Dec 14

Nat Torkington

Nat Torkington

Nature Magazine Checks Out Wikipedia vs Encyclopedia Britannica

As the veracity of Wikipedia is questioned and lawsuits threatened, Nature's Timo Hannay dropped a note to the Radar crew about a test the folks at Nature magazine conducted to see which encyclopedia had more errors. The experiment "sent 50 pairs of Wikipedia and Britannica articles on scientific topics to recognised experts and, without telling them which article came from which source, asked them to count the numbers of errors (mistakes, misleading statements or omissions). Among the 42 replies, Britannica content had an average of just under 3 errors per article whilst Wikipedia had an average of just under 4 errors -- not as much difference, perhaps, as most people would expect." Also see the subject-by-subject breakdown, accompanying editorial (subscribers only), and Timo's blog which has links to Nature's interview with Jimmy Wales about it.

tags:   | comments: 9   | Sphere It

Previous  |  Next

0 TrackBacks

TrackBack URL for this entry:

Comments: 9

  Christopher Allen [12.14.05 02:37 PM]

The link for the "accompanying editorial" appears to be available only to subscribers.

  jc [12.14.05 03:35 PM]

And who, in turn, factchecks those "recognised experts"? Were all 42 replies identical, or did some experts flag mistakes that others did not?

The real issue here is not the number of errors creeping into publications that call themselves encyclopedias. The issue is the deliberate creation of false and misleading entries. It is hard to imagine someone sneaking a piece of malicious fiction about the Kennedy assassination into the Britannica. Until Wikipedia ensures that every new entry and every new edit of existing entries are subject to review and verification before publication, it won't merit our trust.

  Rich Gibson [12.14.05 04:40 PM]

"Until Wikipedia ensures that every new entry and every new edit of existing entries are subject to review and verification before publication, it won't merit our trust."

Huh? I'm confused here. We have a whole lot of 'information' in our news and media that is quite literally false, generated from the top down, reviewed and verified, and jc is worried about Wikipedia, and thinks it doesn't merit our trust?

It may be hard to imagine someone sneaking a piece of malicious fiction about the Kennedy assasination into the Britannica, on the other hand, most of the Britannica suffers from group think and an overwhelming false objectivity.

Our reviewed and verified media is filled with errors. At least with Wikipedia errors and omissions can be corrected.

So, until the mainstream media provides a mechanism for near real time updates and corrections it does not deserve our trust.

  Peter Firminger [12.14.05 07:47 PM]

The problem I see isn't in errors as such, it's in plagiarism. Anyone can come along to Wikipedia and anonymously paste in vast amounts of text direct from existing publications even though they may be breaching author or publisher copyright. As they will be factually correct, they will probably go un-noticed.

Are we ready to accept that all written (or even transcriptions of spoken) human knowledge is up for grabs on a wiki now?

  Rich Gibson [12.14.05 08:46 PM]


Under that view we can not have any user authored content. Clearly that isn't the case. In fact, as in the case of professors who now search the net looking for plagiarism, having content online makes it easier (possibly too easy) to locate copies.

We don't need a wiki for plagiarism, and piracy, to occur on and offline.

  Timo Hannay [12.15.05 03:05 AM]

The editorial link in Nat's post has been made free.

Also, I now see that the subject-by-subject breakdown is available on too, so people might want to use that link (especially this weekend, when the Nacent blog will be down for a bit while we move the server).

  Paul [12.15.05 06:38 AM]

In my opinion, this affair has been blown out of proportion. We are just being reminded of something we have been told a thousand times before: you can't believe everything you read. Behind all of this technology are human beings who make mistakes, view issues with bias, or even inject a little mischief into the process. The answer is to use multiple sources, check your facts, and look for bias or vested interest.

Does this excuse Wikipedia from cleaning up their act? Of course not. It is just to point out that the reader has a responsibility as well. Looking up the answer in an encyclopedia (online or not) has never been a substitute for properly done research.

  GerardM [12.17.05 06:47 AM]

There was a giant that said that science is done by "standing on the shoulders of giants". The modern ideas about copyright are detrimental to the acquisition of knowledge and they are detrimental to science.

When basic information on a topic cannot be made available because people suspect that there might be some copyright violations, it proves how deep we have sunken. When you research how Wikipedia treats copyright violations, you will find that these texts are removed.

One important thing to understand about the content of Wikipedia is, that its content can be represented by a bell curve. There are some really good and there are some really bad articles. As Wikipedia grows, the number of these articles grow as well.

Given that the Wikipedia community actively addresses the problem of quality by introducing measures and by being responsive whenever an issue is raised, Wikipedia deserves that people assume the good intentions of the Wikimedia community.

You will find that the Wikimedia community is responsive to suggestions that have a positive intention. It is truly a situation where it is your choise to be part of the problem or be part of the solution.


  Dann Passoja [01.15.06 02:06 PM]

I'm looking for the email address of Nature Magazine and I thought that you might help
Thank you,
Dann Passoja

Post A Comment:

 (please be patient, comments may take awhile to post)

Type the characters you see in the picture above.