I think this idea would indeed get most of the best of both worlds. People who needed an article they could rely on could look up the most recent version approved by a credentialed source, and make up their own minds about what counts as "credentialed". I also think that in addition to distinguishing between users' claimed credentials, there should also be a flag used to indicate whether someone's credentials have been verified, by for example sending a confirmation mail to their .edu address.

With regard to people's concerns about a backlog of changes that need to be approved, hopefully this would only be an issue for articles about current events where new information is constantly coming out. I would hope that the article on Physics wouldn't change much as the laws are thought to be fairly permanent :)

The only issue this wouldn't resolve would be whether the default article, the "one that everyone sees", would be the bleeding-edge one or the latest approved one. Unfortunately, the instant gratification that people get from seeing their helpful edits go live immediately, is the same that the vandals get from seeing *their* edits go live immediately, so it would seem there's no best-of-both-worlds compromise, because unless the software can distinguish between helpful edits and vandalism, you can't have one without the other. So the question is whether the motivational effect for helpful editors outweighs the motivational effect for vandals. It sounds like George is saying that dealing with vandals is a major time-waster, and people on the Citizendium boards see the same problem.

Personally I don't know why people would be less motivated to make edits that would go live after 1 hour, rather than go live immediately. I wouldn't look at it that way, especially if I knew that the reason for the delay was to prevent vandalism and increase the usefulness of the reference to everybody. I might actually be more motivated to make edits if I thought that the article was being edited by someone who was putting their own credibility on the line, and thus would prevent someone *else* from frivolously overwriting my changes.

-Bennett

At 03:02 PM 3/20/2007 -0500, Sheldon Rampton wrote:

Steve Bennett wrote:

...

Yeah. Which is also bad: "Welcome to Wikipendium, the encyclopaedia that's being edited in real-time. This article is 92 days out of date." Not as bad as Britannica. Not as good as Wikipedia.

I think there's a way to have most of the Citizendium advantages on Wikipedia without losing Wikipedia's advantages.

Wikipedia's advantages are: more inclusive; more editors; articles updated more frequently.

Citizendium's advantage is: expert vetting by people with actual credentials. (And yes, credentials DO mean something.) A problem, though, is that even experts don't always agree, and some people value different sorts of expertise.

One way to have the best of both worlds would be to have a public version of Wikipedia's watchlists feature, with a couple of modifications. Since this feature doesn't currently exist, I'll call it "approval lists" for lack of a better name.

Each user would be able create and maintain his/her own "approval list," which would work just like watchlists, except for the following two changes:

(1) A user's "approval list" would be publicly viewable by everyone, not just the user who creates it.

(2) Rather than marking an ARTICLE for inclusion in the approval list, users would mark a REVISION VERSION of the article.

This information would then be usable in various ways. For example, if user A and B are having an edit dispute, they might prefer to simply put different versions on their approval lists rather than having a revert war. If other users chime in, the effect would be akin to voting on a preferred version between the two. If user A's version gets marked for approval by 200 users while user B's version gets marked by only 10 users, this would provide evidence that A's version is more widely accepted.

Also, the issue of "credentials" could be dealt with by having users who possess credentials "bless" versions of articles that meet their standards. This would not prevent subsequent editing, but it would make it easy to find the latest version that has been vetted and approved by someone with credentials relevant to the topic. "Credentials" would not have to be defined or standardized. They might include "I have a Ph.D in physics" (a credential that I personally respect) or "I represent the John Birch Society" (a credential that I don't). The result would be that people could choose which credentials they personally value and find versions of the article that match their values.

If this functionality existed, I assume that most users would continue to rely primarily on the latest version of article. They would simply find it easier to meaningfully navigate the version history.