Jeremy Smith's blog

Entry Is Labelled

Collaborative Stats on Case Wiki

I did some more mining of the Case Wiki data. I did this before in Tipping Point for a Wiki to Become Self-Correcting. This time, I was looking for the pages that the most different people have edited. Trying to rephrase that, I wanted to see the most collaboratively put together pages.

The #1 was Brief Writing Checklist with 28 different editors. #2 was the lengthly titled Econ395 Richard Shatten Public Policy Case Competition with 15 different editors.

The top 4 breaks down like this:

  1. 28 different editors: Brief Writing Checklist
  2. 15 different editors: Econ395 Richard Shatten Public Policy Case Competition
  3. 12 different editors: Freedman Center
  4. 9 different editors: ToDo

After the top 4, there are 7 pages with 8 different editors. That list includes Email Services and the enigmatic B1IQ6P2_Main page (← and "no," I can't discern really what they are doing and I don't know what "B1IQ6P2" stands for). There are 6 pages with 7 different editors, 17 pages with 6 different editors, so on and so forth. Here is a graph showing this (OpenOffice's graphing capability sucks, by the way):

number_of_pages_by_different_editors.jpg

Because I know Greg is gong to ask for them, here are the SQL queries I used. The query to generate the pages with the most different editors is the following:
SELECT COUNT(DISTINCT(rev_user)) AS editors, page_title
FROM revision
JOIN page ON page_id = rev_page
GROUP BY rev_page
ORDER BY editors DESC

Then, to break the data down for the graph, I just manually compiled it by running the following over and over again subbing in a different number:

SELECT COUNT(DISTINCT(rev_user)) AS editors
FROM revision
JOIN page ON page_id = rev_page
GROUP BY rev_page
HAVING editors = 5,4,3,2,1

Comments