On Wed, Sep 30, 2009 at 6:36 AM, Ken Arromdee <arromdee(a)rahul.net> wrote:
On Tue, 29 Sep 2009, George Herbert wrote:
"Verifiability, not truth" means that sometimes we'll put in something
that's
verifiable but isn't true.
"Verifyable, but untrue" - where
there's evidence to disprove but it's
not compellingly better quality data than the untrue data - is the
hard case. Either walk the narrow line and present both or pick one
and defend using it, staying aware that more info may clarify the
situation into the first case above.
The problem is that the data may actually be better quality (by non-
Wikipedian standards) but not verifiable by Wikipedia standards. (Like the
case of the bridge which was said in a source to have no traffic, and
someone visited it and saw it has traffic. You could make up far-fetched
scenarios of why the reliable source could still be correct, but it's far
more likely that none of those scenarios are and that the source is simply
wrong.)
I believe that primary source evidence, including Wikipedian fact
checking (that they then publish somewhere, i.e. a video on YouTube
etc) would count for impugning the reliability of a source where the
source is demonstrated wrong.
Sources which appear reliable - written by people who should know what
they're doing, published with fact checking by reputable publishers,
etc. - have a presumption of reliability and accuracy. But that's
rebuttable.
Once a valid concern about its accuracy is raised, if you can
demonstrate it's not reliable, it's out. Presumption rebutted.
On the bridge example, at the very least it can demonstrate that the
statement in the book of no traffic was not currently true, though it
might have been at the time of the book's publication.
Some people object to such rebuttals and feel that Wikipedians should
not make judgement calls on whether the sources are reliable. Those
people are wrong. There is plenty of crap information out there in
apparently reliable sources.
WP:RS is not a suicide pact - appearances are sometimes deceiving, and
we have a complete freedom to investigate and throw out sources.
--
-george william herbert
george.herbert(a)gmail.com