% "What determines the size of the population of Wikipedians?"
% Michael Stone
% July 4, 2012

Let $n$ be Wikipedia's age in days. Yesterday, $W(n-1)$ people edited
Wikipedia. Today, $W(n)$ people are editing Wikipedia. Tomorrow, $W(n+1)$ shall
edit. What determines $W(n)$? How are $W(n+30)$ and $W(n+365)$ related to $W(n)$?

Members of the Wikimedia community have been thinking for nearly a decade about
questions like these. For example, a quick search for "*wikipedia more
editors*" yielded these historical references:

  Year      Article
  -----     ------------------------------
  *2003*    [Meta: Wikipedia needs editors](http://meta.wikimedia.org/wiki/Wikipedia_needs_editors)
  *2006*    [Wikipedia: Recruiting Editors Brainstorming](http://en.wikipedia.org/wiki/Wikipedia:Recruiting_Editors_Brainstorming)
  *2006*    [Aaron Swartz: "Making More Wikipedians"](http://www.aaronsw.com/weblog/morewikipedians)
  *2011*    [Sue Gardner: "Nine reasons why women don't edit wikipedia (in their own words)"](http://suegardner.org/2011/02/19/nine-reasons-why-women-dont-edit-wikipedia-in-their-own-words/)
  *2012*    [Signpost: "Foundation restructures to focus on editor retention"](http://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2012-03-26/News_and_notes#Foundation_restructures_to_focus_on_editor_retention)
  *2012*    [NPR: "As Wikipedia gets pickier, editors become harder to find"](http://www.npr.org/blogs/alltechconsidered/2012/07/19/157056694/as-wikipedia-gets-pickier-editors-become-harder-to-find)
  *2012*    [Wikipedia: Editor Engagement](http://en.wikipedia.org/wiki/Wikipedia:Editor_engagement)

Unfortunately, despite the long-standing interest that these questions have
engendered, answers that are both convincing and comprehensive remain few and
far between.

In the following series of posts, I will suggest several ways that I have been
learning to think about systems of questions like these. Some specific
perspectives that I hope to touch on include:

  * [social psychology](../wikipedia-20121006/)
  * graphic & interaction design
  * dynamical systems
  * organizational theory

and perhaps others as I discover them. :-)

<!--

Ideas:

a) What question do we care about?

b) How might we choose to think about the question?

  system / production network
  behavior formation
  epiphenomenon / stat mech

  diff-eq.

  structural / people / political / symbolic

  UX / graphic design

  availability models  + facets

c) Given (b), what data are relevant?



Many of these articles are based on specific hypotheses and axioms about why
people start and stop editing and about how we can best understand reality. For
example, some authors seem to start from axioms like:

  * *Bureaucracy and conflict discourage new editors.*

  * *To find out why population X isn't editing, we should ask a sample of
    members of population X.*

Perhaps if we find the right axioms[^1], a better way forward will become apparent?

[^1]: Alternately, perhaps the problem does not lie solely in ideas. Maybe
      there are equally or more severe problems in execution, coordination,
      consistency of messaging, etc.?


-->