[WikiEN-l] Handling unreferenced but likely-valid material

[charles.r.matthews at ntlworld.com]

"Daniel P. B. Smith" wrote

> a) Most of the unreferenced material in Wikipedia is accurate. What  
> do I mean by "most?" 90%? 95%? 99%? Something like that.

My thinking has shifted a bit after three years at the coal face. Strange errors do get in. And, as one of the world's great historians of science convinced me, the historical truth is just about always stranger than you'd imagine.

But I say citation practices should go 'horses for courses'. The 'course' is per topic area, not per article, naturally. 
 
> c) Everything in Wikipedia should eventually be referenced or  
> removed. 

Nah. There's even an interesting reason, which is that with exponential growth the boundary (wild wiki frontier) is always of size very significantly large with respect to the central 'core' of must-have articles. What we actually need is a _targeted drive_ to bring up the quality of the core articles, and that must include fact-checking and all the other good things. Tagging just everything creates work, and implies work only likely to be done sporadically. 

Wikipedia has been on a billion-second spree (roughly). What is needed is a kind of kneading or churning motion where editors don't just follow the ribbon development.

Charles




-----------------------------------------
Email sent from www.ntlworld.com
Virus-checked using McAfee(R) Software 
Visit www.ntlworld.com/security for more information