What happened in May 2007?
Thanks for both of those comments. But they don't seem to address the sudden change. Up till March 2007 (I've read the graphs a bit more precisely) there was a growth in active Wikipedia editors of about 2,100 per month. After this date there was a decline of 1,400 a month - though it recovered somewhat after that, whilst continuing to decline. That is a HUGE change.
Social networking sites (Facebook, Myspace, Youtube+) must have a longer term effect on Wikipedia but would it be so sudden?
I've never heard of a controversy about user boxes and can't find the controversy. Can you point me to it? -Chris55
- (inline answer for brevity) I think I have the answer about May 2007: students were a major portion of contributors and they left on 3-month school vacation in May 2007, at a time when general interest in Wikipedia began to decline, and more American schools banned Wikipedia use. There were similar, but smaller, drops for school vacations in May 2008 and May 2009, but the user-base seems to have broadened to include more people not leaving on vacations in May-July each year. Many U.S. schools banned use of Wikipedia because the unbalanced (or untrue) POV-pushing of article contents became notorious, or students were unable to focus on their own school work, due to the vast distractions on Wikipedia: it is a bottomless pit of problems and articles not related to typical school subjects. The chart above (as a modified en:negative binomial distribution) shows a similar decline in the article-growth rate, for addition of new articles, rather than new users, each year (also drops in May each year). The original wide-open growth was reduced by widespread news and more school policies against bad problems in Wikipedia (short answer). -Wikid77 12:55, 14 March 2011 (UTC)
- Fascinating. So you would say that the hyper-growth in English wiki between 2005 & 2007 was mainly fuelled by US schoolkids. It would only take about 20,000 of them - I suppose there are no stats available: have wiki editors ever answered anonymized questionnaires about themselves? It would certainly tie in with the suggestion that Facebook was the next fad. I hadn't considered Wikipedians and Facebook users to be similar types, but I can buy it for teenagers.
- My feeling is still that it's only a partial answer. For one thing the decline started in April (though the June decline was the biggest), but also why did the direction of the graph change so much? Even if the last kids were still dropping out a year later the growth should have recovered at some point. (And the long term graph you've included is pure fantasy:) Chris55 13:59, 14 March 2011 (UTC)
Have a look at the WikiEN-l archives in January 2006 for the thread "The userbox fad". That won't tell the whole story, but it will illustrate how established editors viewed an otherwise innocuous activity favored by a new generation of editors. I don't think it would be very hard to see how newcomers reacted to this response.
I could see Yoville, Mafia Wars and Farmville drawing a lot of people, but I agree the drop you're referring to was remarkably sudden. Obama announced his presidential campaign in February 2007, and that campaign attracted a lot of young and tech-using people - iow, the same sort of volunteers as Wikipedia. Still, we may just be seeing a typical peak-then-plateau as the novelty wears off. As for any userbox flap, I never was aware of one. I just copied mine whenever I found one I liked on someone else's page. ;-) Flatterworld 05:40, 14 March 2011 (UTC)
I'm coming to the conclusion that the change in April 2007 was about a change in the way that new editors were treated. It may have been a change in the registration procedure or in how their first edits were treated as many others have suggested in other threads.
Looking at the study, one change is that 57 new admins were appointed in March 2007. This is the highest number in a single month apart from the period March-May 2005 when a total of 175 were appointed. Both of those periods have been highlighted as significant in the study. It might be worth looking more closely at the way this cohort was instructed and what they did in their initial period. It could be either that their actions scared off many potential new editors or that they were appointed because of some other perceived threat, which had a side effect. I'm not even sure these figures apply to the English Wiki.
March 2007 only had 31 successful RFAs, which was a fractionally below average month for RFA as far as 2007 was concerned. Also it came soon after a slight lull in December 2006 and January 2007. There were a couple of spikes in May and November, but neither was as much as double the trend for that year. en:User:WereSpielChequers/RFA_by_month