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