Wikipedia:Pushing to validation/Wikimania

This page outlines the original discussion article validation in Wikipedia, set up to allow us to continue a discussion begun at Wikimania 2006. By article validation is meant a rigorous examination of every fact, reference and assertion in an article. After refinement at the meeting a rough consensus was reached, as described at Wikipedia:Pushing to validation.

Guidelines

This page is only for listing summary points. Please use the talk page for discussing these points, or other ideas and suggestions. Please feel free to add other summary points to this page, or to edit the existing points.

What follows is an outline of the preamble, as well as some points raised during the Wikimania discussion.

---

Two things are needed to achieve successful validation of articles:

1. A process for validation.
2. The appropriate infrastructure to allow validation to be carried out efficiently and effectively.

The validation process

• The system should not only be effective, it should be 'seen to be effective.
• Articles need to be selected for validation in an effective way.
• Suitable teams of reviewers need to be organized, either as permanent groups or on an ad hoc basis. These teams should ideally be able to demonstrate their effectiveness.
• A team would select a good, reasonable stable version of an article and work on this away from the main article space. WARNING: A temporary fork is necessary at this point.
• The team would divide up its responsibilities, and get to work – checking every fact against outside sources, checking references, examining POV and more subtle cultural biases.
• After a set time period (say, one month) the team would then agree on a final validated version of the article. This uneditable version would then be made available for users to view as an alternative to the main (editable) version. THIS ENDS THE FORK.

Infrastructure needed

• Overall organization – A umbrella project (like the Good Articles or 1.0 project) to organize which articles are validated/updated, contact/set up review teams, etc.
• Review teams – self-elected, voted in, or appointed? Would outside reviewers be involved?
• Review process – presumably on a wiki; would reviewers allow open access? Would there be a review namespace for this? MAY NEED SOFTWARE CHANGE.
• Stable versions – once approved, a validated article should not be editable, except perhaps by the original review team and admins. A validated article would be rendered "null and void" by a single unauthorized edit. Probably these need their own namespace. WILL NEED SOFTWARE CHANGE.
• "Validated" tab or equivalent – so users can easily see the validated article when they want to.

To be decided

• Should we do this?
• Can we get the necessary software changes – stable versions, new tabs?
• How should review teams be created? Who will approve the teams? I suggest self-appointed teams from WikiProjects, where the reputable editors are well-known.
• How can validation schemes be coordinated across the different language versions?

Points raised

These represent just some of the main points raised during the discussion.

• Reviews do not need their own namespace, they can be done quiety just off the main article space.
• How will reviewers be selected - based on personal reputation? Qualifications? Will we use outside experts/communities of knowledge?
• Stable versions may well be available soon! Should these be the default page - the questioner thinks emphatically not.
• May be naive to expect people to volunteer for this boring task!

See also

Article validation on Meta