[Foundation-l] Ensuring veracity of articles based on print sources

Tue Oct 10 06:58:50 UTC 2006

Kpjas wrote:
> 
> Hi,
> 
> I've followed closely the thread about verifying content and citing
> (book) sources.
> 
> My comment would be that obviously not all (book) sources that have
> even been proven to exist are credible and reputable.
> 
> As I am a believer in "Work through WikiProjects" motto I think
> WikiProjects could prepare lists of reputable printed sources in
> different fields and eventually (dis)qualify a given source.
> 
> In my opinion we need more WikiProjects that are vivacious and full of
> "experts". All dubious Wikipedia content should be directed towards
> WikiProjects for verification.
> 
> 

This brings up an important issue- that of source criticism.  A 
nominally accurate citation is still not worth much if the underlying 
source is not authoritative in the field.  This can be measured, though, 
by starting to map the authority relationships within a literature, 
capturing not only how many citations are made of a particular work, but 
also what TYPE:

http://meta.wikimedia.org/wiki/WikiTextrose#Citation_Analysis

For example, a work that is cited dozens of times is not necessarily 
more authoritative than one that is cited less frequently if more than 
half of those citations are negative and attack its claims.  Building 
such a database would allow non-experts to quickly orient themselves to 
what is state-of-the-literature and act without the oversight of 
WikiProject experts (not to mention create a resource that, in time, 
could be very useful in its own right).  Not that such WikiProjects 
could not be useful as well, only that they are inherently non-scalable, 
and certainly more inegalitarian in their effect than I think is 
necessary- elevating certain users to the role of oracle rather than mentor.