On Fri, 1 Aug 2003, james duffy wrote:
I think wiki does need as it develops to be able to
have some 'final'
articles that, having reached a clear standard of accuracy, readability etc
can he removed from the editing process. The downside of constant editing is
that some articles that reach a high standard then can lose that as those
who produced the standard leave and someone comes on and rewrites it to a
lower standard. Wiki's open edit policy is its major plus, as it allows us
to evolve and update, but its downside is reliability. Can I be sure if at
8.17pm I read an article /everything/ in it is factual or could I have the
bad luck to read it just after some user either through not knowing what
they were doing or deliberately, mucked it up and added in false
information? For example, Jerusalem's status as the capital of Israel is
disputed. That is stated on wiki (after a battle!). But what if a reader at
8.17 reads a version that says in a POV edit it is an 'undisputed' capital.
Or someone doing an essay on JFK reads an edit at 8.17 that says he was the
33rd not the 35th president?
Keep in mind that as topics gain a density of articles, it will become
harder for a vandal or crank to quickly make edits that lower the quality of
information of that topic. To directly address your example, James, if
someone changes the article on JFK in the manner you mentioned, they would
also have to change the links to previous presidents & future presidents --
as well as to the list of US Presidents. Yes, a determined vandal could
make all of those changes, but not before he was detected & the changes
reverted.
The problem lies in topics that are not as well populated. To throw a
hypothetical example, it would be possible for someone to sneak fabricated
information about a crank theory of Atlantis into existing articles on
Ancient Egypt or Mesopotamia, because there are so few of us working
on articles about ancient history (off the top of my head I could name
three of us, & if given time I might be able to double that number), &
it would take us weeks -- if not months -- to discover the change.
I'd say that as Wikipedia gathers more articles & more contributors, this
problem will recede -- as long as the computer resources scale to handle
the demand.
For all their downsides, the 'centainty of
standard' is the one major plus
that Brittanica, World Book, Encarta has. When you read an article you are
getting a definitive version, not that moment's edit. At some stage wiki is
going to face a credibility barrier where people ask 'but can I be sure that
King Edward VI of England actually died on that day, or is it a bad edit?
How can I be sure W.T. Cosgrave said that? How and when we deal with the
'certainty of standard' issue will mark the moment we go from being a good
secondary source that may give a fascinating insight but which just to be
sure you might want to cross check, just in case, to a /guaranteed/ reliable
primary source.
As someone else remarked, they have to apply a certain degree of scholarly
scepticism to the information. Wiki isn't the first resource that faces this
problem: I recall a book where a scientist, out of curiousity, examined the
articles in EB concerning his area of expertise -- & was appalled at how
out of date the information was! Then again, the EB suffered for decades
from the practice of making only minor revisions to certain articles, adding
other items of current interest, & ignoring topics "nobody" reads (Ancient
and Medieval History comes to mind as examples) until the text was painfully
out of date.
[I'm snipping James' last paragraph because I don't have a response for
the points he raised there.]
Geoff