Essay:RationalWiki's mission and content

The mission statement is one of the key "pillars" of RationalWiki; it purports to define mainspace, governing what our overall purpose is as a wiki. But a question has plagued me for quite some time: just how faithful is the RationalWiki mainspace to its mission statement? Is our philosophy too inclusive of articles and topics which have tenuous or tangential connections to the mission's tenets? Or is it too exclusive? Just right? I feel this is an important discussion to have, or at the very least, to think about. Hopefully, this essay can spark that discussion.

Explanation of terms

Mission: For the purposes of this essay, the term "mission" refers exclusively to the four tenets of the mission statement, not to any precedents, unspoken standards, or guidelines written on various help pages. Articles which exist solely to help explain other on-mission articles are not regarded as "on-mission."

Page: These will all be mainspace articles.

Data

First let me undertake a little experiment. I will press "random page" a total of 80 times, and for each page that comes up, I will classify it in the following way:

Clearly on-mission:

• Relates to pseudoscience, woo, quack medicine, paranormal etc.; example: Eugenics, Sound money
• Relates to authoritarianism or fundamentalism or conspiracy theories; example: Tony Blair, Bible thumper
• Relates to crank ideas or conspiracy theories; example: crop circles, Charles Hapgood
• Relates to creationism or intelligent design or evolution; example: Peer review
• Relates to any of these subjects in the media; example: Newsmax

Debatably on-mission:

• Explains a scientific or mathematical concept; example: Adenosine triphosphate, eukaryote
• Explains a concept in economics that is not related to pseudoscience; example: Dean Baker
• Relates to non-authoritarian, non-fundamentalist, or non-pseudoscientific topic in politics or political science; example: Yucca Mountain, Free Republic
• Relates to a conservative or liberal figure, such as a pundit, not related to authoritarianism; example: Rich Lowry, Lee Doren
• Relates to sociology or psychology; example: Social solidarity, Hindsight bias
• Relates to non-authoritarian philosophy; example: AI-box experiment
• Relates to linguistics, language, or arguments; example: auxiliary language, inflammatory language
• Explains a topic about religion, not one that is anti-science or fundamentalist; example: Panentheism, Druid
• Relates to a non-mission oriented web site or other internet topic; example: Uncyclopedia

Clearly off-mission:

• Relates to geography; example: Dubai, Andorra
• Relates to a non-mission oriented band, composer, etc.; example: Straight edge
• Explains a "historical sidenote"; example: Paracelsus

---

Results
Classification	Number of articles
Clearly on-mission
pseudoscience, woo, quack medicine, paranormal etc.	14
authoritarianism or fundamentalism	12
crank ideas or conspiracy theories	9
creationism or intelligent design or evolution	5
any of these subjects in the media	1
Total	39
Debatable
scientific or mathematical concept	4
concept in economics that is not be related to pseudoscience	3
non-authoritarian, non-fundamentalist, or non-pseudoscientific topic in politics or political science	11
conservative or liberal figure, such as a pundit, not related to authoritarianism	5
sociology or psychology	2
non-authoritarian philosophy	2
linguistics, language, or arguments	5
topic about religion, not one that is anti-science or fundamentalist	3
non-mission oriented web site or other internet topic	2
Total	37
Clearly off-mission
geography	2
non-mission oriented band, composer, etc.	1
"historical sidenote"	1
Total	4

Articles, from this experiment, which should be mission-tagged

• Social solidarity
• Adenosine triphosphate
• Yucca Mountain
• Dubai
• Bigot
• Arnold Schwarzenegger
• World Intellectual Property Organization
• Narco-saint
• Andorra
• Straight edge
• Rugged individualism
• Human Rights Watch
• Bandwagon
• Micronation
• Omnivore
• Paracelsus
• Ribosome

Analysis

49% of the sample were solidly on-mission; 46% were debatably on-mission; and 5% were decidedly off-mission. The ratios between the classifications within each of those categories are unreliable, given the small sample size (see below).

In a mainspace of thousands of articles, it was odd that the article on Tony Blair came up twice; of course, I did not count it twice. Another interesting point is that I encountered two subpages ("sub"-pages, as there are no subpages in mainspace), which I could not count; they were Kent Hovind's doctoral dissertation and a table of the Citizendium sysops.

The listed articles at the bottom of the data section are all ones which I would feel comfortable tagging with the {{mission}} template, even outside the context of this experiment/essay.

Conclusion

This data may just be something to glance at and say "Oh, that's interesting." However, even if that is so, there is a serious discussion that must be had, and I hope this data will aid it.

It is not an option to summarily remove a fair portion of the articles in mainspace, even if they do not pertain to the mission at all (either directly or indirectly). The best course of action, in my opinion, would be to adopt a new guideline or set of guidelines, which first acknowledge that a substantial amount of material in mainspace is unrelated to the four chief tenets of the mission statement, and then clarify which of these articles fall into categories that are acceptable to keep. We should have a discussion about which types of geography articles we should keep, which types of economics articles we should keep, et cetera.

We should also discuss our guiding philosophy on the mission. Right now our philosophy is ultra-inclusive: anything with the remotest, most tenuous connection to the mission is kept. Should that change? Should we focus our efforts on the furthering mission statement, rather than try to maintain a murky web of unrelated topics?

Further research

While 80 is a sizable amount of articles to examine, there are thousands of articles in the RationalWiki mainspace. While this experiment is good to indicate general trends, the specific ratios are most likely unreliable. How large should the sample size be if one were trying to construct a more extensive, far-reaching experiment? 150? 250? 800-1000? If anyone has the time, a wide-ranging analysis of the contents of mainspace would be very enlightening.