Jimmy Wales on Fresh Air

NPR : Jimmy Wales on the User-Generated Generation

This is a link to the Jimmy Wales interview on NPR's Fresh Air. (Jimmy Wales is the founder of Wikipedia.)

Thesis Methodology

Methodology

Research Question: What features contribute to a wiki's growth?
The proposed methodology for this project will be a quantitative analysis that compares the growth rate of different wikis with differnt features.

Growth rate might be measured using the following: (These metrics are taken from Jakob Voss: Measuring Wikipedia)

1. Database size (combined size of all articles including redirects in bytes)

2. Total number of words (excluding redirects and special markup)

3. Total number of internal links (excluding redirects and stubs)

4. Number of articles (at least contain one internal link)

5. Number of active Users (contributed 5 times or more in a given month)

6. Number of very active Users (contributed 100 times or more in a given month)

Possible Hypotheses to Test

Exclusive Content: wikis that provide a place for content that is not easily available from other places will grow faster than wikis that replicate easily available content.

How to measure: Measure the difference in growth rate between:

1. A set of wikis whose common feature is that they offer a unique place for content, with
2. A set of wikis whose common feature is that they replicate already available content.

Competitive magnets: Contests will result in contribution flurry. But do they lead to an increase in overall contributions?

How to measure:

1. Compare the contribution rate of a single page before a contest solicitation, and after the end of the contest.
2. Compare the overall growth rate of the wiki before and after the contest

Seeding: wikis that are seeded with content grow faster than wikis that expect users to create content from scratch.

How to measure: compare the growth rate between:

1. A set of wikis whose common feature is that they were pre-seeded with content.
2. A set of wikis whose common feature is that they were started from scratch.

Possible Thesis

Theory: One of the motivations for participation in human stigmergic systems is competition.

I. Stigmergy: This is a model from the insect world that is increasingly being applied to human collaboration. Stigmergy is a method of communication in complex systems, where the individual parts of the system (agents) communicate with one another indirectly, by modifying their local environment. The result is often a complex artifact, with emergent properties; that is, the artifact has properties that have "emerged" without being centrally planned, and yet have a system of organization. Examples in nature are ant and termite nests: ants and termites communicate by reacting to pheremone trails and build complex artifacts without having a central "architect." This model begs comparison with human collaborative efforts, such as Wikipedia.

A. Motivation: What motivates agents to modify their local environment? The work done by one individual must provide a stimulus for another individual to add to it. Insects have their motivation "cooked in" to them by their genes. What about humans?

1. One theory is that humans are motivated to contribute by competition. Possibilities include competition for social status, as posited by Eric S. Raymond, who talks about open-source hackers motivated by a "gift culture."
a. Some could argue that humans have competition "cooked in" to their genes as well, which is why stigmergy can be so effective.

II. Why this is important: This theory can be a guide to help those who want to build collaborative networks. Many collaborative projects (like wikis) fail to take off.

A. This theory of "collaboration via self-interest" is contrary to some beliefs that cooperation requires altruistic motives. There is a paradox here; selfish behavior results in collaborative products.

III. Approaches:
A. Analysis of successful and unsuccessful projects- I could look at different collaborative projects and compare the stimuli for contribution with the outcomes.