I am familiar with the Citizendium project, and if you think that I try to argue for something similar, then you are way off. The interesting thing is that there are people in the history of that article having absolutely no clue on what they are writing about, and there are people knowing what they writes about. Inspecting the trust coloring it is apparent that the persons without any clue at all has much higher trust metrics than the rest, bumping the trust upwards. To me this does not seem to be what we want. We want to have a system that is much better on identifying the actual experts from those just fuzzing around fixing spelling errors. An article with perfect spelling can be in complete error on the main topic, and if the system does not fix that problem, it isn't a fix at all.
John E
draicone@gmail.com skrev:
On Dec 21, 2007 11:35 PM, John Erling Blad <john.erling.blad@jeb.no mailto:john.erling.blad@jeb.no> wrote:
This is a very central problem to this kind of trust metrics, people are rated according to what they do at some point in time, without taking into account who they relates to and what is they previous history in other contexts. In the history of the Stave Church article, there is an archaeologist anyone able to locate the person? I think it should be possible to identify a person as an expert within a limited field of expertise, but it isn't easy to figure out how this should be done.This is what we started the Citizendium project for. Wikipedia maintains a degree of anonymity that forms the basis for one of the many cultures of the WMF projects. If we start verifying qualifications, we throw this all out.
The middle ground is a peer-based "reputation" system, as found on many forums, where for various actions peers can allocate reputation to a user. This type of ad-hoc, informal peer-review is the only way to achieve verification of authority without taking the Citizendium approach.
Akash
Wikiquality-l mailing list Wikiquality-l@lists.wikimedia.org http://lists.wikimedia.org/mailman/listinfo/wikiquality-l