Tag Archives: wikipedia

Mission accomplished

… also for some people in the field the main paradigma in science. To cite Wikipedia

Bush’s assertion — and the sign itself — became controversial after guerilla warfare in Iraq increased during the Iraqi insurgency. The vast majority of casualties, among both coalition (approximately 98.3% as of October 2008) and Iraqi combatants, and among Iraqi civilians, have occurred after the speech. Due to this fact, “Mission Accomplished” is now a winged word for uncompleted operations with an unclear ending.

goodbye GB!

Search for crystals

First monday has an interesting paper on the 100 most visited Wikipedia pages for the period of September 2006 to January 2007 (Wikipedia is the ninth most visited site in the U.S. with 43 million visitors). The crystal search link in the paper does not work but the table reports that science ranks at place 5 – not too bad.

crystal3.jpgcrystal2.jpgcrystal1.jpg

I agree with everything you said

“I agree with everything you said that was correct, and I disagree with everything you said, that was incorrect” (Adlai Stevenson according to AJRCM 2006;174:1056) – a nice comment that fits every situation.
The German Spiegel has an interview with Tim O’Reilly about the quality of internet resources. It seems that everybody can voice his or her opinion while the final decision about a feature or a patch is done in the “inner circle”. Entry to the inner circle is limited to those who qualify by previous contributions – probably a very similar system in science. O’Reilly talks in this interview also about Jaron Larnier’s warning that Wikipedia may be dangerous for creating mono-culture-knowledge. He agrees that Wikipedia has been abused in the past but believes that the mechanisms behind Wikipedia to identify abuse are much better than in any political system, yea, yea.

Peer production

firstmonday has an interesting article about the limits of self-organization and “laws of quality”. Given 52 million tracks in the Gracenote database, 1 million entries in Wikipedia and 17,000 books in project Gutenberg, Paul Duguid throughly examines the two laws of quality

  • Linus law: “given enough eyeballs, all bugs are shallow” which means that almost every error will be discovered and ultimately fixed
  • Graham law: “people just produce whatever they want; the good stuff spreads, and the bad gets ignored”

Although more professionalized, similar principles operate in science. With these large genetic studies, I have the feeling that most errors occur at the interfaces, during hand-shaking of disciplines. There are certainly only a few people that can design a study, examine a patient, go to the laboratory, analyze and annotate the data and publish them. This means that even many eyeballs can not look around the corner and that it will take many years for the “good stuff to spread”. Yea, yea.