In the past few weeks there’s been much discussion of news stories based on Felipe Ortega’s dissertation; the concern is that Wikipedians are abandoning the online encyclopedia “in droves.” (What is a drove you ask? According to Wikipedia, it is an ancient route by which livestock were herded.) However, Erik Zachte, with the help of Felipe, shows how in such analysis the way that one constructs one’s parameters significantly affects the conclusions one can draw. For example, the alleged drop-off (deaths) of Wikipedia editors may be more the result of when and how the analysis is done. If you assume that an active Wikipedian is someone who did one edit (i.e., someone who was just experimenting), rather than five, or some other number (i.e., actual Wikipedians), this can significantly affect the outcome. Or, if you assume that a “death” is when someone has not been active for a month, you will naturally have a lot of deaths at the end of the analysis period because these people may have been simply “sleeping” for that month, but come back in the next month and you weren’t there to see it. (Like the line from Twin Falls Idaho, a favorite movie of mine, “The sad ending is only because the author stops telling the story. But it still goes on. It’s just untold.”)
Wikimedia’s lesser noted response to the story claims significant efforts are being made to improve the recruitment and retention of users, but on the numbers side:
On the English Wikipedia, the peak number of active editors (5 edits per month) was 54,510 in March 2007. After a more significant decline by about 25%, it has been stable over the last year at a level of approximately 40,000. (See WikiStats data for the English Wikipedia.) Many other Wikipedia language editions saw a rise in the number of editors in the same time period.