[Paper Review] Collaborative Structuring of Knowledge by Experts and the Public

Tom Morris, Daniel Mietchen|arXiv (Cornell University)|Aug 24, 2010

Wikis in Education and Collaboration2 references3 citations

TL;DR

This paper presents Citizendium, a wiki-based knowledge platform that integrates expert oversight with public participation by requiring real-name contributions and implementing a two-tier content model: draft versions for collaborative editing and approved 'clusters' for stable, citable content. The system aims to improve reliability and quality in open knowledge production by combining expert validation with community governance, offering a sustainable model for open science and education.

ABSTRACT

There is much debate on how public participation and expertise can be brought together in collaborative knowledge environments. One of the experiments addressing the issue directly is Citizendium. In seeking to harvest the strengths (and avoiding the major pitfalls) of both user-generated wiki projects and traditional expert-approved reference works, it is a wiki to which anybody can contribute using their real names, while those with specific expertise are given a special role in assessing the quality of content. Upon fulfillment of a set of criteria like factual and linguistic accuracy, lack of bias, and readability by non-specialists, these entries are forked into two versions: a stable (and thus citable) approved "cluster" (an article with subpages providing supplementary information) and a draft version, the latter to allow for further development and updates. We provide an overview of how Citizendium is structured and what it offers to the open knowledge communities, particularly to those engaged in education and research. Special attention will be paid to the structures and processes put in place to provide for transparent governance, to encourage collaboration, to resolve disputes in a civil manner and by taking into account expert opinions, and to facilitate navigation of the site and contextualization of its contents.

Motivation & Objective

• To address the challenge of balancing public participation with expert quality control in open knowledge platforms.
• To create a sustainable, transparent, and reliable alternative to traditional wikis like Wikipedia, particularly for scholarly and educational use.
• To develop a governance model that encourages collaboration, resolves disputes civilly, and integrates expert input without excluding non-experts.
• To support educational integration by enabling structured course-based contributions through initiatives like Eduzendium.
• To explore the potential of codifying Citizendium’s policies as a MediaWiki extension for broader adoption in open knowledge communities.

Proposed method

• Implementing a real-name policy to promote accountability and reduce vandalism and bias in contributions.
• Introducing a dual-content model: draft pages for open editing and approved 'clusters' (with subpages) for stable, citable content.
• Establishing a peer-review-like approval process where experts verify factual accuracy, linguistic clarity, neutrality, and readability for non-specialists.
• Using structured templates and course-specific tools to support educators in assigning and managing student contributions to articles.
• Developing community governance mechanisms that prioritize expert input while maintaining openness and transparency.
• Exploring the possibility of formalizing core policies and collaboration models as a MediaWiki extension to enable adoption by other wikis.

Experimental results

Research questions

• RQ1 How can expert knowledge be effectively integrated into public collaborative knowledge platforms without compromising accessibility or openness?
• RQ2 What governance structures can ensure content quality while maintaining a collegial and transparent environment for contributors?
• RQ3 To what extent can real-name policies and expert oversight reduce vandalism and bias in collaborative knowledge production?
• RQ4 How can educational institutions be meaningfully integrated into the content creation process of open knowledge platforms?
• RQ5 What technical and social innovations are needed to scale a knowledge platform that balances expert authority with public participation?

Key findings

• Citizendium’s dual-content model—drafts for collaboration and approved clusters for citable content—enables both dynamic development and reliable reference use.
• The real-name policy and expert review process significantly reduce vandalism and improve content quality compared to anonymous wikis.
• Educational initiatives like Eduzendium successfully integrated students into content creation, with courses assigned to specific articles and supported by course-specific templates.
• Despite its potential, Citizendium faces challenges in scaling, with only about 15 new articles per day, far below the 150 needed daily to reach 100,000 articles by 2012.
• The platform struggles with contributions on controversial or pseudoscientific topics due to high bias among contributors and low participation from evidence-based experts.
• The project identifies a need for financial support and professionalization to enable growth, multilingual expansion, and the formalization of its governance model into a reusable MediaWiki extension.