Folks,

I hope I am not off-topic to refer this RfC to the mailing list.

http://en.wikipedia.org/wiki/Wikipedia_talk:Requests_for_comment/Roylee

Please direct all discussion to the Talk page above, and not to this list ?

I include a little context below.

Cheers, Andy!

Purpose of this RfC

Personally, I am not interested in sanctioning the editor Roylee. What I think is most important, and accordingly what should be the aim of this RfC, is finding a solution to the underlying social and technical problems of Wikipedia as exposed by this issue. As BanyanTree said here, "it took over a hundred edits before Mark began to reel Roylee in. Is there another user whose made 50 similar edits, who has not been discovered? Are there a hundred such users?"

Simple-minded Linus's law-derivates are not the answer here. Many eyes have looked at the articles that were affected, and did not recognize any problems. More paradoxically, if literally everyone would have assumed good faith, only brute force fact-checking could have detected the problem eventually. This is profoundly worrying. Quite frankly, all this has made me doubt, maybe for the first time, the long-term viability of Wikipedia as a trustworthy resource. -- mark 14:46, 3 December 2005 (UTC)

A clever manipulator will always be able to insert unwarranted material into Wikipedia for a time, but many eyes eventually discover even believable nonsense. This issue is a case in point. The principle of caveat lector should not be restricted to elite education: we should all have been raised as doubters of text, right from the start. Even the Encyclopaedia Britannica. As for me, so far am I from continuing to Assume Good Faith, in the face of bad edits, when I find a vandal I try to check through that IP's contributions, and sometimes discover previously unnoticed vandalism.

The question remains, how many of these bad edits remain in Wikipedia articles? --Wetman 03:05, 4 December 2005 (UTC)

I disagree with your first point ("many eyes eventually discover..."), which is of course a truism among Wikipedians as an extension of the proposition that vandalized articles constitute a fraction of one percent of all articles. This case appears to indicate that "believable nonsense" can remain on Wikipedia for unacceptably long periods. Cases such as the long-lasting misinformation found recently on John Seigenthaler Sr. can perhaps be written off as "believable nonsense" that was missed because the vandal did not edit a number of articles, which has a better chance of rousing suspicion and being reverted (another truism of vandal fighting). This is not the case with Roylee, who created believable nonsense in self-supporting webs of article across the normal Wikipedian topics of specialization over a period of months.

This simply shouldn't have have been possible, and that fact that it did happen indicates that Wikipedia's processes are not as robust as they are advertised. As I mention in the post that Mark links to above, Roylee throws any blanket reassurance given for Wikipedia's credibility into doubt. - BanyanTree 19:59, 4 December 2005 (UTC)

Wikipedia has credibility? Have you seen pages like Nietzsche or Khmer Rouge? Wikipedia has many, many roadblocks to overcome before it has any hint of credibility collectively. I think currently, each article has to prove it's own credibility, it's not simply inherited because it's a Wikipedia article. (Bjorn Tipling 21:23, 4 December 2005 (UTC))

I basically concur (see Wikipedia:Researching with Wikipedia, largely written by me and Dan Keshet), but the goal is to work toward having credibility, and it looks like Roylee's contributions have been a detriment. And that is what this RfC is about, no? -- Jmabel | Talk 03:53, 5 December 2005 (UTC)

This is especially the case as we cannot be sure if he is unique or if his detection is unique, e.g. are there numerous users adding misinformation using similar patterns? I would like to think not, but I don't know how anyone can guarantee it. I had hoped to hear a solution to the problem from people reading this RfC, and the lack makes me think that the problem is structural rather than individual. I would love to be proved wrong on this. - BanyanTree 19:59, 4 December 2005 (UTC)