitoring entries about media personalities connected to the paper's corporate sibling, CTV.)
It is not easy to use WikiScanner to find changes that rise to the level of outrageousness of the cases being collected over at Wired. Many ostensibly "non-neutral" edits appear to be harmless efforts to fix mistakes. They probably shouldn't be happening anyway, but as Wikipedia is used more and more, one can sense the pressure on corporations and public individuals to have their say in the great hullabaloo of edit and counter-edit. And there are plenty of borderline cases: should a staffer for a Cabinet minister ignore wrong information about the order of precedence, even when correcting the information might make his boss look a little better? The safest answer is "Yes" but it comes with a cost to the accuracy of the encyclopedia.
Mr. Griffith's WikiScanner, even though it represents good news for the long-term reliability of Wikipedia, awakens the epistemological issues that have lingered around the site since it came to the attention of the public. Call it the Big Wikiquestion: How is it possible for an encyclopedia that can be edited by anyone to be trustworthy? It induces the same natural nervousness that we feel when we hear that there are towns in Europe and China that have no traffic signals, and learning that such places sometimes have safer roads than our own doesn't help. We possess stubborn biases in favour of centralized planning. It has taken decades to mitigate those biases just a little when it comes to politics, and it will take decades more for us to learn that anarchy or near-anarchy can be a functioning organizational style in narrower social and economic contexts, even though the Internet itself is an obvious everyday example.
No single printed source should ever be taken as gospel, and that goes double for the actual Gospels. But the best answer to the Big Wikiquestion might be "Trustworthy compared to what?" In empirical tests Wikipedia has compared quite favourably to traditional reference books. At one time people had relatively low expectations of neutrality from printed encyclopedias; the great Eleventh Edition of the Britannica (1911) is remembered because its articles were written by pre-eminent thinkers, not necessarily coldly objective ones. Would Kropotkin's survey of "Anarchism," T.H. Huxley's of "Biology" and Einstein's of "Physics" be thought admirable for lack of bias today?
Other great reference books turn out to have a lot in common with Wikipedia when their history is studied; the corpus of citations that still buttresses The Oxford English Dictionary was collected in a rather Wiki-esque fashion, with the editors soliciting the learned public for contributions by means of printed advertisements in literary and scholarly magazines. Is the OED "trustworthy"? After generations of error-checking and revision, it is as trustworthy as works of the human mind can hope to be. But it still depends, as always, on exactly what you need it for and how much time and ability you have to double-check.
[email protected]
Comment