From the Village Pump, found by BanyanTree: the Australian newspaper
The Age is the first to report on Nature's formal comparison of the science coverage in the English Wikipedia and Encyclopaedia Britannica:
http://www.theage.com.au/news/national/online-encyclopedias-put-to-the-test/...
"The exercise revealed numerous errors in both encyclopedias, but among 42 entries tested, the difference in accuracy was not great: the average science entry in Wikipedia contained around four inaccuracies; Britannica, around three"
This can provide us with a good talking point in discussions about Wikipedia's quality: "Wikipedia is not nearly as good as we would like it to be, and you certainly cannot trust all statements in it. But then again, Encyclopedia Britannica is not much better."
I also expect that, once the complete list of evaluated articles is published by Nature, we will find that our articles are longer, so that the error rate *per sentence* might very well be lower in Wikipedia.
Axel
__________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com
If the reporting is accurate, this is a very significant finding, considering the general perception that Wikipedia doesn't cover science well.
On 12/14/05, Tony Sidaway f.crdfa@gmail.com wrote:
If the reporting is accurate, this is a very significant finding, considering the general perception that Wikipedia doesn't cover science well.
Is that really the general perception?I've always been of the belief that Wikipedia covers science very well, especially as the articles are often written by experts with very little edit warring. It's the political, social, historical and biographical articles that I have much less faith in. People who write about, say, [[Analytic combinatorics]] generally know what they're talking about, while the same can't necessarily be said about [[Communism]]...
Anyway, that sounds like an interesting study.
Sam
-- Asbestos http://en.wikipedia.org/wiki/User:Asbestos
There is a general perception that we don't cover science well? I thought we were a lot stronger in the sciences than we are in the humanities. My first thought was that they had picked an area likely to flatter us.
Jason
Quoting Tony Sidaway f.crdfa@gmail.com:
If the reporting is accurate, this is a very significant finding, considering the general perception that Wikipedia doesn't cover science well.
On 12/15/05, jkelly@fas.harvard.edu jkelly@fas.harvard.edu wrote:
There is a general perception that we don't cover science well? I thought we were a lot stronger in the sciences than we are in the humanities. My first thought was that they had picked an area likely to flatter us.
WP's been pretty decent in the sciences, but the weakness has always been in the medicine and anatomy-related articles.
-Andrew (User:Fuzheado)
On 12/14/05, Andrew Lih andrew.lih@gmail.com wrote:
WP's been pretty decent in the sciences, but the weakness has always been in the medicine and anatomy-related articles.
And the readability for the non-scientist; some science articles are very dense and don't lead the reader along well enough. Many math articles are even worse.
-Matt
On 12/14/05, Axel Boldt axelboldt@yahoo.com wrote:
From the Village Pump, found by BanyanTree: the Australian newspaper
The Age is the first to report on Nature's formal comparison of the science coverage in the English Wikipedia and Encyclopaedia Britannica:
http://www.theage.com.au/news/national/online-encyclopedias-put-to-the-test/...
That's a rather good showing for Wikipedia. As soon as the issue is available at Nature's webpage, I'll look for more details.
Bob Mellish
"Bob Mellish" wrote
That's a rather good showing for Wikipedia. As soon as the issue is
available at Nature's webpage, I'll look for more details.
Hang on, let's be scientific. If EB consolidates articles more than WP, because WP uses hypertext freely to break things up, then 'per article' is not necessarily a good metric. And 'per 10000 words' would say more.
Charles
charles matthews wrote:
Hang on, let's be scientific. If EB consolidates articles more than WP, because WP uses hypertext freely to break things up, then 'per article' is not necessarily a good metric. And 'per 10000 words' would say more.
I'd give the authors the benefit of the doubt and assume that the analysis takes things like this into account (at least, I'll assume this until the article is published and we can find out for sure). _Nature_ tends to have fairly rigorous review standards and a high level of article quality, so it's not unlikely that the apparent over-simplifications were introduced by the journalist summarizing the article for _The Age_, which isn't uncommon when reading mainstream-press summaries of scientific articles.
-Mark
The Nature article is now up on the Dec 15th issue on their website (subscribers only, though). Unfortunately it doesn't go into details about which articles they checked, though it mentions [[Dmitri Mendeleyev]] as particularly erronous.
Bob Mellish
On 12/14/05, Bob Mellish bobmellish@gmail.com wrote:
The Nature article is now up on the Dec 15th issue on their website (subscribers only, though). Unfortunately it doesn't go into details about which articles they checked, though it mentions [[Dmitri Mendeleyev]] as particularly erronous.
