The venerable Britannica, founded in Edinburgh in 1768, is demanding that the scientific journal Nature publicly retract its finding that the open-source Wikipedia “comes close” to Britannica in accuracy.

In a 20-page statement on its website, Britannica complained that the Nature study was “fatally flawed”. “We discovered in Nature’s work a pattern of sloppiness, indifference to basic scholarly standards, and flagrant errors so numerous they completely invalidated the results,” it said.

But Nature stuck to its guns, and fired back: “We reject those accusations, and are confident our comparison was fair.” The row goes to the heart of the role of experts in the internet age.

Over the centuries Britannica has earned a reputation as the world’s premier encylopaedia with contributions by leading experts and has charged handsomely. But it is now under threat from the free “user-generated” model of Wikipedia, written by online volunteers.

The Wikipedia, created five years ago, posted its one millionth article in English this month — about Scotland’s Jordanhill railway station. Jimmy Wales, its co-founder, has said that the goal was to “get to Britannica quality, or better”. But it has been embarrassed by revelations that contributors had doctored pieces.

John Seigenthaler, a former newspaper editor and Kennedy Administration official, caused a rumpus last year when he discovered that his Wikipedia biography said “he was thought to have been directly involved in the Kennedy assassinations of both John and his brother, Bobby”. The entry turned out to be a prank.

More recently, Wikipedia has uncovered efforts by American politicians to clean up their image. Staff of Senator Tom Harkin, a Democrat, apparently removed a paragraph from his Wikipedia entry recording his false claim to have flown combat missions over North Vietnam.

Similarly, staff of Senator Norm Coleman, a Republican, rewrote his Wikipedia biography so that he was described merely as an “activist” at university, not a “liberal”. A Wikipedia entry for Senator Tom Coburn of Oklahoma falsely reported that he had been voted “most annoying senator”.

The Nature study, published in the December 15 issue, asked scientists to assess 50 entries on scientific topics ranging from Dolly the Sheep to Dmitry Mendeleev, the Russian 19th-century chemist, without telling them if the articles came from the Britannica or the Wikipedia.

The study found the average Britannica entry contained approximately three inaccuracies, while Wikipedia had four. Only eight “serious errors” were found — four in each encyclopaedia. Britannica objected that, despite Nature's conclusion, the journal’s own figures showed Wikipedia had one third more inaccuracies.

It claimed that in some cases — such as the spelling of Pythagoras’s home in Italy, the Britannica was right and Nature’s experts were wrong.

Britannica also complained that some omissions were wrongly counted as errors.

Which is the more valuable resource: Britannica or Wikipedia? Have your say here.

Explore World News

• Iraq News

• US & Americas News

• Europe News

• Middle East News

• Asia News

• Africa News

Times Recommends

• Valley of death may be $88bn key to life
• Hope fades for the missing as earthquake toll leaps
• Invasion of the computer-eating alien ants