Here is the article list: http://www.nature.com/news/2005/051212/multimedia/438900a_m1.html
here is the full article: http://www.nature.com/news/2005/051212/full/438900a.html
Entry Encyclopaedia Britannica inaccuracies Wikipedia inaccuracies Acheulean industry 1 7 Agent Orange 2 2 Aldol reaction 4 3 Archimedes' principle 2 2 Australopithecus africanus 1 1 Bethe, Hans 1 2 Cambrian explosion 10 11 Cavity magnetron 2 2 Chandrasekhar, Subrahmanyan 4 0 CJD 2 5 Cloud 3 5 Colloid 3 6 Dirac, Paul 10 9 Dolly 1 4 Epitaxy 5 2 Ethanol 3 5 Field effect transistor 3 3 Haber process 1 2 Kinetic isotope effect 1 2 Kin selection 3 3 Lipid 3 0 Lomborg, Bjorn 1 1 Lymphocyte 1 2 Mayr, Ernst 0 3 Meliaceae 1 3 Mendeleev, Dmitry 8 19 Mutation 8 6 Neural network 2 7 Nobel prize 4 5 Pheromone 3 2 Prion 3 7 Punctuated equilibrium 1 0 Pythagoras' theorem 1 1 Quark 5 0 Royal Greenwich Observatory 3 5 Royal Society 6 2 Synchrotron 2 2 Thyroid 4 7 Vesalius, Andreas 2 4 West Nile Virus 1 5 Wolfram, Stephen 2 2 Woodward, Robert Burns 0 3
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
What is going on here? And where did they get permission to use our logo?
http://www.wikipediaclassaction.org/
- - Ryan
On 12/14/05, Ryan Delaney ryan.delaney@gmail.com wrote:
What is going on here? And where did they get permission to use our logo?
See [[QuakeAID]] for details about this.
However, I very much prefer wikipediaclassaction.com (read carefully before screaming).
Mathias
Can anybody say 'asshole litigant?'
Not that there are any direct comparisons between, but considering all recent scandals, IMHO its time to stub [[Wikipedia:Bad press is good press]].
{{Shortcut|[[WP:BPGP]]}}
Stevertigo
--- Mathias Schindler mathias.schindler@gmail.com wrote:
On 12/14/05, Ryan Delaney ryan.delaney@gmail.com wrote:
What is going on here? And where did they get
permission to use our logo?
See [[QuakeAID]] for details about this.
However, I very much prefer wikipediaclassaction.com (read carefully before screaming).
Mathias _______________________________________________ WikiEN-l mailing list WikiEN-l@Wikipedia.org To unsubscribe from this mailing list, visit: http://mail.wikipedia.org/mailman/listinfo/wikien-l
__________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com
Ha ha ha :D I perfer that as well :)
Thanks for the laugh
On 12/14/05, Mathias Schindler mathias.schindler@gmail.com wrote:
On 12/14/05, Ryan Delaney ryan.delaney@gmail.com wrote:
What is going on here? And where did they get permission to use our logo?
See [[QuakeAID]] for details about this.
However, I very much prefer wikipediaclassaction.com (read carefully before screaming).
Mathias _______________________________________________ WikiEN-l mailing list WikiEN-l@Wikipedia.org To unsubscribe from this mailing list, visit: http://mail.wikipedia.org/mailman/listinfo/wikien-l
-- ~Ilya N. http://w3stuff.com/ilya/ (My website; DarkLordFoxx Media) http://en.wikipedia.org/wiki/User:Ilyanep (on Wikipedia)
On 12/14/05, Mathias Schindler mathias.schindler@gmail.com wrote:
On 12/14/05, Ryan Delaney ryan.delaney@gmail.com wrote:
What is going on here? And where did they get permission to use our logo?
See [[QuakeAID]] for details about this.
However, I very much prefer wikipediaclassaction.com (read carefully before screaming).
Mathias
You're right. That one makes much more sense. You might as well shut down all messageboards if you are afraid of defamation...
I also expect that, once the complete list of evaluated articles is
published by Nature, we will find that our articles are >longer, so that the error rate *per sentence* might very well be lower in Wikipedia.
A few people seem to have missed this from the website: "All entries were chosen to be approximately the same length in both encyclopaedias. In a small number of cases some material, such as reference lists, was removed to make the lengths of the entries more similar."
Steve
As for the {{sofixit}} - in the associated podcast the author of the article called on scientists to do just that - find the articles that apply to their area of interest, and work to improve them.
One other thing - there's a subsection entitled 'Challenges of being a Wikipedian' which talks about WMC and his arbcomm cases.
Ian
On 12/14/05, Steve Bennett wiki@stevage.com wrote:
I also expect that, once the complete list of evaluated articles is
published by Nature, we will find that our articles are >longer, so that the error rate *per sentence* might very well be lower in Wikipedia.
A few people seem to have missed this from the website: "All entries were chosen to be approximately the same length in both encyclopaedias. In a small number of cases some material, such as reference lists, was removed to make the lengths of the entries more similar."
Steve
WikiEN-l mailing list WikiEN-l@Wikipedia.org To unsubscribe from this mailing list, visit: http://mail.wikipedia.org/mailman/listinfo/wikien-l